By Todd Hendricks
Apr 17, 2023, 10:45 AM

In its current Term, the United States Supreme Court will decide Merrill v. Milligan,¹142 S. Ct. 1105 (2022). a case involving Black voters challenging Alabama’s 2021 congressional map under the Voting Rights Act of 1965 (“VRA”).²Pub. L. No. 89-110, 79 Stat. 437 (codified as amended in scattered sections of 52 U.S.C.).  This case is the latest entry in a nearly four-decade tradition wherein minority voters challenge discriminatory maps using a legal framework developed by the Supreme Court in 1986’s Thornburg v. Gingles.³478 U.S. 30 (1986).

In Gingles, Justice William Brennan’s majority opinion introduced a three-pronged diagnostic test for minority vote dilution (the Gingles test).⁴Id. at 50–51.  The Gingles test was adopted shortly after Congress amended Section 2 of the VRA to consider the “totality of circumstances” in a given jurisdiction.⁵Commonly referred to as the “Senate Factors,” the U.S. Senate Committee on the Judiciary issued a report that included seven factors “for courts to consider when determining if, within the totality of the circumstances in a jurisdiction, the operation of the electoral device being challenged results in a violation of Section 2.” S. Rep. No. 97-417, at 28–29.  In Gingles, the Court explained that Section 2 of the VRA is violated when, “based on the totality of circumstances, it is shown that . . . a class of citizens . . . [have] less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice.”⁶Gingles, 478 U.S. at 36.  Accordingly, minority plaintiffs bringing Section 2 challenges to districting maps must satisfy the Gingles test by demonstrating (1) that the minority voters are numerous and “geographically compact enough to constitute a majority in a single district,” (2) are “politically cohesive,” and (3) that they likely face consistent defeat by white-bloc voting.⁷Id. at 50–51.

The first prong requires plaintiffs to demonstrate an ability to draw an additional majority-minority district subject to the technical constraints known as traditional redistricting principles.  We will not cover the first prong here.  Rather, this Commentary covers the second and third prongs of the Gingles test.  Together, these prongs work to substantiate the doctrine of Racially Polarized Voting (“RPV”).

RPV analyses are conducted to comply with the VRA.  The second prong is referred to as political cohesion.  Qualitatively, this means that most of the plaintiff group supports the same candidate most of the time.  Cohesion supports a causal attribution to the structure of the election system rather than internal political division.  In practice, political cohesion is discerned by expert witnesses using some form of regression, most often ecological regression or ecological inference.  Cohesion analysis estimates the percentage of the voting population of the plaintiff group that supported their candidate of choice.

Within the same analysis, the statistical model will also produce corresponding estimates of white voting preferences.  That behavior informs the third prong, white-bloc voting, where the question is the degree to which white voters oppose minority-preferred candidates.  Here, courts are interested in the actual frequency of defeat as compared to the expected frequency of defeat given the minority political cohesion that was asserted through the second prong.  Victories for minority cases are not necessarily dispositive; the doctrine is more concerned with long-run performance.  High turnout and hotly contested elections between a minority candidate and a white candidate are of particular interest to the courts in determining white-bloc voting.

Contentiousness is just one quality that gives an electoral contest probative value in an RPV analysis.  In addition to the race of the candidates and the degree of competition, a contest is more probative if the election is endogenous to the jurisdiction and, depending on the context of the complaint, whether the election was a primary, general, or runoff.  Further, although some courts have declared that all election results matter, no matter how old, others contend that “stale” elections (those conducted in the past) are of little value.⁸Compare Nipper v. Smith, 1 F.3d 1171 (11th Cir. 1993) (holding that twenty-year-old election results should be considered in RPV analysis), and McMillan v. Escambia Cnty., 748 F.2d 1037 (5th Cir. 1984) (holding that prior election results demonstrate a history of discrimination and should be considered), with League of Latin Am. Citizens Council No. 4434 v. Clements, 999 F.2d 831 (5th Cir. 1993) (denouncing the probative value of dated election statistics).

In terms of quantitative analyses, there are no hard-and-fast rules when assessing RPV.  Generally, plaintiffs want to have at least three probative elections as evidence.  For congressional redistricting cases, judges will generally want to see more.  Although interpreting electoral statistics is subjective, the strength of an RPV argument often depends on the plaintiff’s ability to show a pattern of both minority political cohesion and white-bloc voting over time.

For more details on the RPV doctrine, I suggest the comprehensively descriptive article written by Professors Chris Elmendorf, Kevin M. Quinn and Marisa A. Abrajanov.  Law Professors Travis Crum and Samuel Issacharoff have also written authoritative articles on the topic.

References:

• 1
  142 S. Ct. 1105 (2022).
• 2
  Pub. L. No. 89-110, 79 Stat. 437 (codified as amended in scattered sections of 52 U.S.C.).
• 3
  478 U.S. 30 (1986).
• 4
  Id. at 50–51.
• 5
  Commonly referred to as the “Senate Factors,” the U.S. Senate Committee on the Judiciary issued a report that included seven factors “for courts to consider when determining if, within the totality of the circumstances in a jurisdiction, the operation of the electoral device being challenged results in a violation of Section 2.” S. Rep. No. 97-417, at 28–29.
• 6
  Gingles, 478 U.S. at 36.
• 7
  Id. at 50–51.
• 8
  Compare Nipper v. Smith, 1 F.3d 1171 (11th Cir. 1993) (holding that twenty-year-old election results should be considered in RPV analysis), and McMillan v. Escambia Cnty., 748 F.2d 1037 (5th Cir. 1984) (holding that prior election results demonstrate a history of discrimination and should be considered), with League of Latin Am. Citizens Council No. 4434 v. Clements, 999 F.2d 831 (5th Cir. 1993) (denouncing the probative value of dated election statistics).