MCCLURE v. JEFFERSON COUNTY COMMISSION (2025)

FINDINGS OF FACT, CONCLUSIONS OF LAW, AND ORDER

The plaintiffs in these consolidated cases have challenged the electoral map that the Jefferson County Commission enacted during the 2021 redistricting process. The plaintiffs contend that the Commission's map is racially gerrymandered in violation of the Equal Protection Clause of the Fourteenth Amendment to the United States Constitution. The Court held a four-day bench trial on the plaintiffs’ claims. (Docs. 172, 173, 174, 175).¹ Having heard the parties’ evidence and considered the parties’ written submissions concerning the evidence, (Docs. 177, 178, 181, 182, 183), the Court issues findings of fact and conclusions of law.

FINDINGS OF FACT

The Parties

1. The McClure plaintiffs are Cara McClure, Greater Birmingham Ministries, the Alabama State Conference of the National Association for the Advancement of Colored People, and the Metro Birmingham NAACP.

2. The Addoh-Kondi plaintiffs are Alexis Addoh-Kondi, Julia Juarez, Cynthia Bonner, Ja'Nelle Brown, Charles Long, Eric Hall, Michael Hansen, William Muhammed, Tammie Smith, Fred Lee Randall, and Robert Walker.

3. Ms. Addoh-Kondi is a lawfully registered Black voter who has resided in District 1 since 2020. (Doc. 171, p. 2, ¶ 6). Ms. Juarez is a lawfully registered white voter who has resided in District 1 since 2020. (See Doc. 171, p. 2, ¶ 7). Mr. Hall is a lawfully registered Black voter who has resided in District 1 since 2020. (Doc. 174, p. 23, tp. 516); (Doc. 171, p. 2, ¶ 11). Mr. Hansen is a lawfully registered white voter who has resided in Jefferson County since 2020. (Doc. 171, p. 3, ¶ 12). He currently resides in District 1. (Doc. 171, p. 3, ¶ 12). Ms. Smith is a lawfully registered Black voter who has resided in District 1 since 2020. (Doc. 171, p. 3, ¶ 14). Mr. Muhammed is a lawfully registered Black voter who has resided in District 1 since 2020. (Doc. 171, p. 3, ¶ 13). Mr. Walker is a lawfully registered Black voter who has resided in District 1 since 2020. (Doc. 171, p. 3, ¶ 16).

4. Ms. McClure is a lawfully registered Black voter who resides in District 2. (Doc. 172, pp. 200-01, tpp. 200-01). Between January 2020 and December 2022, Ms. McClure resided in District 3. (Doc. 171, p. 1, ¶ 1; see Doc. 172, pp. 212-13, tpp. 212-13). Mr. Randall is a lawfully registered Black voter who has resided in District 2 since 2020. (Doc. 171, p. 3, ¶ 15).

5. Ms. Brown is a lawfully registered Black voter who has resided in District 3 since 2020. (Doc. 171, p. 2, ¶ 9).

6. Ms. Bonner is a lawfully registered Black voter who has resided in District 4 since 2020. (Doc. 171, p. 2, ¶ 8).

7. Mr. Long is a lawfully registered white voter who has resided in District 5 since 2020. (Doc. 171, p. 2, ¶ 10).

8. Greater Birmingham Ministries is a multi-racial, multi-faith, 501(c)(3) non-profit organization dedicated to social services, community building, and political participation in Jefferson County and across the state of Alabama. (Doc. 173, pp. 170, 171, tpp. 462, 463). Greater Birmingham Ministries was founded in 1969 by Black and white lay people and clergy in response to the challenges posed by the Civil Rights Movement. (Doc. 173, p. 172, tp. 464). Greater Birmingham Ministries has organizational and individual members. (Doc. 173, pp. 172-73, tpp. 464-65). Its individual members include lawfully registered Black voters who live in Districts 1, 2, 3, and 4. (Doc. 173, p. 175, tp. 467). Scott Douglas is the Executive Director of GBM. (See Doc. 173, p. 170, tp. 462).

9. The Alabama NAACP is the state conference of the National Association for the Advancement of Colored People, Inc. (Doc. 173, p. 141, tp. 433). The Alabama NAACP is a civil rights organization whose mission is to ensure that citizens’ civil, human, and voting rights are protected. (Doc. 173, pp. 142-43, tpp. 434, 435).

10. The Alabama NAACP is comprised of all branches and units of the NAACP in Alabama, including the Metro Birmingham NAACP. (Doc. 173, pp. 142-43, tpp. 434-35). The Alabama NAACP is a membership organization, and everyone who is a member of a local branch or unit within the State of Alabama, including the Metro Birmingham Branch, also is a member of the Alabama NAACP. (Doc. 173, p. 143, tp. 435).

11. Metro Birmingham NAACP operates in Birmingham and in the Greater Birmingham area. (Doc. 173, pp. 142-43, tpp. 434-35). The Metro Birmingham NAACP has identified members who are Black registered voters in Districts 1, 2, 3, 4, and 5. (Doc. 171, p. 2, ¶ 5; see Doc. 173, p. 167, tp. 459).

12. The Jefferson County Commission is the governing body of Jefferson County, Alabama. Ala. Code § 45-37-72(b). Since 1985, the Commission has been comprised of five commissioners, each elected from a single-member district for a four-year term. Ala. Code § 45-37-72(a)-(b); Ala. Code § 11-3-1(c).

13. Pursuant to its duties under Alabama Code § 11-3-1.1(a), in 2021, following the release of the 2020 decennial census, the Commission developed and enacted a redistricting plan. (See generally Doc. 179-3; Doc. 179-5; Doc. 179-8).

14. During the Commission's 2021 redistricting process, Jimmie Stephens was president of the Commission and Commissioner for District 3; Joe Knight was president pro tempore of the Commission and Commissioner for District 4; Lashunda Scales was Commissioner for District 1; Sheila Tyson was Commissioner for District 2; and Steve Ammons was Commissioner for District 5. (See Doc. 179-8, p. 4).²

Witnesses at Trial

15. The Court heard lay testimony from Ms. McClure, Mr. Hall, and Mr. Douglas; Benard Simelton, the President of the Alabama State Conference of the NAACP; and Barry Stephenson, a Jefferson County resident and Chairman of the Jefferson County Board of Registrars. (Doc. 172, p. 200, tp. 200:13-19; Doc. 173, pp. 140–42, 169-71, tpp. 432:24-434:15, 461:3-463:18; Doc. 174, pp. 8, 155-56, tpp. 501:12-20, 648:7-649:7). The McClure plaintiffs called Ms. McClure, Mr. Douglas, and Mr. Simelton. The Addoh-Kondi plaintiffs called Mr. Hall. The Commission called Mr. Stephenson.

16. The Court heard testimony from several expert witnesses. The Addoh-Kondi plaintiffs called Anthony Fairfax as an expert in redistricting demographics and map drawing. (Doc. 172, p. 16, tp. 16:5-9). Mr. Fairfax performed a demographic analysis of the changes the Commission made to the districts in the 2021 plan. (Doc. 172, pp. 19–20, tpp. 19:14-20:15). He also developed an illustrative plan—or an alternative map—to serve as a comparison to the 2021 plan. (Doc. 172, pp. 50–53, tpp. 50:7-53:8).

17. The McClure plaintiffs called William Cooper as an expert in demography and redistricting. (Doc. 172, p. 228, tp. 228:6-9). Mr. Cooper analyzed how the Commission district boundaries evolved as Jefferson County's demographics changed between the 1980s and the 2021 redistricting cycle. (Doc. 172, pp. 229–77, tpp. 229:22-277:9). Mr. Cooper also developed illustrative plans. (Doc. 172, pp. 277-87, tpp. 277:11-287:15).

18. The McClure plaintiffs called Dr. Baodong Liu as an expert in racially polarized voting, regression analysis, and ecological inference. (Doc. 173, p. 78, tp. 370:9-14). Dr. Liu performed a racially polarized voting analysis for Jefferson County and conducted an effectiveness analysis to determine how reducing the Black population in Districts 1 and 2 would impact the ability of Black-preferred candidates to win elections. (Doc. 173, pp. 79, 91–92, tpp. 371:18-23, 383:23-384:8).

19. The McClure plaintiffs called Dr. Cory McCartan as an expert in statistics and redistricting simulation algorithms. (Doc. 174, p. 46, tp. 539:7-12). Dr. McCartan analyzed a simulation created by the Commission's expert witness, Dr. Michael Barber. (Doc. 174, p. 46, tp. 539:17-21). Dr. McCartan conducted a simulation analysis to test the extent to which core retention explained the racial demographics in the 2021 plan. (Doc. 174, p. 47, tp. 540:3-7).

20. The Commission called Dr. Barber as an expert in political science, redistricting, and demographics. (Doc. 175, pp. 9–10, tpp. 745:22-746:2). Dr. Barber evaluated the analyses performed by the plaintiffs’ experts. (Doc. 175, p. 14, tp. 750:12-23). Dr. Barber conducted a precinct-by-precinct analysis of the changes the Commission made during the 2021 redistricting process, (Doc. 179-16, pp. 10–39); analyzed the Commission's movement of precincts in the 2021 plan using a regression model, (Doc. 175, p. 43, tp. 779:5-21); created a simulation to measure how the demographics in the 2021 plan compared to alternative plans; and analyzed Dr. McCartan's simulation analysis, (Doc. 179-16, pp. 44–51; Doc. 179-17, pp. 3–6).

21. In addition to the witness testimony, the Court admitted into evidence public records from the Commission's 2021 redistricting process, including transcripts of the Commission meetings during which the commissioners considered redistricting proposals and a presentation from Mr. Stephenson. (Doc. 179-3; Doc. 179-5; Doc. 179-8).

22. The Court admitted into evidence historical evidence that documents the evolution of the Commission districts from the 1980s through the 2013 redistricting process, including the Commission's 1985, 1993, 2001, 2004, and 2013 preclearance submissions to the United States Department of Justice under Section 5 of the Voting Rights Act. (Doc. 169-2; Doc. 169-3; Doc. 169-4; Doc. 169-5; Doc. 169-6).

Historical Evidence

The 1985 Consent Decree

23. In 1931, the Alabama Legislature established the Commission to govern Jefferson County. (Doc. 169-2, p. 3, ¶ 10).

24. Between 1931 and 1985, by statute, the Commission had three commissioners who were elected at-large by voters in Jefferson County. (Doc. 169-1, p. 2, ¶ 4 (citation omitted); Doc. 169-2, p. 3, ¶ 10).

25. Over that 55-year period, voters in Jefferson County did not elect a Black candidate for commissioner even though commissioners ran for election every four years, and by the 1980s, Black residents comprised approximately one-third of the county's total population. (Doc. 169-2, p. 2, ¶ 7); (Doc. 169-6, p. 7) (explaining that dating to the 1930s, the at-large commissioners for Jefferson County served four-year terms); see Yeldell v. Cooper Green Hosp., Inc., 956 F.2d 1056, 1058 (11th Cir. 1992) (explaining that the first Black commissioner was elected to the Commission in November 1986).

26. On July 2, 1984, Michael Taylor, Willie L. Allen, and Anita Smith sued the Jefferson County Commission, the three county commissioners, the Probate Judge of Jefferson County, and the Alabama Secretary of State. (Doc. 169-1, p. 1; Doc. 169-2, p. 1; Doc. 176-2, pp. 1–3, ¶¶ 3, 5–8).

27. The plaintiffs in that case, Taylor v. Jefferson County Commission et al., No. CA-84-C-1730-S, alleged that the at-large structure of the Commission discriminated against them based on race and diluted minority voting power in violation of Section 2 of the Voting Rights Act and the Fourteenth and Fifteenth Amendments to the United States Constitution. (Doc. 176-2, pp. 5–6, ¶¶ 21-22, 23, 25(a)).

28. The Taylor plaintiffs highlighted the “long history of official purposeful discrimination against and disfranchisement [sic] of qualified [B]lack voters” that had denied those minority voters “the opportunity to register, to vote, and otherwise to participate in the democratic process” in Jefferson County. (Doc. 176-2, p. 3, ¶ 14).

29. The Taylor plaintiffs asserted that voting in Jefferson County had been “racially polarized in elections in which a [B]lack candidate [ran] for office, with white voters generally voting for white candidates and [B]lack [voters] voting for non-white candidates for elective office.” (Doc. 176-2, pp. 3–4, ¶ 15).

30. As a result, the Taylor plaintiffs alleged, though Black candidates had run for office, “no [B]lack candidates ha[d] been elected to the Jefferson County Commission and [B]lack voters ha[d] been denied the opportunity to elect candidates of their choice to elective office.” (Doc. 176-2, pp. 4, 5, ¶¶ 18, 21).

31. The Taylor plaintiffs asserted that single-member districts in Jefferson County could be drawn to allow Black voters an opportunity to elect candidates of their choice. (Doc. 176-2, p. 6, ¶ 24).

32. The Taylor plaintiffs alleged that as of the filing of their complaint:

According to the 1980 Census, the population of Jefferson County [was] 671,324, of which 66% [of the individuals were] white and 33% [were] [B]lack residents of Jefferson County, Alabama.

According to the 1980 Census, the [B]lack population [was] concentrated in the western section of Birmingham and the eastern section of the Bessemer Cut-off․

The [B]lack population of Jefferson County [was] sufficiently numerous and sufficiently concentrated in particular areas of the county, so that if members of the governing body were elected from single member districts, [B]lack [individuals] would be a voting majority in some of those districts and would have a reasonable opportunity to elect candidates of their choice to the county governing body.

(Doc. 176-2, pp. 3, 6, ¶¶ 11, 12, 24).³

33. The parties negotiated a consent decree to resolve the Taylor case. (Doc. 169-1).

34. As part of the consent decree proceedings, the parties filed a “Joint Statement of Principal Facts” that described Alabama's and Jefferson County's systemic discrimination against Black voters, discrimination that impacted the ability of Black citizens to participate in the political process and demonstrated the existence of racially polarized voting in elections. (Doc. 176-2, pp. 8–13, 40–45, ¶¶ 1–13, 111–31).⁴ The polarized voting led to the defeat of Black preferred candidates in Jefferson County. (See Doc. 176-2, pp. 8–13, 40–45, ¶¶ 1–13, 111–31).

35. The parties stated that Jefferson County's “at-large method of selection” had been established in 1819. (Doc. 176-2, p. 58, ¶ 175).

36. The Taylor parties stated that in 1984, 31.3% of the voting age population in Jefferson County was Black, and 72.16% of the Black population in Jefferson County over the age of 18 was registered to vote in the November 1984 election, meaning that 72.16% of the Black Voting Age Population or BVAP was registered to vote. (Doc. 176-2, pp. 48, 52, ¶¶ 146, 152). In Jefferson County in 1984, “a larger portion of the Voting Age Population [was] registered to vote among [B]lacks than among whites.” (Doc. 176-2, p. 12, ¶ 18; see also Doc. 172, p. 244, tp. 244:21–23). The BVAP at the time was only three percentage points less than the total Black population. (Doc. 176-2, p. 13, ¶ 19).

37. The Taylor parties agreed that from 1900 to 1984, no Black person had been elected as a Jefferson County Commissioner even though Black candidates had run for office. (Doc. 176-2, pp. 52–53, ¶¶ 151, 159).⁵

38. The Court entered the consent decree on August 17, 1985. (Doc. 169-1). The consent decree established five single-member districts from which voters would elect five Jefferson County commissioners. (Doc. 169-1, p. 1).

39. Under the decree, Jefferson County had to “promptly file” the decree “with the Attorney General of the United States, along with the necessary supporting material, for preclearance under Section 5 of the Voting Rights Act.” (Doc. 169-1, p. 2, ¶ 6).⁶

40. The parties amended the consent decree on November 1, 1985. (Doc. 169-2, pp. 12–25). The parties attached to the amended consent decree an exhibit that “show[ed] the racial composition of each Commission district by reference to census enumeration tracts” and a map that depicted the boundaries of the five districts. (Doc. 169-2, pp. 13, 19–25).

41. On November 18, 1985, the Commission submitted the consent decree, the amended consent decree, supporting exhibits, and an explanatory letter to the United States Department of Justice pursuant to Section 5 of the VRA. (Doc. 169-2).

42. In the letter to the DOJ, the Commission stated that the consent decree abolished Jefferson County's at-large election of county commissioners and created five districts, two of which were drawn to provide Black voters “with a greater opportunity to elect [B]lack commissioners” to resolve “a minority vote dilution case.” (Doc. 169-2, p. 2, ¶ 7).

43. Districts 1 and 2 were “formed out of the central area of [Jefferson County], primarily within the City of Birmingham.” (Doc. 169-6, p. 7).

44. Under the consent decree, the population of District 1 was 65.6% Black, and the population of District 2 was 66.8% Black. (Doc. 169-2, p. 2, ¶ 8). The decree did not provide explicit racial targets in Districts 1 and 2.

45. In the 1980s, nationally, “there was a general understanding ․ that to have an effective district that would elect a [B]lack candidate of choice, at least in many parts of the country you needed a district that was in the range of 65 percent [B]lack.” (Doc. 172, p. 241, tp. 241:6–13). Nationally, the 65% threshold “took into account lower rates of registration, lower rates of turnout for [B]lack persons of voting age, and also a wider gap between the all ages population and Voting Age Population so that oftentimes the younger population would be 3 or 4 percentage points higher [B]lack than the Voting Age Population.” (Doc. 172, p. 241, tp. 241:13–19).⁷

46. The Commission stated that it anticipated that the “effect” of the new commission structure would be that Black voters would have “a greater opportunity to elect 2/5 or 40% of the County Commission positions.” (Doc. 169-2, p. 2, ¶ 8); see also Yeldell, 956 F.2d at 1058 (citing consent decree and explaining that the decree produced “a five-member district form of government to ensure greater minority representation on the Commission”).

47. In January 1986, the DOJ approved the change to the structure of the Jefferson County Commission. (Doc. 169-6, p. 7).

48. The map below shows the Commission districts as drawn in 1985.

(Doc. 176-3, p. 3).⁸

49. In November 1986, in the first election under the five-district system, voters in Districts 1 and 2 elected Black candidates to serve as commissioners, and voters in Districts 3, 4, and 5 elected white candidates to serve as commissioners. (Doc. 169-6, p. 7).

50. Four years later, in November 1990, voters in Districts 1 and 2 elected Black candidates to serve as commissioners, and voters in Districts 3, 4, and 5 elected white candidates to serve as commissioners. (Doc. 169-6, p. 8).

51. Under Alabama law, after each decennial census, the Commission could alter the boundaries of the districts.⁹

52. Following the 1990, 2000, and 2010 United States censuses, the Commission redrew its electoral map and, pursuant to Section 5, submitted for and received approval from the DOJ for the changes to district boundaries. (See Doc. 169-3; Doc. 169-6; Doc. 169-5).¹⁰

The 1990 Census and Redistricting

53. The Commission's 1993 DOJ preclearance submission begins: “This letter will constitute a submission under Section 5 of the Voting Rights Act regarding the alteration of the Jefferson County Commission single-member district boundary lines, following the 1990 United States Census.” (Doc. 169-3, p. 1).

54. At the time of the 1990 census, the Black population in Jefferson County was nearly 35% of the total population. (See Doc. 169-3, p. 109). Per the 1990 census, District 1's total population was 110,084 and was 72.58% Black, District 2's total population was 122,225 and was 80.40% Black, District 3's total population was 131,711 and was 22.89% Black, District 4's total population was 147,664 and was 7.48% Black, and District 5's total population was 139,781 and was 10.74% Black. (Doc. 169-3, p. 109).

55. As compared to the preceding census, the white population in Districts 1 and 2 had fallen fairly significantly, the white population in District 3 had fallen marginally, and the white population in Districts 4 and 5 had increased marginally. (Doc. 169-3, p. 109). The Black population in District 1 had fallen fairly significantly, the Black population in District 2 had increased fairly significantly, the Black population in District 3 had increased marginally, and the Black population in Districts 4 and 5 had nearly doubled. (Doc. 169-3, p. 109).

56. The 1990 census showed that Districts 1 and 2 were 15.52% and 6.2% below ideal population respectively. (Doc. 169-3, p. 2). Districts 3, 4, and 5 were 1.13%, 13.32% and 7.27% above ideal population respectively. (Doc. 169-3, p. 2).¹¹

57. The Commission reported that the ideal district population was 130,305. (Doc. 169-3, p. 2, ¶ 3; Doc. 169-3, p. 109).

58. After the Commission revised the boundaries of the five districts, District 1's population was 126,847 individuals and was 73.25% Black. (Doc. 169-3, p. 110). District 2's population was 132,635 individuals and was 68.93% Black, District 3's population was 130,166 individuals and was 21.22% Black, District 4's population was 132,792 individuals and was 6.05% Black, and District 5's population was 129,056 individuals and was 6.34% Black. (Doc. 169-3, p. 110).

59. Before redistricting, District 1 consisted of 79,896 Black individuals (72.58%) and 30,188 white individuals (27.42%). (Doc. 169-3, p. 109). In redistricting, the Commission added 13,025 Black individuals to the district to achieve a total Black population of 92,921, or 73.25% of the total population in District 1. (See Doc. 169-3, pp. 109–10). The Commission added 2,926 white individuals to District 1. (See Doc. 169-3, pp. 109–10). Following redistricting, the percentage of white individuals in District 1 fell to 26.1%. (See Doc. 169-3, pp. 109–10).

60. The Commission moved 5,623 Black individuals from District 2 to other districts, leaving a total Black population of 91,423, or 68.93% of the population in District 2. (See Doc. 169-3, pp. 109–10).

61. District 3's Black population fell by more than 2,500 individuals, decreasing from 22.89% to 21.22% of the district's population. (See Doc. 169-3, pp. 109–10). District 3's white population grew by 235 individuals; District 3's percentage of white population increased from 77.11% to 78.25%. (See Doc. 169-3, pp. 109–10).

62. District 4's Black population grew by roughly 1,200 individuals, with a corresponding increase in the district's Black population by about 1%. (See Doc. 169-3, pp. 109–10). District 4 shed nearly 13,000 white individuals. (See Doc. 169-3, pp. 109–10).¹²

63. The Commission stated that its proposed electoral district changes brought “each district close to the ideal district population, without significantly changing the ratio of [B]lack and white population within the districts.” (Doc. 169-3, p. 2, ¶ 3). The Commission stated that the changes it undertook would equalize population “without significantly altering the racial ratios” within the districts. (Doc. 169-3, p. 3). The Commission did not evaluate the need to maintain 65% Black population levels in Districts 1 and 2 using an RPV analysis or an assessment of Black voter registration and turnout. (See Doc. 169-3).

64. The Commission submitted the following district map to DOJ for approval:

(Doc. 169-3, p. 110).

65. As illustrated in the above map, Districts 3, 4, and 5 surrounded Districts 1 and 2. (Compare Doc. 169-97, p. 2, with Doc. 169-3, p. 110). The Commission altered the composition of Districts 1 and 2 so that District 2, while still to the southwest of District 1, extended in a thin fashion eastward above District 5. (Doc. 169-3, p. 110).

66. A newspaper clip attached to the Section 5 documentation and published shortly after the Commission adopted its 1993 redistricting plan stated that federal law required commission districts to have “a fairly equal number of people in each district and that two of the districts have a majority of [B]lack residents.” (Doc. 169-3, p. 130). The article added that following redistricting, District 1, “a majority [B]lack district that include[d] much of Birmingham and the middle of Jefferson County,” was the “smallest” district and was more than 73% Black. (Doc. 169-3, p. 130). In contrast, District 3, which “cover[ed] almost half the county,” was 78% white. (Doc. 169-3, p. 130). The article described District 2 as a “majority [B]lack district stretching from Bessemer to Birmingham's southside.” (Doc. 169-3, p. 130).

67. Another article attached to the Section 5 documentation reported that Commissioner Gary White stated that “moving district lines was necessary to meet federal guidelines keeping districts roughly equal in population and keeping two majority [B]lack districts.” (Doc. 169-3, p. 134). According to the article, “[b]ecause [Commissioner] White's district needed to shrink to lose nearly 9,000 residents and [Commissioner] McNair's needed to grow, the new lines placed much of Birmingham's Southside into [Commissioner] McNair's district.” (Doc. 169-3, p. 134). The article quoted Commissioner White: “We had to swap somewhere, and this was where our (district) lines touched.” (Doc. 169-3, p. 134).

68. The DOJ cleared the Commission's 1993 map. (Doc. 169-3, p. 135).

69. In November 1994, voters in Districts 1 and 2 elected Black candidates to serve as commissioners, and voters in Districts 3, 4, and 5 elected white candidates to serve as commissioners. (Doc. 169-6, p. 8).

70. In November 1998, voters in the five districts re-elected the same five commissioners. (Doc. 169-6, p. 8).¹³

The 2000 Census and Redistricting

71. In 2001, the Commission sought approval from the DOJ “pursuant to Section 5 of the Voting Rights Act” for “the single-member election districts[’] boundary lines altered by the Jefferson County Commission ․ for use in elections beginning in June 2002.” (Doc. 169-6, p. 4).

72. In its 2001 DOJ submission, the Commission noted that the county's Black population had increased and constituted 39% of the county's population. (Doc. 169-6, p. 6). At the time of the 2000 census, Jefferson County had a population of 662,047. (Doc. 169-6, p. 6).

73. The white population in Districts 1 and 2 fell significantly; the white population in District 1 fell nearly 50%, and the white population in District 2 fell slightly more than 30%. (Doc. 169-6, p. 42). The white population in Districts 3 and 5 grew marginally, and the white population in District 4 decreased marginally. (Doc. 169-6, p. 43).

74. The Black population in District 1 fell marginally, the Black population in District 2 grew marginally, the Black population in District 3 increased somewhat, the Black population in District 4 tripled, and the Black population in District 5 doubled. (Doc. 169-6, pp. 42–43).

75. Districts 1 and 2 lost 17% and 9% of their populations respectively, and Black individuals were 82.13% of the population in District 1 and 76.70% of the population in District 2. (Doc. 169-6, p. 11). Districts 3, 4, and 5 were overpopulated by 7.70%, 9.81%, and 8.69% respectively. (Doc. 169-6, p. 11).

76. The Commission reported an ideal district population of 132,409. (Doc. 169-6, p. 10).

77. In its preclearance submission to the DOJ, the Commission described the commissioners representing each district by race and gender and stated that since 1986, two of the five districts had Black populations higher than 65%, and voters in the majority-Black districts consistently had elected Black candidates. (Doc. 169-6, pp. 7–8).

78. The Commission submitted to the DOJ the version of the map it approved, the other two maps the Commission advertised, and a copy of a map of the 1993 districts which contained “an identification of the location of [B]lack populations and white populations.” (Doc. 169-6, p. 13).¹⁴

79. In its 2001 letter, the Commission explained its efforts to redraw district lines:

To re-arrange the boundaries to obtain population for compliance with the one person-one vote rule, [Districts 1 and 2] were expanded somewhat outward, as compared with their previous boundaries. District One expanded westward to include large parts of the Ensley area, and the City of Fairfield. District One's northern boundary also moved, somewhat to the north and took in virtually all of the City of Fultondale. District Two moved west further. It gave up Fairfield and parts of Ensley and expanded its western boundary, especially through most of the City of Bessemer, in a southwesterly direction. District Two also moved some to the south to take in parts of the Southside of Birmingham previously located in District Five. These expansions of the boundaries of District One and District Two leave them located centrally in Jefferson County, although extended somewhat out of the central area.

(Doc. 169-6, p. 12).

80. The Commission stated that it had considered three maps, each of which placed the City of Fairfield, whose population was 90% Black, in a different district. (Doc. 169-6, p. 13). Fairfield drove the three map iterations because the mayor of Fairfield, a Black man, wanted to run for a seat on the Commission. (Doc. 169-6, p. 13). The Commission described its three options this way:

In Plan 1, [Mayor] Langford was located in District 3, which is composed overwhelmingly of white voters. The incumbent in District 3 is white. In Plan 2, Langford was located in District 1, which is composed overwhelmingly of [B]lack voters. The incumbent in District 1 in this plan is [B]lack. In Plan 3, Langford was located in District 5, which is composed overwhelmingly of white voters. The incumbent in District 5 in this plan is white.

(Doc. 169-6, p. 13).

81. Commissioner Germany moved to “scrap” the three maps and to hire a consultant to prepare a map that would “pass legal scrutiny and be fair to the citizens of Jefferson County.” (Doc. 169-6, p. 65).

82. The Commissioners voted 3-2 against the motion and then voted 3-2 to adopt Plan 2. (Doc. 169-6, p. 65).

83. The Commission added 10,019 white individuals and 15,098 Black individuals to District 1, maintaining Black voters’ supermajority in District 1. (See 169-6, pp. 42, 44).

84. In District 2, the Commission added 5,999 white individuals and 4,655 Black individuals, boosting Black individuals’ percentage of the District 2 population to 73.45%. (See 169-6, pp. 42, 44).

85. Under the 2001 plan, the Black population in District 1 was 78% and the Black population in District 2 was 73.45%. (Doc. 169-6, pp. 11, 15, 44).¹⁵

86. In District 3, the Commission removed more than 15,000 Black individuals and added roughly 3,500 white individuals, lowering the total percentage of the population that Black individuals comprised to 17.14%. (See Doc. 169-6, pp. 43–44).

87. Districts 4 and 5 shed both white and Black individuals. (Doc. 169-6, pp. 43, 45). Though the districts lost 9,062 and 10,480 white individuals and 2,624 and 973 Black individuals, respectively, the overall percentage of Black population in each district remained roughly equal to the Black population at the time of the 2000 census—the percentage fell from 16.42% to 15.29% in District 4 and increased from 11.48% to 11.88% in District 5. (See Doc. 169-6, pp. 43, 45).

88. For the first time in 2001, the Commission reported VAP statistics. (Doc. 169-6, pp. 44–45). The statistics showed that under the proposed district lines, District 1 had a BVAP of 74.89%; District 2 had a BVAP of 68.78%; and in Districts 3, 4, and 5, the BVAP was, respectively, 15.83%, 12.92%, and 11.01%. (Doc. 169-6, pp. 44–45).

89. The Commission explained that the new plan was drawn “again with two districts containing African American majorities in excess of 65%.” (Doc. 169-6, pp. 8–11). The Commission stated that its changes brought “each district close to the ideal district population without significantly changing the ratio of [B]lack and white population within the districts.” (Doc. 169-6, p. 11). The Commission anticipated that the “effect of the change on members of racial or language minority groups [was] insignificant.” (Doc. 169-6, p. 11). The Commission did not submit an RPV analysis or an assessment of Black voter registration and turnout with the 2001 Section 5 submission. (See Doc. 169-6).

90. The following maps illustrate the contrast between the district borders in 1993 and the district borders following the 2001 redistricting:

1993 Commission Map (Left) and 2001 Commission Map (Right)

(Doc. 169-3, p. 110); (Doc. 169-4, p. 218) (red boundary and black numbering overlay added for legibility).

91. According to Commissioner Bettye Fine Collins, “all that matter[ed] [was] that the new lines w[ould] [ ] keep the two majority-[B]lack districts and keep about the same number of people in each district.” (Doc. 169-6, p. 48).

92. Commissioner Germany described the plan as “a clear violation of the Voting Rights Act,” and his lawyer “accused [the commissioners] of changing a redistricting plan to maintain a white majority” by pitting Mr. Langford against Commissioner Germany, a Black commissioner, rather than against a white commissioner. (Doc. 169-6, pp. 49–50).

93. Commission candidate Geraldine Bell argued that the plan would dilute the voting strength of Black voters in her neighborhood. (Doc. 169-6, p. 51).

94. The DOJ did not object to the 2001 redistricting. (Doc. 169-6, p. 103).

95. Following the 2001 redistricting, voters in Jefferson County filed a lawsuit in which they asserted that the new district lines diluted the voting strength of Black voters in the county in violation of Section 2 of the VRA. Knott v. Jefferson Cnty. Comm'n, No. 2:02-cv-00030-MHT (N.D. Ala. 2002); (Doc. 169-4, p. 2).

96. The case was dismissed following the Commission's adoption of a plan that adjusted three voting boxes affecting approximately 10,000 people in three of the five districts. (Doc. 169-4, pp. 9, 10-13, 197–98). The adjustments did not affect District 1. (See Doc. 169-4). With respect to District 2, the Commission stated in its Section 5 submission that “[t]he alteration of the boundaries of District 2 ha[d] no significant effect on the racial make-up of District 2” and that each of Districts 1 and 2 retained “majority [B]lack populations in excess of 65% under both the 2001 plan” and after the adjustments. (Doc. 169-4, p. 5). The County did not submit an RPV analysis or an assessment of Black voter registration and turnout with the 2004 Section 5 submission to the DOJ. (See Doc. 169-4).

The 2010 Census and Redistricting?

97. The Commission's final Section 5 DOJ submission in 2013 is similar to the 1985, 1993, and 2001 submissions. As in 1993 and 2001, the Commission sought approval from the DOJ “pursuant to Section 5 of the Voting Rights Act” for “the single member election districts[’] boundary lines ․ altered by the Jefferson County Commission ․ for use in elections beginning in June 2014.” (Doc. 169-5, p. 1075).

98. The Commission noted that the county's Black population had grown to comprise 41% of the population. (Doc. 169-5, p. 1075). Following the 2010 census, the population of Jefferson County was 658,466. (Doc. 169-5, p. 1075).

99. Based on the 2010 census, Districts 1 and 2 were underpopulated by 12.20% and 9.25%, while Districts 3, 4, and 5 were overpopulated by 7.50%, 8.07%, and 5.90% respectively. (Doc. 169-5, p. 1080).

100. The Commission described the commissioners representing each district by their race and by their gender. (Doc. 169-5, pp. 1076–78).

101. The Commission indicated that since 1986, two districts had had Black populations greater than 65% and had elected Black commissioners. (Doc. 169-5, p. 1075).

102. The Commission stated: “In the new plan, the districts are drawn in such a way that incumbent African-Americans represent districts in which the population is majority [B]lack.” (Doc. 169-5, p. 1083).

103. The Commission stated: “The 2013 plan has two [B]lack majority districts, just like the 1993 and 2001 plans. Each of these districts have [sic] majority [B]lack populations in excess of 65%, under the 2013, 2001, and 1993 plan[s].” (Doc. 169-5, p. 1081).

104. The Commission reported that the Black population of District 1 was 76.14% of the district's population, and the Black population of District 2 was 73.39% of the district's population. (Doc. 169-5, p. 1081).

105. The Commission stated that the “change in district boundaries w[ould] bring each district close to the ideal district population without significantly changing the ratio of [B]lack and white population within the districts,” with the “anticipated effect of the change on members of racial or language minority groups [being] insignificant.” (Doc. 169-5, pp. 1080–81).

106. The Commission described the boundary revisions to the DOJ as follows:

[Districts 1 and 2] were expanded somewhat outward, as compared with their previous boundaries. District 1 expanded westward to include large parts of the Forestdale area, and the City of Midfield. District 1's northern boundary also moved, somewhat to the northeast and took in parts of Center Point and East Birmingham. District 2 moved north and south further. It gave up Midfield and moved into central Birmingham and to the south taking in parts of Homewood. These expansions of the boundaries of District 1 and 2 leave them located centrally in Jefferson County, although extended somewhat out of the central area.

(Doc. 169-5, p. 1081).

107. Post-redistricting, although District 1 lost 5,433 Black individuals, the overall percentage of Black population in the district again increased, this time to 76.14%. (Doc. 169-5, p. 1081).

108. District 2 gained 344 Black individuals, and the percentage of Black individuals in District 2 dropped by 0.06% to 73.39%. (Doc. 169-5, p. 1081).¹⁶

109. The Commission did not conduct, consider, or submit an RPV analysis or an assessment of Black voter registration and turnout in adopting the 2013 plan. (See Doc. 169-5; Doc. 174, p. 227, tp. 720:2-4).

110. The DOJ precleared the map on April 26, 2013. (Doc. 169-5, p. 1074).¹⁷

111. Two of the commissioners who adopted the plan at issue in 2021, Commissioner Knight and Commissioner Stephens, were on the Commission in 2013 and in the meeting where the 2013 plan was discussed and eventually adopted. (Doc. 174, p. 219, tp. 712:16–21).

112. Less than two months after the Commission received preclearance for the 2013 plan, the Supreme Court issued Shelby County v. Holder, 570 U.S. 529, 133 S.Ct. 2612, 186 L.Ed.2d 651 (2013). Following that decision, the Commission was relieved of its obligation to submit a Section 5 preclearance letter following the 2020 census. See 570 U.S. at 550, 557, 133 S.Ct. 2612.¹⁸

Demographic Changes in Jefferson County between 1985 and 2021

113. As noted earlier, much of Jefferson County's Black population resided in Birmingham when the Taylor parties finalized their consent decree. (Doc. 172, pp. 240, 261, tpp. 240:10–18, 261:1–6). In the 1985 map, District 1 was based in Birmingham and ran up towards Fultondale and out towards Irondale. (Doc. 172, p. 261, tp. 261:1–3). District 2 covered part of Birmingham and ran into Midfield and south towards Brighton. (Doc. 172, p. 261, tp. 261:5–6).

114. In 1990, the Black population in Jefferson County was heavily concentrated in Birmingham and in some municipalities to the south of the city like Fairfield, Midfield, and Bessemer. (Doc. 172, p. 262, tp. 262:12–15; Doc. 169-73). Only one precinct to the north of downtown Birmingham was majority Black. (Doc. 172, p. 262, tp. 262:16–18; Doc. 169-73). The map below describes these demographics.

(Doc. 169-73).

115. Jefferson County's Black population expanded to the northeast and to the southwest in the 1990s, as described in the map below. (Doc. 173, p. 54, tp. 346:1–3; Doc. 169-116).

(Doc. 169-116).

116. Ms. McClure's and Mr. Douglas's observations of their neighborhoods and communities corroborated these maps. Ms. McClure noted an increase in Black residents moving from Birmingham to Hoover. (Doc. 172, p. 210, tp. 210:9–14). She testified that Homewood increased in Black population when she lived there. (Doc. 172, pp. 211–12, tpp. 211:21–212:4). Black residents left Birmingham and moved northeast to Center Point, Trussville, and Irondale. (Doc. 172, p. 212, tp. 212:5–19). According to maps produced by Mr. Cooper, these areas have increased in Black population over the past three decades. (Doc. 169-113; Doc. 169-114; Doc. 169-116).

117. Mr. Douglas testified that in the 1980s, Powderly, in southwest Birmingham, was majority white and is now “vastly majority [B]lack [with an] increasing number of Latinos.” (Doc. 173, pp. 182–83, tpp. 474:15–475:10). Mr. Douglas testified that neighborhoods like West End and municipalities like Lipscomb, Midfield, Brighton, and Pleasant Grove also have shifted from predominantly white to significantly or majority Black over the last few decades. (Doc. 173, p. 188, tp. 480:15–20).

118. In 1992, Mr. Douglas moved to Huffman, 13 miles northeast of downtown Birmingham. (Doc. 173, p. 184, tp. 476:8–15). His family was the first African-American family in the neighborhood. (Doc. 173, p. 184, tp. 476:21–23). He noted the demographics began to change in the late 1990s. (Doc. 173, p. 184, tp. 476:24–25). Due to white flight, white families moved to places like Trussville. (Doc. 173, p. 185, tp. 477:4–8). Black residents moved to Huffman from north Birmingham areas like the Collegeville and West End neighborhoods. (Doc. 173, p. 185, tp. 477:9–13). Today, Huffman is majority Black. (Doc. 173, p. 184, tp. 476:16–19).

119. Mr. Douglas noted other examples of white flight, stating that Midfield was “vast majority white” when he first moved to the area and now is majority Black and has elected its first Black mayor. (Doc. 173, pp. 188–89, tpp. 480:21–481:11). He testified that neighborhoods in Birmingham like East Lake and Roebuck and cities like Center Point and Pinson have shifted from predominantly white to predominantly Black or mixed race. (Doc. 173, p. 189, tp. 481:12–19). Mr. Douglas corroborated Ms. McClure's testimony that Homewood, Hoover, and Vestavia Hills have increased in Black population. (Doc. 173, p. 190, tp. 482:1–3).

120. From Ms. McClure's and Mr. Douglas's testimonies, the Court finds that demographic shifts in Jefferson County are evident and widely understood by Jefferson County residents, including the commissioners. The Commission placed many of the neighborhoods and suburbs to which Black voters moved between 1990 and 2020 into Districts 1 and 2. (Compare Doc. 169-113, with Doc. 169-114; Doc. 169-116). Over time, District 1's boundaries have shifted to follow Black population growth, extending district lines north towards Tarrant, Center Point, and Pinson. (Doc. 172, pp. 264–65, tpp. 264:23–265:6).

121. In 1990, the area that District 1 currently occupies was 47.7% Black; it was 76.75% Black in 2020. (Doc. 169-75; Doc. 169-113; Doc. 169-114; Doc. 169-115). For the past three decades, the Commission drew district lines that “tracked north and picked up a lot of population that is today predominantly [B]lack that would have been slightly majority white in the 1990s.” (Doc. 172, pp. 264–65, tpp. 264:23–265:3). District 2 would have been 68.93% Black in 1990 and 64.75% Black in 2020. (Doc. 172, p. 265, tp. 265:8–16; Doc. 169-75). The table below captures this demographic data.

(Doc. 169-75).¹⁹

122. The maps below plot the 1990 and 2020 census data against the 2021 plan's lines. (Doc. 169-113; Doc. 169-114; Doc. 169-115).

(Doc. 169-113) (2021 plan with 1990 census data).

(Doc. 169-114) (2021 plan with 2020 census data).

123. For example, the municipalities of Tarrant, Center Point, and Fultondale now have significant Black populations and are in District 1 under the 2021 plan, but these areas were predominantly white and outside of District 1 in the 1990s. (Doc. 172, pp. 264–65, 267, tpp. 264:23–265:6, 267:9–13). Center Point had a Black population that was approximately 5% in 1990; today, according to the 2020 census, Center Point is 75% Black. (Doc. 172, p. 266, tpp. 266:7–9).

2021 Commission Redistricting Process

124. The Commission received 2020 census data in August of 2021. (Doc. 174, p. 160, tp. 653:9–21). Mr. Stephenson and his staff loaded this data into GIS mapping software. (Doc. 174, pp. 158, 160, tpp. 651:16–18, 653:13–14).

125. As of 2020, Jefferson County's overall population had grown 2.2% to 674,721, but Districts 1 and 2 lost population, creating an imbalance in the populations of the Commission's five districts. (Doc. 169-12, pp. 3–7).

126. Between 2010 and 2020, Jefferson County's non-Hispanic white population decreased and the Black and Latino populations increased.

2010 to 2020 Census Population by Race and Ethnicity

(Doc. 169-28). Between 2010 and 2020, the percentage of white population in Jefferson County decreased by 3.61%; the percentage of Black population increased slightly by 0.37%. (Doc. 172, pp. 58–59, tpp. 58:23–59:2; Doc. 169-107, p. 15).

127. According to the 2020 census, Jefferson County's VAP was 50.42% non-Hispanic white, 41.46% Black, and 4.29% Latino. (Doc. 169-29).

128. Per the 2020 census, population shifted away from central Jefferson County in Districts 1 and 2 and into Districts 3, 4, and 5. (Doc. 179-3, pp. 5–6; Doc. 179-16, p. 5; Doc. 173, pp. 14, 15, 16–17, tpp. 306:21–23, 307:16–24, 308:24–309:3).

129. Mr. Stephenson identified a population target of 134,944 individuals per district. (Doc. 169-12, p. 3). The Commission used a plus or minus one percent population variance because the software allowed this level of population accuracy. (Doc. 174, p. 167, tp. 660:7–10).

130. Before redistricting, District 1 contained 122,689 individuals, District 2 contained 121,372 individuals, District 3 contained 142,776 individuals, District 4 contained 142,111 individuals, and District 5 contained 145,773 individuals, reflecting variances of -9.1%, -10.1%, 5.8%, 5.3%, and 8% respectively from the population target. (Doc. 169-12, pp. 5–6).

131. To achieve equal population, the Commission needed to add 12,255 people to District 1 and 13,572 people to District 2. (Doc. 169-12, p. 7). The Commission needed to remove 7,832 people from District 3; 7,167 people from District 4; and 10,829 people from District 5. (Doc. 169-12, p. 7).

132. The commissioners have exercised legislative privilege and have declined to provide testimony on the redistricting process. (Doc. 169-8, p. 1). The Commission has relied on Mr. Stephenson to explain the process. Mr. Stephenson's testimony reflects only public conversations with the commissioners. (Doc. 174, pp. 155–56, tpp. 648:22–649:3).

133. The commissioners prepared three maps for consideration. (Doc. 169-12, pp. 10–16). Mr. Stephenson did not participate in the boundary revisions, but he discussed the plans with the commissioners. (Doc. 174, pp. 209–11, tpp. 702:14–704:8). The three plans varied in their placement of the predominantly Black communities of Ensley and Midfield in Districts 1 and 2. (Doc. 174, p. 173, tp. 666). Districts 3, 4, and 5 did not change in the three maps. (Doc. 174, p. 173, tp. 666:9–24).

134. The commissioners began redistricting with the districts that were in place in August 2021 and used the procedure that the Commission had followed in 2013. (Doc. 174, pp. 160–61, 220, tpp. 653:9–654:11, 713:8–23). Those districts existed when the commissioners were elected in 2018. (See Doc. 174, p. 177, tp. 670:7–17).

135. On October 5, 2021, the Commission held a work session. (Doc. 169-10). Mr. Stephenson presented the three map options. (Doc. 174, pp. 162, 168–69, 171, tpp. 655:9–22, 661:22–662:5, 664:6–19). The public was not able to comment at the Commission's October 5 work session. (Doc. 174, pp. 205–06, tpp. 698:23–699:12). At that session, the commissioners voted to advance the three proposed plans. (Doc. 174, pp. 205–06, tpp. 698:23–699:12).

136. At the commission's October 7, 2021 meeting, Commissioners Knight, Scales, and Stephens approved a resolution to conduct a public hearing on the draft maps on November 4, 2021. (Doc. 169-18, p. 17).²⁰

137. Ahead of the public hearing concerning the maps, Mr. Stephenson maintained the maps for public viewing at the Board of Registrars office per Alabama Code § 11-3-1.1(c). (Doc. 174, p. 169, tp. 662:15–23).²¹

138. The Commission gave public notice that the maps were available for viewing only in the Alabama Messenger. (Doc. 174, pp. 206–07, tpp. 699:24–700:5). The Court takes judicial notice that the Alabama Messenger is a subscription-only publication available in print and online.²² The Alabama Messenger website does not indicate when the publication became available online. The Alabama Messenger advertises that an annual subscription costs $20.00.²³ The “About Us” description for the Alabama Messenger states:

Alabama Messenger is a weekly newspaper dedicated to serving the people, courts, attorneys, and businesses of Jefferson County, Alabama, as an efficient and qualified medium for all legal publications. Published weekly since 1918, Alabama Messenger is now published twice a week, on Wednesdays and Saturdays, to better serve our customers and subscribers.24

The newspaper publishes various “legal notices pertaining to Jefferson County and local municipalities located in Jefferson County,” including notices related to probate court, family court, district court, and circuit court proceedings, and voting information.²⁵

139. The Alabama Messenger notice indicated that the public could view the maps at the Board of Registrars office. (Doc. 174, pp. 169–70, tpp. 662:24–663:2). The notice provided the time, date, and location of the November 4 meeting but did not include images of the proposed maps. (See Doc. 174, pp. 206–07, tpp. 699:24–700:18). The public had to visit the Board of Registrars office to view the maps. (Doc. 174, p. 207, tp. 700:16–22).

140. Six or seven members of the public visited the Board of Registrars office to view the proposed maps and asked Mr. Stephenson questions about the differences among the proposed plans. (Doc. 174, p. 170, tp. 663:13–23).

141. The November 4 hearing was advertised on social media in a Twitter post less than 24 hours before the hearing. (Doc. 174, p. 207, tp. 700:8–14).

142. Ms. McClure was not aware of the November meeting or the redistricting process until after the Commission approved the 2021 plan. (Doc. 172, p. 213, tp. 213:4–8). Mr. Hall did not attend the public hearing hosted by the Commission before the 2021 map was enacted because he was not aware of the hearing. (Doc. 174, p. 26, tp. 519:14–19). Though Ms. McClure and Mr. Hall are community leaders who are engaged with public policy and advocacy, neither received notice of the November meeting. (See Doc. 172, pp. 204–08, tpp. 204:7–208:15; Doc. 174, pp. 14–20, tpp. 507:7–513:15).

143. The commissioners considered the draft plans at a November 4 public hearing. (Doc. 179-8). The Commission heard public comment on the proposals. (Doc. 169-11, pp. 19–20).²⁶ No one raised concerns about the proposed plans being racial gerrymanders. (Doc. 174, p. 175, tp. 668:20–25).

144. By a 4 to 1 vote, the Commission adopted the first plan. (Doc. 169-11, p. 36). Commissioners Ammons, Knight, Tyson, and Stephens voted for plan 1. (Doc. 179-8, p. 37, tp. 35:4–21). Commissioner Scales voted against plan 1. (Doc. 179-8, p. 37, tp. 35:4–21).

145. Mr. Stephenson stated that the Commission did not consider voting tabulation districts in drawing the 2021 plan, (Doc. 174, p. 179, tp. 672:21–22), but Mr. Fairfax testified that Jefferson County's 2020 precincts “matched up pretty much exactly” with the 2020 VTDs provided by the Census Bureau, (Doc. 172, p. 17, tp. 17:4–25).

146. Mr. Stephenson testified that the commissioners did not try to avoid splitting municipalities because municipal annexations are common. (Doc. 174, p. 212, tp. 705:12–16).

147. In the 2021 map, the City of Birmingham is divided among the Commission's five districts, (Doc. 179-16, p. 9). The small parts of Birmingham in Districts 3, 4, and 5 are the result of municipal annexations, such as an area near the Summit that Birmingham annexed. (Doc. 174, pp. 188–89, tpp. 681:21–682:6). The area is surrounded by Vestavia Hills, a municipality in Jefferson County. (Doc. 174, pp. 188–89, tpp. 681:21–682:6).

148. The 2021 plan split 25 census places across Jefferson County. (Doc. 169-107, p. 46).

149. The Commission had racial and ethnic data available during the 2021 redistricting process, but the Commission did not have political data such as party affiliation or elections data. (Doc. 174, p. 212, tp. 705:1–8). The Commission's redistricting software shows the racial make-up of districts but not the partisan preferences of voters because Alabama law does not require voters to register by political party. (Doc. 174, pp. 211–12, tpp. 704:18–705:8). The Commission used the same software in the 2013 redistricting cycle. (Doc. 174, p. 217, tp. 710:9–19).

150. The 2021 plan maintains most of the population from each of the districts in the 2013 district map. Districts 1 and 2 retained approximately 90% of their 2013 populations, and Districts 3, 4, and 5 maintained almost all of their populations from the 2013 plan. (Doc. 179-16, p. 7). Dr. Barber concluded that the 2021 plan retains the “population of the 2013 commission districts to an extremely high degree.” (Doc. 179-16, p. 4).

151. The 2021 plan generally maintains the racial composition of the maps the Commission enacted in each redistricting cycle following the 1990 census. (Doc. 169-4, pp. 191–92; Doc. 169-6, pp. 42–45; Doc. 169-49). The breakdown of Black population for the Commission's five districts over these redistricting cycles is as follows:

(Doc. 169-2, pp. 19–23; Doc. 169-3, p. 110; Doc. 169-6, pp. 44–45; Doc. 169-5, p. 1081; Doc. 169-107, p. 18).

152. The 2021 plan's percentage of Any Part Black population of all ages was 78.27% in District 1, 66.18% in District 2, 27.29% in District 3, 28.45% in District 4, and 14.15% in District 5. (Doc. 169-43).

153. The 2021 plan's BVAP percentages were 76.34% in District 1, 64.11% in District 2, 25.80% in District 3, 25.74% in District 4, and 13.99% in District 5. (Doc. 169-31).

154. The district demographics before and after the 2021 redistricting are as follows:

(Doc. 179-16, pp. 10, 22, 31, 34–35, 37).

Evidence of Compliance with Traditional Redistricting Principles

Plan-Wide Evidence

155. The Commission increased the number of municipality splits in the 2021 plan. The 2013 plan split 22 census places across Jefferson County; the 2021 plan split 25 census places. (Doc. 169-107, p. 46; Doc. 175, p. 148, tp. 884:15–23).

156. The municipality splits in the 2021 plan include Brighton, Fultondale, Homewood, Hoover, Midfield, Irondale, Bessemer, Center Point, Trussville, Leeds, and Tarrant, areas where Black populations have migrated from Birmingham. (See Doc. 169-108, pp. 117–29).

157. The maps below show the municipalities in the 2013 plan and in the 2021 plan. (Doc. 169-108, pp. 16, 22).

2013 District Boundaries with Selected Municipalities

2021 District Boundaries with Selected Municipalities

158. Mr. Fairfax's illustrative plan shows that it was possible to draw a map that respected equal population, contiguity, compactness, and minimized political subdivision splits without the demographic effects of the Commission's 2021 plan. (Doc. 169-107, pp. 44–49).

159. Mr. Fairfax compared his illustrative plan to the 2013 plan and the 2021 map and concluded that his illustrative plan performed better on compactness, contiguity, and splitting census places, as the illustrative plan split only four census places, compared to 25 for the 2021 plan and 22 for the pre-2020 plan. (Doc. 169-107, pp. 44–49; Doc. 172, pp. 51–52, tpp. 51:17–52:17). Mr. Fairfax's illustrative plan splits only the municipalities of Birmingham, Hoover, Irondale, and Vestavia Hills, (Doc. 169-108, p. 188), and the plan keeps whole many of the municipalities that have grown in Black population since the time of the consent decree.

160. Mr. Cooper's illustrative plans performed equal to or better than the 2021 plan on traditional redistricting principles, including contiguity, compactness, respect for political subdivisions, and non-dilution of minority strength. (See Doc. 172, pp. 279–80, tpp. 279:15–280:4; Doc. 173, pp. 43, 59–60, tpp. 335:14–17, 351:17–352:3).

161. Mr. Cooper's illustrative plans A, B, C, and E outperformed the 2021 plan in terms of municipal splits and total splits. These illustrative plans had fewer municipal and total splits than the 2021 plan and reduced the concentration of Black voters in Districts 1 and 2. (Doc. 169-41; Doc. 169-81).

162. Illustrative plan D moved the boundaries of District 1 east, reducing the supermajority of Black voters in District 1 while maintaining a high core retention rate. (Doc. 172, pp. 280–81, tpp. 280:22–281:7). Plan D had five more municipal splits than the 2021 plan, but plan D had nearly the same number of total splits. (Doc. 169-81).

163. Mr. Cooper attributed the higher number of splits in plan D to the plan's high core retention rate. (See Doc. 173, p. 60, tp. 352:12–15).

164. Dr. Barber quantified the core retention scores of each district in the 2021 plan. (Doc. 175, p. 18, tp. 754:24–25). In 2021, District 1 pulled 90% of its population from 2013 District 1, 2.6% of its population from 2013 District 3, and 7.4% of its population from 2013 District 4; District 2 pulled 90.1% of its population from 2013 District 2, 5.2% from 2013 District 3, and 4.7% from 2013 District 5; District 3 pulled 98.9% of its population from 2013 District 3 and 1.1% of its population from 2013 District 4; District 4 pulled 96.4% of its population from 2013 District 4 and 3.6% of its population from 2013 District 5; and District 5 pulled 100% of its population from 2013 District 5. (Doc. 179-16, p. 7).

The table below summarizes Dr. Barber's analysis.

Dr. Barber also quantified “the proportion of the population in each 2013 district that went into a 2021 district,” (Doc. 179-16, p. 7), described in the table below:

Under both methods of calculating core retention described above, each 2021 district had a core retention rate of over 90%. (Doc. 179-16, p. 7).

165. Plan D has an overall core retention rate of 85.74%, compared to the 95.3% core retention rate in the 2021 plan. (Doc. 173, p. 25, tp. 317:4–15; Doc. 179-18; see Doc. 169-26, p. 14). Under plan D, Districts 1 and 2 had any part Black BVAPs of 62.77% and 62.16%. (Doc. 172, p. 286, tp. 286:18–22; Doc. 169-77). Districts 3, 4, and 5 had BVAPs of 24.32%, 42.48%, and 14.64%. (Doc. 172, pp. 286–87, tpp. 286:23–287:3; Doc. 169-77).

166. Mr. Cooper drew plan D with a high core retention rate to show that the Commission could have maintained high core retention without concentrating Black voters in Districts 1 and 2. (Doc. 172, p. 280, tp. 280:14–24).

167. Mr. Cooper testified that the major change between plan D and the 2021 plan was that plan D moved District 1 east. (Doc. 172, p. 281, tp. 281:4–12).

168. Although Mr. Cooper explained that plan D places the incumbents from Districts 1 and 4 in one district, he testified that he could correct the mistake with little impact on core retention. (Doc. 172, pp. 283, 285, 286, tpp. 283:2–10, 285:1–9, 286:4–10).

169. Each of Mr. Cooper's illustrative plans divides portions of Birmingham to a greater degree than the 2021 plan. (Doc. 173, p. 23, tp. 315:11–18; Doc. 169-53; Doc. 169-59; Doc. 169-65; Doc. 169-83; Doc. 169-89). The 2021 plan kept 93.4% of Birmingham in Districts 1 and 2, (Doc. 179-16, p. 9); plan D keeps 81.43% of Birmingham in Districts 1 and 2 and moves another 13.35% of Birmingham into District 4. (Doc. 169-85).

170. Plan D exceeds the 1% population deviation present in the 2021 plan. (Doc. 173, pp. 42–43, tpp. 334:24–335:4; Doc. 169-77).

171. Using Dr. Barber's simulated plans, Dr. McCartan determined that the BVAP in District 1 is higher than 95.1% of the corresponding districts in the simulations. (Doc. 169-26, p. 9; Doc. 175, p. 127, tp. 863:13–16). The BVAP in District 2 is higher than 96.6% of the corresponding districts in the simulations. (Doc. 169-26, p. 9). The BVAP in District 3 was lower than 99.7% of the corresponding districts in the simulations. (Doc. 169-26, p. 9). Dr. Barber ignored statistical significance and did not report that Districts 1, 2, and 3 are statistically significant outliers in their BVAP shares using a standard 0.05 statistical significance level. (Doc. 169-26, p. 9).

172. Dr. McCartan created a summary statistic he called the combined packing-cracking score for his district-by-district and precinct-by-precinct analyses. (Doc. 174, pp. 74, 104–05, tpp. 567:10–23, 597:24–598:21). The “plan-wide” statistic measures a pattern of “packing [B]lack voters into two districts and cracking them in a third.” (Doc. 174, p. 77, tp. 570:8–11). Dr. McCartan undertook this analysis after preliminarily finding that those districts were statistical outliers. (Doc. 169-26, pp. 9–10). Dr. McCartan calculated the score by averaging the BVAP shares of the two districts with the highest BVAPs and then subtracting the BVAP share in the next-highest BVAP district. (Doc. 169-26, p. 10).²⁷

173. Using this summary statistic, Dr. McCartan found that the 2021 plan had a higher packing-cracking score than 99.8% of Dr. Barber's simulated plans. (Doc. 169-26, p. 10; Doc. 174, pp. 144–45, tpp. 637:24–638:3). In testimony about partisan gerrymandering in another case, Dr. Barber stated that he considered a plan that was more extreme than 99.98% of his simulations to be an outlier that represented “a significant deviation from a fair outcome.” (Doc. 175, pp. 90–91, tpp. 826:5–827:20).

174. Dr. McCartan testified that his plan-wide packing and cracking analysis confirmed the pattern he saw in his district and precinct level analyses. (Doc. 174, p. 77, tp. 570:6–12). He concluded that the 2021 plan concentrates Black voters in two districts and removes them from a third. (Doc. 174, p. 77, tp. 570:6–12).

175. Dr. McCartan created three sets of simulations to explore Dr. Barber's opinion that core retention offered a race-neutral explanation for the district lines in the 2021 plan. (Doc. 174, pp. 62-63, tpp. 555:5–556:4). The third set, the “core retention, strong” simulation, generated 120,000 plans with core retention scores between 89.1% and 96.5%. (Doc. 174, pp. 62–63, tpp. 555:5–556:4; Doc. 169-26, pp. 14, 18).

176. Dr. McCartan compared the racial composition of the 2021 plan to the simulated plans. (Doc. 169-26, p. 18). Dr. McCartan repeated the analysis twice: once with the full set of simulations and once using simulated plans with two or more majority-Black districts. (Doc. 169-26, pp. 18–19).

177. Dr. McCartan concluded that core retention could not explain those patterns. (Doc. 174, p. 77, tp. 570:13–24).

178. Dr. McCartan's “strong core retention” simulation set performed as well as or better than the 2021 plan in avoiding municipal splits. (Doc. 169-26, p. 14).

179. Dr. McCartan's strong core retention plan demonstrates that when compared against plans drawn with the strictest core retention specifications, the 2021 plan is an outlier. The strong core retention plan cannot generate the 2021 plan because the 2021 plan split the Commission's 2013 districts too many times to create Districts 1 and 2. (See Doc. 169-26, p. 18). The 2021 plan's combined packing-cracking score was more extreme than 98.49% of randomly generated plans under the strong core retention specification. (Doc. 169-26, p. 19; Doc. 174, p. 84, tp. 577:4–22).

180. Dr. McCartan analyzed his strong core retention simulations at the precinct level. (Doc. 169-26, pp. 15–17). Each simulation assigned each precinct to a hypothetical district with a corresponding BVAP. (Doc. 169-26, p. 16). Dr. McCartan averaged the BVAPs for each hypothetical district in his simulations. (Doc. 169-26, p. 16). He generated a map which shaded each precinct to reflect the average BVAP of the district to which the precinct was assigned, as described in the figure below. (Doc. 169-26, p. 16).²⁸

(Doc. 169-26, p. 16). Dr. McCartan's map demonstrates “that if core retention is prioritized, there will typically be a large gap between the racial composition of the districts covering most of Birmingham and the other districts.” (Doc. 169-26, p. 16). Precincts in 2021 District 1 were assigned to a simulated district with, on average, a roughly 70% BVAP. (Doc. 169-26, p. 16). Non-2021 District 2 precincts surrounding 2021 District 1 were assigned to a simulated district with no more than a roughly 45% BVAP. (Doc. 169-26, p. 16). Simulated precincts in 2021 District 2 were assigned to a simulated district with, on average, a roughly 60% BVAP. (Doc. 169-26, p. 16). Non-2021 District 1 precincts surrounding 2021 District 2 were assigned to a simulated district with no more than a roughly 40% BVAP. (Doc. 169-26, p. 16).

181. Dr. McCartan also visualized the difference in BVAP between a precinct in the 2021 plan and the average BVAP of the districts to which the precinct was assigned in his simulations. (Doc. 169-26, p. 16). In the map below, Dr. McCartan has added hatching where this difference represents a statistically significant result. (Doc. 169-26, p. 16).

(Doc. 169-26, p. 16). Dr. McCartan's map demonstrates that the Commission's choices in assigning new areas to District 1 in 2021 resulted in the placement of those areas into a district with a BVAP nearly 40% higher than the average BVAP of the districts those areas were assigned in Dr. McCartan's simulations. (Doc. 169-26, pp. 16–17). Many of the Commission's choices in assigning new areas to District 2 in 2021 resulted in the placement of those areas into a district with a BVAP more than 20% higher than the average BVAP of the districts those areas were assigned in Dr. McCartan's simulations. (Doc. 169-26, pp. 16–17).

182. Dr. McCartan testified that even including maximal core retention, compactness, municipality splits, and other constraints, the 2021 plan was an outlier compared to his simulations. (Doc. 174, pp. 84–85, tpp. 577:13–578:3). Only 9.28% of Dr. McCartan's simulated district 1s have a higher BVAP than District 1 does in the 2021 plan. (Doc. 169-26, p. 19).²⁹

District 1 Evidence

183. In the 2021 plan, the Commission added two precincts from District 4 to the northeast side of District 1 and two precincts from District 3 to the west side of District 1. (See Doc. 174, p. 179, tp. 672:6–13). The Commission removed from District 1 an area with approximately 100 people. (Doc. 179-16, p. 12).³⁰ The map below depicts the changes to District 1.

(Doc. 179-16, p. 11).

184. The added precincts each contain Birmingham neighborhoods. (Doc. 175, p. 31, tp. 767:4–8; Doc. 172, p. 115, tp. 115:11–21).

185. The 2021 plan moved approximately 13,000 people into District 1. (Doc. 169-107, p. 22). Of those individuals, 77.9% were Black and 15.8% were white. (See Doc. 172, p. 32, tp. 32:13–21).

186. The Commission moved 3,625 individuals, 64% of whom were Black, from District 3 into District 1. (Doc. 169-72, p. 1; Doc. 172, p. 273, tp. 273:17–19). The Commission moved 9,422 individuals from District 4 to District 1, 82.6% of whom were Black. (Doc. 169-72, p. 1; Doc. 172, p. 273, tp. 273:20–21).

187. The Commission split several 2013 precincts to shift populations that were predominantly Black into District 1. Under the Commission's 2013 plan, the East Pinson Valley Precinct was in District 4. (Doc. 169-107, p. 25). The 2021 plan split East Pinson Valley between Districts 1 and 4. (Doc. 169-107, pp. 25–26). Of the East Pinson Valley voters the Commission moved into District 1, 86.18% were Black. (Doc. 172, p. 36, tp. 36:1–25; Doc. 169-107, p. 26). Only 9.79% of voters in the new District 1 portion of East Pinson Valley were white. (Doc. 172, pp. 36–37, tpp. 36:19-37:1; Doc. 169-107, p. 26). The East Pinson Valley voters who remained in District 4 were 56.91% Black and 23.70% white. (Doc. 169-107, p. 26). Had the Commission added the entire East Pinson Valley Center precinct to District 1, the Commission would have lowered the Black population percentage in District 1. (Doc. 169-107, pp. 26–27). The figures below describe this change.

(Doc. 169-107, p. 26).

(Doc. 169-107, p. 37).

188. In addition to splitting a preexisting VTD by breaking off a part of the East Pinson Valley precinct, the Commission split municipal boundaries and cut Center Point in half. (Doc. 169-107, p. 27).

189. The Commission could have split and added census blocks from the Center Point First Baptist Church precinct to District 1, which would have lowered District 1's Black population percentage. (Doc. 169-109, p. 12).

190. The Commission split the Dolomite West Field Community Center precinct, shifting voters from District 3 to District 1. (Doc. 169-109, p. 9). Of the 1,284 voters that the Commission added to District 1 from the old Dolomite West Field Community Center precinct, 86.60% were Black and 8.18% were white. (Doc. 169-109, p. 9).

191. Every whole precinct or VTD the Commission moved into District 1 was majority-Black. (Doc. 172, pp. 38–39, tpp. 38:24–39:1; Doc. 169-109, p. 9).

192. The Commission moved the Minor Fire Station precinct, which was 52.21% Black, from District 3 to District 1. (Doc. 169-109, p. 9).

193. The Commission moved the entire Center Point Community Center precinct in north Jefferson County from District 4 to District 1. (Doc. 169-109, p. 9). In the process, the Commission effectively split the Center Point municipality in half. (See Doc. 169-108, p. 23). Of the 6,202 voters in that precinct, 80.86% were Black and 12.61% were white. (Doc. 169-109, p. 9).

194. The Commission moved the Brookside Community Center precinct out of the already underpopulated District 1 and into District 3. (Doc. 179-16, p. 13). The Brookside Community Center precinct is 24.2% Black. (Doc. 179-16, pp. 11, 13).

195. Dr. Barber testified that, per his regression analysis, race was not a statistically significant predictor of which precincts the Commission added to District 1 and that geographic proximity was the only significant predictor. (Doc. 179-16, pp. 20–21; Doc. 175, p. 783, tp. 783:19–23). Dr. Liu explained that Dr. Barber's regression analysis did not measure the interactive effect of geographic proximity and BVAP percentage at the precinct level and thus could not measure whether the geographically adjacent precincts that were heavily Black were more likely to be moved to District 1 than adjacent precincts that were more white. (Doc. 169-22, p. 3). Dr. Liu performed a multivariate regression with an interaction term to test whether race was a significant factor in the Commission's decision to add an adjacent precinct to District 1. (Doc. 169-22, pp. 3–4). Dr. Liu found that race was a statistically significant predictor of adjacent precinct movement into District 1 at the 1% statistical significance level. (Doc. 169-22, p. 4; Doc. 173, p. 106, tp. 398:16–19).

196. Dr. McCartan's analysis indicates that every precinct assigned to District 1 in the 2021 plan has a statistically significantly higher BVAP as compared to its average district assignment in Dr. Barber's set of simulated maps. (See Doc. 169-26, p. 11). That includes the East Pinson Valley Center, Center Point Community Center, Dolomite West Field Community Center, and Minor Fire Station precincts. (See Doc. 169-26, p. 11).

197. Dr. Barber theorized that the Commission may have made changes to the northeast area of District 1 to increase the overlap between District 1 and Commissioner Scales's former City Council District. (Doc. 175, p. 34, tp. 770:1–23). Dr. Barber did not speak to the commissioners, has no knowledge of their motives, and does not know what criteria they used. (Doc. 175, p. 56, 116, 138, tp. 792:7–10, 874:11–22, 8:52:22–25). Commissioner Scales voted against the 2021 plan. (Doc. 179-8, p. 37, tp. 35:15–16).

198. In the 2021 plan, the Commission added to District 1 areas from the only two adjacent non-District 2 VTDs that were more than 78% Black. (See Doc. 169-107, pp. 34–35). Adding other adjacent VTDs would have lowered the percentage of District 1's Black population. (See Doc. 169-107, pp. 34–35).

District 2 Evidence

199. Though the Commission's changes to District 2 in the 2021 plan reduced District 2's Black population by 2.83%, District 2 is 66.18% Black with a 64.11% BVAP in the 2021 plan. (Doc. 169-107, pp. 39–40; Doc. 169-31).

200. The Commission moved 6,593 people from District 5, the district with the lowest percentage of Black population, into District 2, 20.8% of whom were Black. (Doc. 172, p. 274, tp. 274:11–14; Doc. 169-72, p.1). The Commission shifted 7,008 individuals from District 3 into District 2, 59.8% of whom were Black. (Doc. 169-72, p.1; Doc. 172, p. 274, tp. 274:9–10).

201. The Commission split the Bessemer Civic Center precinct. (Doc. 169-107, p. 29). Before 2021, the Bessemer Civic Center precinct was in District 3. (Doc. 169-107, pp. 29). In the 2021 plan, the Commission moved 2,559 individuals from the old Bessemer Civic Center precinct to District 2. (Doc. 169-107, pp. 290). Of those moved, 80.50% were Black; only 14.34% were white. (Doc. 169-107, p. 29).

202. The Bessemer Civic Center precinct split does not follow municipal boundaries. (Doc. 169-107, p. 30; Doc. 169-108, p. 24). In the 2021 plan, District 2 cuts so deeply into Bessemer that the parts that remain in District 3 no longer are contiguous, as illustrated in the figure below.

(Doc. 169-108, p. 24).

203. The Commission split the Ross Bridge Welcome Center precinct. (Doc. 169-107, p. 30). Under the 2013 plan, Ross Bridge was in District 3. (Doc. 169-107, p. 30). The Ross Bridge precinct is 23.38% Black. (Doc. 169-107, p. 30). Of the 2,068 Ross Bridge voters the Commission moved into District 2, 52.80% were Black. (Doc. 169-107, p. 30). Per the figure below, the portion of the Ross Bridge precinct that remained in District 3 is 13.21% Black. (Doc. 169-107, p. 31).

(Doc. 169-107, p. 31).

204. The Commission moved three entire precincts into District 2. The Afton Lee Community Center precinct, also known as the Rosedale neighborhood, is majority Black. (Doc. 169-107, p. 39). The Homewood and Grant Street Baptist Church precincts are predominantly white. (Doc. 169-107, p. 39).

205. The addition of these precincts to District 2 decreased District 2's BVAP in 2021, (Doc. 169-107, pp. 39–40), but community members testified that these areas are experiencing growth in their Black populations. Mr. Douglas testified that the Homewood area, encompassing both the Homewood and Afton Lee Community Center Precincts, has a growing Black community. (Doc. 173, p. 190, tp. 482:2–3). Mr. Cooper confirmed this demographic change. (Doc. 169-115) (depicting and comparing 1990 to 2020 census data with the 2021 plan boundaries).

206. Dr. McCartan's hatched map shows that, even when prioritizing core retention, the Commission assigned the precincts surrounding Bessemer, Ross Bridge, the Rosedale neighborhood, and the Oxmoor neighborhood to a district with a higher BVAP than simulations with comparable core retention metrics. (Doc. 169-26, p. 16). The Commission assigned these areas to District 2, which has a BVAP higher than 95% of the districts to which those precincts were assigned in Dr. Barber's complete set of simulations. (See Doc. 169-26, pp. 9, 11).

207. Dr. Liu did not perform a regression analysis for District 2. (Doc. 173, p. 134, tp. 426:9–10). Dr. Barber ran the regression analysis with Dr. Liu's interactive variable for District 2 and found that there was not a statistically significant relationship between BVAP and adjacency in District 2. (Doc. 175, p. 47, tp. 783:4–23).

District 3 Evidence

208. The Commission moved 10,550 persons from District 3 into Districts 1 and 2. (Doc. 169-107, p. 41). Black individuals accounted for 61.45% of the individuals who moved out of District 3. (Doc. 169-107, p. 41).

209. As discussed, well over half of the 3,625 individuals the Commission moved into District 1 are Black. (Doc. 169-72, p. 1; Doc. 172, p. 273, tp. 273:17–19). The Commission moved the remaining 7,008 persons, 59.8% of whom were Black, into District 2. (Doc. 169-72, p. 1; Doc. 172, p. 274, tp. 274:9–10).

210. The Commission moved 1,445 people from District 4 into District 3, 3.5% of whom were Black. (Doc. 172, p. 276, tp. 276:2–3; Doc. 169-72, p. 2).

211. The portion of Bessemer Civic Center that the Commission left in District 3 is 10 percentage points whiter than the portion the Commission moved into District 2. (Doc. 169-107, p. 29). By splitting the Ross Bridge precinct, the Commission removed a community that was 52.80% Black from District 3 and kept the portion of Ross Bridge that was 71.69% white in the district. (Doc. 169-107, p. 31). By splitting the Dolomite West Field Community Center precinct and moving 1,284 persons, 86.60% of whom were Black, into District 1, the Commission retained in District 3 less heavily Black neighborhoods. (See Doc. 169-109, p. 9). These moves increased District 3's white population percentage in the 2021 plan. (Doc. 169-109, pp. 17–18).

212. The Commission split the 2013 Warrior City Hall precinct and moved 1,445 persons, 90.52% of whom were white, from District 4 to District 3. (Doc. 169-107, p. 41). By splitting the Dolomite West Field Community Center, the Commission moved 161 more individuals into District 3 than the Commission moved out of District 3. (Compare Doc. 169-107, p. 41, with Doc. 169-108, p. 237).

213. Using Dr. Barber's simulations and a standard statistical significance threshold, District 3 is a statistically significant outlier in terms of its BVAP. (Doc. 169-26, p. 9). District 3 has a BVAP that is lower than 99.72% of all comparable districts in Dr. Barber's simulations and 99.51% of comparable districts in simulated plans with two or more majority Black districts. (Doc. 169-26, p. 19). District 3 was a statistically significant outlier for its exclusion of Black communities under every set of simulation constraints that Dr. McCartan tested. (Doc. 169-26, pp. 19–20). Dr. McCartan illustrated how extreme District 3 is in relation to the distribution of randomly generated plans Dr. Barber produced.

(Doc. 169-119).

214. Dr. McCartan's hatch map analysis corroborates this data. The map shows that the Commission placed most of the precincts in the central and southwestern portions of District 3 in a district that has a lower BVAP than at least 95% of the simulated district's BVAPs. (Doc. 169-26, pp. 9, 11). The precincts that the Commission moved from District 3 into Districts 1 and 2—the Minor Fire Station, Dolomite West Field Community Center, Bessemer area, and Ross Bridge precincts—are in a district with a BVAP that is higher than 95% of the corresponding district assignments in Dr. Barber's simulation. (See Doc. 169-26, pp. 9, 11).

District 4 Evidence

215. The 2021 plan moved 10,752 people from District 4 into Districts 1 and 3. (Doc. 169-107, p. 41). Of the people moved, 72.01% were Black. (Doc. 169-107, p. 41). At the beginning of the 2020 redistricting process, District 4 had a BVAP of 29.5%. (Doc. 179-16, p. 34). Under the 2021 plan, District 4's BVAP fell to 25.7%. (Doc. 179-16, pp. 34–35).

216. The voters moved into District 4 were predominantly white; only 8.6% of the individuals moved into District 4 were Black. (Doc. 172, p. 276, tp. 276:20–21; Doc. 169-72, p. 2).

217. District 5 contributed 4,787 people to District 4, 8.6% of whom were Black. (Doc. 172, p. 276, tp. 276:17–18; Doc. 169-72, p.2).

218. As discussed, the Commission split the East Pinson Valley Center precinct in the 2021 plan. (Doc. 169-107, p. 27). Of the individuals the Commission moved from District 4 to District 1, 86.18% were Black. (Doc. 169-107, pp. 26–27). Of the individuals in the precinct that remained in District 4, 56.91% were Black. (Doc. 169-107, pp. 26–27). The Commission moved the entire Center Point Community Center precinct into District 1, shifting 6,202 people, 80.86% of whom were Black, from District 4 to District 1. (Doc. 169-107, p. 39). In doing so, the Commission eliminated from District 4 the two precincts with the highest Black populations and made District 4 more white. (See Doc. 179-16, p. 36).

219. The Commission removed from District 4 an overwhelmingly white precinct, part of the 2013 Warrior City Hall precinct, and placed the individuals into District 3. (Doc. 169-107, p. 41).

220. The Commission moved several precincts from Districts 4 and 5 into districts in the 2021 plan that are more Black than 95% of the districts they were assigned to in Dr. Barber's simulations. (Doc. 169-26, pp. 9, 11). Those precincts are the East Pinson Valley, Center Point Community Center, and Afton Lee Community Center (Rosewood) precincts discussed above. (See Doc. 169-26, p. 11). Dr. McCartan's map suggests that the portion of the East Pinson Valley precinct that remains in District 4 in the 2021 plan has a lower BVAP than 95% of the districts that precinct was assigned to in Dr. Barber's simulations. (Doc. 169-26, pp. 9, 11). The Commission's decision to split the East Pinson Valley Precinct and move voters from District 4 into District 1 created statistically significant outlier precincts in both districts to which they were assigned compared to Dr. Barber's simulations.

District 5 Evidence

221. When the 2021 redistricting process began, District 5 had a BVAP of 14.1%. (Doc. 179-16, p. 37). After redistricting, District 5 had a BVAP of 14.0%. (Doc. 179-16, p. 37). The Commission's movement of voters made District 5 marginally more white.

222. In the 2021 plan, the Commission moved 7,555 people from District 5 to Districts 2 and 4. (Doc. 169-107, p. 41). Of those moved, 16.10% were Black. (Doc. 169-107, p. 41). The movement of the Afton Lee Community Center and the Oxmoor Valley Community Center accounts for most of the racial changes to District 5 in the 2021 plan. The Afton Lee Community Center precinct has a BVAP of 59.2%, and the Oxmoor Valley Community Center precinct has a BVAP of 27.5%. (Doc. 179-16, p. 38). Keeping either precinct would have given District 5 a higher BVAP.

223. Dr. Barber suggested that because the Oxmoor Valley Community Center was previously split between Districts 2 and 5 and was then “reunited” in District 2 under the 2021 plan, the move was not indicative of racial predominance. (See Doc. 175, pp. 48–49, tpp. 784:17–785:5).

224. Because it moved the McElwain Baptist Church precinct from District 5 to District 4, the Commission could move Black communities from District 4 into District 1, like the trade it made between Districts 1, 3, and 4 using the Warrior City Hall precinct. The McElwain Baptist Church precinct included 4,787 people, 80.74% of whom were white. (Doc. 169-108, p. 244). The Commission moved that group of nearly five thousand people from overpopulated District 5 into overpopulated District 4. (Doc. 179-16, pp. 34–35, 37; Doc. 169-107, p. 41). Meanwhile, the Commission split the East Pinson Valley Center precinct and moved 3,105 voters from District 4 into District 1. (Doc. 169-107, p. 41). Had the Commission kept the East Pinson Valley Center precinct in District 4, and avoided the split precinct, the Commission would not have had to move the McElwain Baptist Church precinct into District 4.

Contemporaneous Statements Made by Commissioners

225. During the Commission's November 4, 2021 public meeting, Commissioner Tyson stated that she drew Plan 1, the plan that the Commission enacted. (Doc. 169-11, pp. 36, 40, tp. 35:4–21, 39:12–15).

226. Immediately before the commissioners voted and selected Plan 1, Commissioner Scales remarked: “We speak of Democratic versus Republican. You figure out what that looks like.” (Doc. 169-11, p. 32, tp. 31:21–22). Commissioner Scales then described several communities by purported political affiliation. She stated: “I am having to take in Center Point which is highly Democratic, I am taking in Dolomite which is highly Democratic, but you ask yourself concerning Homewood, Ross Bridge, Lake Shore, is that a heavily populated Democratic area?” (Doc. 169-11, p. 34, tp. 33:1–5).

227. After the Commission voted to enact the 2021 map, Commissioner Tyson used similar language but explicitly linked political affiliation with race. She remarked that District 2 was underpopulated and stated:

That mean[s] you hold what you got and you pull from the people that's overpopulated. I talked to Commissioner Ammons. I pulled - Rosedale is a 99.2 percent Black community. 99.2 percent, and they all Democrats.

Now, if you think I will draw myself into my demise you got to be crazy. I pulled from Homewood. I already have that portion of Homewood. All I did was pull the other Democratic part of Homewood. Just don't believe everything you hear.

I pulled in the rest of the senior citizen box that I already have. I already have it. All I did was got the other part of that box, which is 89 percent Democratic box.

The other closest addition to me was Commissioner Jimm[ie] Stephens. Yes, Ross Bridge is a part of that, but it's across the street, which is 99 percent Republican. I got Mountain View part. I already have Oxmoor part. Oxmoor is already in my district.

All I got was the Mountain View part which is hooked to Oxmoor. It's a new subdivision which is 89 percent Democratic and [B]lack.

You know how I know? Because I got up and went over there and limped on my leg and knocked on the doors and seen for myself. I did that myself. I went to Bessemer in the President Jimm[ie] Stephens’ district. I got the part that was behind the civic center and the part that's over there by the police department. 99 percent Democratic, 99 percent Black.

Don't believe everything you hear. You know how I know? I got up, walked over there myself, and I looked at the folks in they face. And I know what I'm getting․

(Doc. 169-11, pp. 40–41, tp. 39:16–40:23).

228. Mr. Douglas attended the November 4, 2021 meeting and recalled Commissioner Tyson discussing looking people in the face. (Doc. 173, p. 195, tp. 487:5–7). It was clear to Mr. Douglas that Commissioner Tyson was referring to race. (Doc. 173, p. 195, tp. 487:8–10).

229. Mr. Douglas also recalled conversations at the hearing about the Rosedale neighborhood. (Doc. 173, p. 194, tp. 486:5–23). He explained that these discussions caught his attention because he knows the neighborhood well through his work with GBM. (Doc. 173, p. 194, tp. 486:14–23). He explained that, in observing the map at that meeting, he noticed that Rosedale was captured by “a bump on the lines that stood out.” (Doc. 173, p. 194, tp. 486:19–23). Rosedale is a neighborhood that Commissioner Tyson discussed at the meeting, referring to it as 99.2% Black community that she “pulled” into her district. (Doc. 179-8, p. 41, tp. 39:18–20). This exemplifies how the Commission made granular race-based decisions when deciding what areas to include or exclude in the five districts.

230. Commissioner Scales spoke against Plan 1 at the November 4, 2021 meeting. (Doc. 169-11, pp. 31–40, tp. 30:7–39:2). Commissioner Scales commented that the urban areas in Jefferson County were not underpopulated; they were undercounted. (See Doc. 169-11, p. 33, tp. 32:6–12). She remarked: “because we're not counted, that's the reason why there's a surplus in other districts and there's a minus in others.” (Doc. 169-11, p. 38). Commissioner Scales added: “those areas that have been disproportioned [ ] we gotta do a better job and mak[e] sure that our people are counted.” (Doc. 169-11, pp. 39–40, tp. 38:24–39:1).

231. The other commissioners did not publicly comment on the redistricting proposals before the Commission voted.

232. After the vote, Commissioner Knight thanked the Board of Registrars for its work in the redistricting process, noting the importance of keeping accurate data to meet the one person-one vote standard. (Doc. 169-11, pp. 37–38, tp. 36:16–37:9). He added that the commissioners did not “want to really give up people,” but they had to “to make it balanced and even.” (Doc. 169-11, p. 46, tp. 45:8–10).

233. Commissioner Knight described the redistricting process, noting: “So we – Steve and Ms. Tyson, Jimm[ie], and Ms. Tyson, we go up together. I went up with Ms. Scales. And we look at the maps and say, hey, do you want – here – here, we can give up here. We can – so we generally work together to do this redistricting.” (Doc. 179-8, pp. 46–47, tpp. 44:23–45:3).

234. Commissioner Stephens said change is difficult and that he had “to go now tell some of my good constituents that—that I will no longer represent them.” (Doc. 179-8, pp. 37–38, tpp. 35:21–36:13). He explained that “[u]nderstanding that this is the process we have to go through every 10 years,” but still “no one really likes to change.” (Doc. 179-8, pp. 37, tpp. 35:21–24).

235. Commissioner Scales emphasized her belief that the census did not fully account for urban population. (Doc. 169-11, pp. 38–40, tp. 37:12–39:2). Commissioner Scales said she wanted “to thank the Board of Registrars because [she was] not picking up new areas or territory other than going over there to Dolomite. So it's going to be good for Scales.” (Doc. 179-8, p. 40, tp. 38:6–12).

236. After the November meeting, Commissioner Tyson told Mr. Simelton that she was forced to accept the Commission lines and that the 2021 plan was forced on her by the other commissioners. (Doc. 173, p. 153, tp. 445:4–17). Commissioner Tyson conveyed to Mr. Simelton that the other commissioners would not entertain plans that differed substantially from what they had come up with in their redistricting plan and that she had little opportunity for input. (Doc. 173, pp. 153–55, tpp. 445:4–17, 446:20–447:2). Commissioner Tyson shared with Mr. Simelton her belief that her district was populated with more African Americans than she needed to be elected. (Doc. 173, p. 155, tp. 447:3–8).

237. Mr. Hall recounted a similar meeting with Commissioner Tyson following the November meeting. (Doc. 174, pp. 26–29, tpp. 519:20–522:5). Commissioner Tyson instructed Mr. Hall to contact Judge U.W. Clemon, an attorney for the plaintiffs, to help ensure that “we had a fair process in Jefferson County.” (Doc. 174, pp. 28–29, tpp. 521:22–522:5).

Voting Patterns in Jefferson County

238. In analyzing the extent of racially polarized voting in Jefferson County, Dr. Liu used the ecological inference method. (Doc. 173, p. 82, tp. 374:22–25).

239. Dr. Liu analyzed voting patterns in fourteen elections in which Jefferson County voters had a choice between a white candidate and a Black candidate. (Doc. 173, p. 82, tp. 374:15–19). Those races included elections for Alabama Secretary of State in 2014 and 2022, Alabama Lieutenant Governor in 2014 and 2018, Alabama State Auditor in 2014 and 2018, Alabama Public Service Commission (Place 1) in 2018, Alabama Attorney General in 2022, Alabama Supreme Court Associate Justice (Place 5) in 2022, President in 2008 and 2012, and Jefferson County Sheriff in 2022. (Doc. 169-21, pp. 4–5).

240. Dr. Liu concluded from his RPV analysis that, in each of the elections he analyzed, Black voters provided overwhelming support—over 90%—to Black candidates. (Doc. 169-21, pp. 4–5). Black support for Black candidates was never lower than 92% and was above 95% in all but three elections. (Doc. 169-21, p. 5).

241. White voters did not share Black voters’ preferences. In the elections Dr. Liu analyzed, white support for the Black preferred candidate peaked at 22.4% and was as low as 9.3%. (Doc. 169-21, p. 5). Dr. Liu concluded that Jefferson County elections exhibit consistently extreme levels of RPV because, in each election he reviewed, the vast majority of Black voters (over 90% in every election) and only a small minority of white voters (between 9.3% and 22.4%) cast their votes for the Black candidate. (Doc. 173, pp. 83–84, tpp. 375:20–376:19; Doc. 169-21, pp. 3–5).

242. Dr. Liu considered six recent elections in Jefferson County that included Black and white candidates: Jefferson County Sheriff in 2022, Alabama Governor in 2022, U.S. Senate in 2022, Alabama Secretary of State in 2022, Alabama Attorney General in 2022, and Alabama Supreme Court Associate Justice (Place 5) in 2022. (Doc. 169-21, p. 6). Dr. Liu testified that because these six elections were recent and involved all voters in Jefferson County, these elections were the most probative of likely voting behavior in upcoming Jefferson County Commission elections. (Doc. 173, pp. 93–94, tpp. 385:14–386:9).

243. Dr. Liu analyzed the number of Black voters in each district in the 2021 plan whose votes were unnecessary surplus for the Black-preferred candidate to win an election. (Doc. 169-21, pp. 8–9). He found that there are many more Black voters in Districts 1 and 2 than needed to permit Black voters to elect a candidate of their choice. (Doc. 169-21, pp. 8–9).

244. Dr. Liu testified that Mr. Cooper's illustrative plan B would permit Black voters to elect candidates of their choice in at least two districts while reducing the packing of Black voters in Districts 1 and 2. (Doc. 173, pp. 96–99, tpp. 388:21–391:15).³¹

245. Under the 2021 plan, the BVAP of District 1 is 76.3% and the BVAP of District 2 is 64.1%. (Doc. 179-16, pp. 10, 22). Black-preferred candidates won 87% of the total votes cast in Districts 1 and 2—far more than the voting age population in these districts needs to elect a Black candidate of choice. (Doc. 169-21, pp. 6–7).

246. Mr. Cooper's Plan B reduced the BVAP to 55% and 56% in Districts 1 and 2 respectively. (Doc. 169-37). In Dr. Liu's analysis, the Black-preferred candidates won overwhelming victories with 71% of the total votes cast in District 1 and 74% in District 2. (Doc. 169-21, p. 7).

247. District 1 and District 2 in the 2021 plan unnecessarily pack Black voters as demonstrated by the fact that Black-preferred candidates won almost 90% of the total votes in those districts. (Doc. 173, pp. 98–100, tpp. 390:10–392:15). Though Mr. Cooper's plan B unpacks the heavy concentration of Black voters in Districts 1 and 2, Black-preferred candidates still would win in those two districts. (Doc. 173, pp. 98–100, tpp. 390:10–392:15). With respect to Districts 3 and 4, on average Mr. Cooper's illustrative plans A, B, and C would not result in a change to the partisan composition of the 2021 plan, as described in the table below. (Doc. 169-21, p. 7).

(Doc. 169-21, p. 7).

248. Dr. Liu also compared the “surplus votes” of Black votes in each district under the 2021 plan and Mr. Cooper's plan B. (Doc. 169-21, pp. 8–9). Dr. Liu's analysis shows that under the 2021 plan, 65% of votes for Black-preferred candidates came from Black voters alone—fifteen percentage points greater than needed to elect the Black-preferred candidate. (Doc. 169-21, pp. 8–9). This number does not account for ballots white voters and others cast for Black candidates. (Doc. 169-21, pp. 8–9).

249. The over-concentration of Black voters in Districts 1 and 2 is confirmed by the total vote Black-preferred candidates received in the 2022 Jefferson County elections. Despite the relatively high level of RPV in Jefferson County, Black-preferred candidates won the majority of the county-wide vote in 12 of the 14 elections Dr. Liu analyzed. (Doc. 173, pp. 90–91, tpp. 382:18–383:19).

CONCLUSIONS OF LAW

With one exception, the Commission's challenge to the plaintiffs’ standing does not have merit. In its memorandum opinion and order resolving the parties’ cross motions for summary judgment, the Court held that the plaintiffs have standing to challenge each of the Commission districts as racially gerrymandered under the Equal Protection Clause, with one exception; GBM does not have standing to challenge District 5. (Doc. 164, pp. 35–39). The Court adopts its standing analysis in Doc. 164 here. The factual findings in ¶¶ 1-11 above align with the standing evidence on which the Court relied in the summary judgment opinion in this case. Because plaintiffs have standing to challenge each of the Commission's districts, the Court turns to the merits of the plaintiffs’ challenges.

***

The plaintiffs’ gerrymandering claim arises under the Fourteenth Amendment. The Equal Protection Clause prevents a state and political subdivisions of a state from denying “to any person within its jurisdiction the equal protection of the laws.” U.S. Const., Amend. XIV, § 1; Avery v. Midland Cnty., 390 U.S. 474, 480, 88 S.Ct. 1114, 20 L.Ed.2d 45 (1968). “The central mandate of the Equal Protection Clause of the United States Constitution is ‘racial neutrality in governmental decision making.’ ” Clark v. Putnam Cnty., 293 F.3d 1261, 1266 (11th Cir. 2002) (quoting Miller v. Johnson, 515 U.S. 900, 904, 115 S.Ct. 2475, 132 L.Ed.2d 762 (1995)).³² Consequently, neither a state nor its subdivisions may separate citizens “into different [voting] districts on the basis of race.” Miller, 515 U.S. at 911, 115 S.Ct. 2475. When race is the predominant criterion for line-drawing in a redistricting cycle, a legislative body engages in racial gerrymandering. See Alexander v. S.C. State Conf. of the NAACP, 602 U.S. 1, 7, 144 S.Ct. 1221, 218 L.Ed.2d 512 (2024) (explaining that racial gerrymandering occurs when lawmakers improperly use race in drawing the boundaries of electoral districts by placing voters “ ‘within or without a particular district’ ” because of the voters’ race) (quoting Miller, 515 U.S. at 916, 115 S.Ct. 2475).³³

“The Equal Protection Clause of the Fourteenth Amendment limits racial gerrymandering in legislative districting plans,” Cooper v. Harris, 581 U.S. 285, 291, 137 S.Ct. 1455, 197 L.Ed.2d 837 (2017), because “a racially gerrymandered districting scheme, like all laws that classify citizens on the basis of race, is constitutionally suspect,” Shaw v. Hunt (Shaw II), 517 U.S. 899, 904, 116 S.Ct. 1894, 135 L.Ed.2d 207 (1996). “This is true whether or not the reason for the racial classification is benign or the purpose remedial.” Shaw II, 517 U.S. at 904–05, 116 S.Ct. 1894 (citing Shaw I, 509 U.S. at 642–43, 653, 113 S.Ct. 2816). Therefore, the Equal Protection Clause requires legislative bodies to have a compelling reason for “us[ing] race as the predominant factor in drawing district lines.” Cooper, 581 U.S. at 291, 137 S.Ct. 1455.

Legislative bodies often point to obligations imposed upon them under the Voting Rights Act to justify racial gerrymandering. Section 2 of the VRA prohibits a “ ‘standard, practice, or procedure’ that ‘results in a denial or abridgement of the right ․ to vote on account of race.’ ” Cooper, 581 U.S. at 292, 137 S.Ct. 1455 (quoting 52 U.S.C. § 10301(a)). Section 2's “ban [ ] extend[s] to ‘vote dilution’ ” caused “by the ‘dispersal of [a group's members] into districts in which they constitute an ineffective minority of voters,’ ” Cooper, 581 U.S. at 292, 137 S.Ct. 1455 (quoting Thornburg v. Gingles, 478 U.S. 30, 46 n.11, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986)), such that the minority population has “less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice,” 52 U.S.C. § 10301(b).

To remedy vote dilution, a federal court may “devis[e] voting districts which ensure that all voters have an equally effective opportunity to participate in the electoral process.” Clark, 293 F.3d at 1266 (citing White v. Regester, 412 U.S. 755, 769, 93 S.Ct. 2332, 37 L.Ed.2d 314 (1973)). A remedial redistricting plan enacted in response to a finding of VRA § 2 liability is “ameliorative” and, “[a]s a matter of law,” is “not a statutory violation.” Clark, 293 F.3d at 1274 n.25. Remedial racial gerrymandering implemented under Section 2 of the VRA must be “narrowly tailored to achieve [the] compelling interest” of correcting violations of Section 2 of the Voting Rights Act by giving Black voters the ability to select candidates of their choice. Miller, 515 U.S. at 920, 115 S.Ct. 2475. A legislative body that engages in racial gerrymandering purportedly to cure a Section 2 violation “must establish that it had ‘good reasons’ to think that it would transgress the Act if it did not draw race-based district lines.” Cooper, 581 U.S. at 293, 137 S.Ct. 1455 (italics in Cooper).

Although legislative bodies “enjoy leeway to take race-based actions reasonably judged necessary under a proper interpretation of the VRA,” a court may not “approve a racial gerrymander whose necessity is supported by no evidence and whose raison d'etre is a legal mistake.” Cooper, 581 U.S. at 306, 137 S.Ct. 1455. Where evidence indicates that a legislative body “add[ed] more minority voters than needed for a minority group to elect a candidate of its choice” and “use[d] race, as opposed to other, ‘traditional’ factors” to “achieve an equal population goal,” that evidence may strongly, or even overwhelmingly, demonstrate that “race did predominate as a factor when” the legislative body drew district boundaries. Ala. Legis. Black Caucus v. Alabama, 575 U.S. 254, 260, 273, 135 S.Ct. 1257, 191 L.Ed.2d 314 (2015).

When evaluating claims of racial gerrymandering, courts “must be sensitive to the complex interplay of forces that enter a legislature's redistricting calculus.” Miller, 515 U.S. at 915–16, 115 S.Ct. 2475. “Redistricting legislatures will, for example, almost always be aware of racial demographics; but it does not follow that race predominates in the redistricting process.” Miller, 515 U.S. at 916, 115 S.Ct. 2475. As the Supreme Court noted in Miller v. Johnson:

The distinction between being aware of racial considerations and being motivated by them may be difficult to make. This evidentiary difficulty, together with the sensitive nature of redistricting and the presumption of good faith that must be accorded legislative enactments, requires courts to exercise extraordinary caution in adjudicating claims that a State has drawn district lines on the basis of race.

Miller, 515 U.S. at 916, 115 S.Ct. 2475; see also Alexander, 602 U.S. at 7, 144 S.Ct. 1221 (same). The “presumption of legislative good faith directs district courts to draw the inference that cuts in the legislature's favor when confronted with evidence that could plausibly support multiple conclusions.” Alexander, 602 U.S. at 10, 144 S.Ct. 1221 (citing Abbott v. Perez, 585 U.S. 579, 610–12, 138 S.Ct. 2305, 201 L.Ed.2d 714 (2018)).

A two-part test governs racial gerrymandering claims. First, a plaintiff must show “that race was the ‘predominant factor motivating the legislature's decision to place a significant number of voters within or without a particular district.’ ” Alexander, 602 U.S. at 7, 144 S.Ct. 1221 (quoting Miller, 515 U.S. at 916, 115 S.Ct. 2475). “To make that showing, a plaintiff must prove that the State ‘subordinated’ race-neutral districting criteria such as compactness, contiguity, and core preservation to ‘racial considerations.’ ” Alexander, 602 U.S. at 7, 144 S.Ct. 1221 (quoting Miller, 515 U.S. at 916, 115 S.Ct. 2475). “Racial considerations predominate when ‘[r]ace was the criterion that, in the State's view, could not be compromised’ in the drawing of district lines.” Alexander, 602 U.S. at 7–8, 144 S.Ct. 1221 (quoting Shaw II, 517 U.S. at 907, 116 S.Ct. 1894).

A plaintiff may establish racial predominance “through some combination of direct and circumstantial evidence.” Alexander, 602 U.S. at 8, 144 S.Ct. 1221 (citing Cooper, 581 U.S. at 291, 137 S.Ct. 1455); see also Miller, 515 U.S. at 916, 115 S.Ct. 2475 (to establish racial predominance, a plaintiff may present “circumstantial evidence of a district's shape and demographics or more direct evidence going to legislative purpose”). “Direct evidence often comes in the form of a relevant state actor's express acknowledgement that race played a role in the drawing of district lines.” Alexander, 602 U.S. at 8, 144 S.Ct. 1221. “Such concessions are not uncommon because States often admit to considering race for the purpose of satisfying [ ] precedent interpreting the Voting Rights Act of 1965.” Alexander, 602 U.S. at 8, 144 S.Ct. 1221 (citing Ala. Legis. Black Caucus, 575 U.S. at 259–60, 135 S.Ct. 1257). “In such instances, if the State cannot satisfy strict scrutiny, direct evidence of this sort amounts to a confession of error.” Alexander, 602 U.S. at 8, 144 S.Ct. 1221.

Circumstantial evidence is evidence from which inferences of racial predominance may be drawn. Circumstantial evidence includes “a district's shape and demographics,” Miller, 515 U.S. at 916, 115 S.Ct. 2475, but “ ‘bizarreness’ of shape is not a threshold showing for racial gerrymandering, Clark, 293 F.3d at 1270 (citing Miller, 515 U.S. at 917, 115 S.Ct. 2475). “Shape must be considered in conjunction with racial and population densities.” Clark, 293 F.3d at 1270. Demographic circumstantial evidence correlates race with legislative decisions to split counties, precincts, and other political subdivisions among districts. See Alexander, 602 U.S. at 21–22, 144 S.Ct. 1221. Expert witnesses may offer opinions regarding the role race played in redistricting. See Alexander, 602 U.S. at 24–33, 144 S.Ct. 1221.

Because the “racial predominance inquiry concerns the actual considerations that provided the central basis for the drawn lines, not post hoc justifications the legislature in theory could have used but in reality did not,” a conflict between the adopted plan and traditional redistricting criteria is not a threshold requirement for a plaintiff's claim of racial gerrymandering. Bethune-Hill v. Va. State Bd. of Elections, 580 U.S. 178, 189–90, 137 S.Ct. 788, 197 L.Ed.2d 85 (2017). Where a legislative body considers several maps which appear to comply with traditional redistricting principles and the evidence demonstrates that “race for its own sake” was “the overriding reason” the legislative body “[chose] one map over others, race still may predominate.” Bethune-Hill, 580 U.S. at 190, 137 S.Ct. 788. A plaintiff may establish racial predominance by, for example, demonstrating through circumstantial evidence that a legislative body “used ‘race as a proxy’ for ‘political interest[s].’ ” Alexander, 602 U.S. at 7 n.1, 144 S.Ct. 1221 (quoting Miller, 515 U.S. at 914, 115 S.Ct. 2475).

In reviewing the parties’ evidence, courts must analyze “racial gerrymandering with respect to the individual districts subject to the [plaintiffs]’ racial gerrymandering challenges”—an “undifferentiated [county]wide analysis is insufficient.” Ala. Legis. Black Caucus, 575 U.S. at 264, 135 S.Ct. 1257. That said, plaintiffs “can present [county]wide evidence in order to prove racial gerrymandering in a particular district.” Ala. Legis. Black Caucus, 575 U.S. at 263, 135 S.Ct. 1257 (italics in Ala. Legis. Black Caucus) (citing Miller, 515 U.S. at 916, 115 S.Ct. 2475).

If a plaintiff makes a showing of racial predominance, then the burden shifts to the state “to prove that the map can overcome the daunting requirements of strict scrutiny.” Alexander, 602 U.S. at 11, 144 S.Ct. 1221. The state must show that its “decision to sort voters on the basis of race furthers a compelling governmental interest.” Alexander, 602 U.S. at 11, 144 S.Ct. 1221 (citation omitted). If a state makes this showing, then a district court determines whether the state's “use of race is ‘narrowly tailored’—i.e., ‘necessary’—to achieve that interest.” Alexander, 602 U.S. at 11, 144 S.Ct. 1221.

The Court begins its analysis of the plaintiffs’ gerrymandering claim in this case by presuming that the Commission acted in good faith when it adopted the 2021 plan. Alexander, 602 U.S. at 6, 144 S.Ct. 1221.

***

To rebut the presumption that the Commission acted in good faith in 2021, the plaintiffs rely, in part, on the Commission's preclearance submissions between 1985 and 2013. The plaintiffs characterize this historic evidence as direct evidence of racial predominance. Though the Court has not found a case in which the Supreme Court or the Eleventh Circuit has addressed evidence of Section 5 preclearance submissions in the context in which the plaintiffs present them here, the Court believes these appellate courts would decline to accept preclearance submissions relating to previous redistricting legislation as direct evidence that race predominated in the Commission's selection of the boundaries of the five single-member districts in the 2021 plan.

The Supreme Court has set a high bar for direct evidence. See Cooper, 581 U.S. at 318, 137 S.Ct. 1455 (providing as an example of direct evidence in a racial gerrymandering case “scores of leaked e-mails from state officials instructing their mapmaker to pack as many black voters as possible into a district, or telling him to make sure its BVAP hit 75%”); Bush v. Vera, 517 U.S. 952, 116 S.Ct. 1941, 135 L.Ed.2d 248 (1996) (citing as “direct evidence of the legislature's racial motivation” in drawing district lines the legislature's Section 5 submission to DOJ concerning the map at issue and drawing attention to the legislature's statement that its proposed redistricting plan increased “black voting strength” in Harris County “by increasing the population to assure that the black community [might] continue to elect a candidate of its choice”); see also Jefferson v. Sewon Am., Inc., 891 F.3d 911, 921 (11th Cir. 2018) (defining direct evidence as “evidence, that, if believed, proves [the] existence of [discriminatory intent] without inference or presumption”) (quotation omitted); Harris v. Pub. Health Tr. of Miami-Dade Cnty., 82 F.4th 1296, 1301 (11th Cir. 2023) (noting that, in employment discrimination cases, “only the ‘most blatant remarks,’ whose intent could be nothing other than to discriminate, constitute direct evidence of discrimination” and explaining that “[i]f the alleged statement suggests, but does not prove, a discriminatory motive, then it is circumstantial evidence—not direct”) (internal quotation marks and quotation omitted).

As these cases illustrate, direct evidence constitutes clear evidence of the role that race played in legislative action. Here, because the Commission did not submit a preclearance letter to the DOJ in 2021, and because there are no other statements from commissioners that explicitly link the 2021 plan to earlier VRA compliance letters, the plaintiffs may establish that the Commission purposefully maintained, for example, racial thresholds identified in Section 5 preclearance submissions in the 2021 Enacted Plan only by inference. Because the Court's consideration of preclearance evidence from 1985, 1993, 2001, 2004, and 2013 requires “inference or presumption” to establish racial predominance in the Commission's redistricting work in 2021, the Section 5 preclearance submissions do not constitute direct evidence with respect to the 2021 plan.

Though they are not direct evidence of racial predominance concerning the 2021 plan, the Commission's preclearance submissions provide powerful circumstantial evidence of racial gerrymandering in 2021. As the Supreme Court explained in Abbott, while not necessarily predictive of the predominant considerations underlying the enactment at issue, the “ ‘historical background’ of a legislative enactment” is “ ‘one evidentiary source’ relevant to the question of intent” that courts weigh together with other circumstantial evidence of the Commission's purpose. Abbott, 585 U.S. at 603–04, 607, 138 S.Ct. 2305 (quotation omitted); see also Vill. of Arlington Heights v. Metro. Hous. Dev. Corp., 429 U.S. 252, 267, 97 S.Ct. 555, 50 L.Ed.2d 450 (1977) (“The specific sequence of events leading up to the challenged decision also may shed some light on the decisionmaker's purposes.”).³⁴ In examining the Commission's predominant purpose in adopting new district boundaries in 2021, the intent of earlier Commissions is relevant to the extent that it “naturally give[s] rise to—or tend[s] to refute—inferences” regarding the predominant purpose of the Commission in 2021. Abbott, 585 U.S. at 607, 138 S.Ct. 2305. Evidence from earlier redistricting cycles may illumine whether the Commission “reenact[ed] the plan previously passed by its 201[3] predecessor” and “carried forward the effects of any discriminatory intent on the part of the” 2013 Commission in the 2021 plan. Abbott, 585 U.S. at 604, 138 S.Ct. 2305.³⁵

A court's consideration of historical background “should be focus[ed] ․ on the specific sequence of events leading up to the challenged decision rather than providing an unlimited lookback to past discrimination.” League of Women Voters of Fla., Inc. v. Fla. Sec'y of State (LOWV), 32 F.4th 1363, 1373 (11th Cir. 2022) (brackets in LOWV) (internal quotation marks and quotation omitted). Here, the Court considers the Commission's decisions in several previous redistricting cycles to determine whether prior line-drawing sheds light on the Commission's predominant purpose in 2021. See Jacksonville Branch of NAACP v. City of Jacksonville, 635 F. Supp. 3d 1229, 1285–86 (M.D. Fla. 2022) (“The Court makes no attempt to study or recount the history of race discrimination in Jacksonville ․ But, because the [legislature] in 2022 expressly decided to maintain the lines that were drawn in 2011, the Court finds the history of what occurred in 2011 is part of the ‘specific sequence of events’ that led to the 2021 plan.”).³⁶

The relevant historical background for the Commission's redistricting decisions in 2021 traces its roots to 1985. In November 1985, the Commission sought approval from the DOJ for the first iteration of Jefferson County's new five-member Commission. (Doc. 169-2). The Commission explained in its letter to the DOJ that the Commission sought to move from a “commission form consisting of three (3) at-large positions” to a “five (5) member County Commission elected from districts within Jefferson County” to resolve “a minority vote dilution case.” (Doc. 169-2, p. 2, ¶ 3, 7). The Commission stated that two of the five districts were drawn “to provide [B]lack[ ] [voters] with a greater opportunity to elect [B]lack commissioners.” (Doc. 169-2, p. 2, ¶ 7). Specifically, District 1 “contain[ed] 65.6% [B]lack[ ]” individuals, and District 2 “contain[ed] 66.8% [B]lack[ ]” individuals, giving Black residents of Jefferson County “a greater opportunity to elect 2/5 or 40% of the County Commission positions.” (Doc. 169-2, p. 2, ¶ 8). The racial thresholds established by the Commission in 1985 reflected the conventional wisdom at the time that Black voters needed a 65% Black electoral district to elect candidates of their choice. (See Doc. 172, p. 241, tp. 241:6–13). As noted, the Eleventh Circuit has recognized that the Taylor consent decree produced “a five-member district form of government to ensure greater minority representation on the Commission.” Yeldell, 956 F.2d at 1058.³⁷ Thus, from its inception, the five-member Commission was designed to concentrate Black voters in two districts to give Black voters the opportunity to elect candidates of their choice. In the first election after the consent decree, voters in Districts 1 and 2 elected Black commissioners. (Doc. 169-6, pp. 7–8).

In 1993, the Commission submitted its first letter following the implementation of the Taylor consent decree. (Doc. 169-3). The Commission reported that 1990 census data indicated that the populations of Districts 1 and 2 had decreased-significantly. (Doc. 169-3, p. 2, ¶ 3). The Commission stated that it redrew district lines to “bring each district close to the ideal district population, without significantly changing the ratio of [B]lack and white populations within the district.” (Doc. 169-2, p. 2, ¶ 3). Because the population in Districts 1 and 2 had fallen as of 1990, to achieve ideal population distribution, the Commission had to increase the population in Districts 1 and 2 and reduce the population in Districts 3, 4, and 5.

The revised district boundaries that the Commission submitted to DOJ in 1993 demonstrate that the Commission accomplished its goal by moving more Black population into District 1. As of the 1990 census, District 1 consisted of 79,896 Black individuals (72.58%) and 30,188 white individuals (27.42%). (Doc. 169-3, p. 109). In redistricting, the Commission added 13,025 Black individuals to the district, for a total Black population of 92,921 or 73.25% of the population in District 1. (See Doc. 169-3, pp. 109–10). The Commission added only 2,926 white individuals such that white individuals composed 26.1% of the population in District 1 after redistricting. (See Doc. 169-3, pp. 109–10). The Commission added more white population to District 2 and decreased the Black population in the district, but the Black population after redistricting was 68.93% of the population in District 2, even though the Commission moved 5,623 Black individuals from the district to another district. (See Doc. 169-3, pp. 109–10).³⁸

An article published shortly after the Commission adopted its 1993 redistricting plan stated that following redistricting, District 1, the “smallest” district, was more than 73% Black. (Doc. 169-3, p. 130). In contrast, District 3, which “cover[ed] almost half the county,” was 78% white. (Doc. 169-3, p. 130). There is no evidence that Black citizens required supermajorities in Districts 1 and 2 to elect the candidate of their choice in those districts. See Cooper, 581 U.S. at 293, 137 S.Ct. 1455.

The pattern of increasing the Black population in Districts 1 and 2 continued in the redistricting cycle that followed the 2000 census. In a November 2001 letter to DOJ, the Commission wrote that “[t]he 2001 plan has two [B]lack majority districts, just like the 1993 plan. Each of these districts has majority [B]lack populations in excess of 65%, under both the 2001 plan and the 1993 plan.” (Doc. 169-6, pp. 11). Under the 2001 legislation, the Black population in District 1 increased from 73.25% to 78% of the population, and the Black population in District 2 increased from 68.93% to 73.45% of the population. (Doc. 169-6, p. 11). The Commission explained that in rearranging “the boundaries to obtain population for compliance with the one person-one vote rule,” it expanded the boundaries of Districts 1 and 2 “somewhat outward, as compared with their previous boundaries.” (Doc. 169-6, p. 12). Through municipal splits of Ensley, Fairfield, and Fultondale in District 1, and Bessemer in District 2, (Doc. 169-6, p. 12), the Commission tracked Black population that expanded beyond Birmingham's municipal boundaries and added to District 1 the parts of communities that had become predominately Black, producing a District 1 population that was 78% Black—a population that significantly exceeded the population needed for Black voters to select the candidates of their choice in District 1. Notably, in 2001, the Commission placed the City of Fairfield in District 1 rather than in District 3 or District 5 because the popular mayor of Fairfield, who was Black, intended to run for a seat on the Commission. (Doc. 169-6, p. 13).

The data that the Commission submitted with its 2001 preclearance letter follows the trends established in the 1993 redistricting. (See Doc. 169-6, pp. 42–45). The data shows that while the white population in District 1 had fallen nearly 50%, and the white population in District 2 had fallen slightly more than 30% over 10 years, (Doc. 169-6, p. 42), in redistricting, the Commission drew lines that replaced the white population lost from those districts largely with Black population. (See Doc. 169-6, pp. 42–43). The Commission added 10,019 white individuals and 15,098 Black individuals to District 1, maintaining Black voters’ supermajority in District 1. After redistricting, Black individuals comprised 73.45% of the district's population. (See 169-6, pp. 42, 44). In District 2, the Commission added 5,999 white individuals and 4,655 Black individuals, moving Black individuals’ percentage of the District 2 population from 68.93% in 1990 to 73.45%. (See 169-6, pp. 42, 44).³⁹

The Commission repeated to the DOJ that in redistricting in 2001, the Commission had maintained “two of the districts containing African-American majorities in excess of 65%.” (Doc. 169-6, pp. 11). The Commission again stated that its changes brought “each district close to the ideal district population without significantly changing the ratio of [B]lack and white population within the districts.” (Doc. 169-6, p. 11). In 2001, VAP statistics showed that under the new district lines, District 1 had a Black VAP that was 74.89%. (Doc. 169-6, p. 44). The statistics showed that under the new district lines, District 2 had a Black VAP that was 68.78%. (Doc. 169-6, p. 44). In Districts 3, 4, and 5, the Black VAP was, respectively, 15.83%, 12.92%, and 11.01%. (Doc. 169-6, pp. 44–45). There is no evidence that the Commission had information in 2001 that suggested that Black voters required a supermajority in Districts 1 and 2 to elect the candidates of their choice. See Cooper, 581 U.S. at 293, 137 S.Ct. 1455.

The Commission's final Section 5 submission to DOJ in 2013, reflecting boundary adjustments following the 2010 census, demonstrates that the Commission repeated the pattern of redistricting in 1993 and 2001 in the 2013 plan. (Doc. 169-5). The Commission stated: “The 2013 plan has two [B]lack majority districts, just like the 1993 and 2001 plans. Each of these districts have [sic] majority [B]lack populations in excess of 65%, under the 2013, 2001, and 1993 plan[s].” (Doc. 169-5, p. 1081). The Commission again noted that the county's Black population had grown, comprising 41% of the county's total population. (Doc. 169-5, p. 1075). The Commission again stated that changes in district boundaries would “bring each district close to the ideal district population without significantly changing the ratio of [B]lack and white population within the districts.” (Doc. 169-5, pp. 1080). The Commission noted that, in “the new plan, the districts [we]re drawn in such a way that incumbent African-Americans represent[ed] the districts in which the population is majority [B]lack.” (Doc. 169-5, p. 1083). The Commission stated that in the years since Districts 1 and 2 were created under the Taylor consent decree, “each of these two majority [B]lack districts has elected a [B]lack candidate to the Commission.” (Doc. 169-5, p. 1075).

In terms of demographic changes, following the 2010 census, Districts 1 and 2 again lost population, while Districts 3, 4, and 5 gained population. (Doc. 169-5, p. 1080). Although District 1 lost 5,433 Black individuals from 2000 to 2010, post redistricting, the percentage of Black population in the district remained above 75% at 76.14%. (Doc. 169-5, p. 1081). District 2 maintained its Black population, and the percentage of Black individuals in its population dropped by 0.06% to 73.39%. (See Doc. 169-5, p. 1081).

In sum, the Commission's Section 5 preclearance letters to the DOJ between 1985 and 2013 are direct evidence of the Commission's intent in 1985, 1993, 2001, 2004, and 2013. As this discussion of the preclearance letters demonstrates, the letters are powerful circumstantial evidence of the continuation of a pattern of using race to set the boundaries of the Commission's districts in 2021. See Miller, 515 U.S. at 919, 115 S.Ct. 2475 (citing statement from state official to DOJ during preclearance process as “powerful evidence that the legislature subordinated traditional districting principles to race”). The submissions disclose the Commission's purpose of redrawing district boundaries after each census to aggregate the BVAP in Jefferson County in Districts 1 and 2 and the pattern the Commission established of bolstering BVAP in Districts 1 and 2 over consecutive redistricting cycles. The history of the Commission's consistent report to DOJ of the effort to create and maintain two majority Black voting districts in Jefferson County permits the inference that the Commission continued its decades-long practice in 2021. Abbott, 585 U.S. at 607, 138 S.Ct. 2305. Moreover, by “invoking core retention ․ as the predominant motive behind the shape of the Challenged Districts, the [Commission] makes the historical foundation for these districts particularly relevant.” City of Jacksonville, 635 F. Supp. 3d at 1286.

The DOJ submissions reveal the Commission's intent not only to create and maintain two majority-minority districts in Jefferson County but also to maintain a specific racial threshold in those districts. As the Commission stated in its 2013 letter: “The 2013 plan has two [B]lack majority districts, just like the 1993 and 2001 plans. Each of these districts have [sic] majority [B]lack populations in excess of 65%, under the 2013, 2001, and 1993 plan[s].” (Doc. 169-5, p. 1081). In Bethune-Hill, the Supreme Court recognized that express racial targets may be evidence of racial gerrymandering. Bethune-Hill, 580 U.S. at 192, 137 S.Ct. 788; see Cooper, 581 U.S. at 300–01, 137 S.Ct. 1455 (holding that district court did not clearly err in finding that race predominated in redistricting where legislative body used an “announced racial target” in redistricting). Here, the evidence presents a pattern of decennial increases in Black population in Districts 1 and 2 through redistricting, consolidating BVAP in Districts 1 and 2 and diminishing BVAP in other districts so as to maintain a Black population in Districts 1 and 2 in excess of 65%.

(Doc. 169-2, pp. 19–20; Doc. 169-3, p. 110; Doc. 169-6, p. 44; Doc. 169-5, p. 1081; Doc. 169-107, p. 18).⁴⁰ In Cooper, the Supreme Court found that the district court properly considered that the “State's preclearance submission to the Justice Department indicated a similar determination to concentrate [B]lack voters in District 12.” Cooper, 581 U.S. at 311, 137 S.Ct. 1455.

In short, the Section 5 preclearance materials from 1985, 1993, 2001, 2004, and 2013 contain evidence of the Commission's purpose in each of those redistricting cycles. In each, the Commission redistricted to achieve ideal population across all districts and to maintain two Black majority districts with the Black population in each constituting more than 65% of the total population. The preclearance data from each redistricting cycle demonstrates that though the population fell in Districts 1 and 2 in the 1990, 2000, and 2010 censuses, following each redistricting cycle, the Commission exceeded the 65% racial target in Districts 1 and 2 by, on average, 10%. Clark, 293 F.3d at 1269-70 (“[A]lthough malapportionment was the ‘why’ of the redistricting plan, race was the ‘how.’ ․ The evidence is overwhelming that the County decided at the outset to maintain its two [B]lack voting districts and to assign as much of the [B]lack voting age population as possible to those districts.”) (italics in Clark). In each redistricting cycle, there is no evidence that indicates that Black voters required Black supermajorities in Districts 1 and 2 to select candidates of their choice. See Cooper, 581 U.S. at 293, 137 S.Ct. 1455.

***

Against the backdrop of historical evidence, the Commission's purported reliance on traditional redistricting criteria to explain the Commission's 2021 plan is not credible. The Commission argues that in 1985, when the Commission transitioned to five districts, it anchored Districts 1 and 2 in the City of Birmingham, maintained Districts 1 and 2 in Birmingham in subsequent redistricting cycles, and continued adding parts of Birmingham to Districts 1 and 2 in the 2021 plan. (Doc. 177, pp. 129–33). The Commission argues that there was a racial effect of that line-drawing, but this effect was not the equivalent of race-based redistricting. (Doc. 177, pp. 133–34). According to the Commission, the preexisting population of Birmingham in 1985 necessarily meant that the Commission's split of the city into two districts produced two majority-minority districts. (See Doc. 177, pp. 129–33).

Here, the Commission's reimagining of its restructuring in 1985, characterizing race as incidental to its line drawing, cannot be reconciled with the historical record or with precedent. In its 1985 submission to DOJ, the Commission reported that it deliberately designed two of its five districts “to provide [B]lack[ ] [voters] with a greater opportunity to elect [B]lack commissioners.” (Doc. 169-2, p. 2). The redistricting did not happen in a vacuum, and the Commission did not share its plans with DOJ on its own accord. Rather, the Commission reported its redistricting to the DOJ per a consent decree that resolved alleged VRA and constitutional violations concerning vote dilution in Jefferson County. Though the City of Birmingham was the building block for the Commission's two majority Black districts, the Commission over consecutive redistricting cycles tracked Black citizens who moved from Birmingham and added Black communities outside of the City of Birmingham to Districts 1 and 2 to concentrate BVAP in those districts and to minimize BVAP in Districts 3, 4, and 5.

Under the Taylor consent decree, to comply with the VRA, the Commission purposefully created majority Black populations in Districts 1 and 2. The Taylor consent decree compelled this result.⁴¹ Contrary to the Commission's position, the Commission's 1985 preclearance submission shows that the geographic centering of District 1 and 2 around Birmingham represents the effect of its decision to draw district lines based on race. To achieve 65% Black districts, the Commission had to create majority-Black districts where the county's Black population lived. The Commission's initial anchoring of Districts 1 and 2 in Birmingham and its expansion of those districts in thin, awkward juts and protrusions through a series of municipal splits evinces this same intent in subsequent redistricting cycles.

The Commission argues that “[c]ore preservation is a legitimate, nonracial redistricting priority” that explains the 2021 plan. (Doc. 177, pp. 125–29) (citing Alexander, 602 U.S. at 7, 144 S.Ct. 1221). Mr. Stephenson stated that the three maps the Commission designed in 2021 were “least-changes plans from the prior redistricting plan.” (See Doc. 174, pp. 225–26, tpp. 718:17–719:11). He testified that the Commission used the procedure in drawing the 2021 plan that the Commission used to draw the 2013 plan. (Doc. 174, pp. 160–61, 220, tpp. 653:9–654:11, 713:8–23).⁴² Dr. Barber opined that “[t]he 2021 enacted map seeks to retain the population of the 2013 commission districts to an extremely high degree. All districts retain more than 90% of their population and the overall shift in population as a result of the 2021 redraw is less than 5% of the population of the county.” (Doc. 179-16, pp. 4, 52). Dr. Barber reasoned that “[i]t is clear by the incredibly small shift in population” in the 2021 redistricting “that the primary objective of the commission when drawing the new maps was to retain the old districts to a very high degree while also bringing the districts in line with population equality.” (Doc. 179-16, p. 5).

As discussed, by “invoking core retention ․ as the predominant motive behind the shape of the Challenged Districts, the [Commission] makes the historical foundation for these districts particularly relevant.” City of Jacksonville, 635 F. Supp. 3d at 1286. Had the Commission simply readopted the 2013 lines with slight modifications to add population to Districts 1 and 2 without regard to the race of the voters who changed districts, the Commission's arguments might hold water. But the evidence shows that is not what happened. Therefore, the Commission cannot rely on core retention to shield itself from the plaintiffs’ racial gerrymandering claims.

***

As mentioned, the evidence demonstrates that the Commission consistently adjusted the lines for Districts 1 and 2 to follow Black population growth beyond the City of Birmingham. At the time of the 1985 consent decree, much of Birmingham's Black population resided in Birmingham, and in the 1985 map, District 1 largely overlapped Birmingham. (Doc. 172, pp. 240, 261, tpp. 240:10–18, 261:1–6). As demonstrated by Mr. Cooper's map below, the county's Black population remained concentrated in the Birmingham area in 1990, with only one majority-Black precinct to the north of downtown. (Doc. 172, p. 262, tp. 262:12–18; Doc. 169-73).

(Doc. 169-73).

By the time of the 2000 census, the county's Black population had begun expanding northeast, as depicted in Mr. Cooper's map reproduced below.

(Doc. 169-116).

The evidence shows, for example, that Huffman, a community 13 miles northeast of downtown Birmingham, was majority white in 1992. (See Doc. 173, pp. 169–70, 184, tpp. 461:13–462:5, 461:14–16, 476:21–23; Doc. 171, p. 2, ¶ 2). Today, Huffman is majority Black. (Doc. 173, p. 184, tp. 476:16–19). Communities like Midfield and Center Point also have experienced demographic shifts from predominantly white to predominantly Black or mixed race. (Doc. 173, pp. 188–89, tpp. 480:21–481:19). Over time, the Commission has incorporated these communities into District 1. (Doc. 169-115) (comparing Doc. 169-113 with Doc. 169-114); (compare Doc. 176-3, p. 3, with Doc. 169-108, p. 22).

The Commission achieved its goal of maintaining a supermajority Black District 1 by tracking Black population growth beyond the corporate boundaries of the City of Birmingham. With Jefferson County's 2020 demographics, District 1 would have been 47.7% Black in 1990, but it was 76.75% Black in 2020. (Doc. 169-75; Doc. 169-113; Doc. 169-114; Doc. 169-115). As the Commission has tracked expansion of Black population beyond Birmingham's 1985 municipal boundaries, District 1 has developed an irregular shape. (Compare Doc. 176-3, p. 3, with Doc. 169-108, p. 22); see Miller, 515 U.S. at 916, 115 S.Ct. 2475 (noting shape of district relevant to racial gerrymandering analysis).

As with District 1, the Commission has expanded District 2 to the southwest to capture the expanding Black population in the Bessemer area. District 2 also has taken in parts of Homewood, which has experienced an increase in Black population. (Doc. 172, pp. 211–12, tpp. 211:21–212:4; Doc. 173, p. 190, tp. 482:1–3). The maps below depict these changes.

(Doc. 169-73).

(Doc. 169-116).

(Doc. 169-113) (2021 plan with 1990 census data).

(Doc. 169-114) (2021 plan with 2020 census data).

(Doc. 169-108, pp. 16, 22).

Though not as pronounced as the Commission's changes to District 1 over time, the Commission's changes to District 2 have ensured that the district's Black population has remained near the 65% threshold in the consent decree. District 2 would have been 68.93% Black in 1990 and 64.75% Black in 2020. (Doc. 172, p. 265, tp. 265:8–16; Doc. 169-75). The Commission structured District 2 to have a 66.8% Black population in 1985, a 68.93% Black population in 1993, a 73.45% Black population in 2001, a 73.39% Black population in 2013, and a 66.18% Black population in 2021. (Doc. 169-2, pp. 19–23; Doc. 169-3, p. 110; Doc. 169-6, pp. 44; Doc. 169-5, p. 1081; Doc. 169-107, p. 18). Before the 2021 redistricting, District 2 had a BVAP of 66.7%, and the district had a BVAP of 64.11% following redistricting. (Doc. 179-16, p. 22; Doc. 169-31).

To follow the movement of Black citizens from the City of Birmingham and place those citizens in Districts 1 and 2, the Commission departed from traditional redistricting criteria. The Commission increased the number of census place splits in the 2021 plan, splitting 25 census places as compared to 22 in the 2013 plan. (Doc. 169-107, p. 46; Doc. 175, p. 148, tp. 884:15–23). The increase in census place splits suggests that the Commission did not prioritize the integrity of political subdivisions when redistricting in 2021. Municipalities split in the 2021 plan include Brighton, Fultondale, Homewood, Hoover, Midfield, Irondale, Bessemer, Center Point, Trussville, Leeds, and Tarrant, areas where Black populations have migrated from Birmingham. (See Doc. 169-108, pp. 117–29). The Commission placed the Black population of these cities in Districts 1 and 2. This evidence allows the inference that, in deciding to split municipalities in the 2021 plan, the commissioners did so predominantly because of race. Alexander, 602 U.S. at 7, 144 S.Ct. 1221 (explaining that racial gerrymandering occurs when lawmakers place voters “ ‘within or without a particular district’ ” because of the voters’ race (quoting Miller, 515 U.S. at 916, 115 S.Ct. 2475)).

As discussed, Mr. Fairfax's illustrative plan shows that it was possible to draw a map that respected equal population, contiguity, compactness, and minimized political subdivision splits without the demographic effects of the Commission's 2021 plan. (Doc. 169-107, pp. 44–49). Mr. Fairfax's illustrative plan performed better than the 2021 plan on the traditional redistricting criteria of compactness, contiguity, and splitting census places, as the Fairfax plan split only four census places. (Doc. 169-107, pp. 44–49; Doc. 172, pp. 51–52, tpp. 51:17–52:17).

Mr. Cooper's illustrative plans performed equal to or better than the 2021 plan in terms of contiguity, compactness, respect for political subdivisions, and non-dilution of minority strength. (See Doc. 172, pp. 279–80, tpp. 279:15–280:4; Doc. 173, pp. 43, 59–60, tpp. 335:14–17, 351:17–352:3). Mr. Cooper's illustrative plans A, B, C, and E outperformed the 2021 plan in terms of municipal splits and total splits. Together, the plaintiffs’ illustrative plans demonstrate that if the Commission had adhered to traditional redistricting principles, the Black population in Districts 1 and 2 would have fallen, but the Black population in those districts would not have fallen so much as to prevent Black voters from electing candidates of their choice. The Commission's decision to maintain supermajority-Black populations in Districts 1 and 2 at the expense of traditional redistricting criteria allows the inference that race predominated in the 2021 redistricting process. Ala. Legis. Black Caucus v. Alabama, 575 U.S. at 260, 273, 135 S.Ct. 1257 (evidence that indicates that a legislative body “add[ed] more minority voters than needed for a minority group to elect a candidate of its choice” and “use[d] race, as opposed to other, ‘traditional’ factors” to “achieve an equal population goal,” demonstrates that “race did predominate as a factor when” the legislative body drew district boundaries).

***

Remarks made by Commissioner Scales and by Commissioner Tyson underscore the predominant role of race during the 2021 redistricting cycle. During the Commission meeting at which the Commission enacted the 2021 map, Commissioner Scales remarked: “We speak of Democratic versus Republican. You figure out what that looks like.” (Doc. 169-11, p. 32, tp. 31:21–22). Commissioner Scales then described several communities, including those the Commission added to her district, by purported political affiliation. She stated: “I am having to take in Center Point which is highly Democratic, I am taking in Dolomite which is highly Democratic, but you ask yourself concerning Homewood, Ross Bridge, Lake Shore, is that a heavily populated Democratic area?” (Doc. 169-11, p. 34, tp. 33:1–5).

The Commission did not have partisan political data available to it when redistricting, but its mapping software included racial data. (Doc. 174, pp. 211–12, tpp. 704:18–705:8). Mr. Douglas testified that he understood similar comments Commissioner Tyson made to refer to voters’ race. (Doc. 173, p. 195, tp. 487:8–10). In context, Commissioner Scales's remarks support the inference that the Commission intentionally carved Black population from several communities and moved the predominately Black populations of cities and precincts into District 1 while leaving the predominately white populations in those cities and precincts outside of District 1.

Commissioner Tyson's remarks during the 2021 redistricting process underscore the Commission's goal of maintaining District 2 as a supermajority Black district. Commissioner Tyson stated that in the 2021 redistricting, District 2 gained Rosedale, “a 99.2% Black community.” (Doc. 169-11, p. 40, tp. 39:18–20). She explained that District 2 pulled in Mountain View, an “89 percent Democratic and [B]lack” community. (Doc. 169-11, p. 41, tp. 40:10–12). District 2 obtained a “99 percent Democratic, 99 percent Black” part of Bessemer. (Doc. 169-11, p. 41, tp. 40:15–19).

While Commissioner Tyson referred to these areas in part based on their purported political affiliation, Mr. Douglas testified that it was clear to him that Commissioner Tyson was discussing race. (Doc. 173, p. 195, tp. 487:8-10). Commissioner Tyson acknowledged as much when she stated that she knew “what [she was] getting” in redistricting because she “got up, walked over there [ ], and [ ] looked at the folks in they face.” (Doc. 169-11, p. 41, tp. 40:20–23). Following redistricting, Commissioner Tyson told Mr. Simelton and Mr. Hall that she believed that District 2 had more Black individuals than necessary for her to be elected. (See Doc. 173, p. 155, tp. 447:3–8; Doc. 174, pp. 26–29, tpp. 519:20–522:5). Commissioner Tyson's comments provide additional evidence of racial predominance in the drawing of District 2.

***

The specific changes to each Commission district in 2021, coupled with the analyses performed by the plaintiffs’ experts, demonstrate how the Commission's use of race predominated during the 2021 redistricting cycle.

District 1

As discussed, the Commission split several 2013 precincts to shift predominantly Black populations into District 1. For example, in 2013, the Commission placed the entire East Pinson Valley precinct in District 4. (Doc. 169-107, p. 25). The 2021 plan split East Pinson Valley between Districts 1 and 4. (Doc. 169-107, pp. 25–26). Of the East Pinson Valley citizens the Commission moved into District 1, 86.18% are Black. (Doc. 172, p. 36, tp. 36:1–25; Doc. 169-107, p. 26). Only 9.79% of voters in the new District 1 portion of East Pinson Valley are white. (Doc. 172, pp. 36–37, tpp. 36:19–37:1; Doc. 169-107, p. 26). The East Pinson Valley voters who remained in District 4 are 56.91% Black and 23.70% white. (Doc. 169-107, p. 26). Adding the entire East Pinson Valley Center precinct to District 1 would have had the effect of lowering the Black population percentage in District 1. (Doc. 169-107, pp. 26–27).⁴³

Every whole precinct or VTD the Commission moved into District 1 was majority-Black. (Doc. 172, pp. 38–39, tpp. 38:24–39; Doc. 169-109, p. 9). For example, the Commission moved the Minor Fire Station precinct, 52.21% Black, from District 3 to District 1. (Doc. 169-109, p. 9). The Commission moved the Center Point Community Center precinct in north Jefferson County from District 4 to District 1. (Doc. 169-109, p. 9). In the process, the Commission effectively split the city of Center Point in half. (See Doc. 169-108, p. 23). Of the 6,202 voters in that precinct, 80.86% are Black and 12.61% are white. (Doc. 169-109, p. 9).

The 2021 plan added the only adjacent non-District 2 precincts that are above 78% Black to District 1. (Doc. 169-107, p. 35). Adding other adjacent precincts would have lowered District 1's Black population percentage, as demonstrated in the map below. (Doc. 169-107, p. 35).

(Doc. 179-16, p. 11).

Though the Commission's stated goal in redistricting in 2021 was to add voters to District 1 to equalize population in the Commission's five districts, the Commission moved the Brookside Community Center precinct out of District 1. (Doc. 179-16, p. 13). That precinct is 24.2% Black. (Doc. 179-16, pp. 11, 13). This change is particularly telling. Dr. Barber asserts that in 2021, the areas contiguous to District 1 were predominately Black, so any changes the Commission could have made would not have had the effect of lowering the district's Black population significantly. (See Doc. 175, pp. 41–42, tpp. 777:2–778:7). Setting aside the data discussed that undermines this argument, the Commission's decision to eliminate from District 1 a precinct that was 24% Black instead of all or parts of precincts that were more Black, (see Doc. 169-107, p. 34; Doc. 179-16, p. 11), supports the plaintiffs’ contention that race was the factor that could not be compromised. For example, looking to the northern boundary of District 1 as depicted in the map below, the Commission could have extracted from District 1 parts of precincts that contained BVAPs of 55.4%, 89.2%, 81.4%, 83.3%, 72.1%, 98.1%, 82.7%, 79.7%, and 67.1%.

With a BVAP of 76.34%, the BVAP in District 1 is higher than in 95.1% of the corresponding District 1s in Dr. Barber's 100,000 simulated plans. (Doc. 169-26, p. 9; Doc. 175, p. 127, tp. 863:13–16). This is a statistically significant result. (Doc. 169-26, p. 9). Dr. Barber hypothesized that the Commission may have made changes to District 1 to increase the overlap between District 1 and Commissioner Scales's former City Council District. (Doc. 175, p. 34, tp. 770:1–23). Dr. Barber has never spoken to the commissioners, has no knowledge of their motives, and does not know what criteria they used. (Doc. 175, p. 56, tp. 792:7–10). Commissioner Scales noted in the public record that neither she nor her office drew the three plans the Commission considered. (Doc. 179-8, p. 33, tp. 31:2–9). Commissioner Scales voted against the 2021 plan, (Doc. 179-8, p. 37, tp. 35:4–21). This fact cuts against Commissioner Scales having participated meaningfully in the map-drawing process. Therefore, Dr. Barber's testimony does not find support in the record and does not provide a race-neutral explanation for the changes to District 1.

Dr. Barber testified that his regression analysis revealed that race was not a statistically significant predictor of which precincts were added to District 1 and that geographic proximity was the only significant predictor. (Doc. 179-16, pp. 20–21). Dr. Liu explained that Dr. Barber's regression did not measure the interactive effect of geographic proximity and BVAP percentage at the precinct level and thus could not assess whether the geographically adjacent precincts that were also heavily Black were more likely to be moved to District 1 than were other adjacent precincts that were more white. (Doc. 169-22, p. 3). Dr. Liu performed a multivariate regression with an interactive term to test whether race was a significant factor in the Commission's decision to add an adjacent precinct to District 1. (Doc. 169-22, pp. 3–4). Dr. Liu found that race was a statistically significant predictor of adjacent precinct movement into District 1 at the 1% statistical significance level. (Doc. 169-22, p. 4).

Mr. Cooper's illustrative plan D shows that the Commission could have drawn District 1's boundaries while maintaining a high core retention rate in the plan. In plan D, District 1 had a 62.77% Black population and an overall core retention rate of 85.74%, compared to 78.27% and 95.3%, respectively, in the 2021 plan. (Doc. 172, p. 286, tp. 286:18–22; Doc. 169-77; Doc. 173, p. 25, tp. 317:4–15; Doc. 179-18; see Doc. 169-26, p. 14).⁴⁴

Several precincts at the boundaries of District 1 were assigned to districts with particularly high BVAPs. (Doc. 169-26, p. 16). In Dr. McCartan's simulated plan, these high-BVAP precincts were assigned to districts that are as much as 40 percentage points less Black than the districts they belong to under the 2021 plan. (Doc. 169-26, p. 16). Dr. McCartan noted that many of these high-BVAP precincts were the same precincts Dr. Barber stated the Commission pulled into District 1 for reasons other than race. (Doc. 169-26, pp. 16–17). Dr. McCartan found that every precinct assigned to District 1 in the 2021 plan has a statistically significant higher BVAP as compared to its average district assignment in Dr. Barber's set of simulated maps. (See Doc. 169-26, p. 11).⁴⁵

In sum, the historic drawing of District 1's boundaries to create Black supermajorities, the demographic evidence from changes made to District 1 in the 2021 redistricting cycle, and the expert analyses of those changes support the conclusion that race predominated when the Commission altered District 1's boundaries in 2021. As explained, the Commission's series of preclearance submissions undermine the Commission's reliance on core retention as a race-neutral reason for the Commission's changes. The other explanations offered by the Commission do not find support in the public record. The Commission cannot rely on the post-hoc speculation of Dr. Barber and Mr. Stephenson to justify its map. See Bethune-Hill, 580 U.S. at 189–90, 137 S.Ct. 788. Therefore, the Court finds that race predominated in the drawing of District 1.

District 2

While not as stark as the Commission's adjustments to District 1, the Commission's precinct-level changes to District 2 in 2020 also evidence a continuation of the Commission's efforts to maintain Black supermajorities in the two districts instituted in the 1985 consent decree to allow Black citizens to elect the candidates of their choice. (See Doc. 169-107, pp. 39–50). In 2020 as in previous redistricting cycles, the Commission maintained this practice without analyzing whether Black voters required District 2's 64.11% BVAP to elect a candidate of their choice.

As in District 1, in District 2, the Commission maintained its historical practices and departed from traditional redistricting practices to move Black population to District 2 to increase the total population there. For example, in splitting the Bessemer Civic Center precinct between Districts 2 and 3, the Commission moved an 80.5% Black segment of the precinct into District 2. (Doc. 169-107, p. 29). This precinct split does not follow municipal boundaries. (Doc. 169-107, p. 20; Doc. 169-108, p. 24). As demonstrated by Mr. Fairfax, in the 2021 plan, District 2 cuts so deeply into Bessemer that the parts of Bessemer remaining in District 3 no longer are contiguous. (See Doc. 169-108, p. 24).⁴⁶ Overall, of the 7,008 people moved into District 2 from District 3, 59.8% were Black. (Doc. 169-72, p.1; Doc. 172, p. 274, tp. 274:9–10).

Of the three whole precincts the Commission placed in District 2, only one, the Afton Lee Community Center precinct, or the Rosedale neighborhood, is majority Black. (Doc. 169-107, p. 39). The other two precincts, the Homewood and Grant Street Baptist Church precincts, are majority white, (Doc. 169-107, p. 39), but the addition of these precincts did not meaningfully alter District 2's heavily majority Black demographic. The Homewood and Rosedale areas have growing Black communities, (Doc. 173, p. 190, tp. 482:2–3; Doc. 169-115), meaning that the addition of these areas will increase District 2's Black population in the coming years.

The BVAP in District 2 is higher than 96.6% of the corresponding districts in Dr. Barber's simulations. (Doc. 169-26, p. 9). Dr. Liu did not analyze the relationship between BVAP and adjacency for the precincts the Commission moved into District 2 in 20221, (Doc. 175, p. 47, tp. 783:4–23), but Dr. McCartan's hatch map shows District 2's BVAP, inclusive of the Bessemer, Ross Bridge, Rosedale, and Homewood precincts moved into District 2, is more extreme than 95% of the districts those precincts were assigned to in Dr. Barber's simulations, (Doc. 169-26, pp. 9, 11).

In sum, historical racial packing in District 2, the demographic evidence from changes made to District 2 in the 2021 redistricting cycle, the expert analyses of those changes, and Commissioner Tyson's remarks (discussed above) support the conclusion that race predominated when the Commission redrew District 2's boundaries in 2021. As with District 1, the Commission's explanations for the changes to District 2 do not find support in the record, and Dr. Barber's and Mr. Stephenson's post-hoc speculations are not credible in the face of the objective and expert evidence of racial predominance, see Bethune-Hill, 580 U.S. at 189–90, 137 S.Ct. 788. Therefore, the Court finds that race predominated in the drawing of District 2.

District 3

The Commission redrew the boundaries of Districts 3, 4, and 5 to remove Black voters from these districts and move these Black voters to Districts 1 and 2. In 2020, the BVAP of District 3, as redrawn in 2013, was 28.6%. (Doc. 169-6, p. 44; Doc. 179-16, p. 31).⁴⁷ In 2021, the Commission's revisions of the boundaries of District 3 reduced the district's BVAP to 25.8%. (Doc. 179-16, p. 31). Though the Black population in District 3 is not substantial numerically, of the individuals the Commission removed from District 3 in 2021, 51.45% are Black, and approximately 90% of the individuals the Commission moved into District 3 are white. (Doc. 169-107, p. 41; Doc. 172, p. 276, tp. 276:9–11). As discussed, by splitting the Dolomite West Field Community Center precinct, the Commission retained in District 3 the less heavily Black neighborhoods in the original precinct while moving an 86.6% Black segment of the precinct into District 1. (Doc. 169-109, p. 9). The Commission also moved a majority Black precinct, the Minor Fire Station precinct, from District 3 to District 1. (Doc. 169-109, p. 9). The Minor Fire Station precinct was 52.21% Black. (Doc. 169-109, p. 9). The portion of the Bessemer Civic Center precinct that the Commission left in District 3 is roughly 10% more white than the portion of the precinct the Commission moved into District 2. (Doc. 169-107, p. 29). By splitting the Ross Bridge precinct, the Commission kept in District 3 a 71.69% white portion of the precinct. (Doc. 169-107, p. 31).

Tellingly, though the Commission needed to remove population from District 3 to achieve the Commission's equal population goal, the Commission added a segment of the Warrior City Hall precinct to District 3 from District 4, another overpopulated district. (Doc. 169-107, p. 41). This change added 1,445 individuals to District 3. (Doc. 169-107, p. 41). Of those individuals moved into District 3 from District 4, 90.52% were white. (Doc. 169-107, p. 41). The Commission also moved the 24.2% Black Brookside Community Center precinct from District 1 to District 3 so that the Commission could add to District 1 precincts that were more than 50% Black. (Doc. 179-16, p. 13). Race predominated in the changes to District 3 because the individuals the Commission moved in and out of District 3 facilitated the Black supermajority in District 1 and the near supermajority in District 2.

Dr. McCartan's analysis shows that District 3's BVAP in the 2021 plan is lower than the BVAP in 99.7% of the corresponding districts in Dr. Barber's simulation, a statistically significant result. (Doc. 169-26, p. 9). Dr. McCartan's hatch map shows that the Commission placed most of the precincts in the central and southwestern portions of District 3 into a district that had a lower BVAP than 95% of the corresponding simulated districts. (Doc. 169-26, pp. 9, 11).

The Commission had other options. For instance, the Commission could have altered District 1 by moving the Minor Fire Station, Adamsville Baptist Church, and Mulga Town Hall precincts from District 3 to District 1 instead of adding portions of East Pinson Valley and the Dolomite West Field Community Center precincts and the Center Point Community Center precinct. (Doc. 172, pp. 198–99, tpp. 198:1–199:11; see Doc. 179-13). If the Commission had done so, it would have achieved its equal population goal while removing fewer Black voters from District 3. (Doc. 172, pp. 198–99, tpp. 198:1–199:11).

In sum, as in previous redistricting cycles, the Commission removed Black voters from District 3 to maintain the district's white majority and to sustain Black majorities in Districts 1 and 2. The Commission removed more population than necessary to reach equal population to make room to add majority white precincts to District 3. These changes follow the Commission's redistricting pattern of restricting the Black population in District 3 as evidenced in historical preclearance submissions from 1993, 2001, and 2013. Dr. Barber's and Mr. Stephenson's post-hoc justifications for the changes made to District 3 do not find support in the record. See Bethune-Hill, 580 U.S. at 189-90, 137 S.Ct. 788. Therefore, the Court finds that race predominated in the drawing of District 3.

District 4

The Commission followed suit in District 4. In 2013, the BVAP in District 4 was 29.5%. (Doc. 169-6, p. 45; Doc. 179-16, pp. 34).⁴⁸ In 2021, the Commission's revisions of the boundaries of District 4 reduced the district's BVAP to 25.7%. (Doc. 179-16, pp. 34–35). The redrawing of District 4 to decrease District 4's BVAP supports the inference that the Commission altered District 4 to maintain supermajority Black Districts 1 and 2 while limiting the Black population growth in District 4. To achieve this result, the Commission excised significantly majority Black areas from District 4 and moved those areas into District 1 without regard to municipal or precinct boundaries. The portion of the East Pinson Valley precinct moved out of District 4 is 86.18% Black; the portion of the precinct retained in District 4 is 56.91% Black. (Doc. 172, p. 36, tp. 36:1–25; Doc. 169-107, p. 26). The Commission also moved the Center Point Community Center precinct out of District 4 and into District 1. (Doc. 169-109, p. 9). This precinct is 80.86% Black. (Doc. 169-109, p. 9). These changes removed from District 4 all or part of the district's two most Black precincts. (See Doc. 179-16, p. 36). Dr. McCartan's analysis shows that the Commission placed the East Pinson Valley and Center Point Community Center areas in a district that had a greater Black population than 95% of the districts the areas were assigned in Dr. Barber's simulations. (Doc. 169-26, pp. 9, 11).

Though the Commission needed to remove only 7,167 people from District 4 to achieve the Commission's equal population goal, (Doc. 169-12, p. 7), the movement of population from District 4 to District 1 took approximately 9,300 individuals out of District 4, (see Doc. 169-109, p. 9; Doc. 172, p. 36, tp. 36:19–21). The removal of the Warrior City Hall precinct from District 4 took approximately 1,500 more individuals out of District 4. (Doc. 169-107, p. 41). To offset these changes, the Commission added approximately 4,800 people from District 5 to District 4, (Doc. 172, p. 276, tp. 276:17–18; Doc. 169-72, p. 2). Only 8.6% of these individuals are Black. (Doc. 172, p. 276, tp. 276:17–18; Doc. 169-72, p. 2).

As in District 3, the Commission had other options. For example, the Tarrant City Hall and Center Point Community Center precincts had similar populations, but the Tarrant City Hall precinct was only 50.72% Black. (See Doc. 169-107, p. 33; Doc. 179-13). Moving the Tarrant City Hall precinct into District 1 from District 4 instead of moving the Center Point Community Center precinct would have removed fewer Black individuals from District 4 and would have eliminated the split of Center Point in the 2021 plan. (See Doc. 169-107, p. 33; Doc. 169-108, p. 23; Doc. 179-13).

In sum, to facilitate the Black supermajorities in Districts 1 and 2, with no statistical evidence to the support the supermajorities, the Commission removed Black voters from District 4 with a corresponding boost to District 4's white majority. Though the Commission needed to remove voters from District 4 to achieve equal population, the Commission added majority white areas to the district. These changes replicate the Commission's historical redistricting pattern in District 4 as evidenced in the Commission's preclearance submissions in 1993, 2001, and 2013. Dr. Barber's and Mr. Stephenson's post-hoc justifications for the changes made to District 4 do not find support in the record. See Bethune-Hill, 580 U.S. at 189–90, 137 S.Ct. 788. Therefore, the Court finds that race predominated in the drawing of District 4.

District 5

In 2001, the BVAP in District 5 was 11.01%. (Doc. 169-6, p. 45). By 2020, the BVAP in District 5 increased to 14.1%. (Doc. 179-16, p. 37). Following the 2021 redistricting, the BVAP in District 5 declined slightly to 14%. (Doc. 179-16, p. 37). Overall, District 5's Black population increased from 6.3% in 1985 to 11.88% in 2001 to 14.15% following the 2021 redistricting. (Doc. 169-2, p. 23; Doc. 169-6, p. 45; Doc. 169-107, p. 18).

To maintain District 5's racial composition, the Commission removed from District 5 precincts adjacent to District 2 with comparably higher BVAPs than the other adjacent precincts in District 5. (See Doc. 179-16, p. 38). The Commission moved out of District 5 the district's only precinct with a BVAP above 50%, the Afton Lee Community Center. (Doc. 179-16, p. 38). The Commission moved the Afton Lee Community Center precinct to a district that was 95% more Black than the districts to which Dr. Barber's simulations assigned the precinct. (See Doc. 169-26, p. 11). Because the BVAP in District 5 already was very low at 14.1%, the Commission's decision to move the Oxmoor Valley Community Center precinct with its BVAP of 27.5% reduced the district's Black population further. (See Doc. 179-16, pp. 37, 38). The Oxmoor Valley Community Center precinct had the second-highest BVAP of any District 2-adjacent precinct in District 5. (See Doc. 179-16, p. 38).⁴⁹ In addition, the movement of the McElwain Baptist Church precinct, an 80.74% white and 8.56% Black precinct, (Doc. 169-107, p. 41), from District 5 to District 4 boosted District 4's white population without significantly altering District 5's supermajority white population.

In sum, the Commission removed from District 5 areas containing comparably higher BVAPs than other areas in District 5 and the only concentrated Black populations in District 5. The Commission's removal of a significantly majority white precinct from District 5 allowed the Commission to maintain District 5's racial composition while boosting District 4's white population. These changes follow the Commission's pattern of restricting the Black population in District 5 as evidenced in the Commission's preclearance submissions in 1993, 2001, and 2013. The record does not support Dr. Barber's and Mr. Stephenson's post-hoc justifications for the Commission's changes to District 5. See Bethune-Hill, 580 U.S. at 189–90, 137 S.Ct. 788. Therefore, the Court finds that race predominated in the drawing of District 5.⁵⁰

***

Because the plaintiffs have established that race predominated in the Commission's revisions to its five districts following the 2020 census, the Commission must prove that the 2021 plan “can overcome the daunting requirements of strict scrutiny” and that the Commission has “narrowly tailored” its use of race to advance a “compelling governmental interest.” Alexander, 602 U.S. at 11, 144 S.Ct. 1221 (internal quotation marks omitted). The Commission has not attempted to make this showing in this case. (See, e.g., Doc. 183, pp. 216–17). Therefore, the Court finds that the Commission's 2021 plan violates the Fourteenth Amendment.

CONCLUSION

Because the 2021 plan violates the Fourteenth Amendment's protection against racial gerrymandering, the Court permanently enjoins the Commission and its agents from using the 2021 plan in Jefferson County Commission elections. Within 30 days, the parties shall please file a joint report on the development of a remedial redistricting plan.

DONE and ORDERED this September 16, 2025.

FOOTNOTES

1.   Citations refer to the docket in case number 2:23-cv-00443-MHH, McClure et al. v. Jefferson County Commission.

2.   See also Jefferson County Commission, Jefferson Cnty. Ala., https://www.jccal.org/Default.asp?ID=963&pg=About+% 2D+Jefferson+County+Commission [https://perma.cc/EE9S-764N] (last visited Feb. 7, 2025). In 2023, Mr. Ammons resigned from the Commission. Jefferson County Election Commission Resolution Calling Special Election, Jefferson Cnty. Prob. Ct. (June 8, 2023), https://www.jccal.org/Default.asp?ID=993&pg=News&action=view&aid=365&title=Election+C ommission+Resolution+Calling+Special+Election+To+Fill+Jefferson+County+Commission+Di strict+Five+Vacancy [https://perma.cc/5NE5-23BB]. In a special election in July 2023, Mike Bolin was elected Commissioner for District 5. Biography – Commissioner Mike Bolin (District 5), Jefferson Cnty. Ala., https://www.jccal.org/Default.asp?ID=581&pg=District+5+% 2D+Mike+Bolin [https://perma.cc/9FDW-57KQ] (last visited Feb. 7, 2025). The Court takes judicial notice of this governmental website. Fed. R. Evid. 201(b)(2); see also R.S.B. Ventures, Inc. v. Fed. Deposit Ins. Corp., 514 Fed. Appx. 853, 856 n.2 (11th Cir. 2013) (taking judicial notice of information on FDIC website).

3.   The Taylor plaintiffs amended their complaint two weeks after they filed their lawsuit. (Doc. 176-1, pp. 1, 7). The amendment did not affect the plaintiffs’ substantive allegations.

4.   The parties described other instances of systemic de jure segregation and racial discrimination in Alabama and in Jefferson County. (Doc. 154-1, pp. 13–39, ¶¶ 20-107).

5.   In addition to the parties’ joint statement of facts, the Taylor plaintiffs and defendants separately submitted additional facts. (Doc. 176-2, pp. 61–73 (defendants’ additional facts); Doc. 176-2, pp. 74–78 (plaintiffs’ additional facts)). The defendants asserted, for example, that “Black persons ha[d] qualified as candidates for the Jefferson County Commission in five of the [preceding] eight elections to the Commission, and the defendants twice asserted that Black candidates “ha[d] a fair chance of being elected at large to the Jefferson County Commission.” (Doc. 176-2, p. 63, ¶¶ 1203, 1204, 1206). In contrast, the plaintiffs asserted that Black candidates “ha[d] almost no chance of being elected at large to the Jefferson County Commission. (Doc. 176-2, p. 77, ¶ 2203).

6.   Under Section 5 of the VRA, DOJ approval for revisions of district boundaries was called preclearance. “Under § 5, a covered jurisdiction [had to] seek preclearance either by filing a declaratory judgment action in the United States District Court for the District of Columbia, or by submitting the election law change to the United State [sic] Attorney General for administrative preclearance. The Voting Section of the Civil Rights Division of the DOJ [was] charged with administering preclearance on behalf of the Attorney General.” Clark v. Putnam Cnty., 293 F.3d 1261, 1264 n.11 (11th Cir. 2002). “Jurisdictions covered by § 5 include[d] those which in 1965 had substantial histories of intentional disenfranchisement of black voters.” Clark, 293 F.3d at 1264 n.11.

7.   Mr. Fairfax testified that map drawers created electoral districts with at least a 65% Black population “to ensure that [B]lack voters could elect candidates of choice.” (Doc. 172, p. 29, tp. 29:11-19). The 65% threshold derived from a theory “that there was a 5 percent decrease between the total population and the Voting Age Population, another 5 percent decrease between the Voting Age Population and the registered voters, and a third decrease between the registered voters and turnout between [B]lack and white” voters. (Doc. 172, p. 29, tp. 29:20-25). The Commission objected to Mr. Fairfax's testimony. (Doc. 172, p. 30, tp. 30). The Court overruled the objection. The national 65% historical standard is context; the Black populations of District 1 and of District 2 in 1985 is undisputed.

8.   The Birmingham Public Library provided the 1985 map in this split format. The library's map spread the map across two pages. The Court has merged the pages together for the purpose of this opinion; the map pieces do not fit perfectly.

9.   See Ala. Code § 11-2-1.1(a) (“Following the release of any federal decennial census, any county commission of this state which is at that time electing its members from single-member districts, pursuant to either state or local law or a court order, may, by resolution, alter the boundaries of its districts.”).

10.   In 2004, the Commission sought and received DOJ approval for adjustments made to the district lines drawn in 2001 in compliance with the terms of a settlement agreement resolving a lawsuit filed following the 2001 redistricting. (See Doc. 169-4).

11.   “Ideal population size of a district is the quotient of the population of a county divided by the number of its electoral districts.” Clark, 293 F.3d at 1263 n.6. “Reapportionment every ten years ensures that electoral districts contain approximately equal populations, thus ensuring one-person, one-vote, as the Constitution requires.” Clark, 293 F.3d at 1264 n.7 (citing Reynolds v. Sims, 377 U.S. 533, 583-84, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964)).

12.   The Court cannot analyze accurately the demographic changes in District 5 because of the condition of the scanned documents in Doc. 169-3. (See Doc. 169-3, pp. 109–10).

13.   In the interim between the 1993 and 1998 elections, the Alabama Legislature codified the Commission's single-member district structure. (Doc. 169-6, p. 8); see Ala. Code § 45-37-72.

14.   The maps are not part of the DOJ submission that appears in the record.

15.   On page 44 of the 2001 DOJ submission, the Commission misreported the Black population in District 1 as 73.45%. (See Doc. 169-6, p. 44). Dividing the District 1 Black population of 105,275 by the District 1 population of 134,968 yields the 78% Black population statistic reported elsewhere in the submission. (Doc. 169-6, pp. 11, 15, 44).

16.   The Commission provided information concerning demographic changes to Districts 3, 4, and 5 electronically; the data is not in the Commission's preclearance letter. (See Doc. 169-5, pp. 1081-82).

17.   Mr. Stephenson testified that then-District 1 Commissioner George Bowman voted against the plan that the Commission adopted; he had submitted an alternative plan for consideration. (Doc. 174, p. 221, tp. 714:10-18). The Commission's 2013 preclearance submission did not mention the substance of Commissioner Bowman's objections to the 2013 plan, nor did the submission mention the alternative plan that Commissioner Bowman submitted. (See Doc. 169-5). In their proposed findings of fact, the plaintiffs stated that Mr. Stephenson “acknowledged that Commissioner Bowman had introduced an alternative plan that Commissioner Bowman believed would be fairer to the community.” (Doc. 178, pp. 16-17, ¶ 54). Mr. Stephenson testified only that he remembered Commissioner Bowman introducing an alternative plan and that Commissioner Bowman “was not happy that day.” (Doc. 174, pp. 221, 224, tpp. 714:14-18, 717:14-17). Even after counsel for the McClure plaintiffs refreshed Mr. Stephenson's recollection of the meeting where Commissioner Bowman discussed his objections and his alternative plan, Mr. Stephenson did not remember “the overall thrust of” Commission Bowman's objections as Commissioner Bowman's belief that the plan enacted in 2013 “was not fair to the Black community.” (Doc. 174, pp. 222–24, tpp. 715:2-717:18). The plaintiffs did not offer other evidence concerning Commissioner Bowman's objections and alternative plan, so the Court finds only that Commissioner Bowman voted against the 2013 plan and submitted an alternative map for consideration.

18.   In Shelby County v. Holder, 570 U.S. 529, 133 S.Ct. 2612, 186 L.Ed.2d 651 (2013), the Court held that § 4(b) of the Voting Rights Act, the formula that determines which jurisdictions are covered by the § 5 preclearance requirement, is unconstitutional. See 570 U.S. at 550, 557, 133 S.Ct. 2612. Although the Court did not find the preclearance requirement itself unconstitutional, see 570 U.S. at 556, 133 S.Ct. 2612, the § 5 requirements are inoperable without a valid coverage formula to determine when the requirements apply. See Shelby Cnty., 570 U.S. at 587, 133 S.Ct. 2612 (Ginsberg, J., dissenting).

19.   The growth in Black populations in Districts 3, 4, and 5 tracks the overall growth in Black population in Jefferson County over the past several decades.

20.   The minutes indicate that Commissioner Tyson was absent for the vote, and the Commission excused Commissioner Ammons from voting. (Doc. 169-18, p. 17).

21.   Alabama Code § 11-3-1.1(c) requires county commissions to “advertise[ ] in a newspaper of general circulation in the county for at least two consecutive weeks the time and place of the meeting at which the [map] shall be considered” before adopting a redistricting plan. Ala. Code § 11-3-1.1(c).

22.   Subscription, Ala. Messenger, https://www.alabamamessenger.com/membership-account/membership-levels/ (last visited Feb. 10, 2025).

23.   Subscription, Ala. Messenger, https://www.alabamamessenger.com/membership-account/membership-levels/ (last visited Feb. 10, 2025).

24.   About Us, Ala. Messenger, https://www.alabamamessenger.com/about-us/ (last visited Feb. 10, 2025).

25.   About Us, Ala. Messenger, https://www.alabamamessenger.com/about-us/ (last visited Feb. 10, 2025); see, e.g., Jefferson County Official Voters List 2024, Ala. Messenger, https://www.alabamamessenger.com/jefferson-county-official-voters-list-2024/ (last visited Feb. 10, 2025).

26.   The record does not indicate how many individuals attended the November 4 meeting. David Russell from the Alabama Democratic Executive Committee spoke first. He stressed his desire for districts with equal population to ensure that all citizens “have the equal opportunity to cast their vote.” (Doc. 169-11, p. 21). He remarked that his organization “would love for the County Commission [to] just do the right thing.” (Doc. 169-11, p. 22).George McCall, the President of the Ensley, Alabama community and neighborhood, indicated that his group had held meetings on the draft plans, discussed the plans with Commissioners Scales and Tyson, and voted to ask the Commission to keep the Ensley community in District 2 per the first proposal because Ensley appreciated Commissioner Tyson's representation of the area. (Doc. 169-11, pp. 22–23).Dr. Tyree Anderson, the pastor of the First Baptist Church in Ensley, commented that he believed underpopulation in Districts 1 and 2 resulted in large part from economic injustice. (Doc. 169-11, p. 24). He advocated for the Commission to adopt the second proposal, reasoning that changing Ensley's district “might actually bring economic development into the community.” (Doc. 169-11, pp. 24–25). Ensley resident and President of the Ensley Business Alliance, Bryan K. Rice, emphasized his perception of the economic injustice impacting Ensley. (Doc. 169-11, pp. 26–28). Alice Westry, co-chair of the Community Affairs Committee, and Johnny Gunn, president of the Belview Heights Neighborhood Association and vice president of the Five Points West Community, indicated their support of the first plan. (Doc. 169-11, pp. 28–30). Mr. Gunn noted that under the first proposal, his community would remain in the same district. (Doc. 169-11, pp. 29–30).

27.   Dr. McCartan explained that all simulations analyses involve translating a plan into a number using a summary statistic to quantify that plan's properties. (Doc. 174, pp. 75, 107, tpp. 568:13–24, 600:8-12). Dr. McCartan testified that he has not used this specific cracking and packing score in his prior work, but his methodology is common among academics in the computational redistricting field and that he has used a similar methodology in prior peer reviewed work. (Doc. 174, pp. 75, 144, tpp. 568:13–24, 637:6–15). Dr. McCartan explained that he has averaged BVAP shares across a range of districts to create summary statistics for an article about census privacy protections. (Doc. 174, p. 75, tp. 568:13–24). That article was peer-reviewed and published in Science Advances. (Doc. 174, pp. 75–76, tpp. 568:13–569:2).Summary statistics like the combined packing-cracking score are context-dependent because they are used to answer a particular question. (Doc. 174, p. 144, tp. 637:16–23). Summary statistics are useful only in the context of an entire analysis, not in isolation or in other contexts. Dr. McCartan creates summary statistics to answer questions particular to a specific analysis. (See Doc. 174, p. 144, tp. 637:6–15).

28.   For instance, take a precinct X. In a hypothetical three-simulation analysis, assume precinct X was assigned to districts with BVAPs of 25%, 40%, and 62%. If Dr. McCartan's map contained precinct X, precinct X would be shaded in a dark blue-green color, reflecting its assignment to a district with, on average, a BVAP of 43.33%.

29.   Nearly 50% of Dr. McCartan's simulated district 2s would have a higher BVAP than District 2 has in the 2021 plan. (Doc. 169-26, p. 19).

30.   The data cited by the parties’ experts concerning the number of individuals moved between the districts differs slightly. For example, Dr. Barber reported that the Commission moved 12,944 people into District 1 and removed 89 people from District 1. (Doc. 179-16, pp. 11, 14, 15). Mr. Fairfax reported that the Commission added 13,073 people to District 1 and removed 149 people from District 1. (Doc. 169-107, p. 22). These slight differences do not impact the overall analysis of the Commission's changes to the Jefferson County Commission districts during the 2021 redistricting cycle.

31.   Dr. Liu refers to the illustrative plan he compared to the 2021 plan as the “plaintiffs’ plan.” (See, e.g., Doc. 173, p. 95, tp. 387:17; Doc. 169-22, p. 7). The Court understands Dr. Liu to refer to Mr. Cooper's illustrative plan B as the plaintiff's plan in his testimony, as the numbers Dr. Liu reports for each district's BVAP most closely match those in Mr. Cooper's analysis. (Compare Doc. 169-22, p. 7; Doc. 169-20, p. 8, with Doc. 169-35; Doc. 169-37; Doc. 169-39).

32.   In Shaw v. Reno, the Supreme Court put it this way: the Fourteenth Amendment's “central purpose is to prevent the States from purposely discriminating between individuals on the basis of race.” Shaw v. Reno (Shaw I), 509 U.S. 630, 642, 113 S.Ct. 2816, 125 L.Ed.2d 511 (1993) (citing Washington v. Davis, 426 U.S. 229, 239, 96 S.Ct. 2040, 48 L.Ed.2d 597 (1976)). The Equal Protection Clause prohibits race discrimination because “[c]lassifications of citizens solely on the basis of race ‘are by their very nature odious to a free people whose institutions are founded upon the doctrine of equality.’ ” Shaw I, 509 U.S. at 643, 113 S.Ct. 2816 (quoting Hirabayashi v. United States, 320 U.S. 81, 100, 63 S.Ct. 1375, 87 L.Ed. 1774 (1943)). The classifications “pose the risk of lasting harm to our society,” and “reinforce the belief, held by too many for too much of our history, that individuals should be judged by the color of their skin.” Shaw I, 509 U.S. at 657, 113 S.Ct. 2816.

33.   The Supreme Court first recognized a claim of racial gerrymandering in Shaw I.

34.   Per Abbott, “past discrimination cannot, in the manner of original sin, condemn governmental action that is not itself unlawful.” 585 U.S. at 603, 138 S.Ct. 2305. But under Abbott, the Section 5 preclearance evidence is relevant circumstantial evidence that the Court may consider with all of the relevant evidence to determine whether the plaintiffs may overcome the presumption of legislative good faith that attends the 2021 plan. 585 U.S. at 605, 607, 138 S.Ct. 2305.Because Abbott is a Section 2 VRA case, not a racial gerrymandering case, the Supreme Court's analysis in Abbott focuses on discriminatory intent rather than predominant purpose. The general principles in Abbott concerning historical evidence guide the Court's analysis of the Section 5 preclearance evidence in the record in this case.

35.   In Abbott, the Supreme Court answered this question in the negative because the 2011 Texas Legislature had enacted district boundaries that a Texas federal court developed “pursuant to instructions” from the Supreme Court “ ‘not to incorporate any legal defects.’ ” Abbott, 585 U.S. at 604, 138 S.Ct. 2305 (quoting Perry v. Perez, 564 U.S. 388, 394 (2012)). On this record, the Supreme Court easily concluded that the 2013 Texas Legislature could not have carried forward discriminatory intent from the 2011 legislation. This case is factually different.

36.   The district court in League of Women Voters surveyed Florida's history of racial discrimination, “beginning immediately after the Civil War” and continuing through “acts of ‘terrorism’ and ‘racial violence’ that occurred during the early and mid-1900s.” LOWV, 32 F.4th at 1373 (quotations and citations omitted). The Eleventh Circuit found that “the district court's inquiry d[id] not seem appropriately “focus[ed]” or “[ ]limited.” LOWV, 32 F.4th at 1373 (brackets added and in LOWV). A similar historical survey appears in the parties’ joint statement of facts in the Taylor case. (Doc. 172-2, pp. 13-39). The Court does not rely on that survey of racial discrimination in Alabama in evaluating the predominant purpose of the 2021 Commission.

37.   At times in this litigation, the Commission has argued that the consent decree was not related to race. This argument ignores the precedent in Yeldell.

38.   In District 3, both the Black population and the Black population as a percentage of the district's population fell. (See Doc. 169-3, pp. 109–10). District 3's Black population decreased by over 2,500 individuals, falling from 22.89% to 21.22% of the district's population. (See Doc. 169-3, pp. 109–10). District 3's white population grew by 235 individuals, but District 3's percentage of white population increased from 77.11% to 78.25%. (See Doc. 169-3, pp. 109–10). The demographic changes in District 4 did not follow this trend, as the district's Black population grew by roughly 1,200 individuals, with a corresponding increase in the district's total population by about 1%. (See Doc. 169-3, pp. 109–10). At the same time, District 4 shed nearly 13,000 white individuals. (See Doc. 169-3, pp. 109–10). The Court cannot analyze accurately the demographic changes in District 5 given the condition of the scanned documents in Doc. 169-3. (See Doc. 169-3, pp. 109–10).

39.   In District 3, on the other hand, the Commission removed more than 15,000 Black individuals and added roughly 3,500 white individuals, lowering to 17.14% the percentage of the population that Black individuals comprised. (See Doc. 169-6, pp. 43–44). Despite Districts 4 and 5 losing 9,062 and 10,480 white individuals and 2,624 and 973 Black individuals, respectively, the percentage of Black population in each district remained roughly equal to that at the time of the 2000 census—the percentage fell from 16.42% to 15.29% in District 4 and increased from 11.48% to 11.88% in District 5. (See Doc. 169-6, pp. 43, 45).

40.   For example, as in 1990, 2000, 2001, and 2013, by the 2020 census, Jefferson County had grown in overall population, but District 1 had lost population. (Doc. 169-12, pp. 3–7). With only 122,689 individuals in 2020, District 1's population reflected a -9.1% variance from the Commission's 134,944 population goal in each district. (Doc. 169-12, pp. 5–6). The Commission needed to add 12,255 people to District 1. (Doc. 169-12, p. 7). The Commission added five times more Black population than white population to District 1 in the 2021 plan. (Doc. 172, p. 32, tp. 32:11–21). The 2021 plan moved approximately 13,000 into District 1. (Doc. 169-107, p. 22). Of those people, 77.9% were Black and 15.8% were white. (See Doc. 172, p. 32, tp. 32:13–21). The 2021 plan moved 9,422 persons from District 4 to District 1, 82.6% of whom were Black. (Doc. 172, p. 273, tp. 273:20–21; Doc. 169-72).

41.   In Yeldell, the Eleventh Circuit did not view these districts as a mere by-product of a districting procedure that began with the county's largest city. Instead, the Eleventh Circuit recognized that the Commission designed a “form of government to ensure greater minority representation on the Commission.” Yeldell, 956 F.2d at 1058.

42.   Two of the Commissioners—Commissioner Stephens and Commissioner Knight—participated in the 2013 redistricting process. (See Doc. 169-11, p. 37, tp. 36:16–25).

43.   Dr. Barber reported that any choice of precincts in the East Pinson Valley area would have moved substantially Black populations from District 4 to District 1, (Doc. 179-17, pp. 16–17), but Dr. Barber did not explain why the Commission had to tap East Pinson Valley rather than other areas from District 3 to avoid creating a bigger hole in District 4.The Commission also split the Dolomite West Field Community Center precinct in the 2021 plan, shifting voters from District 3 to District 1. (Doc. 169-109, p. 9). Of the 1,284 voters that the Commission added from the old Dolomite West precinct to District 1, 86.60% are Black and 8.18% are white. (Doc. 169-109, p. 9).

44.   Citing Alexander, the Commission argues that the core retention score in plan D does not match that of the 2021 plan. (Doc. 177, p. 148, ¶ 489) (citation omitted). True, the Supreme Court faulted an expert in Alexander for not accounting for core retention in his simulations to the same degree the legislature had. See Alexander, 602 U.S. at 26–27, 144 S.Ct. 1221. But as the Court has explained, core retention in this case does not represent a race-neutral redistricting criterion given the history of the 2013 map. If the Commission sought to maintain the 2013 district cores to a high degree, then the Commission effectively sought to reenact the 2013 map, which the Commission enacted with a race-based purpose. The Commission cannot immunize itself from a racial gerrymandering challenge by repackaging its race-based decision-making under the guise of a race-neutral redistricting criterion. On another record, a record that resembles the record in Alexander, the difference between core retention rates in Mr. Cooper's plan D and the 2021 plan might mean that the Court could not “rule out core retention as another plausible explanation for the difference between the 2021 plan and” plan D. See Alexander, 602 U.S. at 27, 144 S.Ct. 1221. The record here is different from the record in Alexander in significant respects.Mr. Cooper mistakenly drew the incumbents of both District 1 and District 4 into District 1 in plan D, but he testified that he could correct this mistake with little impact on core retention. (Doc. 172, pp. 283, 285, 286, tpp. 283:2–10, 285:1–9, 286:4–10). Plan D exceeds the 1% population constraint present in the 2021 plan, though Mr. Cooper indicated that this fact had a “de minimis” impact on his analysis. (Doc. 173, pp. 13, 42–43, tpp. 305:7–12, 334:24–335:4; Doc. 169-77). While relevant, the Court gives Mr. Cooper's plan D less weight than other substantial evidence in the record of racial predominance.

45.   Roughly 10% of the race-neutral maps in Dr. McCartan's strong core retention simulation had a higher BVAP than the 2021 plan's District 1, not a statistically significant result, (Doc. 169-26, p. 19; Doc. 174, p. 148, tp. 641:20–22), but when analyzing Dr. Barber's simulation, Dr. McCartan found that District 1's BVAP was more extreme than 95.1% of the corresponding districts in Dr. Barber's simulations—a statistically significant result. (Doc. 169-26, p. 9; Doc. 174, p. 65, tp. 558:11–18; Doc. 175, p. 127, tp. 863:13–16).

46.   As another example, under the 2013 plan, the Ross Bridge precinct was 23.38% Black. The Commission split this precinct in 2021 and brought a 52.8% Black portion into District 2 and left a 13.21% Black portion in District 3. (Doc. 169-107, pp. 20, 30, 31).

47.   District 3 had a Black population of 22.2% in 1985 and 21.22% in 1993. (Doc. 169-2, p. 21; Doc. 169-3, p. 110). Though District 3's Black population declined to 17.14% following the 2001 redistricting, the district's Black population grew to 27.29% following the 2021 redistricting. (Doc. 169-6, p. 45; Doc. 169-107, p. 18).

48.   District 4 grew from 5% Black in 1985 to 6.05% Black in 1993, 15.29% Black in 2001, and 28.45% following the 2021 redistricting. (Doc. 169-2, p. 22; Doc. 169-3, p. 110; Doc. 169-6, p. 45; Doc. 169-107, p. 18).

49.   Dr. Barber suggested moving the Oxmoor Valley Community Center precinct into District 2 was not evidence of racial predominance because this change reunited a precinct split in the 2013 plan, (see Doc. 175, pp. 48–49, tpp. 784:17–785:5), but Dr. Barber's theory conflicts with Mr. Stephenson's testimony that the commissioners did not focus on municipal splits when redistricting because municipal annexations were common, (Doc. 174, p. 212, tp. 705:12–16).

50.   The results in this case might be different if there was evidence that the Commission had information before it that indicates that Black voters needed a Black VAP higher than 64% to select the candidates of their choice. See Cooper, 581 U.S. at 293, 137 S.Ct. 1455. The Commission has offered no such evidence.

MADELINE HUGHES HAIKALA, UNITED STATES DISTRICT JUDGE