Patino et al v. City of Pasadena et al, No. 4:2014cv03241 - Document 151 (S.D. Tex. 2017)

Court Description: MEMORANDUM AND OPINION entered: The City of Pasadena, Texas is ordered to conduct its May 2017 City Council elections using the map and plan in force in May 2013. The court retains jurisdiction under 52 U.S.C. § 10302(c) to review before enfor cement any voting qualification, prerequisite, practice, or procedure different from the map and plan in use in May 2013. Any new City Council voting map, plan, or procedure may be enforced before court review only if it has first been submitted to the United States Attorney General and the Attorney General has not interposed an objection within 60 days after submission. The parties must submit a proposed form of injunction and a proposed period for continued jurisdiction no later than January 13, 2017.(Signed by Chief Judge Lee H Rosenthal) Parties notified.(leddins, 4)
Download PDF
Patino et al v. City of Pasadena et al Doc. 151 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 1 of 113 United States District Court Southern District of Texas ENTERED IN THE UNITED STATES DISTRICT COURT FOR THE SOUTHERN DISTRICT OF TEXAS HOUSTON DIVISION ALBERTO PATINO, et al., Plaintiffs, VS. CITY OF PASADENA, Defendant. § § § § § § § § § January 06, 2017 David J. Bradley, Clerk CIVIL ACTION NO. H-14-3241 MEMORANDUM AND OPINION SETTING OUT FINDINGS OF FACT AND CONCLUSIONS OF LAW TABLE OF CONTENTS Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 I. The Law Governing the Court’s Inquiry, Findings, and Conclusions. . . . . . . . . . . . . . . . . 5 A. Section 2 of the Voting Rights Act and the Fifteenth Amendment. . . . . . . . . . . . . 6 B. The Fourteenth Amendment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 II. Findings of Fact. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 A. Procedural Background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 B. The Record Evidence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1. Fact Witnesses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 2. Expert Witnesses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 C. Pasadena’s Election Maps and Plans. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 D. Pasadena’s History.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 1. Texas Laws Affecting Pasadena. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 2. Racially Discriminatory Ordinances and Official Activities in Pasadena . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 3. Recent Private Racial Discrimination in Pasadena. . . . . . . . . . . . . . . . . . 25 E. Pasadena’s Demographics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 1. The Latino Population in Pasadena. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 2. Latino Citizen Voting-Age Population in Pasadena. . . . . . . . . . . . . . . . . 29 3. The Eight Single-Member District Plan.. . . . . . . . . . . . . . . . . . . . . . . . . . 33 4. The Six Single-Member District and Two At-Large Plan. . . . . . . . . . . . . 34 5. Racial Cohesion in Pasadena City Council Elections. . . . . . . . . . . . . . . . 34 a. The 2015 At-Large Race for Place G.. . . . . . . . . . . . . . . . . . . . . . 35 b. The 2015 At-Large Race for Place H.. . . . . . . . . . . . . . . . . . . . . . 36 1 Dockets.Justia.com Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 2 of 113 c. d. e. F. G. H. III. The November 2013 Special Election on Proposition 1. . . . . . . . 37 The 2015 District-Level Elections. . . . . . . . . . . . . . . . . . . . . . . . . 37 Exogenous Elections. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 (1). Countywide Elections. . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 (2). Primary Elections.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 f. Findings on Racially Cohesive Voting in Pasadena Elections.. . . 41 6. Socioeconomic Conditions and the Effects on Pasadena Voting. . . . . . . 41 a. Income, Education, and Employment. . . . . . . . . . . . . . . . . . . . . . 41 b. Living Conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 c. The Effects on Voting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 d. Findings on the Effect of Socioeconomic Conditions on Latino Voting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 The Circumstances Surrounding the Change to Pasadena’s City Council Election Map and Plan. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 1. Recent § 5 Preclearance Objections to Converting to At-Large Voting in Other Texas Municipalities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 2. The Shelby County Decision. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 3. The Shift from a Special Election to Approve a Council Bond Proposal to Amending the City Charter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 4. Council Consideration of the Charter Amendment. . . . . . . . . . . . . . . . . . 52 5. The Campaign to Enact the Charter Amendment to Change to a Six Single-Member, Two At-Large District Map and Plan for City Council Elections.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 a. Improper Use of City Resources to Mobilize Anglo Voters in South Pasadena. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 b. The Use of Race and Party as Proxies for Each Other.. . . . . . . . . 58 c. The Result. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 6. Council Approval of the 6–2 Map and Plan. . . . . . . . . . . . . . . . . . . . . . . 59 The Impact of Pasadena’s New Election Map and Plan. . . . . . . . . . . . . . . . . . . . 62 1. The Previous 8–0 Map and Plan. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 2. The 2015 Election under the Current 6–2 Map and Plan.. . . . . . . . . . . . . 64 Summary on Findings of Fact. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Conclusions of Law on § 2 of the Voting Rights Act. . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 A. Gingles Step One. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 1. The First Two Conditions: The Minority is Sufficiently Numerous and Geographically Compact. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 2. The Third Condition: The Majority Votes as Bloc Sufficiently to Defeat the Minority’s Preferences. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 B. Gingles Step Two.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 1. History of Discrimination. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 2. Racially Polarized Voting and Dilutive Measures. . . . . . . . . . . . . . . . . . . 77 3. Racial Appeals in Political Campaigns. . . . . . . . . . . . . . . . . . . . . . . . . . . 78 4. Successful Elections of Minority Candidates. . . . . . . . . . . . . . . . . . . . . . 79 2 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 3 of 113 C. 5. Officials’ Responsiveness to Minority Concerns.. . . . . . . . . . . . . . . . . . . 80 6. Tenuousness of the Policy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 7. Proportionality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 8. Conclusion of Law on the Totality of the Circumstances. . . . . . . . . . . . . 85 Findings and Conclusions on Intentional Discrimination. . . . . . . . . . . . . . . . . . . 85 1. Guidance from the Fifth Circuit: Veasey v. Abbott . . . . . . . . . . . . . . . . . 86 2. The Arlington Heights Factors.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 a. The Historical Background of the Decision.. . . . . . . . . . . . . . . . . 90 b. The Sequence of Events Leading Up to the Decision; Legislative History. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 c. Departures, Both from the Normal Procedural Sequence and Substantive; Legislative History and Preenactment Statements by Proponents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 d. Meeting the Burdens: The Plaintiffs’ Showing that Racial Discrimination was a Substantial Factor in Enacting the New Electoral Map and the Defendants’ Failure to Demonstrate that the Law Would Have Been Enacted Without this Factor. . . . . . . . . . 96 3. Conclusion: The City Intended to Dilute Latino Voting Strength. . . . . 103 IV. Remedy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 V. Conclusion and Order. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 Appendix A: Demonstrative Maps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Appendix B: Time Line. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Introduction This suit is one of many filed over the years to protect minority voting rights by ensuring “equal opportunity to participate in the political process.” Thornburg v. Gingles, 478 U.S. 30, 44 (1986) (internal quotation and citation omitted). This suit is one of the first involving redistricting done shortly after, and because, the Supreme Court decided Shelby County, Alabama v. Holder, 133 S.Ct. 2612, 2631 (2013), which removed the federal Department of Justice preclearance requirement under § 5 of the Voting Rights Act. The plaintiffs are Latinos in Pasadena, Texas who are citizens of voting age. They ask this court to find that Pasadena’s 2014 change from eight single-member 3 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 4 of 113 districts for electing City Council members to six single-member districts and two at-large districts dilutes Latino voting strength and violates § 2 of the Voting Rights Act and the Fourteenth and Fifteenth Amendments of the United States Constitution. The plaintiffs ask this court for a declaratory judgment finding Latino vote dilution, an injunction against using the mixed singlemember and at-large map and plan, an order that the 2017 City Council elections be held under the eight single-member map and plan in place in 2013, and an order that Pasadena submit to preclearance by the Department of Justice before making future changes to its voting map or plan for electing City Council members. The defendant, the City of Pasadena, opposes the requests. The City asserts that in recommending the redistricting, Pasadena’s Mayor and City Council intended to promote two legitimate goals unrelated to Latino vote dilution: first, the nonpartisan goal of making the Council more broadly representative and responsive to the concerns of all Pasadena residents; and second, the partisan goal of enhancing Republican, rather than Anglo, votes. The City argues that the change from a single-member to a mixed map and plan did not cause the Latino-vote dilution necessary to find a § 2 violation; and, if dilution resulted, that the City lacked the intent necessary to find a constitutional violation. This suit has received the careful court review the subject matter commands. After discovery and rulings on motions to dismiss and for summary judgment, the court held a seven-day bench trial, at which 16 witnesses testified and the court admitted 468 exhibits into evidence. The parties presented closing arguments at a three-hour hearing on December 2, 2016. After carefully considering the pleadings, the evidence, the arguments of counsel, and the law that applies, the court issues this Memorandum and Opinion setting out its Findings of Fact and Conclusions of Law under 4 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 5 of 113 Federal Rule of Civil Procedure 52(a).1 For the reasons stated in detail below, the court finds and concludes that Pasadena’s 2014 change from an eight single-member district map and plan to a six single-member district and two at-large position map and plan for electing its City Council dilutes the votes of its Latino citizens, in violation of § 2 of the Voting Rights Act. The court also finds and concludes that the change to the mixed map and plan was intended to dilute those votes because they were cast by Latino voters, in violation of the Fourteenth Amendment. Pasadena is enjoined from using the 2014 mixed map and plan in the 2017 City Council elections or subsequent elections. Instead, Pasadena must conduct the 2017 City Council election using the eight single-member district map it used in the May 2013 City Council elections. Pasadena must also submit to federal Department of Justice preclearance before implementing future redistricting changes. The reasons for these rulings are explained in detail below. I. The Law Governing the Court’s Inquiry, Findings, and Conclusions Although great progress has been made, “voting discrimination still exists; no one doubts that,” and § 2 of the Voting Rights Act remains a crucial “permanent, nationwide ban,” Shelby County, 133 S.Ct. at 2619, on “even the most subtle forms of discrimination,” Chisom v. Roemer, 501 U.S. 380, 406 (1991) (Scalia, J., dissenting). Federal courts have a vital role in protecting the right “to participate equally in the political process.” Gingles, 478 U.S. at 80. In requiring federal 1 “[T]he resolution of a voting dilution claim requires close analysis of unusually complex factual patterns, and because the decision of such a case has the potential for serious interference with state functions, . . . district courts [must] explain with particularity their reasoning and the subsidiary factual conclusions underlying their reasoning.” Westwego Citizens for Better Gov’t v. City of Westwego, 872 F.2d 1201, 1203 (5th Cir. 1989) (quoting Velasquez v. City of Abilene, 725 F.2d 1017, 1020 (5th Cir. 1984)). Any findings of fact that are also, or only, conclusions of law are so deemed. Any conclusions of law that are also, or only, findings of fact are so deemed. 5 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 6 of 113 courts to consider “the totality of circumstances,” 52 U.S.C. § 10301(b), Congress has made clear that, again in the Court’s words, “whether the political processes are equally open depends upon a searching practical evaluation of the past and present reality and on a functional view of the political process.” Gingles, 478 U.S. at 45 (internal quotations and citation omitted). At the same time, federal courts are reluctant to interfere with legislative decisions, especially when they are decisions by state or local legislative bodies, and when the decisions concern issues as sensitive as those regarding who votes, how they vote, and what districts they vote in. See Wise v. Lipscomb, 437 U.S. 535, 540 (1978). A. Section 2 of the Voting Rights Act and the Fifteenth Amendment Congress passed the Voting Rights Act to enforce the Fifteenth Amendment and prevent “an inequality in the opportunities enjoyed by [racial minority and majority] voters to elect their preferred representatives.” Gingles, 478 U.S. at 47. Although the Supreme Court held in 1980 that a § 2 plaintiff had to prove discriminatory intent, City of Mobile, Alabama v. Bolden, 446 U.S. 55 (1980), Congress amended the Voting Rights Act in 1982 so that a § 2 plaintiff need only show that a particular voting practice produces a discriminatory result. 52 U.S.C. § 10301 (formerly 42 U.S.C. § 1973(a)); see Gingles, 478 U.S. at 35 (“Congress substantially revised § 2 to make clear that a violation could be proved by showing discriminatory effect alone and to establish as the relevant legal standard the ‘results test,’ applied by this Court in [White v. Regester, 412 U.S. 755 (1973)], and by other federal courts before Bolden.”). The 1982 amendment created a “results-based” test to analyze vote-dilution claims. S. Rep. No. 97-417, at 40 (1982), reprinted in 1982 U.S.C.C.A.N. at 218 (“S. Rep.”). In Gingles, the Supreme Court established a two-step analysis for vote-dilution claims. 478 6 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 7 of 113 U.S. at 50–51. The minority group must first demonstrate: (1) “that it is sufficiently large and geographically compact to constitute a majority in a single-member district;” (2) “that it is politically cohesive;” and (3) “that the white majority votes sufficiently as a bloc to enable it—in the absence of special circumstances . . . —usually to defeat the minority’s preferred candidate.” Id. The court’s task is then to consider the “totality of the circumstances and to determine, based upon a searching practical evaluation of the past and present reality whether the political process is equally open to minority voters.” Id. at 79 (citations and internal quotation marks omitted). That analysis is informed by the Senate Judiciary Committee’s Report accompanying the 1982 amendments to the Voting Rights Act, which sets out a nonexhaustive list of factors bearing on whether the challenged practice impermissibly impairs a minority group’s ability to elect their preferred representatives. These factors include: (1) the extent of any history of official discrimination in the state or political subdivision that touched the right of the members of a minority group to register, to vote, or otherwise to participate in the democratic process; (2) the extent to which voting in the elections of the state or political subdivision is racially polarized; (3) the extent to which the state or political subdivision has used unusually large election districts, majority vote requirements, anti-single shot provisions, or other voting practices or procedures that may enhance the opportunity for discrimination against the minority group; (4) if there is a candidate slating process, whether the members of the minority group have been denied access to that process; (5) the extent to which members of the minority group in the state or political subdivision bear the effects of discrimination in such areas as education, employment and health, which hinder their ability to participate effectively in the political process; (6) whether political campaigns have been characterized by overt or subtle racial appeals; 7 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 8 of 113 (7) the extent to which members of the minority group have been elected to public office in the jurisdiction. S. Rep. at 28–29. Additional factors are “whether there is a significant lack of responsiveness on the part of elected officials to the particularized needs” of the minority group and “whether the policy underlying the . . . use of such voting qualification, prerequisite to voting, or standard, practice or procedure is tenuous.” S. Rep. at 29. A court must also consider whether the number of “majority-minority districts,” that is, districts in which a racial minority in a jurisdiction makes up the majority population of a district within that jurisdiction, “[is] in substantial proportion to the minority’s share of voting-age population.” Johnson v. De Grandy, 512 U.S. 997, 1013 (1994). There is no requirement that every factor be met, that “any particular number of factors be proved, or that a majority of them point one way or the other.” S. Rep. at 29. “The courts ordinarily have not used these factors . . . as a mechanical ‘point counting’ device . . . . Rather, the provision requires the court’s overall judgment, based on the totality of circumstances and guided by those relevant factors in the particular case, of whether the voting strength of minority voters is . . . ‘minimized or canceled out.’” Id. at 29 n.118. The totality-of-the-circumstances analysis is “local in nature.” League of Women Voters of North Carolina v. North Carolina, 769 F.3d 224, 243 (4th Cir. 2014). A court must determine “upon a searching practical evaluation of the past and present reality whether the political process is equally open to minority voters.” Gingles, 478 U.S. at 79 (citations and internal quotation marks omitted). “This determination is peculiarly dependent upon the facts of each case, and requires an intensely local appraisal of the design and impact of the contested electoral mechanisms.” Id. (citation and quotation marks omitted). The factual finding of vote denial or abridgement rests on 8 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 9 of 113 the court’s “particular familiarity with the indigenous political reality. . . .” Id. B. The Fourteenth Amendment “In decision after decision, [the Supreme] Court has made clear that a citizen has a constitutionally protected right to participate in elections on an equal basis with other citizens in the jurisdiction.” Dunn v. Blumstein, 405 U.S. 330, 336 (1972) (collecting cases). “[A] vote dilution claim alleges that the [City] has enacted a particular voting scheme as a purposeful device to minimize or cancel out the voting potential of racial or ethnic minorities, an action disadvantaging voters of a particular race” in violation of the Equal Protection Clause. Miller v. Johnson, 515 U.S. 900, 911 (1995) (internal citation and quotation marks omitted). At-large, “multimember districts violate the Fourteenth Amendment if ‘conceived or operated as purposeful devices to further racial discrimination’ by minimizing, cancelling out or diluting the voting strength of racial elements in the voting population.” Rogers v. Lodge, 458 U.S. 613, 617 (1982) (quoting Whitcomb v. Chavis, 403 U.S. 124, 149 (1858)). Fourteenth Amendment “[c]ases charging that multimember districts unconstitutionally dilute the voting strength of racial minorities are thus subject to the standard of proof generally applicable to Equal Protection Clause cases.” Id. (citing Village of Arlington Heights v. Metro. Hous. Dev. Corp., 429 U.S. 252 (1977); Washington v. Davis, 426 U.S. 229 (1976)). “Proof of racially discriminatory intent or purpose is required to show a violation of the Equal Protection Clause.” Arlington Heights, 429 U.S. at 265. But “[r]acial discrimination need only be one purpose, and not even a primary purpose,” of an official action for a violation to occur. United States v. Brown, 561 F.3d 420, 433 (5th Cir. 2009) (citation omitted). A court must make a “sensitive inquiry into such circumstantial and direct evidence of intent as may be available.” Arlington Heights, 429 U.S. at 266. 9 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 10 of 113 In Arlington Heights, the Supreme Court set out five factors to determine whether a decision affecting minority voting strength was made with a discriminatory purpose. The list is not exhaustive. “Those factors include: (1) the historical background of the decision, (2) the specific sequence of events leading up to the decision, (3) departures from the normal procedural sequence, (4) substantive departures, and (5) legislative history, especially where there are contemporary statements by members of the decision-making body.” Overton v. City of Austin, 871 F.2d 529, 540 (5th Cir. 1989) (citing Arlington Heights, 429 U.S. at 267–68). “Legislators’ awareness of a disparate impact on a protected group is not enough: the law must be passed because of that disparate impact.” Veasey v. Abbott, 830 F.3d 216, 231 (5th Cir. 2016) (en banc) (citing Pers. Adm’r of Mass. v. Feeney, 442 U.S. 256, 279 (1979)). The challengers bear the burden to show that racial discrimination was a “‘substantial’ or ‘motivating’ factor behind enactment of the law”; if they meet that burden, “the burden shifts to the law’s defenders to demonstrate that the law would have been enacted without this factor.” Hunter v. Underwood, 471 U.S. 222, 228 (1985) (citation omitted). II. Findings of Fact A. Procedural Background After extensive discovery, the City moved for summary judgment in July 2016. (Docket Entry No. 71). The court denied the motion because the plaintiffs satisfied the first step of showing undisputed facts in the record that, as a matter of law, could prove minority vote dilution under Thornburg v. Gingles, 478 U.S. at 47. (Docket Entry No. 87). The court held a seven-day bench trial and heard five expert witnesses and eleven other witnesses testify. The court admitted hundreds of pages of documents, including maps showing before- and after-redistricting lines. (Docket Entry Nos. 109–114, 116). 10 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 11 of 113 In response to the plaintiffs’ argument that the evidence proves both Latino vote dilution and the intent to achieve this result, the City argues that using at-large districts is a legitimate government objective and that the change to the mixed single-member and at-large map and plan does not as a matter of fact or law dilute Latino voting power in Pasadena. (Docket Entry No. 148). The City argues that because there is no vote dilution, there cannot be intentional discrimination, or, alternatively, that the evidence does not show that Pasadena officials acted with racially discriminatory intent. (Id.). Instead, the City argues that at most, Pasadena officials intended to dilute Democratic votes or depress Democratic turnout in a partisan move to bolster Republican support in Pasadena. The City insists that the evidence of intentional partisan discrimination is not evidence of intentional racial discrimination. The court considers both sides’ arguments based on the large record; the applicable law; counsels’ arguments presented at lengthy hearings on the summary judgment motions and at the conclusion of trial on December 2, 2016; and the trial testimony and documents. B. The Record Evidence The parties submitted 468 exhibits into evidence; 16 witnesses testified, and many more were deposed before trial. The parties agree that two maps, Plaintiffs’ Exhibits 152-1 and 252-1, accurately represent the change in Pasadena’s voting map and plan from eight single-member districts (the 8–0 map and plan) to six single-member districts and two at-large places (the 6–2 map and plan). Those maps are reproduced in Appendix A. The parties rely on the same sources for Pasadena’s demographic data and, with the exception of the current number for the citizen votingage Latino population of Pasadena, they arrive at the same population figures, with the inevitable minor variances. (Docket Entry No. 94 ¶ 19). The parties agree on the basic chronology of the 11 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 12 of 113 change to the 6–2 map and plan. The parties sharply disagree about the legal significance of the facts, events, and statements the record establishes. The court reviews the factual record in light of the appropriate legal framework to resolve the disagreements. 1. Fact Witnesses The fact witnesses testified about elections in Pasadena and the events leading up to and following the change to the 6–2 voting map and plan. The witnesses and their testimony are briefly summarized below. • Mayor Johnny Isbell. Mayor Isbell has been involved in Pasadena politics for thirty-four years, serving a number of terms as a City Council member. He was elected mayor in 2009 and reelected in 2013 for another four-year term. Mayor Isbell is approaching his term limit and cannot run again for mayor in 2017. Mayor Isbell proposed the change to the 6–2 map and plan for electing City Council members. In the summer of 2013, Mayor Isbell commissioned the law firm of Bickerstaff, Heath, Delgado and Acosta, LLP to prepare and review three redistricting maps that mixed single-member districts with seats elected at-large. Def. Exs. 13–14; Bench Trial Tr. 3:19. • Council member Ornaldo Ybarra. Mr. Ybarra was born and raised in Pasadena and attended college there. He is currently a sergeant in the Pearland Police Department. Other than his four years of military service, Mr. Ybarra has lived in the same Pasadena district he now represents on the Council. In 2009, Mr. Ybarra became the first Latino in nearly two decades elected to the City Council. He has been reelected in both the old and the new District A. He, too, is at the end of his term and cannot run again for District A’s City Council seat in 2017. As a Council member, Mr. Ybarra consistently voted against the Mayor’s positions 12 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 13 of 113 on resource-allocation issues that divided the Council. These divisive issues typically involved deciding what resources would go to predominately Latino, older, and less prosperous North Pasadena and what resources would go to the predominately Anglo, newer, and wealthier South Pasadena. Mr. Ybarra consistently voted to send more resources to the north side, where the infrastructure is in worse shape and the amenities are significantly fewer, than on the south side. When these resource-allocation issues divided the Council, the Mayor and his allies would generally cast votes that Mr. Ybarra opposed. Mr. Ybarra also voted against the Mayor’s proposal to change to the 6–2 map and plan for City Council elections. • Council member Pat Van Houte. Ms. Van Houte moved to Pasadena in 1980 and worked for the Texas Workforce Commission investigating unemployment claims. In 2009, Ms. Van Houte was elected to represent District D under the 8–0 single-member district map and plan. She has been continuously reelected to the City Council. Most recently, in 2015, she was elected to represent the new at-large Place G. Ms. Van Houte is not Latino but was the Latino-preferred candidate in the 2015 at-large election in which she defeated another Anglo candidate by 143 votes. As a Council member, Ms. Van Houte consistently voted against the Mayor’s positions on resource-allocation issues. She also voted against the Mayor’s proposal to change to the 6–2 map and plan for City Council elections. • Council member Cody Ray Wheeler. Mr. Wheeler attended elementary school in Pasadena and returned to the City during college after four years in the Marine Corps. He teaches kindergarten in the Aldine Independent School District. In 2013, Mr. Wheeler was elected to represent District E under the 8–0 single-member district map and plan. In 2015, he was 13 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 14 of 113 reelected to represent the new District D under the 6–2 map and plan, a district with substantially similar boundaries as the former District E. Mr. Wheeler’s mother is Latino and his father is half-Latino. Mr. Wheeler identifies as Latino. Bench Trial Tr. 5:96. As a Council member, Mr. Wheeler consistently voted against the Mayor’s positions on resourceallocation issues. He also voted against the Mayor’s proposal to change to the 6–2 map and plan for City Council elections. • Former Council member Don Harrison. Mr. Harrison moved to Pasadena when he was eleven years old and has lived there continuously for nearly sixty-five years. Before retiring, Mr. Harrison was an appraiser for U.S. Army Corps of Engineers. Mr. Harrison has served on the City Council for approximately nineteen years. His most recent continuous term ran from 2007 to 2015. During that time, Mr. Harrison represented District C under the 8–0 single-member district map and plan. As a Council member, Mr. Harrison consistently voted against the Mayor’s positions on resource-allocation issues. He also voted against the Mayor’s proposal to change to the 6–2 map and plan for City Council elections. • Council member Cary Bass. Mr. Bass is a Pasadena businessman who has contracts with the City. He is a political ally of, and has received civic board appointments from, Mayor Isbell. In 2015, Mr. Bass was elected to represent the new District E under the 6–2 map and plan. Mr. Bass’s 2015 campaign received financial support from Mayor Isbell’s political action committee, Citizens to Keep Pasadena Strong. Bench Trial Tr. 5:7–9, 5:13–14; Pls. Ex. 154. Mr. Bass has consistently voted with the Mayor and his allies on resource-allocation issues when they divided City Council. • Council candidate Oscar Del Toro. Mr. Del Toro is a Pasadena businessman who ran for the 14 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 15 of 113 new at-large Place H in 2015. Mr. Del Toro, a Latino and the Latino-preferred candidate, lost the election to an Anglo opponent. He received 39.1% of the citywide vote. (Docket Entry No. 94 ¶ 64). Before running for City Council, Mr. Del Toro was an ambassador for the Pasadena Chamber of Commerce and worked on greeting new businesses established in the City. He volunteered with the local school district, belonged to several local political organizations, and had worked on several local political campaigns in Pasadena. • Director of Community Relations Richard Scott. Mr. Scott is a lifelong Pasadena resident, except for his Army service. Before retiring, he worked for Western Waste Industries. Mr. Scott is a longtime friend of Mayor Isbell, who appointed Mr. Scott as the City’s Director of Community Relations when Mayor Isbell returned to office in 2008. Mr. Scott supervises five managers who control the City’s public outreach. • Director of Financial Planning Andy Helms. Mr. Helms has been a Pasadena resident for nearly thirty years. He has held a number of City offices, including director of traffic, purchasing director, planning director, assistant to the mayor for public works, and associate municipal judge. He is currently Pasadena’s Director of Financial Planning. Mr. Helms is Mayor Isbell’s chief advisor and a personal friend and has worked on Mayor Isbell’s political campaigns for many years. Bench Trial Tr. 5:23–25. • Bianca Gracia. Ms. Gracia has resided in Pasadena for twelve years. She is the president of the Village Grove East Homeowner’s Association in South Pasadena. • Victor Villarreal. Mr. Villarreal is a businessman who has lived in Pasadena for fourteen years. He has served as a board member for several nonprofit organizations in Pasadena and is a former president of the City’s Hispanic Business Association. In 2013, Council member 15 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 16 of 113 Ybarra recommended, and Mayor Isbell appointed, Mr. Villarreal to a citizens’ committee to review bond proposals. Shortly after the Bond-Review Committee was put in place, the Mayor changed it to a committee to consider amending Pasadena’s Charter provisions on City Council elections to the 6–2 voting map and plan. 2. Expert Witnesses The plaintiffs presented Mr. David Ely to testify under Rule 702 of the Federal Rules of Evidence on his specialized field of demography. Mr. Ely holds degrees in engineering and social science from the California Institute of Technology. He owns and is president of Compass Demographics, a consulting firm that builds specialized data sets using the United States decennial census and other sources. The plaintiffs also presented Dr. Richard Engstrom to testify under Rule 702. His field is statistics, and he testified on applying statistical analysis to issues raised by election redistricting. Dr. Engstrom has a doctorate in political science and is on the graduate faculty in political science at Duke University. Dr. Engstrom served as an advisor for the Mississippi legislature on gaining United States Department of Justice preclearance for state redistricting plans before the Supreme Court held in Shelby County that § 5 of the Voting Rights Act no longer required it. He has testified regularly and published many articles and scholarly works on redistricting and the Voting Rights Act. Finally, the plaintiffs offered the Rule 702 testimony of Dr. Andres Tijerina, whose field is Texas history and the historical experience of Latinos in Texas. Dr. Tijerina holds a doctorate in American history and is a professor of history at Austin Community College. Dr. Tijerina has published award-winning work on Latino and Tejano history. The City offered the Rule 702 testimony of Dr. John Alford, whose work applies statistical 16 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 17 of 113 analysis to issues raised by election districting. Dr. Alford holds a doctorate in political science and a master’s degree in public administration. He is on the Rice University political science faculty. He has appeared as an expert witness in many federal court proceedings involving the Voting Rights Act and has frequently published in his field. The City also offered the testimony of Dr. Bill Rives as an expert in demography. Dr. Rives holds a Ph.D. in economics with a concentration in demography and is a senior lecturer in the Fisher College of Business at the Ohio State University. He has held postdoctoral research fellowships in demography, economics, and statistics at the Woodrow Wilson School at Princeton University and the Census Bureau’s Office of Population Research. The court finds that Drs. Engstrom and Alford meet the Rule 702 requirements to testify in applying statistical analysis to issues raised by election districting; that Mr. Ely and Dr. Rives are qualified to testify on demography; and that Dr. Tijerina is qualified to testify on Texas history and the historical experience of Latinos in Texas. C. Pasadena’s Election Maps and Plans Pasadena is a Texas home rule city. (Docket Entry No. 94 ¶ 2). The City Charter, in place since 1942, sets up a “mayor-council government” with a strong mayor. (Id. ¶ 3). Pasadena’s powers are vested in an elected Council made up of the Mayor and, since 1992, eight Council members. The City Charter empowers the Council to enact legislation, adopt budgets, and determine City policies. The Charter empowers the Mayor to recommend ordinances for the Council’s consideration and to vote on matters before the Council. (Id.). Pasadena holds nonpartisan elections to elect the Mayor and the eight City Council members, with a majority vote required for election. (Id. ¶ 7). 17 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 18 of 113 The City Charter adopted in 1964 provided for a Mayor and six-member Council, all elected at-large. Four Council members had to live in designated districts. (Id. ¶ 4); Bench Trial Tr. 3:55. In 1992, Pasadena amended its Charter to enlarge the City Council from six to eight members and to elect all eight from single-member districts. (Docket Entry No. 94 ¶ 5); Bench Trial Tr. 3:56. From 1992 to 2013, Pasadena used an eight single-member district map and plan to elect Council members and elected its Mayor at-large. (Docket Entry No. 94 ¶¶ 5, 6). In 2011, Pasadena’s City Council voted in favor of adopting a redistricting map that produced four majority Spanish-surnamed registered-voter districts and four majority Anglo registered-voter districts. The City held one City Council election under that map and plan in May 2013. On June 25, 2013, the Supreme Court issued its opinion in Shelby County, Alabama v. Holder, 133 S.Ct. at 2612, holding that § 5 preclearance under the Voting Rights Act was no longer required in many states, including Texas. On June 27, two days later, Mayor Isbell sent the City Council a memo calling for a committee to consider bond proposals for a November 2013 special election. During the Committee’s deliberations, Mayor Isbell recommended amending the City Charter to create at-large voting districts and proposed maps that created a 4–2 map and plan, a 7–1 map and plan, and a 6–2 map and plan. Bench Trial Tr. 3:6–7; Pls. Ex. 246. He turned the Bond Review Committee into a committee to consider redistricting. The Committee recommended against amending the City Charter to create the mixed single-member and at-large election districts. Pls. Ex. 21. Mayor Isbell nevertheless proposed the Charter amendment to the City Council in August 2013. With the Mayor’s tie-breaking vote in favor, the Charter amendment proposal was put on the November 2013 special-election ballot. (Docket Entry No. 94 ¶¶ 48, 50). On November 5, 2013, Pasadena voters 18 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 19 of 113 cast 3,292 votes to approve Proposition 1, changing the districts to elect City Council members to six single-member and two at-large districts. (Id. ¶ 51). The vote was very close; 3,213 voted against it. (Id.). Latinos opposed the proposal with an estimated 99.6% of their votes. (Id. ¶ 63). In April 2014, the City Council passed the ordinance adopting the redistricting proposal that moved to a 6–2 map and plan. During the debate on the proposed redistricting, Council member Van Houte, elected as a Latino-preferred candidate, spoke against it. The Mayor ordered police officers to escort her out of the meeting when she exceeded a recently enacted rule limiting speech on a single topic to three minutes. (Id. ¶ 54). Pasadena’s City Council approved the Charter amendment by a 5–4 vote, with Mayor Isbell breaking the tie in favor of the change. (Id. ¶¶ 54–55). For the 2015 election, the 6–2 map and plan produced one less Latino-majority single-member district than under the 8–0 single-member district map and plan. (Id. ¶ 56). The two at-large positions on the City Council are elected by numbered place. Under a place system, seats are contested separately. Candidates for an at-large position on the Council file for only one place and compete only with the other candidates filing for that same place. Each voter may cast only one vote between the candidates for each place. (Id. ¶ 13). At-large voting using numbered places precludes a single-shot voting strategy. (Id. ¶ 14); Bench Trial Tr. 1:111. With single-shot voting, group members can cast one vote for the candidate the group favors and not cast any of their remaining votes for another candidate. By withholding their remaining votes from the candidates competing with their preferred choice, minority voters have a better chance of seeing their preferred representative finish among the top candidates and win one of the seats. Id. at 1:111–12; Pls. Ex. 198 at 4. The loss of single-shot voting disadvantages minority voters because their preferred candidates who finish in second place are eliminated. Bench 19 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 20 of 113 Trial Tr. 1:111–12. The Pasadena election plan uses a majority-vote requirement for each City Council seat. (Docket Entry No. 94 ¶¶ 7, 12). A successful candidate must win a majority of the votes cast in citywide elections. If no candidate receives a majority, a runoff election is held between the top two vote recipients. A majority-vote requirement also disadvantages minority voters by forcing their preferred candidate into a head-to-head contest with only one other candidate and eliminating the chance for a plurality victory if multiple candidates in the majority group divide the vote. Bench Trial Tr. 1:112; Pls. Ex. 198 at 4. D. Pasadena’s History Pasadena has a long history of discrimination against minorities, including Latinos, as well as a history of more recent events that are more probative. This history is “one evidentiary source, particularly if it reveals a series of official actions taken for invidious purposes,” Arlington Heights, 429 U.S. at 267, but as the Fifth Circuit recently recognized in Veasey, “unless historical evidence is reasonably contemporaneous with the challenged decision, it has little probative value.” Veasey, 830 F.3d at 232 (quoting McCleskey v. Kemp, 481 U.S. 279, 298 n.20 (1987)). 1. Texas Laws Affecting Pasadena “Texas has a long, well-documented history of discrimination that has touched upon the rights of African-Americans and Hispanics to register, to vote, or to participate otherwise in the electoral process. Devices such as the poll tax, an all-white primary system, and restrictive voter registration time periods are an unfortunate part of this State’s minority voting rights history. The history of official discrimination in the Texas election process—stretching back to Reconstruction—led to the inclusion of the State as a covered jurisdiction under Section 5 in the 20 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 21 of 113 1975 amendments to the Voting Rights Act.” League of United Latin Am. Citizens v. Perry, 548 U.S. 399, 439–40 (2006) (LULAC) (quoting Vera v. Richards, 861 F.Supp. 1304, 1317 (S.D. Tex. 1994)). The political, social, and economic legacy of past discrimination for Latinos in Texas also hinders their current ability to participate effectively in the political process. Id. The 1903 Terrell Election Law imposed a poll tax in Texas. The Act’s sponsor explained that the law was intended to close “the flood gates for illegal voting as one person could buy up the Mexican and Negro votes.” Pls. Ex. 201 at 12. The poll tax was eliminated in Harper v. Virginia Board of Elections, 383 U.S. 663 (1966). In response, the first Senate bill of the first 1966 Texas legislative session required voters to register annually. The annual registration requirement was invalidated in 1971. Beare v. Smith, 321 F.Supp. 1100, 1108 (S.D. Tex. 1971), aff’d sub nom. Beare v. Briscoe, 498 F.2d 244 (5th Cir. 1974). The Texas State Democratic Executive Committee established a White Man’s Primary Association in 1904, requiring an oath declaring “I am a white person and a Democrat” for membership. Pls. Ex. 201 at 12. In Nixon v. Herndon, 273 U.S. 536 (1927), the Supreme Court struck down the Texas white primary law under the Fourteenth Amendment. The Texas Legislature then passed a law authorizing political parties to set their own voter qualifications, and the Democratic Party enacted a rule that only whites could vote in the primary. The Supreme Court struck down this law in 1932. Nixon v. Condon, 286 U.S. 73 (1932). Texas enacted a law in 1918 to eliminate interpreters at the polls. Act of March 23, 1918, 35th Leg., 4th C.S. Ch. 30 (H.B. 104); Pls. Ex. 201 at 12. The following year, Texas enacted a requirement that election officials could communicate only in English in the polling place. Act of March 13, 1919, 36th Leg. Ch. 55 (S.B. 244), 1919 Tex. Gen. Laws p. 94. In 1975, the U.S. House 21 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 22 of 113 of Representatives Committee on the Judiciary report accompanying H.R. 6219 referred to “overwhelming evidence of voting discrimination against language minorities” and stated that “it is not surprising that the registration and voting statistics of language minorities are significantly below those of the Anglo majority. In 1972, for example, only 44.4 percent of persons of Spanish origin were registered compared to 73.4 percent for Anglos.” S. Rep. No. 94-295, at 30 (1975), reprinted in 1975 U.S.C.C.A.N. 774, 797. The 1974 percentages indicated similar disparity of 34.9 percent for Mexican-Americans to 63.5 percent registered Anglos. The Mexican-American voting rate was half of the voting rate for Anglos in 1974. See id. In 1972, a federal district court explained that a “cultural and language impediment, conjoined with the poll tax and the most restrictive voter registration procedures in the nation have operated to effectively deny Mexican-Americans access to the political processes in Texas even longer than the Blacks were formally denied access by the white primary.” Graves v. Barnes, 343 F. Supp. 704, 731 (W.D. Tex.), aff’d sub nom. Archer v. Smith, 409 U.S. 808 (1972), and aff’d in part, rev’d in part sub nom. White v. Regester, 412 U.S. 755 (1973). As set out in greater detail below, it is only in recent years, as Pasadena’s Latino population has swelled, that Latinos have attained a meaningful political voice. In 2012, a three-judge panel of the federal district court for the Western District of Texas ordered the creation of a Latino-majority congressional district in Harris County that includes the portion of Pasadena north of Spencer Highway. Perez v. Perry, Civil No. 11-360, Docket Entry No. 682 (W.D. Tex. Feb. 28, 2012). The court ordered this interim remedy after finding that the plaintiffs were likely to succeed on a § 2 vote-dilution claim. The court found that the State’s redistricting reduced the Latino citizen voting-age population in the district in order to “mak[e] the 22 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 23 of 113 district safer for its Republican incumbent,” violating § 2. Id., Docket Entry No. 690 at 8. 2. Racially Discriminatory Ordinances and Official Activities in Pasadena Dr. Tijerina credibly testified about the history of Pasadena’s treatment of its Latino residents. Dr. Tijerina documented the use of restrictive covenants and exclusionary laws and practices targeting Latinos in housing, education, City ordinances, and the workplace. Pls. Ex. 201. When Pasadena was incorporated in 1942, its City Charter complied with a Texas law imposing segregation and outlawing Spanish-language instruction, as well as Bohemian and German. The 1942 City Charter incorporated segregation in Article VIII, Section 4, titled “Segregation of Races.” Pls. Ex. 201 at 30. Restrictive housing covenants were in place until the 1940s. Bench Trial Tr. 1:174–76; Pls. Ex. 203. The pattern of housing segregation, now de facto, remains. Most Latino residents live in North Pasadena, which is separated from the predominately Anglo South Pasadena by the Spencer Highway. South Pasadena’s streets, sewage, recreation areas, and other basic infrastructure elements and amenities are much better than North Pasadena’s.2 As late as 1980, the Pasadena Independent School District excluded undocumented immigrant students. Pls. Ex. 201 at 30–31; see In re Alien Children Ed. Litig., 501 F.Supp. 544, 550 n.6 (S.D. Tex. 1980). In 1987, the federal government successfully sued the School District for failing to hire African-American teachers and administrators. Pls. Ex. 201 at 30; see also United States v. Pasadena Indep. Sch. Dist., Civil No. 83-5107, 1987 WL 9919 (S.D. Tex. Apr. 18, 1987). Pasadena was the Texas headquarters of the Ku Klux Klan. Pls. Ex. 201 at 29. The Klan targeted Mexican-Americans, including by burning crosses, publicly carrying high-powered rifles, and maintaining a public presence. Bench Trial Tr. 1:184. Former Council member Don Harrison 2 See Part II.E.6 below. 23 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 24 of 113 recalls seeing Klan members outside their headquarters building, which was located on a major thoroughfare. Id. at 4:173–74. In the late 1980’s, Council member Van Houte recalls seeing Klan members outside the building lining the street and wearing their white robes. Id. at 4:6–7. In the mid-1980s, Latino ad hoc committees met with the Pasadena Police Chief and City Council to protest police brutality and harassment. Pls. Ex. 201 at 31. In 1986, the Federal Bureau of Investigation came to Pasadena to investigate the alleged beating of a Latino man in police custody. Id. In the 1990s, there were continuing incidents of police mistreatment of Latinos and complaints about a close relationship between the Pasadena Police Department and the Ku Klux Klan. Id. Complaints of police antipathy toward Latino residents have continued to the present. In the past several years, Latino residents in North Pasadena have repeatedly complained to Council members that they are more likely to be targeted for traffic stops by police than Anglo residents. Bench Trial Tr. 2:14, 4:7. Former Council member Harrison recalled a 2008 conversation he had with Mayor Isbell, in which the Mayor said that he had told the Pasadena police department that if officers found a Hispanic driving without car insurance, the officers should impound the car. Id. at 4:199–200. When Mr. Harrison asked the Mayor what the police were to do if an Anglo was driving without insurance, the Mayor responded that impoundment should be left to the officer’s discretion. Id. Mr. Harrison credibly testified that he understood this as a racially biased comment. Id. Witnesses testified to continuing racially tinged incidents. Council member Ybarra testified that approximately three years ago, an Anglo City human resources employee made comments to a group of Latino employees in the City Water Department about “Mexicans making tacos.” Id. at 24 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 25 of 113 2:185–86. When the Water Department employees complained to Mr. Ybarra, the employee was transferred but continued to work for the City. Id. Mr. Ybarra credibly testified that he has witnessed instances of official racist attitudes and acts during his City Council service. “I have been in Pasadena for 38 years. The whole time I have been there, I never experienced racism or prejudice to what I have seen in my time serving on council.” Id. at 2:185. 3. Recent Private Racial Discrimination in Pasadena Private acts of discrimination are less probative of discriminatory effects or intentions than official acts. See Veasey, 830 F.3d at 306 (Jones, J., dissenting) (citing Frank v. Walker, 768 F.3d 744, 755 (7th Cir. 2014) (Easterbrook, J.)). But “under the results standard of section 2, pervasive private discrimination should be considered, because such discrimination can contribute to the inability of [minorities] to assert their political influence and to participate equally in public life.” United States v. Marengo Cty. Comm’n, 731 F.2d 1546, 1567 n.36 (11th Cir. 1984) (Wisdom, J.); see also Gomez v. City of Watsonville, 863 F.2d 1407, 1418 (9th Cir. 1988) (rejecting the argument that only discrimination by the defendant is relevant to a § 2 vote-dilution case); Solomon v. Liberty Cty., Florida, 899 F.2d 1012, 1032 (11th Cir. 1990) (en banc) (Tjoflat, J., concurring) (“Congress . . . revised [§ 2] to prohibit election practices that accommodate or amplify the effect that private discrimination has in the voting process.” (quoting David L. Eades, Recent Developments, Section 2 of the Voting Rights Act: An Approach to the Results Test, 39 VAND. L. REV. 139, 172 (1986))). And in this case, private discrimination in Pasadena is not entirely private. As the City notes, because the City Charter can be amended only by a vote of the citywide electorate, the official, “ultimate decision-maker on the charter amendment is the entire electorate.” (Docket Entry No. 148 ¶ 51). 25 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 26 of 113 The court finds that Pasadena’s current Latino Council members and a recent candidate for City Council credibly testified about private discrimination among Pasadena’s electorate, particularly in South Pasadena. On multiple occasions while campaigning in 2009, Anglo residents of Pasadena slammed the door shut on Mr. Ybarra, telling him that they “weren’t going to vote for a wetback” and would not vote for a Hispanic. Bench Trial Tr. 2:24. As a candidate in 2013 and 2015, Mr. Wheeler did not advertise his Latino ethnicity. Id. at 5:96. If Mr. Wheeler was talking to an Anglo homeowner who asked about his background, he would answer that he is a little bit of everything but would emphasize his military background and service to the country and to the community to ease their concerns. Id. at 5:97. Mr. Wheeler credibly testified that it would have been harder for him to be elected in old District E if he had been Spanish-surnamed. “In my first election [in 2013], I won by 32 votes. If my name was something other than Wheeler, if it was Ramirez or Sanchez, I don’t believe that I would have won that election.” Id. at 5:138. After Mr. Wheeler won the 2013 election in District E, his Anglo opponent, Leroy Stanley, filed an election contest and stated publicly that Mr. Wheeler had won because of “illegal votes.” Id. at 5:114. Mr. Stanley later dismissed the lawsuit. Mr. Wheeler understood Mr. Stanley’s comments about illegal voting, in conjunction with the fact that “he was going around telling people that I was Mexican, making sure that people knew that,” as a subtle appeal to racism. Id. at 5:152. Oscar Del Toro ran for Pasadena City Council’s new at-large Place H in 2015 against an Anglo opponent. Mr. Del Toro focused his door-to-door campaign on South Pasadena. Id. at 5:173–75. In North Pasadena, residents received him well and were proud that a Mexican immigrant was running for City Council. Some asked him questions in Spanish. Id. at 5:174–75. In South 26 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 27 of 113 Pasadena, he was not well-received. One woman told him, “I’m not going to vote for you. You know why?” She did not explain her remark, but Mr. Del Toro understood it as a racial reference. Id. at 5:175. Another told him that Pasadena is a “good ole boy town,” and asked him to drop his materials on the porch rather than hand them to her. Id. Dr. Tijerina concluded that Latinos in Pasadena continue to lag behind politically and economically in the face of persistent racial prejudice and a structure of discrimination. Pls. Ex. 201 at 3–4. This recent background is more probative of the impact of, and intent behind, the 2014 change from an 8–0 single-member to a mixed 6–2 single-member and at-large election map and plan in Pasadena. E. Pasadena’s Demographics 1. The Latino Population in Pasadena Pasadena’s population is approximately 149,285 persons. (Docket Entry No. 94 ¶¶ 20–21). The parties agree that the Latino citizen voting-age population of the City has risen in recent decades. (Id. ¶ 22). In 1990, the U.S. Census reported that the Hispanic share of Pasadena’s citizen voting-age population was 18.7%. (Id.). The American Community Survey reported that the Hispanic share of Pasadena’s citizen voting-age population was 48.2% for the period 2011–2015.3 (Docket Entry 3 The parties presented several different figures from the American Community Survey showing the Hispanic share of Pasadena’s citizen voting-age population. Mr. Ely’s initial report used the American Community Survey five-year data from 2009 to 2013 and reported that the Hispanic share of Pasadena’s citizen voting-age population was 45.87%. Pls. Ex. 194; (Docket Entry No. 94 ¶ 29). In Mr. Ely’s supplemental report, he used the five-year data from 2010 to 2014 and found that the Hispanic share of Pasadena’s citizen voting-age population was 47.1%. Pls. Ex. 195 at Table 1; Bench Trial Tr. 1:71–73. At trial, defense counsel notified the court that the 2011 to 2015 American Community Survey data would be released a week after the trial ended and asked for leave, without objection from plaintiffs, to supplement the record with that updated data. Bench Trial Tr. 6:46–47. Mr. Ely filed another supplemental report using the five-year data from 2011 to 2015 and reported that the Hispanic share of Pasadena’s citizen voting-age population was 48.23% with a margin of error of 1.63%. (Docket Entry No. 124, Ex. 1). 27 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 28 of 113 No. 124, Ex. 1). According to the 2010 Census, the City of Pasadena is 62.1% Hispanic, and Pasadena’s Hispanic voting-age population, both citizen and noncitizen, is 55.78%. Pls. Ex. 195; (Docket Entry No. 94 ¶ 29). Pasadena’s Spanish-surnamed voter registration is 41.97% as of September 2016. (Id. ¶ 32). Using Spanish-surnamed voter registration to estimate Latino voter registration reasonably balances errors of omission and commission, particularly in jurisdictions with a significant Latino population. Bench Trial Tr. 1:55–56. Spanish-surnamed voter registration is a reliable proxy for Latino registered voters in Pasadena. Id. Pasadena “geocoded” individual registered voter’s addresses to the Pasadena map. The geocoding was the basis for determining how many Spanish-surnamed registered voters lived in specific districts. The result is more accurate than estimating Spanish-surnamed voter registration at the block level using data from larger areas like census precincts. Id. at 6:18–19. The decennial census does not include a question on citizenship. Id. at 1:65. Citizenship is requested on the American Community Survey form, an extensive survey the U.S. Census Bureau sends annually to two percent of United States households. Id. at 1:57–58, 65. As a result, “[t]he sole source of citizenship data published by the Census Bureau now comes from the American Community Survey (‘ACS’).” Cisneros v. Pasadena Indep. Sch. Dist., No. 4:12-CV-2579, 2014 WL 1668500, at *5 (S.D. Tex. Apr. 25, 2014).4 The Census Bureau tabulates the American Community Survey results from a five-year period for the Justice Department. Both sides used this information to estimate the citizen voting-age population in Pasadena City Council districts. Bench Trial Tr. 1:65–67. The five-year tabulation of American Community Survey data from 2010–2014 was 4 “ACS data is Census data. It is produced and promulgated by the Census Bureau, and it is intended to replace the long form in the decennial Census. . . . ACS data can be—and eventually must be—relied upon; the Census Bureau will be utilizing ACS data in lieu of long form survey data beginning in 2010.” Benavidez v. City of Irving, Tex., 638 F.Supp.2d 709, 729 (N.D. Tex. 2009). 28 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 29 of 113 available to analyze citizenship of the residents in each Pasadena City Council district at the time of trial. Id. at 1:67–68. The citywide data from 2011–2015 is also now available.5 (Docket Entry No. 124, Ex. 1). 2. Latino Citizen Voting-Age Population in Pasadena The Census provides two estimates of the 2015 Hispanic citizen voting-age population in Pasadena. The first is a Hispanic citizen voting-age population of 48.2% and is based on the American Community Survey five-year survey for 2011 to 2015. The second is a Hispanic citizen voting-age population of 50.6% and is based on the American Community Survey one-year survey for 2015. The five-year survey is the more reliable estimate of citizen voting-age population. The parties agree that the American Community Survey five-year estimate is statistically more reliable than the one-year estimate. The parties also agree that the one-year estimate is more current than the five-year estimate because it is not averaging in data from three or four years ago. Bench Trial Tr. 1:68–69, 6:21–23. The Census Bureau advises that “[e]stimates for large geographic areas benefit from the increased sample of [multiyear surveys] resulting in more precise estimates of population and housing characteristics, especially for subpopulations within those areas” and that “single-year estimates . . . have higher sampling variability.” A COMPASS FOR UNDERSTANDING AND USING AMERICAN COMMUNITY SURVEY DATA at A-3 (2009). The margin of error for the five-year data is less than half that of the one-year data.6 Bench Trial Tr. 1:72. Similar to decennial census estimates 5 District-level data for the 2011 to 2015 five-year survey will not be available until late January or February 2017. Bench Trial Tr. 6:46. 6 “Larger margins of error do not indicate that the data are unreliable for all purposes, but the margins of error must be taken into account nonetheless, and the purposes for which the data may be used 29 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 30 of 113 of citizen voting-age population, the American Community Survey’s five-year estimate is presumptively correct because it is more statistically reliable. That presumption applies unless different information and numbers are “thoroughly documented, have a high degree of accuracy, and [are] clear, cogent and convincing to override the presumptive correctness of the prior [data].” Valdespino v. Alamo Heights Indep. Sch. Dist., 168 F.3d 848, 853–54 (5th Cir. 1999).7 The American Community Survey five-year estimate applies to all years in the five years it covers. Pls. Ex. 319 at A-2. According to the Census-published guidelines, if the difference between one point estimate and a later point estimate is smaller than the difference between the margins of error for the two estimates, the more recent estimate should not be used to infer a change between the two periods. Id. at A-19; Bench Trial Tr. 1:68–69. A higher one-year point estimate provides some evidence of a change, but not enough for statistical significance. Id. When the difference between the point estimates is smaller than the difference between the margins of error, the balance between statistical reliability and currency tips in favor of using the more reliable figure.8 Id. must be limited accordingly.” Benavidez v. Irving Indep. Sch. Dist., Tex., 690 F.Supp.2d 451, 458 (N.D. Tex. 2010). 7 The district court in the most recent § 2 challenge against the Pasadena Independent School District relied on American Community Survey five-year estimates of Hispanic citizen voting-age population. Cisneros, 2014 WL 1668500, at *9. Another district court in a recent § 2 challenge to redistricting in Harris County relied on American Community Survey five-year estimates of Hispanic citizen voting-age population. Rodriguez v. Harris Cty., Tex., 964 F.Supp.2d 686, 728 (S.D. Tex. 2013), aff’d sub nom. Gonzalez v. Harris Cty., Tex., 601 F.App’x 255 (5th Cir. 2015) (“On this record, the Court concludes that the five-year aggregated ACS citizenship data is sufficiently probative on the issue of citizen voting-age population and Plaintiffs may rely upon this data in establishing the first Gingles precondition.”). Similarly, in a § 2 challenge to the City of Irving, Texas, the court relied on three-year American Community Survey data to determine Hispanic citizen voting-age population when one-year data was unstable. See Benavidez, 638 F.Supp.2d at 720–21. 8 > 1.645, where X1 and X2 are the point estimates and SE1 an To be precise, if: 30 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 31 of 113 In Pasadena, the American Community Survey five-year estimate of the Hispanic share of total citizen voting-age population was 48.2% with a margin of error of 1.6%. (Docket Entry No. 124, Ex. 1). The margin of error is based on 90% statistical confidence, which is the Census Bureau standard for American Community Survey margin-of-error estimates. This yields a 90% interval estimate running from 46.6% to 49.8%. (See Docket Entry No. 94 ¶ 31). The American Community Survey one-year estimate of the citywide Hispanic share of total citizen voting-age population for 2015 is 50.6%, with a margin of error of approximately 3.3%. The 90% interval estimate for the citywide Hispanic share of total citizen voting-age population runs from 47.3% to 53.9%. (Id.). Mr. Ely performed the Census Bureau’s recommended procedure to evaluate whether the difference between the American Community Survey one-year and five-year estimates was statistically significant.9 He determined that the difference between the margins of error was larger than the difference between the point estimates.10 As a result, the two estimates were not statistically SE2 are the standard errors for the point estimates (defined as the positive margin of error for each estimate divided by 1.645 for a 90% confidence interval), then the difference between the point estimates is statistically significant at a 90% level of confidence. If the result is smaller than 1.645, the difference between the estimates is not statistically significant. COMPASS FOR USING AMERICAN COMMUNITY SURVEY DATA at A-19. 9 The Census advises caution when comparing single-year estimates with multiyear estimates, “as they are measuring the characteristics of the population in two different ways, so differences between such estimates are difficult to interpret.” COMPASS FOR USING AMERICAN COMMUNITY SURVEY DATA at A-19. The Census does not prohibit such comparisons, but rather speaks in terms of “ideal” comparisons. Id. The court finds that Mr. Ely properly accounted for the recommended Census procedures to choose the more reliable figure when the ideal—in this case, a five-year survey ending in 2016—was not available for his report. The one-year American Community Survey estimates vary above and below the 50% line and do not evidence a clear trend that would sufficiently support a finding that Latinos have recently become the majority of citizen voting-age population in Pasadena. The 2012 one-year figure was 46.8%; the 2013 oneyear figure was 50.5%; the 2014 one-year figure was 48.1%; and the 2015 one-year figure was 50.6%, all with a margin of error around or above 3%. Def. Ex. 40 at Ex. 5-A. 10 Because of the slight differences in their computed margins of error, Mr. Ely used his margins of error, Dr. Rives’s margins of error, and the compromise that they agreed on for the Admissions in the Proposed Joint Pretrial Order. Using any of those measures, the difference in the margins of error was 31 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 32 of 113 significantly different. For that reason, Mr. Ely concluded that the more statistically reliable fiveyear estimate is the best source for measuring Pasadena’s Citywide Latino voting-age population. Bench Trial Tr. 1:68–69. Mr. Ely gave a second reason to rely on the five-year and not the one-year data. Because only the American Community Survey five-year estimate is available for estimating citizen voting-age population in individual Council districts, using the American Community Survey five-year estimate of Citywide citizen voting-age population ensures consistency across the analysis of single-member districts compared to at-large voting, avoiding comparing “apples to oranges.” Id. at 1:67–68. The City did not rebut Mr. Ely’s reasoning for relying on the American Community Survey’s five-year estimate of citizen voting-age population in Pasadena. The testimony of the City’s expert, Dr. Rives, which endorsed the one-year figure of 50.6% Hispanic citizen voting-age population, was not credible. Dr. Rives testified that he “wasn’t interested” in whether the differences between the one-year and five-year surveys were statistically significant. Bench Trial Tr. 6:72. Dr. Rives presented no margins of error for any of the figures he discussed in his report. Id. at 6:59. In a previous case, Dr. Rives concluded that the effect of Latinos over-reporting their citizenship on surveys likely meant American Community Survey figures for Latino citizen voting-age population were inflated. Dr. Rives testified that he offered those conclusions in the previous case because he was “tasked” to show Latinos had not attained a majority in that case, but in this case “I’m not on greater than the difference in the point estimates. Bench Trial Tr. 1:96–97. Mr. Ely performed his calculations on the 2010–2014 five-year data, as that was the most current data at the time of his initial report. The court finds that the updated five-year data from 2011–2015 does not substantially change the analysis, as the relative margins of error and relative differences between the five-year and one-year figures remain similar. The difference between the point values remains significantly smaller than the difference between the margin of errors for those point values, according to the formula provided in note 8 above. 32 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 33 of 113 the side that’s trying to establish a majority.” Id. at 6:94. The court finds that Dr. Rives’s inconsistent research methods and lack of statistical rigor make his testimony about the reliability of the American Community Survey one-year data not credible. The court concludes that Latinos make up 48.2% of the citizen voting-age population citywide, and that they likely made up even less of the citizen voting-age citywide population when the 6–2 map and plan was proposed and enacted. 3. The Eight Single-Member District Plan Under the single-member district map and plan in place in May 2013, when the last City Council election was held, Latinos made up the majority of Pasadena’s citizen voting-age population and the majority of registered voters in four of eight districts: Districts A, B, C, and D. (Docket Entry No. 94 ¶¶ 25, 26). Table 1 Demographic Profile of Pasadena’s 8–0 Map and Plan As It Would Be Today District A B C D E F G H Total 2010 Census Population 17,915 18,384 18,353 17,986 19,002 19,449 19,547 18,649 149,285 Latino VotingAge Population 80.0% 73.7% 76.3% 64.1% 57.2% 37.1% 54.3% 14.8% 55.7% Latino Citizen Spanish-Surname Voting-Age Registered Voters Population 69.8% 68.6% 66.2% 62.4% 64.9% 63.5% 59.7% 56.0% 45.8% 44.7% 34.9% 30.0% 43.9% 40.5% 17.5% 13.2% 48.2% 42.0% Sources: 2010 U.S. Census; 2010–2014 and 2011–2015 five-year American Community Surveys; Pls. Ex. 195-3. Latinos would constitute the majority of the citizen voting-age population and the majority of registered voters in the same four of eight districts in the 8–0 map and plan today. (Docket Entry 33 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 34 of 113 No. 94 ¶ 26); Pls. Ex. 195. 4. The Six Single-Member District and Two At-Large Plan When the 2015 election was held under Pasadena’s 6–2 map and plan, Latinos made up the majority of the citizen voting-age population and the majority of registered voters in three of six single-member districts: Districts A, B, and C. (Docket Entry No. 94 ¶ 29). Table 2 Demographic Profile of Pasadena’s 6–2 Map and Plan As It Is Today 2010 Census Population District A B C D E F Total (Places G & H) 24,607 25,035 24,737 24,800 24,815 25,291 149,285 Latino VotingAge Population 82.8% 72.4% 65.2% 59.1% 41.2% 21.3% 55.8% Latino Citizen Spanish-Surname Voting-Age Registered Voters Population 71.5% 71.0% 62.9% 59.5% 59.4% 56.5% 46.9% 45.9% 36.9% 33.0% 21.2% 16.7% 48.2% 42.0% Sources: 2010 U.S. Census; 2010–2014 and 2011–2015 five-year American Community Surveys; Pls. Ex. 195-3. The 6–2 map eliminated District D as a Latino-majority district and relocated the District D identifier to an Anglo-majority district. Pls. Exs. 152, 153. The 6–2 map, because it divided Pasadena’s population into six instead of eight single-member districts, also enlarged each single-member district by approximately 6,220 people. Pls. Ex. 195-3. 5. Racial Cohesion in Pasadena City Council Elections Dr. Engstrom, the plaintiffs’ expert witness on statistics, used ecological inference to determine the candidate preferences of Latino and non-Latino voters in racially contested elections. He analyzed the racial composition of, and the votes cast in, each of the City Council district 34 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 35 of 113 elections under the earlier and the current plans. (Docket Entry No. 94 ¶ 58); Bench Trial Tr. 1:112– 15. The parties agree that racially contested elections, which present voters with a choice between Latino and non-Latino candidates, are generally more probative for assessing raciallypolarized voting than elections in which two Latinos or two Anglos are running against each other. (Docket Entry No. 94 ¶ 62). The parties agree that Latino and non-Latino voters demonstrated different preferences in the 2015 at-large race for Place G, the 2015 at-large race for Place H, and the 2013 Proposition 1 race. Bench Trial Tr. 1:128, 6:175; Pls. Ex. 198. Dr. Engstrom also analyzed the three local district elections in May 2015 that were racially contested. Dr. Engstrom testified, and the court finds, that because each district only represents one-sixth of Pasadena’s population, the citywide elections are more probative of racially polarized voting in Pasadena. Bench Trial Tr. 1:128–29. a. The 2015 At-Large Race for Place G In the May 2015 at-large election for Place G on the Pasadena City Council, former Council member Van Houte, an Anglo candidate, defeated her opponent, Steven Cote, another Anglo Council incumbent, by 143 votes. Pls. Ex. 239. Latino voters supported Ms. Van Houte with an estimated 70.6% of their votes. Def. Ex. 54. Non-Latino voters supported her with 41.6% of their votes. Id. Because that race for Place G involved only Anglo candidates, the racial polarization in that election is less clear than an election between an Anglo and a Latino candidate, but it has some probative value because Ms. Van Houte was clearly identified as aligned with Latino voters and was their preferred candidate. The court finds that the 2015 voting for Place G under the 6–2 map and plan was racially polarized. 35 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 36 of 113 Dr. Alford testified that Ms. Van Houte would not have won without crossover support from Anglo voters, and that she needed a higher percentage of crossover votes, “something in the high 30’s,” than a Latino candidate, because she was not as dominant a candidate among Latino voters as a Latino candidate. Bench Trial Tr. 6:124–25, 6:129–30. Dr. Alford agreed that Ms. Van Houte almost lost her election despite receiving 70% of the Latino vote. She could not have slipped much lower in Anglo support than the 41.6% she received and still have won. Id. at 6:138–39. Dr. Alford conceded that the single largest factor between Ms. Van Houte’s at-large victory in Place G and the defeat of the Latino candidate in Place H, described in the next subsection, was the level of crossover voting by non-Latino voters. Id. at 6:175. In short, the election of the Latino-preferred candidate depended on significant and unusually high Anglo crossover voting. b. The 2015 At-Large Race for Place H In the May 2015 at-large election for City Council Place H, pitted a Latino candidate, Oscar Del Toro, against an Anglo candidate, Darrell Morrison, who had represented District H in the 8–0 map and plan. Mr. Del Toro received an estimated 39.1% of the citywide vote and 87.3% of the Latino vote. (Docket Entry No. 94 ¶ 64). He lost to his Anglo opponent, who received 71.7% of the Anglo votes. (Id.). Only 28.3% of the Anglo voters voted for Mr. Del Toro. (Id.); Bench Trial Tr. 1:118–19; Pls. Ex. 255. Mr. Del Toro testified at trial. Before the election, he did a by-precinct turnout analysis. The analysis showed that predominately Anglo South Pasadena would have greater turnout rates than the more heavily Latino North Pasadena. He concentrated 70% of his door-to-door campaign on South Pasadena. Bench Trial Tr. 5:164–65, 5:170, 5:174. Mr. Del Toro encountered racial hostility while 36 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 37 of 113 campaigning in South Pasadena.11 Despite his concentrated and concerted efforts, few Anglos voted for him. Council member Wheeler testified that there were no policy differences between Ms. Van Houte, who narrowly won her at-large race, and Mr. Del Toro, who resoundingly lost. Id. at 5:138–39. Dr. Engstrom concluded, and this court finds, that the 2015 at-large race for Place H under the challenged 6–2 plan was racially polarized; and that Oscar Del Toro lost the election as a consequence. Id. at 1:119; Pls. Ex. 255. c. The November 2013 Special Election on Proposition 1 Dr. Engstrom examined the results of the November 2013 special election on Proposition 1, the proposal to change from the 8–0 single-member map and plan to the mixed 6–2 map and plan to elect City Council members. Dr. Engstrom found that Latino voters opposed Proposition 1 with an estimated 99.6% of their votes. (Docket Entry No. 94 ¶ 63). Non-Latino voters supported Proposition 1 with an estimated 60.2% of their votes. (Id.). Proposition 1 prevailed. (Id.); Bench Trial Tr. 1:120–21; Pls. Ex. 255. Dr. Engstrom concluded, and the court finds, that the vote in the 2013 Proposition 1 election was racially polarized, and that Proposition 1 succeeded because of Anglo bloc voting. Bench Trial Tr. 1:121. d. The 2015 District-Level Elections Three single-member district elections held in May 2015 were racially contested, in Districts A, B, and D. In the District A election, Council member Ybarra, the longtime incumbent, was reelected. Latinos supported Mr. Ybarra with 97% of their votes, and Anglos supported him with 55.4% of their votes. (Docket Entry No. 94 ¶ 75); Pls. Ex. 255. In the District B election, Celestino 11 See Part II.D.3. above. 37 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 38 of 113 Perez, a Latino and the Latino-preferred candidate, was defeated by a slender 34-vote margin. Latinos supported Mr. Perez with 82.0% of their votes, but only 36.3% of Anglos crossed over to vote for him. (Docket Entry No. 94 ¶ 76); Pls. Ex. 255. In the District D election, Council member Wheeler, the incumbent for most of the territory making up the new District D, was reelected. Latinos supported Mr. Wheeler with 89.5% of their votes, and Anglos supported him with 55.1% of their votes. (Docket Entry No. 94 ¶ 77); Pls. Ex. 255. Dr. Engstrom concluded, and the court finds, that the District B election was racially polarized. Bench Trial Tr. 1:129. The racially polarized vote in that district was significant and caused the narrow defeat of the Latino-preferred candidate. Council member Ybarra testified that in his opinion, Mr. Perez would have won the District B election if it had been held under the 8–0 map and plan, because the district under that plan would have held 6,000 fewer residents and required less campaign time and expense to turn out the vote. Id. at 2:128–30. District B has a Latino citizen voting-age majority of 58%. (Docket Entry No. 94 ¶ 76). Mr. Perez lost the election by 34 votes because of lower Latino turnout and Anglo bloc voting. e. Exogenous Elections The parties agree that endogenous elections held within Pasadena are most probative and that exogenous elections are supplemental. Id. at 1:116–17, 6:171–72. But exogenous election results can be helpful in determining whether Anglos typically bloc vote to defeat a Latino-preferred candidate. See Rodriguez v. Bexar Cty., Tex., 385 F.3d 853, 863, 865 (5th Cir. 2004). “Reconstituted election analysis [of exogenous elections] is a relatively simple method that extracts actual election results from a variety of statewide and local races that subsume the area being analyzed and determines, precinct-by-precinct within the new district, the racial composition of the 38 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 39 of 113 vote and the ‘winner’ within the new district. This method of aggregation allows a researcher to determine how an individual candidate performed within the boundaries of the target district even though the actual election covered a different geographical area.” Id. at 861. Pasadena is one of thirty-four municipalities in Harris County, which is among the largest counties in the nation. Dr. Engstrom analyzed countywide election results limited to votes cast only in Pasadena precincts. Bench Trial Tr. 1:121–22. He also analyzed primary elections, again limited to votes cast in Pasadena precincts. The pattern was clear. (1). • Countywide Elections In the 2008 general election race for Harris County Sheriff, Adrian Garcia, the Democratic nominee, received 98.2% of Latino votes and 28.6% of non-Latino votes. (Docket Entry No. 94 ¶ 68); Pls. Ex. 255. He lost to the Anglo Republican opponent. • In the 2012 general election race for Harris County Sheriff, Adrian Garcia ran again as the Democratic nominee. He received 95.9% of Latino votes and 17.6% of non-Latino votes, and again lost to the Anglo Republican opponent. (Docket Entry No. 94 ¶ 66); Pls. Ex. 255. • In the 2010 general election race for County Commissioner, Silvia Garcia ran as a Democrat and received 98.4% of Latino votes and 20.9% of non-Latino votes. (Docket Entry No. 94 ¶ 67); Pls. Ex. 255. She lost to an Anglo opponent. Pls. Ex. 255. • In the 2010 general election for County Treasurer, Billy Briscoe, a Democrat, was the Latino-preferred candidate. He received 82.2% of the Latino votes cast, but only 20.1% of the non-Latino votes. Pls. Ex. 255. He lost to Orlando Sanchez, a 39 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 40 of 113 Republican. Id. Dr. Engstrom concluded that the Latino-preferred candidate in this race was defeated by Anglo bloc voting. Bench Trial Tr. 1:123–24. The court finds that Pasadena’s countywide general elections were polarized along Latino and Anglo lines. None of the Latino-preferred candidates won their elections despite Latino cohesive support because Anglo crossover voting was so low. Id. at 1:122. (2). Primary Elections Primaries are less probative than general elections for detecting racially polarized voting in an at-large district because general elections present the same candidate pool to every voter, while primary elections limit voters to one party’s candidates. Id. at 1:127–28. Dr. Engstrom examined two Democratic and two Republican primaries and found racially polarized voting in both. He also found that party affiliation corresponded to, and was a proxy for, voters who voted with the Anglo or Latino blocs. Latino-preferred candidates won both Democratic primary races and were eliminated only in the two Republican primaries. Id. at 1:124; Pls. Ex. 255. • In the multi-candidate primary election for sheriff, Ruben Monzon ran in the Republican primary against seven non-Latino candidates. Latino voters supported Mr. Monzon by 38.6% (a plurality) and spread 61.4% of their support across seven other non-Latino candidates. Non-Latino voters gave 8.7% of their votes to Mr. Monzon. (Docket Entry No. 94 ¶ 71); Bench Trial Tr. 1:125; Pls. Ex. 255. He lost the primary. Pls. Ex. 255. • The 2010 Republican primary for County Commission Precinct 2 results were, according to Dr. Engstrom, uninformative because only a small number of Latinos voted in that primary. Bench Trial Tr. 1:126–27; Pls. Ex. 255. 40 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 41 of 113 • In the two 2008 and 2012 Democratic primaries for Sheriff, Adrian Garcia won as the Latino-preferred candidate, despite Anglo bloc voting against him. He lost to the Anglo-preferred candidate in the general elections. (Docket Entry No. 94 ¶¶ 70, 72); Bench Trial Tr. 1:125–26; Pls. Ex. 255. f. Findings on Racially Cohesive Voting in Pasadena Elections The court agrees with Dr. Engstrom, and finds, that in citywide elections other than primaries, which are less informative because voters are limited to the candidates of a single party, voting is typically racially polarized. In all the elections examined, Anglo or other non-Latino bloc voting usually defeated the Latino-preferred candidate. Bench Trial Tr. 1:128. At the district level, Anglo or non-Latino bloc voting defeated the Latino-preferred candidate in District B. Id. at 1:129. Under the challenged 6–2 redistricting map and plan, in Place G, the Latino-preferred candidate could not win without significant and unusually high Anglo crossover votes, and in Place H, did not win without Anglo crossover votes. Id. at 1:119; Pls. Ex. 255. Successful Latino-preferred candidates in the 2015 election were either aided by incumbency, aided by running in a non-racially contested race, aided by an Anglo identity or Anglo-sounding surname, or aided by a combination of these factors. In all of these elections, whether successful or not, Pasadena Latinos voted cohesively. 6. Socioeconomic Conditions and the Effects on Pasadena Voting a. Income, Education, and Employment The median income in Pasadena is $39,354 for Latino households and $55,896 for Anglo households. The citywide household median income is $46,058. Bench Trial Tr. 1:60–61; Pls. Ex. 196. The differences are even greater between Latino and Anglo per capita income, largely because 41 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 42 of 113 Latino households tend to have more people than Anglo households. The overall per capita income is $20,146, which breaks down to $13,984 for Latinos and $32,051 for Anglos. Bench Trial Tr. 1:60–61; Pls. Ex. 196. Citywide, about 51% of Latino households are renters, in contrast to 33% of Anglo households. Bench Trial Tr. 1:61; Pls. Ex. 196. Under the federal standards, only 1% of Anglos live in overcrowded housing compared to 16% of Latinos. Bench Trial Tr. 1:62–63; Pls. Ex. 196. Twenty-seven percent of Latinos as compared to 11% of Anglos live in poverty, a disparity that cuts across all age groups. Bench Trial Tr. 1:62; Pls. Ex. 196. Only 6% of Latinos have a bachelor’s degree or higher, in contrast to 22% of Anglos. Pls. Ex. 196. Only 11% of Anglos have less than a high school diploma, compared to 45% of Latinos. Pls. Ex. 196. Council member Ybarra testified that for District A residents, the income level is lower than the middle income range and the average educational attainment level is a high school degree or less. Bench Trial Tr. 2:12. b. Living Conditions Witnesses described different circumstances and conditions between the predominantly Latino North Pasadena and Anglo South Pasadena. Across North Pasadena, the older part of town, the sewers, streets, sidewalks, and water lines were neglected for more than thirty years. The sidewalks are cracked and uneven; some areas that need them do not have them. Id. at 2:8 (Ybarra); 4:13 (Van Houte); 4:178 (Harrison); 5:102–05 (Wheeler). Mr. Del Toro, an unsuccessful 2015 candidate for an at-large City Council place, testified that the streets and sidewalks on the north side of Spencer Highway are in poor condition. Id. at 5:163–64. During his campaign, residents told him that their children could not safely walk to and from school because of the sidewalk conditions, or absence. The streets are full of potholes. Poor drainage produces frequent flooding. Id. at 5:173. 42 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 43 of 113 Council member Cody Ray Wheeler testified that current District D would greatly benefit from work on major thoroughfares and that many roads need to be repaved. Id. at 5:100–02. Water and sewer lines in District A need repair. Id. at 2:8–9. Mr. Wheeler continues to receive calls from voters in current District D reporting that they cannot flush commodes when it rains and that lawns are flooding. Id. at 5:103. Mr. Wheeler testified that getting the City to fix streets on the north side “is always a fight. It is always a large effort . . . . When you call work orders in, things don’t get done as quickly on that side of town.” Id. at 5:101–02. In one instance, the lack of sidewalks meant that Keller Middle School students had to walk in the street to get to and from school. Teachers petitioned at City Council meetings to ask for sidewalks. The City did not respond until media reports described the risks to the children. The City then built sidewalks. Id. at 5:103. In another instance, a street project on a major street in North Pasadena planned for the first year Mr. Wheeler served on the City Council had “yet to be done because the funding always gets moved or another project comes up.” Id. at 5:101–02. Former Council member Harrison testified that since Mr. Isbell has been Mayor, there has been little repair work on North Pasadena’s infrastructure unless it was funded by external sources, such as Harris County. Id. at 4:176–77. In Anglo South Pasadena, by contrast, the streets are relatively smoothly paved and more quickly repaired than in North Pasadena. Streets are well lit. There are sidewalks on both sides of most streets, and they are generally in good repair. Rains that produce floods in North Pasadena do not in South Pasadena. Neighborhood associations in South Pasadena help maintain the infrastructure. Council member Wheeler testified that “it is almost another way of life in those parts of town.” Id. at 5:104. Mr. Del Toro testified that in South Pasadena, where he currently lives, there 43 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 44 of 113 are better neighborhoods, better infrastructure, better houses, and better streets and sidewalks. Id. at 5:165. c. The Effects on Voting Based on these socioeconomic factors, several witnesses with long experience in Pasadena elections gave unrebutted testimony that Latino voters in Pasadena are less likely to be familiar with and participate in Pasadena elections than Anglo voters. Because more North Pasadena residents rent rather than own, they are more transient and it is more difficult for candidates to communicate with them. Id. at 2:102–03. Mr. Del Toro believes that Pasadena Latinos are less aware of how to vote and many fear going to the polls. Id. at 5:181–83. Council member Wheeler testified that the Latino voters in his district—old District E, and new District D—are often wary or distrustful of elections and of the City Council itself. Although Mr. Wheeler campaigned to bring more services to the district he serves, many Latino voters believe that “the system is rigged against us.” Id. at 5:147–48. Council member Ybarra, who was born and raised in Pasadena, testified about campaigning in North Pasadena in 2009 and seeing some of his friends’ parents vote in municipal elections for the first time: [They had] never voted before in a municipal election[,] had no idea what a municipal election was—they didn’t know who the council members were or who the mayor was. For the first time, you saw those parents going to vote. The first time to realize, I can call my council member. If I have an issue with a department head or a city department, I have a voice. And now those same people, who never got involved because there wasn’t somebody like them up there, now get involved, now call you and tell you, ‘Hey, there is an election coming up. Which candidates do you think we should vote for?’ It has had an effect, and I think that was, along with other issues, they like to see minority candidates get elected. Id. at 2:94. 44 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 45 of 113 d. Findings on the Effect of Socioeconomic Conditions on Latino Voting The socioeconomic conditions in North Pasadena dampen Latino voter participation and contribute to making it more difficult for Latinos to elect their candidates of choice, even though when Latinos vote, they vote cohesively. The reasons include a lack of awareness of elections and a skepticism and cynicism about them, driven by the past history and present effects of discrimination, the continuing presence of socioeconomic problems facing North Pasadena, and the continuing problems with North Pasadena’s infrastructure. F. The Circumstances Surrounding the Change to Pasadena’s City Council Election Map and Plan The Pasadena City Council often votes in unison. Id. at 5:115–17. However, in recent years, as Latino North Pasadena voters elected more of their candidates of choice to Council seats, votes divided more often, particularly over resource-allocation issues. Id. at 5:116–17. The Mayor and Anglo Council members representing South Pasadena districts usually voted as a bloc to defeat the measures proposed or preferred by the Latino Council members and their allies representing North Pasadena districts. These issues included the 2012 termination of bus service in Pasadena, id. at 2:49–50, 4:180–81; the 2013 appointment of an ally of the Mayor to the City’s economic development board, id. at 2:68, 5:116–17; and the construction of a $10 million water feature in a public park located just north of Spencer Highway that residents would have to pay to use, id. at 2:66–67. Although the Mayor testified that the City Council had over the years voted in favor of certain infrastructure improvement projects in North Pasadena, these projects were far short of what was needed. At the same time, City Council voted to fund major improvements not only to basic infrastructure needs but also to recreational facilities in South Pasadena. Id. at 2:65–66, 4:179–81, 4:185. 45 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 46 of 113 Resource-allocation issues are among the most important decisions any municipal governing body must make. By 2013, Latinos in Pasadena were becoming more politically active and had elected Latino-preferred candidates to the City Council in three of the four existing Latino majority districts—Districts A, C, and D. The election of Cody Ray Wheeler, a Latino with an Anglo surname, in an Anglo-majority district, and the population changes that made this possible, made it more likely that Latinos would elect a majority of Council positions in the 2015 election. Pasadena’s “Latino voters were poised to elect their candidate of choice” and “threatened to oust the [Anglo] incumbent” in District B and deprive the Mayor of a majority of Council votes on the divisive issues that mattered most. See LULAC, 548 U.S. at 438. 1. Recent § 5 Preclearance Objections to Converting to At-Large Voting in Other Texas Municipalities Until the Supreme Court decided Shelby County in June 2016, Mayor Isbell and the Pasadena City Council knew that if they proposed a change from the eight single-member districts to a mixed single-member and at-large map and plan, they would have to go through the § 5 preclearance process with the federal Department of Justice. Mayor Isbell believed that the Department would likely block the City from implementing such a plan. Bench Trial Tr. 2:221–22, 3:61. The Department had consistently objected to similar proposals made by other cities in the recent past. On December 21, 2012, the Department of Justice objected under § 5 to a Beaumont Independent School District proposal to change the districts it used to elect school board members from seven single-member to five single-member and two at-large districts. The Department noted that it had previously blocked a Beaumont ISD proposal to consolidate with another school district and elect school board members using at-large voting. See December 21, 2012 Objection Letter from US DOJ Civil Rights Division to Beaumont Independent School District. 46 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 47 of 113 On October 3, 2011, the Department declined to withdraw its § 5 objection to a City of Galveston, Texas proposal to change its method of electing City Council members from six singlemember to four single-member and two at-large districts. The Department noted that in 1992, it had objected to a proposed map and plan of six single-member and two at-large districts. See March 5, 2012 Objection Letter from US DOJ Civil Rights Division to Galveston County. On August 12, 2002, the Department objected under § 5 to a City of Freeport, Texas proposal to change how it elected City Council members from four single-member districts to four at-large seats with numbered positions. See August 12, 2002 Objection Letter from US DOJ Civil Rights Division to City of Freeport. The Department concluded that the change would have a discriminatory effect on Latino voters by leading “to a retrogression in the position of members of a racial or language minority group (i.e., will make members of such a group worse off than they had been before the change) with respect to their opportunity to exercise the electoral franchise effectively.” Id. (citing Reno v. Bossier Parish School Board, 528 U.S. 320, 328 (2000); Beer v. United States, 425 U.S. 130, 140–42 (1976)). Before Shelby County, it was reasonable to expect the Department of Justice to object to Pasadena’s redistricting proposal because it would dilute Latino voting strength and would at least delay the ability of Latinos to elect the candidates of their choice until Latinos grew to a secure majority of the City’s voting-age population. 2. The Shelby County Decision On June 25, 2013, the Supreme Court invalidated the geographic coverage formula of § 5 of the Voting Rights Act. After the decision, Texas and its political subdivisions, including Pasadena, 47 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 48 of 113 no longer needed federal government approval before implementing changes to their voting systems. (Docket Entry No. 94 ¶ 33). Mayor Isbell gave an interview around November 4, 2013 to SCOTUSblog. He stated that he proposed changing to the 6–2 mixed map and plan promptly after the Shelby County decision “because the Justice Department can no longer tell Pasadena what to do.” Bench Trial Tr. 3:33.12 At trial, Mayor Isbell admitted that he knew that the Department of Justice sought to protect minority voters and that it had objected to changing to at-large systems that diluted minority voting strength. Id. at 3:61–62. Mayor Isbell thought that it likely that the Justice Department would have rejected the 6–2 map and plan for diluting the voting strength of Latino voters. Id. at 3:62. 3. The Shift from a Special Election to Approve a Council Bond Proposal to Amending the City Charter Two days after the Shelby County decision, on June 27, 2013, Mayor Isbell sent a memorandum to the City Council calling for a committee to consider bond proposals for a November 2013 special election and asking each Council member to nominate two candidates to serve on the committee. Pls. Ex. 8. Mayor Isbell testified that he does not remember if he had redistricting in mind when he called for the Bond-Review Committee in 2013. Bench Trial Tr. 2:219–20. The Bond-Review Committee held its first meeting, which was public, on July 18, 2013, one month after Shelby County. Pls. Ex. 10. The Committee held its second meeting, at which it discussed bond proposals, on July 23, 2013; the meeting was public. Pls. Ex. 11. At some point before July 24, Mayor Isbell asked City staff to draw up maps to see how a 6–2 district map might look. Bench Trial Tr. 3:5. The maps Mayor Isbell requested included a 4–2, a 7–1, and a 6–2 plan. 12 Mayor Isbell’s interview is available online at http://www.scotusblog.com/2013/11/after-shelby- county/. 48 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 49 of 113 Id. at 3:5–7. By July 24, 2013, the Mayor had at the ready the 6–2 map that the City would adopt in 2014. Id. at 3:11–13; Pls. Exs. 245, 246, 247. At the Bond-Review Committee’s third meeting, on July 25, 2013, Mayor Isbell proposed specific amendments to the City Charter, including converting one or more of the eight single-member districts to at-large districts. (Docket Entry No. 94 ¶¶ 34–36). Mayor Isbell made the Bond-Review Committee a Charter-Amendment Committee. The meeting was public. The Mayor distributed copies of the 6–2 map to the Committee members but did not give copies to Council members or the public. Bench Trial Tr. 3:13–15, 4:34–35, 4:41–42. Several members of the public spoke in opposition to the redistricting proposal. Id. at 4:42. Council member Ybarra stated that the map would result in losing a Latino district and that although § 5 was no longer in effect, the City still had to comply with § 2 of the Voting Rights Act. Id. at 4:39. One Committee member, Victor Villarreal, spoke against the mixed plan and in favor of maintaining the 8–0 single-member district plan. Another Committee member, Bill Welch, objected that the 6–2 plan “dilutes” Latino voting power. He pointed out that South Pasadena turned out more voters, which made dilution in North Pasadena more likely. Id. at 4:40–42. Mayor Isbell responded that 70% of Pasadena’s Latino population were illegal aliens, a figure that witnesses testified was greatly exaggerated. Id. at 2:72, 3:16–17, 4:39–40, 4:43. Mayor Isbell claimed that after the meeting, he learned that the 70% figure was wrong. Id. at 3:17. The court finds that Mayor Isbell knew of increased and increasing Latino voting strength, and therefore of larger numbers of Latino citizens, when he made the comment. Four days later, on July 29, what had been formed as the Bond-Review Committee and then turned into the Charter-Amendment Committee held its fourth and final meeting. The meeting was 49 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 50 of 113 closed. Police officers were posted at the door. Members of the public who had first heard about the Charter amendment proposal at the July 25 meeting came prepared to speak, but they were not allowed into the room. Id. at 4:44–46; Pls. Ex. 14. The Committee’s Final Recommendations document, which was sent to the Mayor and Council members, listed the Committee’s recommended bond projects and amounts for each project. Pls. Ex. 21. The document also discussed several proposed Charter amendments that might be taken up at a later date. The document did not discuss, much less recommend, the 6–2 redistricting proposal. Bench Trial Tr. 2:79–80, 3:17–18, 4:51–52; Pls. Ex. 21. According to the testimony of Victor Villarreal, a Committee member, the Committee voted 10 to 1 against recommending the 6–2 Charter amendment proposal. According to Mr. Villarreal, only Chairman Roy Mease supported the 6–2 redistricting proposal; several other witnesses also testified that Mr. Villarreal or other Committee members had told them the same thing. Bench Trial Tr. 6:6–7, 2:78–79, 5:123–124. Mr. Mease is allied and aligned with Mayor Isbell, who had nominated Mr. Mease to sit on the Pasadena Second Century Economic Development Corporation Board, which he chaired, and to replace the term-limited Port Commissioner. Id. at 3:48–49, 4:47–48. Mayor Isbell testified at trial that Mr. Mease had told him before he recommended enacting the 6–2 redistricting plan that the Bond-Review, then Charter-Amendment, Committee had supported adopting that plan in its final meeting; that every Committee member had expressed an opinion; and that 8 members were for the change and 3 against. Id. at 2:128–33. But Mayor Isbell testified in his deposition that he did not recall that anyone had told him about the Committee’s deliberations and positions. The Mayor did not correct his testimony until trial, although he read his 50 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 51 of 113 deposition well before trial and made other corrections. Id. The credible version is that Mayor Isbell at least did not know of any Committee vote in favor of changing the City Council election districts, but he nonetheless reported the support of that Committee. It is more credible that Mayor Isbell knew that except for the Chair, the Committee was against amending the City Charter to change the election map. Mayor Isbell’s testimony that he believed the Committee supported the Charter amendment was not credible. Cary Bass, a Pasadena businessman with close political and business ties to Mayor Isbell, testified. Bench Trial Tr. 4:252–54. Mr. Bass is a current Council member from District E. He was appointed to the Bond-Review turned Charter-Amendment Committee by Mayor Isbell. Id. at 4:255. He recalled little about the Committee meetings he attended; what opinions Committee members, Council members, or the Committee Chair expressed at the Committee meetings; or of any votes taken, straw or otherwise. Id. at 4:257–261. He does not remember whether members of the public were at any meetings, does not remember any discussion about how changing to the 6–2 map and plan would affect Latino voters, and does not remember receiving maps from the Mayor. Id. Mr. Bass does remember Mayor Isbell speaking in favor of the 6–2 redistricting plan. Id. at 4:262–65. Mr. Bass testified that no one on the Committee had a strong opinion about the redistricting Charter amendment. He recalled little to no dissent. Id. at 4:257–65. Mr. Bass recalled only that “it made common sense to me and probably one other that might have been in there.” Id. at 5:20–21. His testimony was not credible. His failure to remember the Committee’s discussions on changing to the 6–2 map and plan, and his failure to remember what any member specifically said, leads the court to find that Mr. Bass’s testimony that the Committee was in favor of the redistricting proposal was not credible. Compare Bench Trial Tr. 4:263–65 with 5:20–21. 51 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 52 of 113 4. Council Consideration of the Charter Amendment At a Council meeting on August 13, 2013, Mayor Isbell introduced the ordinance calling for a November 2013 election on bonds. The Council voted 5–3 on first reading in favor of putting bonds on the November ballot. Council member Ybarra was not present. Pls. Ex. 30. Council member Van Houte examined the bond projects and concluded that the four districts represented by Council members who do not agree with the Mayor on resource-allocation issues—Districts A, C, D, and E, all in North Pasadena—would get only 42.6% of the bond money, approximately $45 million, despite much greater needs than South Pasadena. The districts represented by Council members who consistently agreed with Mayor Isbell on resource-allocation issues—Districts B, F, G, and H, all in South Pasadena—would get over 50% of the bond money, approximately $60 million. Bench Trial Tr. 4:58–59. Council member Van Houte voted against the bond proposal. Id. at 4:56. Mayor Isbell sent the City Council a memo dated August 15, explaining that he was withdrawing the bond ordinance because he believed that the lack of unanimous Council support meant a failed bond election. Instead, he proposed an ordinance that would place the redistricting Charter amendment on the November 2013 special-election ballot. Pls. Ex. 22. Shortly after that, a Committee member, Larry Peacock, emailed the Mayor to say that the Bond-Review turned Charter-Amendment Committee had decided against the redistricting proposal. Pls. Ex. 23. This email appears inconsistent with Mayor Isbell’s testimony that he did not know of the Committee’s opposition to changing the map and plan for electing Council members by mid-August 2013, before the Council met to consider it. At the August 20 City Council meeting, the Council passed on first reading Ordinance 52 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 53 of 113 2013–126, which called for a Charter-amendment election to include the 6–2 redistricting proposal. The vote was 5–4, with Mayor Isbell, Steve Cote, Phil Cayten, Bruce Leamon, and Darrell Morrison—all except Mr. Leamon (District B) representing districts in South Pasadena—voting in favor of the ordinance. Council members Van Houte, Harrison, Wheeler, and Ybarra—all representing districts in North Pasadena—voted against. Bench Trial Tr. 2:84–85, 4:62, 5:126; Pls. Ex. 172. The discussion by City Council and members of the public at the August 20, 2013 Council meeting focused on how the 6–2 redistricting plan would make it harder for Latino candidates to get elected. Id. at 2:85–86. Bond-Review turned Charter-Amendment Committee member Victor Villarreal spoke against the Charter revision at the meeting, calling it divisive for the community and reporting the Committee’s 10 to 1 vote against the proposed Charter revision. Id. at 6:7–8. Council member Van Houte recorded in her notes of the meeting that the comments by members of the public all opposed the new plan. Id. at 4:60–61. The Council passed Ordinance 2013–126 on second reading. The vote divided on the same lines as before. Id. at 2:86; Pls. Ex. 173. Former Council member Harrison stated in the meeting that he thought the change would violate the Voting Rights Act. Id. at 4:193–94. Council member Van Houte also thought the change to a 6–2 plan was potentially illegal. Id. at 4:62. 5. The Campaign to Enact the Charter Amendment to Change to a Six Single-Member, Two At-Large District Map and Plan for City Council Elections a. Improper Use of City Resources to Mobilize Anglo Voters in South Pasadena In the campaign leading up to the Charter-amendment election, the Mayor’s appointees improperly used City employee time and other City resources to campaign for, and turn out votes 53 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 54 of 113 in favor of, Proposition 1, the proposal to amend the Charter to replace the 8–0 plan with the 6–2 redistricting map and plan. Mayor Isbell funded a specific-purpose political action committee known as Citizens for Positive Change to urge a “yes” vote on Proposition 1 in the November 2013 Charteramendment election. Id. at 2:88, 3:24–25. Richard Scott was the organization’s secretary, and Cary Bass was its treasurer. Id. at 4:222; Pls. Ex. 7. As treasurer, Mr. Bass was in charge of receiving and depositing donations and paying the invoices billed to the PAC. Bench Trial Tr. 4:267–69. Mr. Bass did not inform himself about the PAC’s fundraising activities or donor solicitations before he paid the bills. Mr. Bass testified that he did not ask where the money was coming from, did not initially know that the committee was formed and funded by the Mayor’s political action committee, did not know what a political action committee was, and did not research the laws or regulations that applied and that he had to comply with. Id. at 4:270–276, 4:279. He testified that he just relied on Mr. Scott’s assurances that there would be enough money to cover expenses. Id. at 4:274–75. Mr. Scott—the Mayor’s appointee as Director of Community Relations—used City resources to organize an event at a restaurant on October 3 to promote Proposition 1. Mayor Isbell’s political action committee paid for the event. Id. at 4:224–25, 4:276–277. Mr. Scott called the meeting a “kick-off” event for the Proposition 1 campaign. Id. at 4:220–22; Pls. Ex. 329. Mr. Scott sent letters and emails on his City work time and using his City email address to invite people to the October 3 event and to solicit donations for the Mayor’s Citizens for Positive Change committee, which backed Proposition 1. Id. at 4:220–22; Pls. Ex. 329. He also used his City email during work hours to send Proposition 1 mailers to be distributed to the Pasadena Chamber of Commerce, to review a draft of a yard sign, and to do other work mustering support for Proposition 1. Bench Trial Tr. 4:219–20; Pls. Ex. 44. Mr. Scott’s testimony that he worked on the Proposition 1 campaign during 54 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 55 of 113 his City work hours, using City employees working on City time to help him, was laudable and credible, given his knowledge that what he had done was improper and would, at a minimum, expose him to harsh and public criticism. Bench Trial Tr. 3:31, 4:214–16, 4:249–50; Pls. Exs. 33–36, 38, 40–42, 44, 51, 143, 327–36. Mayor Isbell was deeply involved in the Citizens for Positive Change political action committee work. Bench Trial Tr. 4:269–70, 4:274; Pls. Ex. 32. The committee reported $39,500 in expenditures by Mayor Isbell’s campaign account between July 16, 2013 and January 15, 2014, representing all of his account’s spending for that period. Bench Trial Tr. 3:24–25. Mayor Isbell commented on, and approved, drafts of campaign materials, including mailers, yard signs, and robo-call scripts, for Citizens for Positive Change. Id. at 3:25–28; Pls. Exs. 113–16. The Mayor and his appointees also used Pasadena’s Neighborhood Network Grants Program to target and mobilize voters from South Pasadena to vote for Proposition 1. The Neighborhood Network Division is a department within the City that helps neighborhoods form associations, helps with deed restrictions, and distributes grants to neighborhoods. Bench Trial Tr. 2:113–14; Pls. Ex. 347 at 32–35. Neighborhood Network matching grants are given to neighborhood groups for recreation centers, pools, basketball courts, and beautification projects. Bench Trial Tr. 2:114; Pls. Ex. 347 at 32–35. The average annual budget for the Neighborhood Network matching grants is between $75,000 and $100,000. Pls. Ex. 347 at 67. Most of the Neighborhood Network neighborhood and homeowners’ associations are in South Pasadena. Bench Trial Tr. 2:115; Pls. Ex. 31. Council member Van Houte testified that she is aware of several North Pasadena neighborhood groups that are finding it hard to get the formal City recognition needed to qualify for the matching grants program. Ms. Van Houte testified that 55 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 56 of 113 part of the problem is that the Director of the Neighborhood Networks Program devotes more time and attention and is more responsive to South Pasadena than to North Pasadena. Bench Trial Tr. 4:15–16, 4:157–58. Council member Wheeler testified that there are only two active neighborhood associations in current District D and that most of the homeowners’ associations are in South Pasadena. Id. at 5:141. Mr. Wheeler testified that after he was first elected, he contacted the Director of the Neighborhood Network Program and asked for help forming neighborhood associations in his district. Id. at 5:141–42. The Director met with Mr. Wheeler in her office and offered to send letters out and help arrange a meeting of residents. Id. Mr. Wheeler did not hear from the Director again. There is still no new neighborhood association in the district. Id. The most recent North Pasadena neighborhood association formed was in District C. This association was formed after the plaintiffs filed this suit and propounded discovery about the disparities between the North and South Pasadena neighborhood associations. Id. at 3:127; Pls. Ex. 242. Council members Van Houte and Ybarra testified that funding from the Neighborhood Network Program does not flow in equal amounts to the north side and the south side of Pasadena. South Pasadena has more neighborhood programs recognized by the City, and those programs receive more grants. Bench Trial Tr. 4:15, 2:115. From November 4–8, 2013, the Neighborhood Grants Program issued checks totaling $99,532.22 to neighborhood associations south of Spencer Highway, and $776.23 to neighborhood associations north of Spencer Highway. Id. at 4:63–69; Pls. Exs. 89, 346 at 10. Issuing the checks requires prior approval by the Grant Review Board and by the City Council. Bench Trial Tr. 4:164. There are no payments described as “Neighborhood Grants” in the City’s online check registry in the twelve months before the November 4 to 8, 2013 campaign period. (Docket No. 147). The election was held on November 5, 2013. Pls. Ex. 308. 56 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 57 of 113 The court finds that the City and Mayor used Pasadena’s Neighborhood Network Program to promote voting for the candidates and issues they favor. Bench Trial Tr. 2:115. The City and Mayor gain support from predominately Anglo South Pasadena in part by sending significant grant money and lending substantial City organizational and resource support to that part of town. Id. at 5:142–43. Homeowners’ associations are told at Neighborhood Network Program meetings and events which candidates and positions they should support. Id. at 5:86–88, 4:226. The October 3 event at a Pasadena restaurant promoting support for Proposition 1 was one such meeting. Bianca Gracia is a homeowners’ association president from South Pasadena and founder of the Latino Trump Coalition. Id. at 5:84–85. She attended the October 3 event at the invitation of the Director of the Neighborhood Network Program. The invitation billed the event as a homeowners’ association Board of Directors appreciation social. Id. at 5:85–86. When Ms. Gracia arrived, Mayor Isbell and Council members Darrell Morrison, Phil Cayten, and Steve Cote were asking the attendees to support Proposition 1, to take yard signs to display and hand out, and to talk to neighbors to promote votes for Proposition 1. Id. at 5:86–88. Council members Ybarra, Wheeler, Harrison, and Van Houte, all Latino-preferred Council members and all representing primarily North Pasadena districts, were not invited. Id. at 2:120. Council member Wheeler testified that the Neighborhood Network Program was used to help campaign for Proposition 1 and other issues. The Neighborhood Network Program has the email addresses of homeowners’ association leaders and can mobilize them easily. The Neighborhood Network Program hosts award ceremonies and gives favored neighborhoods money. Council member Wheeler testified that the message to South Pasadena residents is, “We have $10,000 for you right here. You raise this money and we are going to match it, but we need you to keep electing 57 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 58 of 113 us.” Id. at 5:144–45. Mr. Wheeler testified that the sentiment among voters in his district in 2013 was “a feeling that they were left behind, a feeling that our district wasn’t a concern, a feeling like you just see—you look at Fairmont and see new roads, new streets. You see neighborhood associations. You see so many things over there and you look at our district and we weren’t getting those things . . . like they kind of forgot about north Pasadena where Pasadena started.” Id. at 5:112–13. b. The Use of Race and Party as Proxies for Each Other Throughout the Proposition 1 campaign, Pasadena officials used partisan terms as proxies for race or racial terms. Ms. Gracia testified that at the October 3 Neighborhood Network Program meeting at the restaurant, she was offended when Council member Steve Cote said that if the voters did not vote for Proposition 1, Pasadena would “turn blue.” Id. at 5:87–88, 5:90–91. A Republican activist in the 2016 presidential campaign, she nonetheless understood the comment to mean that Pasadena would fall under the power of Latinos, coded as “Democrats.” She was the only Latino in the room. Id. When Ms. Gracia went to the next City Council meeting to voice her concerns about using the Neighborhood Network Program to promote Proposition 1, Mayor Isbell called her to his seat and asked her repeatedly, “Who sent you here?” Id. at 5:89–90. In response to the Mayor’s question, Ms. Gracia pulled up her sleeve, pointed to her skin, and asked the Mayor whether he thought the color of her skin meant that she had no voice and could not come to City Council and speak. Id. She later met with Steve Cote, who apologized for making the remark. Id. at 5:90–91. In one of his City emails campaigning for Proposition 1, Mr. Scott forwarded a Houston Chronicle article from another Pasadena employee to a Citizens for Positive Change vendor on September 6, 2013. Id. at 4:217–18. He stated that the Proposition 1 campaign “could use [the 58 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 59 of 113 article] in one of our brochures, Republicans will really like.” Id. The Chronicle article attached to the email discussed opposition to the change to the 6–2 redistricting plan as diluting Latino representation. The article did not mention Democrats or Republicans. Id.; Pls. Ex. 35. In preparing a mailing list to target voters to receive campaign materials in favor of changing to the 6–2 map and plan, Mr. Scott wrote a campaign vendor and recommended using Mayor Isbell’s campaign list from a previous campaign but asked the vendor first to “pull out Hispanic names” from the list. Pls. Ex. 34. At trial, Mr. Scott testified that when he wrote “pull out Hispanic names,” he meant to direct the vendor to pull out the names of Democratic voters. Bench Trial Tr. 4:240. When asked by the court why he said “Hispanics” if he meant “Democrats,” Mr. Scott testified that he did not know, but then testified that he thought of “Hispanic” as a proxy for Democratic voters and “Anglos” as a proxy for Republican voters. Id. at 4:242. Mr. Scott testified that he did not know how many Latinos received the mailers he sent through Citizens for Positive Change. Id. at 4:216–17. c. The Result On November 5, 2013, Pasadena voters approved Proposition 1 by a vote of 3,292 to 3,213. The voting was racially polarized.13 (Docket Entry No. 94 ¶ 51). 6. Council Approval of the 6–2 Map and Plan On February 18, 2014, the City Council passed on second reading a proposal to amend City ordinances to limit debate time in Council meetings to three minutes per Council member on each issue. Pls. Ex. 191. Mayor Isbell initiated the debate time-limit proposal, which passed by a 5–4 vote with his tie-breaking vote in favor. Id.; Bench Trial Tr. 5:130–31. Council members Van 13 See Part II.E.4.c. above. 59 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 60 of 113 Houte, Harrison, Wheeler, and Ybarra opposed the measure. Bench Trial Tr. 4:72, 4:194–95, 5:130–31; Pls. Ex. 191. Mr. Wheeler testified that he believed the Mayor spearheaded the debate time limit because “council members [we]re making some very good arguments that were in opposition to the mayor’s policies. And often they went over two minutes and I think the mayor did not like hearing opposition to him. I also think that he was about to propose a plan that was very controversial, very complex and required a lot of information, so I think he wanted to muffle council members and also reduce the amount of speech that we had during public comment.” Bench Trial Tr. 5:131–32. The City held a public “Redistricting Workshop” on March 4, 2014. The City Council discussed the proposed redistricting maps and plans the Mayor commissioned from the law firm of Bickerstaff, Heath, Delgado and Acosta, LLP. Pls. Ex. 187. Council member Van Houte testified that she would have liked, but was not permitted by the Mayor, to discuss the maps and plans with the lawyers and consultants who prepared and reviewed them for the Mayor. Bench Trial Tr. 4:70. At a March 18, 2014 City Council public hearing on redistricting, Council members and members of the public raised concerns about the legality of the 6–2 maps. (Docket Entry No. 94 ¶ 53). Mayor Isbell brought a pistol to that hearing. Council member Cody Ray Wheeler saw the pistol when the Mayor dropped it. Bench Trial Tr. 5:132–35. City rules forbid Council members or others to bring guns to Council meetings. Id. at 5:134. The Mayor testified at trial that the weapon was a pellet gun; that it was a pellet gun that was broken; and that he brought it because he had received a threatening message at the company where he worked. Id. at 3:37–41. The credible evidence, including Council member Wheeler’s testimony, is that the gun was a 9-millimeter Beretta. Id. at 5:132–35. Mr. Wheeler not only grew up around firearms, he also served for four years in the 60 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 61 of 113 Marine Corps, where he worked in the Armory. He credibly testified that the 9-millimeter Beretta is Marine standard issue, that he both shot this weapon and saw hundreds of them during his military service, and that the weapon the Mayor dropped was the same firearm. Id. at 5:132–35. Both Mayor Isbell and Council member Wheeler testified that after the Mayor dropped the gun, he quickly picked it up and left the room. Id. at 3:41–42 (Isbell), 5:134 (Wheeler). Mr. Wheeler thought it likely that Mayor Isbell had brought the gun to this meeting because the topic was the controversial one of redistricting. “It is my personal belief that it was, that he knew what he was proposing was controversial [and] he had altered city rules.” Id. at 5:135. This is a more credible account of why the Mayor brought the weapon to the meeting than the Mayor provided. The Mayor testified at trial that he carried it because a threatening message from a woman against him and his family was left at the oil company where he worked. Id. at 3:37–41. In his deposition, the Mayor testified that he did not know the source of the threats that caused him to carry a gun on March 18, and did not know if it was a male or female. Id. at 3:40. In short, neither the trial testimony describing the type of gun he carried or the reason he did so are credible, although this evidence is only marginally probative of the issues before the court. At the April 1, 2014 City Council meeting, Council member Pat Van Houte expressed concerns over the legality of the redistricting plan. Council member Van Houte exceeded her threeminute debate-time limit. Mayor Isbell ordered police officers to remove her from the Council chambers. (Docket Entry No. 94 ¶ 54); Bench Trial Tr. 2:112–13, 4:196; 5:136. When Council member Van Houte was removed, Council members Harrison, Wheeler, and Ybarra also walked out in protest. Id. Mr. Ybarra walked out because he felt that Mayor Isbell had given more time to Council members and others speaking at meetings before the redistricting proposal, on topics far less 61 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 62 of 113 important. Bench Trial Tr. 2:112. Mr. Wheeler explained that he “felt that everything up to this point, the [election contest] lawsuit, the negative campaigning, the proposal for the redistricting, the weapon being brought, the way that we are limited in speech and then he kicks her out of the meeting, the whole process felt like a sham. And the fact that even the charter committee voted against it, everything felt like it was a sham. And it didn’t matter what anybody did or what the people wanted, the mayor was going to get his way. And I walked out in support of Van Houte and walked out in protest of the whole process.” Id. at 5:136. The 6–2 redistricting map and plan was passed on first reading with only the four remaining Council members and the Mayor in attendance. (Docket Entry No. 94 ¶ 54). On April 15, 2014, Council’s second-reading vote on the 6–2 map and plan divided 5 in favor and 4 against. Mayor Isbell and Council members Leamon, Morrison, Cayten, and Cote voted “aye,” and Council members Ybarra, Wheeler, Van Houte, and Harrison voted “nay.” Id. at 2:113; Pls. Ex. 185. G. The Impact of Pasadena’s New Election Map and Plan 1. The Previous 8–0 Map and Plan Pasadena adopted its previous 8–0 single-member redistricting map and plan in 2011 and held one election under them in May 2013. In that municipal election, Latinos elected their candidates of choice in three of the four Latino-majority districts: A (Ybarra), C (Harrison), and D (Van Houte). In Anglo-majority District E, a Latino candidate with an Anglo surname, Cody Ray Wheeler, was elected by a 32-vote margin. Bench Trial Tr. 5:122. District E in the 8–0 plan was not a Latino-majority district.14 It consisted of a 45.8% 14 “Latino-majority district” indicates that a district has a Latino citizen voting-age population greater than 50%. 62 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 63 of 113 Hispanic voting-age population and 44.6% Spanish-surnamed registered voters. Pls. Ex. 195-3. In his 2013 race for City Council District E, Cody Ray Wheeler’s campaign increased total voter turnout to 837 from 458 (in the District E 2011 runoff election). Bench Trial Tr. 5:107–08; Pls. Exs. 237, 238 at 21. Council member Wheeler testified that his Anglo surname helped him get Anglo crossover votes. Bench Trial Tr. 5:110. Other experienced politicians in Pasadena believe that Mr. Wheeler’s victory in 2013 was the result of his Anglo-sounding name and the Anglo votes that attracted. Id. at 2:124–26, 4:196–97. In 2013, District B elected an Anglo-preferred candidate, Bruce Leamon, who aligned with the Mayor and who had the support of the Mayor’s political action committee. Id. at 5:118–19. Council members Ybarra, Wheeler, and Van Houte testified that, given District B’s strong majority of Latino citizen voting-age population and Spanish-surnamed registered voters, the District was likely to elect a Latino-preferred candidate in the 2015 election. Id. at 2:128–130, 5:118–19, 4:141. They also testified that, if the 8–0 district lines remained, all four Latino-majority Council districts would elect a Latino-preferred candidate in 2015 and the incumbent Cody Ray Wheeler would be reelected in Anglo-majority District E. Id. As a result, they believed that in 2015, Latino-preferred candidates would hold a majority of seats on the Pasadena City Council for the first time in the City’s history. Id. at 5:118–19. The court finds that Mayor Isbell and the Pasadena City Council believed this as well. The court finds that when the Mayor proposed changing to the 6–2 redistricting map and plan, when the City Council voted to put it on the November 2013 ballot, and when the Council voted to approve its enactment, City officials knew that unless the district lines were changed, a majority of the City Council members would be Latino-preferred candidates by 2015 or shortly 63 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 64 of 113 thereafter. 2. The 2015 Election under the Current 6–2 Map and Plan In the 2015 City Council election held under the 6–2 redistricting map and plan, Latinomajority District A elected Ornaldo Ybarra and Latino-majority District C elected Sammy Casados. Pls. Ex. 239. Latino-majority District B once again elected the Anglo-preferred candidate, Bruce Leamon, this time by 34 votes. Id. Anglo-majority District D elected Cody Ray Wheeler, who ran as an incumbent in the parts made up of the old District E. Id. In the 2015 at-large race for Place G, the Latino-preferred candidate, Pat Van Houte, defeated Steven Cote by a slim 143-vote margin. Id. Both candidates in the Place G race were Anglo and former City Council incumbents. In the 2015 at-large race for Place H, the Latino-preferred candidate, Oscar Del Toro, was defeated by Darrell Morrison. Id. Mr. Del Toro is Latino and Mr. Morrison is Anglo. Council members Wheeler and Van Houte relied on significant Anglo crossover voting to win election: 55.1% of nonLatino voters supported Mr. Wheeler in District D and 41.6% of non-Latino voters supported Ms. Van Houte in Place G. (Docket Entry No. 94 ¶¶ 65, 77). After the change to the 6–2 redistricting map and plan, the number of Latinos voting in Pasadena elections from November 2013 to May 2015 decreased. Despite rising Latino voterregistration rates, the number of Spanish-surnamed registered voters who cast ballots in 2013 was 1,161, compared to 1,015 in 2015. Pls. Exs. 344, 345. Council member Wheeler testified that the change in the redistricting map and plan from 8–0 to 6–2 was disheartening to many Pasadena Latinos, who were already wary or distrustful of the City’s governance. Bench Trial Tr. 5:147–48. Although Mr. Wheeler had campaigned to bring more services to his district, seeing the Mayor redraw the lines “unfortunately [] confirms their fears that the system is rigged against us. We win 64 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 65 of 113 an election, but, you know what, they move the field goal back and move it back farther and farther and they don’t have the same support.” Id. Latino voter registration in Pasadena dramatically rose from 2015 to 2016. The City’s expert Dr. Alford testified that an uptick in registration is normal in presidential election years, but the increase in Latino registration in 2016 was “dramatically larger” than expected. Id. at 6:30. From 2011 to 2016—that is, from one presidential election to another—non-Spanish-surname voter registration decreased by an estimated 1,434 voters. Over the same period, Spanish-surname voter registration increased by 6,255 voters citywide. Def. Ex. 43. Dr. Alford believed that the unusual features of the 2016 presidential campaign, including a candidate perceived as especially hostile to immigration, drove these changes. Bench Trial Tr. 6:30–31. He also believed that the 2016 campaign and the consequent rise in Latino registration was unforeseen and unforeseeable when Pasadena adopted the 6–2 voting map and plan in 2013 and 2014. Id. at 6:31–32. Despite the recent and dramatic rise in Latino registration, Dr. Alford and Mr. Ely agreed that the district-by-district majorities remain unchanged. Under the 6–2 map, Spanish-surnamed registered voters account for a majority in three districts and around 45% of the registered voter population in a fourth district. Def. Ex. 48. Were the 8–0 map in place today, Spanish-surnamed voters would account for a majority in four districts and around 45% of the registered voter population in a fifth district. Bench Trial Tr. 1:75; Pls. Ex. 195. H. Summary on Findings of Fact The court finds that Pasadena has shared in Texas’s long history of voter discrimination; has a more recent, and more probative, history of official discrimination in City administration and policing; and has had recent Council election campaigns marked by racial discrimination. 65 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 66 of 113 The Latino population and the Latino registered-voter population have been rising in recent years. But Latinos do not yet constitute a majority of Pasadena’s citizen voting-age population or registered-voter population. Latinos make up 48.2% of the citywide citizen voting-age population. City Council elections in Pasadena are racially polarized, and in districts in which Latinos do not make up the majority of the citizen voting-age population, Latinos need a significant—more than 30%—Anglo crossover vote to elect their preferred candidates. In most elections, Anglo citizens vote sufficiently as a bloc to deny Latinos the election of their preferred candidates. Pasadena’s Latino population remains socioeconomically depressed relative to the Anglo majority population, in part as the result of lingering effects of official and private discrimination in the past and the effects of official unresponsiveness to infrastructure needs in the present. Poverty, overcrowding, lack of education, lack of job opportunity and advancement, and especially the feeling that City officials ignore or undermine voting and voter-mobilization efforts, further depress Latino voter turnout. Pasadena’s Mayor initiated the change to the 6–2 voting map and plan immediately after the Supreme Court’s Shelby County decision removed the requirement for Pasadena to obtain federal Department of Justice preclearance. Mayor Isbell knew that the Department did not preclear at-large systems that dilute the voting strength of minorities, had not precleared nearby jurisdictions that had proposed dilutive at-large systems, and likely would not preclear Pasadena’s map and plan because of its dilutive effect. Despite opposition from the public, four City Council members, and the Committee tasked with reviewing the proposed change to the voting map and plan, Mayor Isbell persisted in putting the 6–2 map and plan on the November 13 ballot through 5 to 4 votes on the City Council, proposals for which Mayor Isbell cast the tie-breaking votes. 66 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 67 of 113 During the November 2013 campaign, the Mayor’s appointees improperly used City resources to promote Proposition 1. Richard Scott, the Director of Community Relations, used City email and directed City employees on City time to work on the Proposition 1 campaign. The Mayor and four Council members used Pasadena’s Neighborhood Network Program to mobilize voters and resources in the predominantly Anglo South Pasadena to support Proposition 1. Through his political action committee, Mayor Isbell contributed at least $39,000 to helping the 6–2 redistricting map and plan pass, which it did by a narrow vote. Latino voters opposed the 6–2 map and plan with more than 99% of their votes, but Anglo bloc voting and turnout were sufficient to defeat the clear Latino preference. Pasadena officials used race and political party as proxies for one another in the November 2013 election. Mr. Scott forwarded a newspaper article about racial vote dilution, yet described the article in terms of party affiliation. On another occasion, he directed a subordinate to “pull out the Hispanic names” from a campaign list, explaining at trial that by “Hispanic” he meant “Democrat.” A homeowner’s association president understood the threat of Pasadena “turn[ing] blue” to be a reference to the need to stop the growth of Latino political power. The Council sessions that led to the enactment of the 6–2 map and plan were marked by procedural irregularities. A measure to limit speaking time in City Council meetings was passed shortly before the redistricting proposal was likely to come before the Council. A Council member speaking in opposition to the 6–2 map and plan was removed by police officers for speaking longer than three minutes. After three other Council members departed in protest, the Mayor and remaining Council members proceeded to pass the 6–2 map and plan. The Mayor brought a firearm to one Council session at which the voting change was being debated. 67 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 68 of 113 The result of the change in the voting map and plan was the elimination of a Latino-majority district and the creation of two at-large places in which Latinos are in the minority of citizen votingage population and registered voters. Under the previous 8–0 single-member district map and plan, Latinos were the citizen voting-age majority in four districts and nearing 46% in a fifth district. Under the 6–2 map and plan, Latinos are the citizen voting-age majority in three districts and nearing 47% in a fourth district.15 Without significant crossover Anglo voting, Latinos cannot elect the candidates of their choice in proportion to their population. In the 2015 elections, Latino-preferred candidates who were incumbent Council members with Anglo surnames secured sufficient crossover voting to win reelection to the Council. The at-large Latino-preferred candidate with a Spanish surname fell far short of victory (39%), despite large support by Latino voters (87%).16 In short, Pasadena’s elections are racially polarized. The City’s 2013 racially polarized vote in favor of the 6–2 redistricting map and plan and the Council’s 2014 vote to approve the change were narrowly decided. The effect was to dilute Latino voting strength. That effect was foreseeable and foreseen. III. Conclusions of Law on § 2 of the Voting Rights Act A. Gingles Step One 1. The First Two Conditions: The Minority is Sufficiently Numerous and Geographically Compact The parties agree, and the court finds, that the first two conditions of Gingles step one are satisfied. (Docket Entry No. 94 ¶¶ 18, 57). Latinos in Pasadena are in the minority but are sufficiently numerous and live in areas that are sufficiently compact to make up a majority in single15 See Part II.E.3.–4. above. See also Pls. Ex. 195-3. 16 See Part II.E.5. above. 68 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 69 of 113 member districts, and Pasadena’s Latino minority is politically cohesive. See Gingles, 478 U.S. at 48–49. The experts agree. Pls. Ex. 198 at 10 (Dr. Engstrom); Bench Trial Tr. 6:171 (Dr. Alford). 2. The Third Condition: The Majority Votes as Bloc Sufficiently to Defeat the Minority’s Preferences The parties agree that Latinos make up nearly half of Pasadena’s citizen voting-age population, and that representation in the City Council proportional to these numbers would be four out of eight seats. Under the first condition of Gingles step one, proportionality in this context is a simple calculation: only “minority-majority” districts, or districts in which a racial minority in a jurisdiction makes up the majority of the district’s citizen voting-age population, count in the proportionality analysis, and proportionality is measured against the minority’s citizen voting-age population. Bartlett v. Strickland, 556 U.S. 1, 13–20 (2009) (plurality). The City argues that this proportionality measure does not apply to the third condition of the Gingles first step, because under that condition, courts are to consider whether the majority votes as a bloc to defeat the minority’s preferred candidates. The City argues that Anglo crossover voting for Latino-preferred candidates in Pasadena turns a district that does not have a Latino citizen voting-age or Spanish-surnamed registered voter population into a minority-opportunity district for the purpose of finding proportionality and precluding a dilution finding. The plaintiffs prevail under either the Court plurality’s bright-line rule or the defendants’ suggested approach to the legal standard for the Gingles third condition. First, the bright-line rule the Court plurality uses for the first Gingles condition. The parties agree that under the 8–0 singlemember district map and plan in place from 2011 to 2014, Latinos made up, and today would make up, the majority of citizen voting-age population and the majority of Spanish-surnamed registered voters in four districts—Districts A, B, C, and D. The parties and experts agree that under the 69 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 70 of 113 current 6–2 map and plan, Districts A, B, and C are Latino-majority and -opportunity districts. (Docket Entry No. 94 ¶ 56). That is one fewer than under the single-member district plan, and that is dilution. The City relies on the American Community Survey’s 2015 one-year estimate that Pasadena has a Hispanic citizen voting-age population of 50.6%, with a margin of error of 3.3%, to argue that the two at-large places in the 6–2 map and plan are also Latino-majority and -opportunity districts. But the court finds that the clearly more reliable figure for Pasadena’s Latino citizen voting-age population is the American Community Survey’s 2011–2015 five-year estimate of 48.2%, with a margin of error of 1.6%.17 Under the previous 8–0 single-member district map and plan, Latinos were the citizen voting-age majority in four districts and nearing 46% in a fifth district (District E). Under the 6–2 map and plan, Latinos are the citizen voting-age majority in three districts and nearing 47% in a fourth district (District D).18 Again, by simple arithmetic, this is dilution. The City argues that sufficient Anglo crossover voting in District D and the two at-large places means Latinos have three additional opportunity districts under the 6–2 plan, for a total of six opportunities out of eight to elect their Council members of choice. Although the legal standard is electoral opportunity, not success, the City points to the election of Latino-preferred candidates in 2015 in District D (Wheeler), and at-large Place G (Van Houte) as evidence of proportional—and therefore undiluted—representation. The City argues that these elections prove that Anglos do not sufficiently vote as a bloc to defeat Latino preferences, and therefore the third condition of Gingles is not satisfied. 17 See Part II.E.2. above. 18 See Part II.G. above. 70 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 71 of 113 The court disagrees with the City’s reasoning and case-law interpretation and application. In Gingles, the Supreme Court found that majority voting against minority-preferred candidates was legally significant when it ranged between 51% and 72% in general elections. See Gingles, 478 U.S. at 59. The amount of white bloc voting that can generally ‘minimize or cancel,’ [minority] voters’ ability to elect representatives of their choice . . . will vary from district to district according to a number of factors, including the nature of the allegedly dilutive electoral mechanism; the presence or absence of other potentially dilutive electoral devices, such as majority vote requirements, designated posts [i.e., a place system], and prohibitions against bullet voting [i.e., single-shot voting]; the percentage of registered voters in the district who are members of the minority group; the size of the district; and, in multimember districts, the number of seats open and the number of candidates in the field. Id. at 56 (internal citations omitted). Several of the dilutive devices described by the Gingles Court are present here, including majority-vote requirements, the use of a place system in the at-large districts, and the prohibition of single-shot voting.19 The 6,200-person increase to the district sizes that resulted from changing to the 6–2 map and plan is also dilutive.20 The record also clearly supports finding and concluding that Anglo voters consistently oppose Latino preferences in Pasadena City Council and other elections by voting as a bloc. In the citywide vote on Proposition 1 to create the 6–2 map and plan, Latinos opposed the measure by 99.6%. But 60% of Anglo voters approved the measure, which passed. In Oscar Del Toro’s race in at-large Place H against an Anglo opponent, Mr. Del Toro had 87.3% of the Latino votes but 71.7% of the Anglo vote went to his Anglo opponent, who won. In the at-large Place G race, Pat Van Houte, the Latino-preferred candidate and incumbent Council member, received 70.6% of the 19 See Part II.C. above. 20 See Part II.E.4. above. 71 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 72 of 113 Latino vote but 41.6% of the Anglo vote went to her Anglo opponent. She won by only 143 votes out of 4,152 votes cast.21 The Anglo voting patterns established on this record do not avoid a finding and conclusion of dilution, as the City argues. To the contrary, the record clearly shows that Pasadena’s Anglo voters usually vote as a bloc to defeat Latino-preferred candidates and policies. When a Latinopreferred candidate does win in a district that is not a Latino-majority district, it is because of serendipitous factors—such as incumbency and an Anglo-sounding last name (Cody Ray Wheeler’s victory in 2015 in District D), or the pairing of two Anglo candidates and incumbency (Ms. Van Houte’s narrow victory in at-large Place G in 2015). See Gingles, 478 U.S. at 76 (“Where multimember [i.e., at-large] districting generally works to dilute the minority vote, it cannot be defended on the ground that it sporadically and serendipitously benefits minority voters.”). The Gingles test looks to whether “the white majority votes sufficiently as a bloc to enable it—in the absence of special circumstances, . . . usually to defeat the minority’s preferred candidate.” Gingles, 478 U.S. at 51 (emphasis added). The City relies on two successful races of Latinopreferred candidates to defeat finding dilution. Both races clearly involved special circumstances that prevented the defeat of the Latino-preferred candidate. In both the 2015 District D and Place G races, the Latino-preferred candidate was an incumbent with an Anglo surname—Mr. Wheeler and Ms. Van Houte. In the same election, the at-large Latino-preferred candidate with a Spanish surname, Oscar Del Toro, lost with 71.7% of Anglo voters in opposition. Both parties’ experts testified that Mr. Wheeler and Ms. Van Houte needed a large share of Anglo crossover voting to win their seats. The evidence shows that both earned that share in part because of their incumbency and 21 See Part II.E.5. above. 72 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 73 of 113 in part because of their Anglo surnames.22 And Ms. Van Houte won by the proverbial hair. If Latinos in Pasadena can—barely—elect their candidates of choice only if those candidates are incumbents with Anglo surnames, then Latinos in those districts do not have an equal opportunity to elect candidates of their choice. See LULAC, 548 U.S. at 428–29 (“Since the redistricting prevented the immediate success of the emergent Latino majority in District 23, there was a denial of opportunity in the real sense of that term.”); Texas v. United States, 831 F.Supp.2d 244, 261 (D.D.C. 2011) (“ability” to elect is not interchangeable with “opportunity” to elect); cf. Bartlett, 556 U.S. at 13–20 (plurality) (courts should not consider crossover voting to find minority-opportunity districts under the first condition of Gingles step one). Pasadena argues that even if Latinos have lost one majority-opportunity district in the move from an 8–0 to a 6–2 map and plan, the plaintiffs have not stated a claim under § 2 of the Voting Rights Act because § 2 forbids only dilution, not retrogression, and because demographic trends in the City put Latinos on the cusp of gaining a citizen voting-age majority in District D and the two at-large places. These arguments are not persuasive. The City is correct that § 2 does not prohibit retrogression. Dilution under § 2 is measured against a hypothetical undiluted practice rather than against a municipality’s prior practice (as it would be in a § 5 analysis). See Reno v. Bossier Par. Sch. Bd., 520 U.S. 471, 480–81 (1997); see also Little Rock Sch. Dist. v. Pulaski Cty. Special Sch. Dist. No. 1, 56 F.3d 904, 910 (8th Cir. 1995) (district court erred in a § 2 case by comparing a challenged redistricting plan to the prior plan, mistaking retrogression for dilution). But in this case, both dilution and retrogression analyses lead to the same result. From 2013—the date of the last election under Pasadena’s 8–0 map and plan—to 22 See Part II.E.5. and II.G.2. above. 73 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 74 of 113 the present, Latinos have made up just under half of Pasadena’s citizen voting-age population. Proportionality would require an opportunity to elect Latino candidates of choice for four out of eight seats on the City Council—or at least four out of nine, if one includes the Mayor’s seat. Under the 8–0 map and plan, Latinos had that opportunity because they made up a clear majority of citizen voting-age population in four of the eight districts. Under the 6–2 map and plan, they have similar opportunities—because of similar majorities—in only three out of eight districts.23 The change to Pasadena’s voting map and plan drops Latinos’ proportional opportunity for representation from around 44%, a close approximation of Latinos’ citizen voting-age population in the City, to 33% (both figures counting the Mayor’s seat on the Council). The plaintiffs are not impermissibly pursuing a retrogression claim. They have properly identified dilution as the result of changing to a 6–2 mixed single-member and at-large map and plan to elect City Council members. The City relies on its statistics expert, Dr. Alford, to argue that Latinos will, if demographic trends hold, be poised to elect a majority within the next few years. That reasoning does not avoid finding impermissible dilution. At a minimum, the change away from a single-member districting plan to the mixed plan delayed the emergence of a Latino-preferred Council majority in Pasadena. The fact that the dilutive effects of the 2014 change will some time in the future be overcome by population growth trends does not defeat finding dilution. Under § 2, a municipality cannot defend diluting a minority’s voting power with the argument that demographic shifts might or even will eventually tilt the balance in the other direction. Nor does the fact that the Latino population in Pasadena continues to grow avoid a finding and conclusion of dilution and racially polarized Anglo 23 See Part II.E.3.–4. and Part II.G. above. 74 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 75 of 113 bloc voting that generally defeats Latino-preferred candidates. If potential or speculative demographic shifts in the future could defeat a § 2 claim, “voting rights cases [could] be defeated at the outset by the very barriers to political participation that Congress has sought to remove.” Perez v. Pasadena Indep. Sch. Dist., 958 F.Supp. 1196, 1220–21 (S.D. Tex. 1997), aff’d, 165 F.3d 368 (5th Cir. 1999) (quoting Clark v. Calhoun Cty., Miss., 88 F.3d 1393, 1398 (5th Cir. 1996)) (internal quotation marks omitted). Polarized voting is to be judged “primarily on the basis of the voting preferences expressed in actual elections.” Gomez v. City of Watsonville, 863 F.2d 1407, 1415 (9th Cir. 1988). Pasadena Latinos were poised to elect a majority of the City Council for the first time in the City’s history under the 8–0 map and plan. That imminent power shift away from the Mayor and the Anglo majority, and the end of the § 5 preclearance requirement, led the Mayor and City Council to propose and approve changing from the 8–0 single-member district plan to the mixed 6–2 map and plan. Cf. LULAC, 548 U.S. at 428–29. The three conditions of Gingle’s first step are satisfied. Pasadena Latinos are a large, geographically compact, and politically cohesive minority. Pasadena Anglos usually vote as a significant bloc to defeat Latino candidates, absent special circumstances. B. Gingles Step Two “If the Gingles three-part threshold is met, plaintiffs must then also show that under the ‘totality of the circumstances,’ plaintiffs do not possess the same opportunities to participate in the political process and elect representatives of their choice enjoyed by other voters.” Perez, 958 F.Supp. at 1201 (citations omitted). But “it will be only the very unusual case in which the plaintiffs can establish the existence of the three Gingles factors but still have failed to establish a violation 75 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 76 of 113 of § 2 under the totality of the circumstances.” Clark v. Calhoun Cty., Miss., 21 F.3d 92, 97 (5th Cir. 1994) (internal quotation marks and citation omitted). While there is no requirement that “any particular number of [Senate] factors be proved, or that a majority of them point one way or the other,” S. Rep. at 29, the court finds and concludes that nearly all the Senate factors show that the plaintiffs do not have the same opportunities to participate in the political process and elect representatives of their choice as Anglo voters in Pasadena. 1. History of Discrimination Two Senate factors touch on historical discrimination: (1) history of official discrimination affecting the right of a member of a minority group to participate in the democratic process; and (5) the extent to which minority group members in the state or political subdivision bear the effects of past discrimination in areas such as education, employment, and health, which hinder their ability to participate effectively in the political process. Both factors weigh in favor of the plaintiffs. In LULAC, the Supreme Court noted that the “‘[t]he long history of discrimination against Latinos and Blacks in Texas’ . . . may well ‘hinder their ability to participate effectively in the political process’” and found a § 2 violation under the totality of the circumstances. LULAC, 548 U.S. at 439–40, 442 (quoting Session v. Perry, 298 F.Supp.2d 451, 473 (E.D. Tex. 2004) and Gingles, 478 U.S. at 45). Texas in general and Pasadena in particular have a long history of discriminating against Latinos. As Dr. Tijerina concluded, Pasadena’s history of discrimination in voting and segregation in housing, education, and employment have left a “legacy of fear, alienation and a lower participation in voting and other practices of democracy.” Bench Trial Tr. 1:182–83. The history of discrimination provides not only context but also evidence of lingering effects. The court’s focus must be on these present effects and on continuing contemporaneous 76 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 77 of 113 discrimination. See Veasey, 830 F.3d at 232. The court credits Council member Ybarra’s testimony that many of his Latino constituents were first-time voters in his 2009 campaign and, because of past discrimination and exclusion from the political process, were unfamiliar with election procedures and candidates for office.24 The court credits testimony by Dr. Tijerina, Council member Ybarra, and former Council member Harrison that Pasadena officials from the 1990s to the present have engaged or permitted discriminatory conduct in law enforcement, including Mayor Isbell’s direction to police to automatically impound cars driven by Latinos without proof of insurance but use their discretion when the driver is Anglo.25 The court credits testimony by Council member Wheeler, Council member Ybarra, and Mr. Del Toro that their election campaigns in 2009, 2013, and 2015 respectively were marked by racial animus, although far from widespread, in the same electorate that narrowly approved changing to the 6–2 map and plan.26 The continuing effects of past and contemporaneous official and private discrimination are among the complex reasons for lower voter registration and turnout among Latinos in Pasadena. 2. Racially Polarized Voting and Dilutive Measures Two Senate factors functionally repeat the third condition of the Gingles threshold analysis: (2) the extent to which the political body’s voting is racially polarized; and (3) the extent to which the body uses voting practices or measures, such as majority-vote requirements or anti-single-shot provisions, to enhance the opportunity for discrimination against the minority group. For the reasons explained above, these factors weigh in favor of the plaintiffs. 24 See Part II.E.6.c. above. 25 See Part II.D.2. above. 26 See Part II.D.3. above. 77 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 78 of 113 Pasadena’s elections have been, and continue to be, racially polarized. Anglo bloc voting usually defeats Latino preferences in districts in which Latinos do not make up the majority of citizen voting-age population or the Spanish-surnamed registered voter population.27 Pasadena’s 6–2 map and plan has a majority-vote requirement, uses a place system, and prohibits the use of single-shot voting.28 The extent of racially polarized voting and the use of dilutive election procedures and devices contribute to the finding and conclusion that Latino voters in Pasadena do not yet have equal opportunities for political participation. 3. Racial Appeals in Political Campaigns The sixth Senate factor requires the court to ask whether political campaigns have been characterized by overt or subtle racial appeals. This factor weighs in the plaintiffs’ favor, although slightly. The court credits the testimony of Council member Wheeler that Leroy Stanley, his opponent in the 2013 election in District E, used overt racial appeals in his door-to-door campaigning. Mr. Stanley’s campaign labeled Mr. Wheeler as “a Mexican” and warned voters not to be “fooled by his name” into thinking he was Anglo.29 Director of Community Relations Richard Scott’s instruction to a campaign vendor to “pull out Hispanic names” from the list of those receiving Proposition 1 campaign materials, while not a direct appeal, is also relevant.30 27 See Part III.A.2. above. 28 See Part II.C. above. 29 See Part III.D.3. above. 30 See Part II.F.5.b. above. 78 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 79 of 113 The evidence indicates that overt racial appeals in campaigns were neither frequent nor routine. Even subtle appeals are uncommon. But they are occasionally a feature. Most recently and most importantly, these incidents occurred in a racially polarized election in District E, where Mr. Wheeler ran against an Anglo opponent, and in the racially polarized Proposition 1 campaign. These campaigns contribute, although not heavily, to the finding and conclusion that Latino voters do not have equal opportunities of political participation in Pasadena. 4. Successful Elections of Minority Candidates Two Senate factors concern the past success of minority candidates: (4) whether minoritygroup members have been denied access to a candidate-slating process, if there is one; and (7) the extent to which minority-group members have been elected to public office in the political body. These factors neither weigh in favor of the plaintiffs nor the defendants. Pasadena does not use a formal slating process, but Mayor Isbell’s political action committee, Citizens to Keep Pasadena Strong, slates candidates for City Council. In 2013 and 2015, Citizens to Keep Pasadena Strong slated a combined eight candidates; two of them, Rick Guerrero (endorsed in 2013) and Emilio Carmona (endorsed in 2015) were Latino. Bench Trial Tr. 3:73–76. Before 2009, only two Latino candidates were elected to City Council, Roy Ybarra and Emilio Carmona. Since 2009, under the 8–0 single-member map and plan, Pasadena elected two City Council candidates who identify as Latino (Ornaldo Ybarra in District A; Cody Ray Wheeler in District E). In 2015, under the current 6–2 map and plan, Pasadena elected another Latino candidate to City Council, Sammy Casados, who ran against Emilio Camona for District C. Id. at 3:76–77. Latino success in Pasadena elections has been slow, slight, and disproportionately lower than 79 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 80 of 113 Pasadena’s Latino population. But Latino success has increased in recent years. Indeed, the plaintiffs’ theory of this case, which the court finds credible and amply supported by reliable record evidence, is that Pasadena changed to the 6–2 map and plan precisely because Latinos were becoming more successful at winning City Council seats.31 Those successes mean this factor does not weigh significantly in the plaintiffs’ favor. At the same time, the long dearth of Latino representatives on the City Council, and the steps taken to reduce the number of Latino and Latinopreferred candidates, means that this factor cannot weigh in the City’s favor either. 5. Officials’ Responsiveness to Minority Concerns Besides the enumerated factors, the Senate Report requires a court to ask whether there is a significant lack of responsiveness by elected officials to the minority group members’ particularized needs. This factor weighs in favor of the plaintiffs. The “provision[] of municipal services to neighborhoods populated by minority group members” is relevant to this factor. David v. Garrison, 553 F.2d 923, 929 (5th Cir. 1977). Council members Pat Van Houte, Orlando Ybarra, and Cody Ray Wheeler, and former Council member Don Harrison, all testified that the City’s Neighborhood Network Program department has not been responsive to North Pasadena requests to help establish associations there, although there is a budget, an organization, and a program for doing so. Documents show that the Neighborhood Grants Program issued checks for $99,532 to South Pasadena associations from November 4 to 8, 2013, just before the election, while distributing $776.23 to North Pasadena associations. These were the only Neighborhood Grants Program checks issued in the twelve months before the 2013 City Council 31 See Part II.G.2 above. 80 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 81 of 113 election.32 The timing of the issuance of the Neighborhood Grants Program checks, and the fact that almost all went to predominately Anglo South Pasadena, support the finding and conclusion that the City Council and City discriminate against North Pasadena in ways that reinforce the inequitable distribution of resources and that are politically advantageous to the Mayor and his Council allies. The City is not as responsive to requests to make repairs in North Pasadena as it is in South Pasadena, despite the greater need for repairs in that part of town. The Neighborhood Network Program sends most of its funding to South Pasadena projects, despite greater need in the north part of town.33 The record also demonstrates the Mayor and Council’s use of the Neighborhood Network Program for political campaigns to reward relatively resource-rich and predominately Anglo South Pasadena for political gain. The timing of the release of the checks to South Pasadena, right before the 2013 election, is one indication. More importantly, the record shows that the Mayor and those running for office in South Pasadena used the Neighborhood Network Grants Program and homeowners’ associations to organize voters and turn out the vote, particularly in municipal elections that are typically low-turnout such as the Proposition 1 election. The Neighborhood Network Program organizes meetings of the homeowners’ associations and uses those meetings to drum up support for specific candidates and positions.34 Council votes on resource allocation between North and South Pasadena are most likely to 32 See Part II.F.5.a. above. 33 See Part II.E.6. above. 34 See Part II.F.5.a. above. 81 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 82 of 113 divide the Council. The Mayor and the City Council members representing the South Pasadena districts are most likely to vote as a bloc in opposition to the members representing the North Pasadena districts. It was when the Latino-preferred Council members representing North Pasadena threatened to achieve a majority position on the City Council that Mayor Isbell and his Council allies proposed and approved Proposition 1, changing to a mixed 6–2 map and plan and delaying the shift in Council power. The record shows that Proposition 1’s dilutive impact on Latino power was evident and clearly understood. Latinos and Latino-preferred Council members were unified and vocal in opposing Proposition 1. The Mayor and South Pasadena Council members were unified and vocal in supporting it, using City money, time, employees, organization, and resources to do so. In a series of meetings, including the meeting of the Bond-Review turned Charter-Amendment Committee at which Mayor Isbell announced his proposal for a 6–2 map and plan, the Mayor offered no response to repeated and vigorous objections that the 6–2 map and plan would dilute Latino voting strength, other than to claim—incorrectly—that 70% of Latinos in Pasadena were illegal aliens who could not vote.35 Ignoring clear and supported objections about the racially disparate impact of a proposed law is probative of a lack of responsiveness to minority concerns. See Veasey, 830 F.3d at 262 (“The district court noted that minority legislators and constituents testified about the likely disparate impact of SB 14, yet their amendments to ameliorate that impact were rejected without explanation. . . . While this does not necessarily prove improper intent on the part of those legislators, it nonetheless supports a conclusion of lack of responsiveness.”) (internal citations omitted). 35 See Part II.F.3–4. above. 82 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 83 of 113 The City’s lack of responsiveness to minority concerns contributes to the finding and conclusion that Latinos in Pasadena do not have equal opportunities for, or access to, political participation. 6. Tenuousness of the Policy A remaining Senate factor requires the district court to examine the stated reasons for the political subdivision’s use of the challenged voting plan and ask whether it is a valid policy goal or a tenuous one. Mayor Isbell testified that he preferred an at-large system to elect City Council because this was the system under which he first served on the Pasadena City Council and because it encouraged members to account for and serve the interests of the City as a whole. Bench Trial Tr. 3:63–64, 5:25. Dr. Alford testified to the benefits, in general, to electing council members in a mixed system of at-large seats and single-member districts. Id. at 6:160–62; Def. Ex. 53 at 12. Fostering responsiveness to the overall needs of a city is certainly legitimate. But the City presented no evidence to support the connection between that interest and the specific design or likely (and, as explained below, intended) impact and operation of the 6–2 map and plan in Pasadena. Electing Council members with a citywide constituency does not require using a majorityvote standard or a place system, both of which prohibit single-shot voting. Indeed, by arbitrarily dividing at-large candidates into separate tranches, the place system undermines the purported interest in making candidates responsive to citywide interests. And this interest is neither served by, nor requires, drawing the 6–2 map to entirely eliminate a Latino-majority district. Ample credible testimony showed that only one part of the City—the existing political majority—favored the change, and did so over the vigorous opposition that it came at the expense of the growing power 83 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 84 of 113 of the political minority.36 At-large districts may serve the legitimate interest of encouraging elected officials’ responsiveness and accountability to overall citywide interests rather than individual districts. But the design and implementation of this plan in Pasadena is tenuously connected to that interest. That tenuous connection reinforces the finding and conclusion that Latinos in Pasadena do not have equal opportunities for, or access to, political participation. 7. Proportionality Finally, whether Latinos make up the citizen voting-age majority in a number of districts proportional to their population “is a relevant fact in the totality of circumstances.” De Grandy, 512 U.S. at 1000. De Grandy’s instruction to courts is to consider, when assessing the totality of circumstances, whether there exists a number of “majority-minority districts in substantial proportion to the minority’s share of voting-age population.” Id. at 1013. As explained above, Pasadena’s change from 8–0 single-member districts to a mixed 6–2 map and plan reduced the number of Latino “majority-minority districts” from four to three. Four Latino-majority districts out of eight would be proportional to Latino citizen voting-age population in Pasadena. The court finds and concludes that the 6–2 map and plan is one district less and falls one district short of proportionality. That is dilution.37 Even if there were some way to work the numbers to find proportionality under the 6–2 map and plan, “that consideration would not overcome the other evidence of vote dilution for Latinos.” LULAC, 548 U.S. at 438. “[T]he degree of probative value assigned to proportionality may vary 36 See Part II.F.3–4. above. 37 See Part III.A.2. above. 84 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 85 of 113 with other facts.” Id. (quoting De Grandy, 512 U.S. at 1020). And “[t]he Court’s instruction to employ a case-by-case approach counsels against a per se rule that a protected class, that is also a registered voter majority, cannot experience vote dilution through use of an at-large district.” Salas v. Sw. Tex. Jr. Coll. Dist., 964 F.2d 1542, 1551 (5th Cir. 1992). In LULAC, even though Latinos were “two districts shy of proportional representation,” the Supreme Court noted that even if the “disproportionality were deemed insubstantial, that evidence would not overcome the other evidence of vote dilution for Latinos.” 548 U.S. at 438. The Supreme Court found a § 2 violation in part because “Latino voters were poised to elect their candidate of choice” and “were becoming more politically active” precisely at the time the State altered the voting system in Congressional District 23. Id. Those considerations apply here. Latinos in Pasadena were poised to elect a majority of the City Council for the first time in its history. Latinos had recently become more politically active, electing three Latino candidates over the course of five years in a city that had no more than two Latino Council members in its history. 8. Conclusion of Law on the Totality of the Circumstances Under the totality of the circumstances, based on a “practical evaluation of the past and present reality and on a functional view of the political process,” Gingles, 478 U.S. at 45, this court has found, and now concludes, that changing to the 6–2 mixed electoral map and plan for Pasadena’s City Council diluted Latino voting strength and violates § 2 of the Voting Rights Act. C. Findings and Conclusions on Intentional Discrimination In the 1982 amendments to § 2 of the Voting Rights Act, Congress “repudiated” a requirement that plaintiffs prove intentional discrimination to succeed on a claim that a challenged action violates the Voting Rights Act. Gingles, 478 U.S. at 44. The 1982 Senate Report factors and 85 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 86 of 113 the Court’s analysis in Gingles established a “results test” by which evidence of a discriminatory disparate impact may prove a violation of the Voting Rights Act. Id. at 46. Courts need not reach a finding on intentional discrimination unless “the remedy to which Plaintiffs would be entitled for a discriminatory intent violation is potentially broader than the remedy the district court may fashion for the discriminatory impact violation.” Veasey, 830 F.3d at 230 n.11 (citing City of Richmond v. United States, 422 U.S. 358 (1975)). Indeed, if under § 2, “Plaintiffs will be entitled to the same relief they could access if they prevailed on . . . Fourteenth Amendment claims,” the court must decline to decide the Fourteenth Amendment claim under the canon of constitutional avoidance. Id. at 265 (citing Escambia Cty. v. McMillan, 466 U.S. 48, 51 (1984); Ketchum v. Byrne, 740 F.2d 1390, 1409–10 (7th Cir. 1984)). In this case, the plaintiffs request that Pasadena submit future changes to its electoral map and plan to preclearance by the Department of Justice under § 3(c) of the Voting Rights Act. See 52 U.S.C. § 10302(c). Section 3 permits this remedy following a finding “that violations of the fourteenth or fifteenth amendment justifying equitable relief have occurred within the territory of such State or political subdivision.” Id. Under § 2 alone, this court could enjoin the use of the 6–2 map and plan, but it could not order submission to Department of Justice preclearance. The plaintiffs have stated a claim of intentional discrimination that, if proven, deserves a broader remedy than a violation of § 2 standing alone. The court proceeds to analyze the intentional discrimination claim. 1. Guidance from the Fifth Circuit: Veasey v. Abbott The Fifth Circuit has recently explored the legal standard for determining whether, if legislation impacts voting in a way that violates § 2, it was enacted with discriminatory intent. In 86 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 87 of 113 Veasey v. Perry, individuals and advocacy groups challenged the constitutionality of Senate Bill 14 (SB 14), the 2013 Texas voter-identification law. 71 F.Supp.3d 627, 632–33 (S.D. Tex. 2014). The district court found that the voter ID law violated § 2 by denying or abridging minorities’ right to vote, and that the Texas Legislature had enacted the law with the intent to discriminate against those minority voters. Id. On appeal, the panel opinion held that the district court committed legal errors in the discriminatory intent analysis and remanded for further proceedings. Veasey v. Abbott, 796 F.3d 487, 493 (5th Cir. 2015), reh’g en banc granted, 815 F.3d 958 (5th Cir. 2016), and on reh’g en banc, 830 F.3d 216 (5th Cir. 2016). The panel opinion affirmed the district court’s discriminatory effect finding. Id. Sitting en banc, the Fifth Circuit affirmed in part, vacated in part, and remanded. Veasey v. Abbott, 830 F.3d at 227 (en banc). The majority affirmed the finding that the voter-identification law had a discriminatory effect on minority voters and remanded for the district court to consider the remedy. Id. at 272. The court reversed in part the district court’s judgment that Texas passed the Senate Bill with a racially discriminatory purpose and remanded for reconsideration, including reweighing the evidence. Id. The en banc court recognized that “drafters and proponents of SB 14 were aware of the likely disproportionate effect of the law on minorities, and that they nonetheless passed the bill without adopting a number of proposed ameliorative measures that might have lessened this impact.” Id. at 236. The en banc court also recognized that the record contained evidence supporting the district court’s finding of discriminatory intent, including “a ‘seismic demographic shift,’ as minority populations rapidly increased in Texas, such that the district court found that the party currently in power is ‘facing a declining voter base and can gain partisan advantage’ through” changing the voting methods or procedures. Id. at 241. The change in Veasey 87 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 88 of 113 was enacting a strict voter ID law. The change here is shifting from an 8–0 single-member district plan to a mixed 6–2 map and plan. A plurality of the en banc court found that “some of the evidence . . . was infirm,” especially evidence of Texas’s long-past history of enacting discriminatory laws, evidence of speculation by the bill’s opponents about the proponents’ motives (as opposed to objective evidence of their statements and actions), and evidence of post-enactment justifications by the bill’s proponents. Id. at 232–34, 241. The plurality nonetheless found that “there remains evidence to support a finding that the cloak of ballot integrity could be hiding a more invidious purpose.” Id. at 241. The plurality remanded because it did “not know how much the evidence found infirm weighed in the district court’s calculus.” Id. The en banc Veasey court remanded the case for the district court to reweigh the evidence on intentional discrimination. One concurring opinion emphasized the validity of the district court’s § 2 analysis, including its examination of the totality of the circumstances, understood to include historical as well as recent official discrimination and pervasive private discrimination. Id. at 277–78 (Higginson, J., concurring). Several dissenting opinions would have reversed the district court’s judgment that the voter identification law had discriminatory effects that violated § 2, as well as the judgment finding discriminatory intent. Id. at 280–318 (Jones, J., dissenting); 318–319 (Smith, J., dissenting). A final dissent would have affirmed finding intentional discrimination, including historical practices of enacting voting restrictions in response to potential increases in minority voting strength. Id. at 328–336 (Costa, J., dissenting). This final dissent helpfully clarified that finding intentional discrimination, “although one of grave importance, is not tantamount to a finding that the law had a ‘racist motivation.’” Id. at 335–36 (Costa, J., dissenting) (quoting id. at 88 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 89 of 113 326 (Elrod, J., dissenting)).38 The Veasey plurality opinion noted that “evaluating motive, particularly the motive of dozens of [legislators], is a difficult enterprise. We acknowledge the charged nature of accusations of racism, particularly against a legislative body, but we must also face the sad truth that racism continued to exist in our modern American society despite years of laws designed to eradicate it.” Id. at 231. Unlike the Veasey case, which posed the problem of finding intent within the 181-member Texas Legislature, this case involves a much smaller legislative body of 8, headed by a single Mayor who all parties agree initiated and drove the change to Pasadena’s electoral map and plan. Although the plurality opinion identified legal infirmities in some of the evidence the district court used in its Arlington Heights analysis, the Veasey majority found that “the record also contained evidence that could support a finding of discriminatory intent.” Id. at 234–35. The majority noted that the extensive trial record did not contain “direct evidence that the Texas Legislature passed SB 14 with a racially invidious purpose,” but this “[did] not mean there [was] no evidence that support[ed] a finding of discriminatory intent.” Id. at 235 (citing Rogers, 458 U.S. at 618 (“[D]iscriminatory intent need not be proved by direct evidence.”)). The Veasey court’s analyses of discriminatory intent and how to apply the Arlington Heights framework to properly considered 38 See also Garza v. Cty. of Los Angeles, 918 F.2d 763, 778 (9th Cir. 1990) (Kozinski, J., concurring) (“[T]here is no indication that what the district court found to be intentional discrimination was based on any dislike, mistrust, hatred or bigotry against Hispanics.”); id. (“official engag[ing] in the single-minded pursuit of incumbency” can support a finding of discriminatory intent); Ketchum, 740 F.2d at 1408 (“[M]any devices employed to preserve incumbencies are necessarily racially discriminatory.”); N. Carolina State Conference of NAACP v. McCrory, 831 F.3d 204, 221–23 (4th Cir. 2016) (“Using race as a proxy for party may be an effective way to win an election. But intentionally targeting a particular race’s access to the franchise because its members vote for a particular party, in a predictable manner, constitutes discriminatory purpose.”). 89 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 90 of 113 record evidence provides guidance for this court’s analysis of discriminatory intent. 2. The Arlington Heights Factors Each Arlington Heights factor supporting finding discriminatory intent that the Veasey majority identified is present in this case, with equal or greater force. The (1) “historical background of the decision,” (2) “specific sequence of events leading up to the challenged decision,” (3) “departures from the normal procedural sequence,” (4) “substantive departures,” and (5) “legislative history” form the nonexhaustive list of factors the district court must consider. Arlington Heights, 429 U.S. at 267–68. Evidence in the record supports finding discriminatory intent under each factor. a. The Historical Background of the Decision The background of the decision, including recent or contemporaneous evidence of official discriminatory actions, particularly in the voting-rights context, is useful here, though much less significant than the other factors. The Veasey majority found that “circumstantial evidence of discriminatory intent is augmented by contemporary examples of State-sponsored discrimination in the record.” Veasey, 830 F.3d at 239. The majority juxtaposed this “contemporary” history to the “long-ago” history on which the district court placed undue weight. Examples of contemporary history included Texas’s attempt to purge minorities, including Latinos, from voter rolls in 1975, and the fact that “[i]n every redistricting cycle since 1970, Texas has been found to have violated the [Voting Rights Act] with racially gerrymandered districts.” Id. at 239–40 (quoting Veasey v. Perry, 71 F.Supp.3d. at 636 & n.23). This evidence is noted but, for the reasons the Veasey plurality provided, given little weight. The Veasey majority also made it clear that official discrimination in areas other than the challenged voting procedure is appropriately considered. Id. at 239–40. The evidence here on police 90 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 91 of 113 treatment of Latinos in Pasadena is as contemporary as the evidence considered by the Veasey court.39 Id. (treating intentional § 2 violations in redistricting laws running from 1980 to the present as “contemporary examples of State-sponsored discrimination in the record”). And the record shows that official discrimination persists in other departments of Pasadena’s government. Mayor Isbell told Council member Harrison that police officers finding Latinos driving without proof of insurance were to impound the vehicles, but that officers would have discretion to decide what to do when Anglo drivers could not produce proof of insurance.40 b. The Sequence of Events Leading Up to the Decision; Legislative History The “specific sequence of events leading up to the decision” and the legislative history weigh in favor of finding intentional discrimination here. The Veasey majority emphasized the record evidence showing “that the drafters and proponents of [the challenged voting measure] were aware of the likely disproportionate effect of the law on minorities, and that they nonetheless passed the bill without adopting a number of proposed ameliorative measures that might have lessened this impact.” Id. at 236. The present record similarly shows that members of the Bond-Review turned Charter-Amendment Committee, City Council members, and members of the public objected repeatedly to the Mayor and the City Council that the 6–2 map and plan diluted Latino voting strength, would not have passed Justice Department preclearance, and was likely illegal.41 The Mayor, the Committee, and the Council nonetheless proceeded to enact the 6–2 map and plan as the Mayor proposed them, without significant amendment and apparently without considering possible 39 See Part II.D.2. above. 40 See Part II.D.2. above. 41 See Part II.F.3.–4. above. 91 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 92 of 113 ameliorative measures, such as eliminating the place system, majority-vote requirement, or the prohibition on single-shot voting.42 The Veasey majority noted that “the extraordinary measures accompanying the passage of SB 14 occurred in the wake of a seismic demographic shift, as minority populations rapidly increased in Texas, such that the district court found that the party currently in power is facing a declining voter base and can gain partisan advantage through a strict voter ID law,” adding that “[i]ntentions to achieve partisan gain and to racially discriminate are not mutually exclusive. . . . [A]cting to preserve legislative power in a partisan manner can also be impermissibly discriminatory.” Id. at 241 & n.30 (internal quotations and citations omitted). Pasadena’s Latino population has grown rapidly, from 18.7% of the citizen voting-age population in 1990 to 48.2% today. Latinos appeared to be on the verge of electing a Council majority for the first time in Pasadena’s history in the 2015 election.43 But between the Supreme Court’s Shelby County decision and the winter of 2014, half of the City Council, led by the Mayor and with the Mayor’s tie-breaking vote, eliminated a Latino-majority district by creating an at-large seat and made it more difficult for the Latino-preferred candidate to succeed in all the districts except the ones in which the Latino citizen voting-age and Spanish-surnamed registered voter population have a decisive and secure majority.44 The major precipitating event for converting Pasadena’s electoral map and plan to eliminate a Latino-majority district and create two at-large places that are clearly not Latino opportunity 42 See part II.F.6. above. 43 See Part II.E.1 and Part II.G.2. above. 44 See Part II.F.–G. above. 92 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 93 of 113 districts was the Supreme Court’s Shelby County decision. This allowed the Mayor and City Council to redistrict with the clear impact of diluting Latino voting strength without the obstacle of Department of Justice preclearance, which the Department clearly—given the recent experience of nearby cities who had failed to get preclearance for redistricting proposals—would have refused. Mayor Isbell drafted the proposal to create the Committee that would ultimately review the proposed Charter changes two days after Shelby County. The Mayor had City employees prepare redistricting maps changing to a mixed single-member and at-large plan, including a map of the 6–2 map and plan that was ultimately enacted, within a month after Shelby County. The Mayor stated in a media interview that he proposed changing to a 6–2 map and plan when he did “because the Justice Department can no longer tell us what to do.” Mayor Isbell testified that he understood the purpose of preclearance was to prevent the dilution of minority voting strength and that the Department of Justice likely would not have precleared Pasadena’s proposed change for that reason.45 This evidence supports not only the Mayor’s and the Council’s knowledge of the dilutive impact, but of their intent to achieve it. c. Departures, Both from the Normal Procedural Sequence and Substantive; Legislative History and Preenactment Statements by Proponents The Veasey majority noted that the voter-identification legislation “was subjected to radical departures from normal procedures,” providing some indication that “the Legislature’s race-neutral reason of ballot integrity . . . is pretextual.” Id. at 237. The present record is similar. Pasadena’s Mayor and Council substantially departed from usual legislative procedures. The Mayor established a Committee for the single purpose of reviewing a bond proposal. When the bond proposal was 45 See Part II.F.2. above. 93 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 94 of 113 opposed by the Latino-preferred Council members, the Mayor withdrew it but used the Committee to pursue his intent to redraw the City Council election map. He had been preparing to achieve that end since at least July 2013, when City employees began the work that produced three alternative mixed single-member and at-large district maps.46 The Mayor’s plan to pursue a Charter amendment to redraw the map for electing Council members was publicly revealed on July 25, 2013, when the Bond-Review Committee held its third and final public meeting and its members—but not members of the public—received the proposed 6–2 map. The next meeting, the Committee’s final gathering, was closed to the public. The Committee did not recommend redrawing the map, but the Mayor and the City Council majority pursued it.47 The Mayor prepared the maps and devised the procedures for holding City Council elections starting in 2015 using the 6–2 mixed plan without oversight by or input from the Bond-Review turned Charter-Amendment Committee. When that Committee decisively recommended against the redistricting plan, the Mayor disregarded that recommendation, testifying without credibility that he was not informed of that fact or did not recall it.48 The Mayor presented the 6–2 proposal to the public and to the Council despite the near-unanimous opposition of the Charter-Amendment Committee.49 The Mayor introduced an ordinance to place the redistricting Charter amendment on the 46 See Part II.F.3. above. 47 See Part II.F.4. above. 48 See id. 49 See Part II.F.3. above. 94 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 95 of 113 November 2013 ballot. The City Council passed it on first reading in August 2013. Rather than place the proposal on the ballot for the next City Council election in May 2015—when Latinos were expected to win a majority of Council seats—the Mayor rushed the map-drawing and Committee review process to put the proposal on the ballot for the November 2013 Harris County elections. This also supports finding not merely knowing that the redistricting plan would dilute Latino voting strength and would have a disparate impact, but an intent to pass the law because of—in order to achieve—the dilution and disparate impact.50 The amendment converting the City Council electoral plan to the 6–2 map and plan passed in the November 2013 election. The Council held a public hearing at which it heard significant opposition from Latinos fearing voter dilution. The Council did nothing in response. It passed the ordinance with the Mayor’s tie-breaking vote, after Ms. Van Houte, a Latino-preferred Council member who spoke against the change, was ushered out of the meeting by police.51 The City’s approach to the campaign for Proposition 1 also deviated from the usual procedures. Pasadena’s Director of Community Relations, Richard Scott, an ally of the Mayor, directed those working to promote Proposition 1 to “pull out Hispanic names” from the mailing lists used for the campaign, and used City time, City employees, and City organizational resources on the campaign. Mayor Isbell’s political action committee and the Neighborhood Network Program devoted significant resources to mobilizing support for Proposition 1. The campaign excluded predominately Latino North Pasadena from the campaign events, mailings, and outreach.52 50 See Part II.F.4. above. 51 See Part II.F.6. above. 52 See Part II.F.5. above. 95 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 96 of 113 Another deviation from the usual procedures accompanied the passage of Proposition 1. The Mayor’s tie-breaking vote instituted a new three-minute-debate rule, limiting the time to speak on a particular topic. This rule was new. It deviated from prior practice. It was introduced when the divisive nature of the proposed new map and plan to elect City Council members had become clear. This new rule was applied to the opponents of the 6–2 map and plan. Council member Van Houte’s violation of the rule led the Mayor to order her removed from the meeting. Three other Council members departed in protest, yet the remaining Council, guided by the Mayor, persisted to pass the 6–2 map and plan in their absence. And the Mayor deviated from ordinary procedures and violated City ordinances by bringing a firearm to one of the Council meetings at which the 6–2 map and plan was being debated.53 This evidence provides a more compelling record than was present in Veasey to find that the proponents of the 6–2 map and plan not only knew it would have a dilutive, discriminatory impact on Latinos, but that the legislative body and its leader, the Mayor, intended that result. The fact that this legislative body is considerably smaller than the Texas legislature makes the intent analysis more straightforward as well. Cf. Veasey, 830 F.3d at 233. d. Meeting the Burdens: The Plaintiffs’ Showing that Racial Discrimination was a Substantial Factor in Enacting the New Electoral Map and the Defendants’ Failure to Demonstrate that the Law Would Have Been Enacted Without this Factor “Legislators’ awareness of a disparate impact on a protected group is not enough: the law must be passed because of that disparate impact.” Veasey, 830 F.3d at 230–31 (citation omitted). The Mayor and a majority of City Council proposed, approved, and, with the help of City employees 53 See Part II.F.6. above. 96 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 97 of 113 working on City time and using City resources, campaigned for, and enacted Proposition 1, knowing that the results would be to dilute the voting strength of Latino voters. The dilutive effect was lower than those acting in 2013 and early 2014 reasonably expected. The unexpected nature of the 2016 presidential campaign led Latino voter registration to rise, an increase that was not foreseen in 2013.54 But the credible and properly considered evidence is that, when the 6–2 plan was enacted, Pasadena’s Mayor and City Council knew that the proposed change to the 6–2 redistricting plan would significantly dilute Latino votes and would have a discriminatory impact on Pasadena’s Latinos. The issue is whether the evidence meets the plaintiffs’ burden of showing racially discriminatory intent, not merely knowledge that there would be a racially disparate and discriminatory impact. The court finds and concludes that the plaintiffs have met their burden of showing that a substantial reason—the intended purpose—of enacting Proposition 1 was to dilute Latino voting strength. This is not a case in which the legislative body selected a particular course of action “in spite of” its predictable and expected adverse effect of diluting the votes of citizen-voting-age Latinos and reducing the number of City Council seats that Latino-preferred candidates had an equal opportunity to secure. Rather, this is a case in which the legislative body acted “because of” the adverse effect, in order to dilute the voting strength of Latinos. See Veasey, 830 F.3d at 285 (quoting Feeney, 442 U.S. at 279 (rejecting claim by women that legislation creating lifetime state employment preference for veterans was intentional discrimination against women)). The intent was to delay the day when Latinos would make up enough of Pasadena’s voters to have an equal opportunity to elect Latino-preferred candidates to a majority of City Council seats. Recent 54 See Part II.G.2. above. 97 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 98 of 113 population shifts and growth in the Latino citizen voting-age and Spanish-surnamed registered voter population made it clear that this power shift was about to occur. That day was expected to occur by 2015 unless the 8–0 single-member district map and plan was changed. It was changed, with the intent to dilute Latino votes and at least delay and defer the shift in power away from the Anglo majority. The City argues that this court cannot find discriminatory intent because: (1) there is no evidence that the Mayor or City Council members or other actors acted out of personal dislike of or bias against Latinos because of their race or ethnicity; (2) Mayor Isbell perceived that City elections were becoming more partisan, and his intent was only to mobilize Republican voters against Democrats, not Anglos against Latinos; (3) the Mayor’s support of Latino candidates and policies, and his directive not to draw the 6–2 map to achieve the fewest possible Latino-majority districts defeat, are inconsistent with finding an intent to dilute Latino votes; and (4) the court cannot reliably discern the legislative intent because the decisionmaker was the entire Pasadena electorate, which by a narrow majority, divided on racial lines, approved the redistricting. (Docket Entry No. 148). The City argues for a higher standard than the law requires. While enacting a law motivated by lawmakers’ individual and personal racial animus would be strong evidence of a discriminatory intent, courts are not called on to plumb the depths of lawmakers’ psychological reasons or subjective motives for acting. See Richard H. Fallon, Constitutionally Forbidden Legislative Intent, 130 HARV. L. REV. 523, 541–54 (2016). The court need not and does not find that individual lawmakers were motivated by reasons of personal dislike or bias against Latinos because of their ethnicity or race. Mayor Isbell has supported and continues to support Latino candidates for City Council. The record contains no evidence of racial epithets during Mayor Isbell’s long service to 98 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 99 of 113 Pasadena. But the plaintiffs’ burden does not require them to show that individual legislators’ subjective personal racism toward Latinos was the motive for proposing, supporting, or enacting legislation that has a disparate and dilutive impact on Latinos. The standard is objective evidence of an intent or purpose to discriminate by taking a course of action that will cause a racially disparate adverse impact, at least in part in order to achieve that impact. See Veasey, 830 F.3d at 241 & n.30. The plaintiffs’ burden is to show that objective evidence. The Supreme Court has observed the difference between analyzing the subjective motivations of individual legislators and actors, which courts are ill-equipped to do, and objective indications of legislative intent, which courts are used to detecting and addressing. In Rogers v. Lodge, the Court affirmed a district court’s finding that a multimember district in Georgia, although “racially neutral when adopted, [was] being maintained for invidious purposes.” 458 U.S. at 622. The Court ruled that the objective circumstantial evidence in the record was enough to find discriminatory intent, including a past history of discrimination in the jurisdiction, the depressed socioeconomic status of minorities, the size of the district, and a majority-vote requirement. Id. at 622–27. These factors showed that the Georgia district purposefully “submerge[d] the will of the minority” and “den[ied] the minority’s access to the system.” Id. at 627. Objective indicia of dilutive intent are present here as well. The objective indicia include the timing of the redistricting proposal so soon after Shelby County removed Department of Justice preclearance; the number and extent of procedural irregularities used in enacting the redistricting map and plan; the clear dilutive impact the map and plan would have; the clear knowledge that if a mixed redistricting map and plan was not put into place, the 2015 elections would produce a Latinopreferred candidate majority of City Council members for the first in Pasadena’s history; and the use 99 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 100 of 113 of such dilutive procedures as the place system and majority-vote requirement for the at-large Council positions. These provide strong objective indicia of a legislative intent to dilute Latino voting strength. The plaintiffs’ evidence is consistent with, and supported by, cases that have found objective indicia of intentional discrimination in redistricting cases. In Shaw v. Reno, 509 U.S. 630 (1993), the Court reversed a three-judge panel that dismissed a challenge to a North Carolina redistricting plan on the ground of intentional discrimination. In the absence of direct evidence of subjective intent, the Court nonetheless held that the reapportionment scheme could objectively be understood only as an intentional effort to segregate voters into separate districts on the basis of race. Id. at 658; see also Gomillion v. Lightfoot, 364 U.S. 339 (1960) (changing a square district to a 28-sided district objectively demonstrated a racially discriminatory intent). The Court based its reasoning in part on the objective expressive content of the law. Shaw, 509 U.S. at 647. The court here similarly credits testimony that Pasadena Latinos reasonably inferred from the change to the City’s electoral map and plan that, as Council member Wheeler stated, “it confirms their fears that the system is rigged against us. We win an election, but you know what, they move the field goal back and move it back farther and farther.” Bench Trial Tr. 5:147–48. In LULAC v. Perry, the Court reasoned that polarized voting and recent demographic shifts provided strong objective evidence that a redistricting law was motivated by a racially discriminatory intent. 548 U.S. at 427–28. The Court found that “[i]n old District 23 the increase in Latino voter registration and overall population, the concomitant rise in Latino voting power in each successive election, the near-victory of the Latino candidate of choice in 2002, and the resulting threat to the” incumbent representative motivated the controlling party to dilute the minority vote. Id. at 428 100 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 101 of 113 (citation omitted). Although the Court grounded its holding on the § 2 results test, which does not require proof of intentional discrimination, the Court noted that the challenged legislation bore “the mark of intentional discrimination.” Id. at 440. The surging—but not quite majority—Latino population in Pasadena and the City’s decision to change its electoral map and plan shortly after the Shelby County decision removed the requirement for preclearance similarly bears “the mark of intentional discrimination.” Id. All these are objective measures or indications of an intent to change to a 6–2 map and plan in order to dilute Latino votes, not merely knowing that this could occur. Framing the reasons for enacting legislation in terms of enhancing partisanship rather than reducing or diluting minority votes is not persuasive on this record. Partisan goals can be a pretext for invidious intent behind legislation. See Veasey, 830 F.3d at 236, 241 n.30 (majority opinion) and 335–36 (Costa, J., dissenting). When, as here, the party in power stands to gain partisan advantage by diluting Latino votes, pursuing partisan advantage can—and here does—mean intending to dilute Latino votes. A recent Fourth Circuit case, decided nine days after Veasey, helps make this clear. In North Carolina State Conference of NAACP v. McCrory, 831 F.3d at 204, the appellate court reversed a ruling that the record of how a North Carolina omnibus voting bill was enacted failed to show discriminatory intent. North Carolina argued that the evidence showed only partisanship, not racial animus. The appellate court ruled that using race and party affiliation as proxies for one another established discriminatory intent. “Using race as a proxy for party may be an effective way to win an election. But intentionally targeting a particular race’s access to the franchise because its members vote for a particular party, in a predictable manner, constitutes discriminatory purpose. 101 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 102 of 113 This is so even absent any evidence of race-based hatred and despite the obvious political dynamics.” Id. at 222–23. The same reasoning applies here. Indeed, Pasadena’s Director of Community Relations Richard Scott testified that in Pasadena, referring to “Republican” voters was a proxy for Anglo voters, and referring to “Democrat” or “Democratic” voters was a proxy for Latino. Bench Trial Tr. 4:240–42. Bianca Gracia, the Latina president of a Pasadena homeowners’ association and a Republican activist, also credibly testified that she understood campaign rhetoric about Pasadena “turn[ing] blue” as a reference to Latino political power.55 Mr. Scott’s directive to “pull out the Hispanic names” from campaign lists for Proposition 1 provides a clear example. The City attempts to characterize Mr. Scott’s statements as race-neutral because they were “premised on [the] belief that most Hispanics in Pasadena vote as Democrats and that Democratic groups were opposing the charter change.” (Docket Entry No. 148 ¶ 133). That explanation is not inconsistent with, and does not lessen, the evidence showing the intent to discriminate against Latinos in changing the district map. The word Mr. Scott chose was not Democrats but Hispanics. Pasadena officials supporting Proposition 1, and doing so on behalf of Mayor Isbell and at the direction of his appointed officials, understood race and party as interchangeable proxies. By clearly and explicitly intending to diminish Latinos’ voting power for partisan ends, Pasadena officials intentionally discriminated on the basis of race. The court has carefully considered and weighed the evidence that the Mayor had in the past supported, and continues to support, some Latino and Latino-preferred candidates; that he and the City Council had in the past supported, and continue to support, funds and projects to improve North Pasadena; that the Mayor and City Council did not try to minimize the number of Latino-opportunity 55 See Part II.F.5.b. above. 102 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 103 of 113 districts. But the disparate and discriminatory dilutive impact of changing to the 6–2 map and plan was and is clear. While this impact, and the knowledge that it would occur, are not enough on their own to find intentional discrimination, they provide objective evidence that, combined with other evidence, provide ample support for finding discriminatory intent. Disproportionate impact cannot be “the sole touchstone” of an intentional discrimination claim. Davis, 426 U.S. at 242. But “[s]howing disproportionate impact, even if not overwhelming impact, suffices to establish one of the circumstances evidencing discriminatory intent.” North Carolina State Conference of NAACP, 831 F.3d at 231. Other facts establish circumstances evidencing discriminatory intent. These include the timing of proposing the change to the 6–2 map and plan, the sequence of events leading up to the placement of Proposition 1 on the November 2013 ballot, the deviation from usual procedures and standards in the course of the City Council’s enactment of the new map and plan, the clear discriminatory adverse impact on Latino voters, and the fact that unless the change was made, the City Council would sooner, rather than later, shift from the long history of Anglo-majority rule to Latino-preferred candidates holding the majority vote on divisive issues. The evidence that the City of Pasadena intended to dilute Latino voting strength when it changed from the eight single-member district to the six single-member and two at-large seat map and plan is far stronger and far clearer than was true in Veasey. 3. Conclusion: The City Intended to Dilute Latino Voting Strength Based on the ample evidence of objective factors evidencing discriminatory intent—the intent to disproportionately and discriminatorily dilute Latino minority voting strength and acting in order to achieve that goal—the court has found, and now concludes, that the City violated the Fourteenth Amendment’s guarantee of equal protection by intentionally discriminating against 103 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 104 of 113 Latinos and disproportionately diluting their voting power in changing to the 6–2 electoral map and plan to elect the Pasadena City Council. IV. Remedy “When devising a remedy to a § 2 violation, the district court’s ‘first and foremost obligation . . . is to correct the Section 2 violation.’” Brown, 561 F.3d at 435 (quoting Bone Shirt v. Hazeltine, 461 F.3d 1011, 1022 (8th Cir. 2006)). A remedy “should be sufficiently tailored to the circumstances giving rise to the § 2 violation.” Id. “[T]o the extent possible, courts should respect a legislature’s policy objectives when crafting a remedy . . . .” Veasey, 830 F.3d at 269 (citations omitted). But “[i]n fashioning a reapportionment plan or in choosing among plans, a district court should not preempt the legislative task nor intrude upon state policy any more than necessary.” Upham v. Seamon, 456 U.S. 37, 41–42 (1982) (per curiam) (quoting White v. Weiser, 412 U.S. 783, 794–795 (1973)). “There are times when a court might give a state legislature an opportunity to cure the infirmities in the statute before permitting the district court to fashion a remedy.” Veasey, 830 F.3d at 269 (citing Wise, 437 U.S. at 540). “However, the Supreme Court and [the Fifth Circuit] have acknowledged that when it is not practicable to permit a legislative body this opportunity because of an impending election, it becomes the unwelcome obligation of the federal court to devise and impose a remedy pending later legislative action.” Id. (internal quotations and citations omitted). Pasadena is facing an impending election under a voting map and plan that violates § 2 of the Voting Rights Act and intentionally discriminates against the City’s Latinos by diluting their voting strength. Pasadena holds elections for City Council in May of odd-numbered years. Candidates must register for the May 2017 election less than a month from the date of this opinion. 104 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 105 of 113 Because the City’s Charter requires amendments to be put to a citywide vote, there is no time for legislative action to cure the infirmities in the 6–2 map and plan before the next election, which would be the earliest opportunity for Pasadena voters to approve a City Charter amendment. The court must balance the requirement to give as much force as possible to the legislature’s legitimate intentions with the requirement to rectify the § 2 violation without over-intruding on the legislature’s province and prerogatives. The use of at-large seats to encourage responsiveness to city-wide interests among elected officials is a legitimate government policy. Although there may be remedies available that would correct the dilution of Latino voting strength while preserving one or more at-large places on the City Council—perhaps by redrawing the maps or restructuring the place system to permit single-shot voting and elective victory by plurality—none has been suggested. And devising these remedies requires hard and delicate work and expertise that cannot be done in the time available. This work is better left to the legislative process, subject to court review if invoked. The less intrusive remedy that cures the § 2 dilution in time to allow Pasadena to prepare for the next election, while giving the most deference to the local legislature, is to restore the previous 8–0 single-member district map and plan the City used in its May 2013 elections. The districts for that map were drawn in 2011, based on the most recent decennial census. The 8–0 plan gave Latino voters an equal opportunity to elect their preferred candidates without guaranteeing success. Indeed, Latinos in 2013 did not elect their candidate of choice in District B, even with a majority of citizen voting-age population and registered voters. Curing the § 2 violation requires that the map provide Latino voters an equal opportunity to elect their candidates of choice. See Wise, 437 U.S. at 540 (“Among other requirements, a court-drawn plan should prefer single-member districts over 105 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 106 of 113 multimember districts, absent persuasive justification to the contrary.” (collecting cases)). Section 3(c) of the Voting Rights Act of 1965, 52 U.S.C. § 10302(c), authorizes a federal court, following a finding “that violations of the fourteenth or fifteenth amendment justifying equitable relief have occurred within the territory of such State or political subdivision,” to order a jurisdiction to obtain the Justice Department’s preclearance of its election changes under § 5 of the Voting Rights Act. Because the court finds that Pasadena officials intentionally discriminated against Latinos in diluting their voting strength, the court grants the plaintiffs’ request under § 3(c) to require Pasadena to submit future changes to its electoral map and plan to the Department of Justice for preclearance. The court also grants the request for an order under § 3(c) to retain jurisdiction to review, before it is enforced, any “voting qualification or prerequisite to voting or standard, practice, or procedure with respect to voting different from that in force or effect” from the map and plan in use in the May 2013 election. Any new City Council voting map or procedure may be enforced if it has first been submitted to the United States Attorney General and the Attorney General has not interposed an objection within 60 days after submission. Section 3(c) requires the court to define a period to retain jurisdiction. The parties have not made any suggestions about how long that should last. The court invites the parties to do so, no later than January 13, 2017. As a starting point, five years, or through the 2021 election, might be appropriate, because it is likely enough time for demographic trends to overcome concerns about dilution from redistricting. Cf. Allen v. City of Evergreen, Civil No. 13-107, Docket Entry No. 82 (S.D. Ala. Jan. 13, 2014) (six-year preclearance period). The parties must also submit a proposed form of injunction for this court to consider. No bond is required. See Dynamic Sports Nutrition, Inc. v. Roberts, No. CIV. A. H-08-1929, 2009 WL 106 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 107 of 113 136023, at *2 (S.D. Tex. Jan. 16, 2009). The parties’ proposals as to the period for continued jurisdiction and the form of injunction are due no later than January 13, 2017. V. Conclusion and Order In the Supreme Court’s foundational decision on vote dilution, Reynolds v. Sims, 377 U.S. 533 (1964), the Court ruled that dilution was tantamount to infringing the very right to vote. “Overweighting and overvaluation of the votes of those living here has the certain effect of dilution and undervaluation of the votes of those living there. The resulting discrimination against those individual voters living in disfavored areas is easily demonstrable mathematically. Their right to vote is simply not the same right to vote as that of those living in a favored part of the [jurisdiction].”56 Id. at 563. In Pasadena, Texas, Latino voters under the current 6–2 map and plan do not have the same right to vote as their Anglo neighbors. “Legislative bodies should not leave their reapportionment tasks to the federal courts; but when those with legislative responsibilities do not respond [to constitutional violations], or the imminence of a state election makes it impractical for them to do so, it becomes the ‘unwelcome obligation’ of the federal court to devise and impose a reapportionment plan pending later legislative action.” LULAC, 548 U.S. at 415 (quoting Wise, 437 U.S. at 540). Pasadena’s next City Council elections are imminent. The most expedient and deferential remedy is a return to the May 2013 election map and plan in use before the dilutive 6–2 plan. The City of Pasadena, Texas is ordered to conduct its May 2017 City Council elections using 56 Reynolds is the origin of the Court’s famous statement on the right to vote: “Undoubtedly, the right of suffrage is a fundamental matter in a free and democratic society. Especially since the right to exercise the franchise in a free and unimpaired manner is preservative of other basic civil and political rights, any alleged infringement of the right of citizens to vote must be carefully and meticulously scrutinized.” 377 U.S. at 561–62. 107 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 108 of 113 the map and plan in force in May 2013. The court retains jurisdiction under 52 U.S.C. § 10302(c) to review before enforcement any voting qualification, prerequisite, practice, or procedure different from the map and plan in use in May 2013. Any new City Council voting map, plan, or procedure may be enforced before court review only if it has first been submitted to the United States Attorney General and the Attorney General has not interposed an objection within 60 days after submission. The parties must submit a proposed form of injunction and a proposed period for continued jurisdiction no later than January 13, 2017. SIGNED on January 6, 2017, at Houston, Texas. ______________________________________ Lee H. Rosenthal Chief United States District Judge Appendix A: Demonstrative Maps Appendix B: Time Line 108 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 109 of 113 Appendix A: Demonstrative Maps 201 610 526 55 146 225 225 45 8 3 146 45 8 3 45 2351 PL 152- 1 109 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 110 of 113 8 201 610 526 55 146 225 225 225 45 146 45 8 8 3 45 2351 110 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 111 of 113 Appendix B: Time Line Summer 2011 Pasadena’s City Council approves the redistricting plan for the 8–0 map and plan to account for the 2010 decennial census. December 21, 2012 The U.S. Department of Justice objects under § 5 of the Voting Rights Act to a Beaumont Independent School District proposal to change from single-member districts to a mixed system of singlemember and at-large districts. May 2013 Pasadena holds elections for City Council under the 2011 map for the 8–0 plan. Latino-preferred candidates are elected in Districts A, C, D, and E, but not in District B, a Latino-majority district. June 25, 2013 The Supreme Court issues its decision in Shelby County, Alabama v. Holder, removing Pasadena from § 5 preclearance procedures. June 27, 2013 Pasadena’s Mayor Johnny Isbell calls for a Bond-Review Committee. July 18, 2013 The Bond-Review Committee holds its first public meeting. July 23, 2013 The Bond-Review Committee holds its second public meeting. July 24, 2013 Mayor Isbell finalizes alternative districting maps proposing a 4–2 plan, a 7–1 plan, and the 6–2 plan that would ultimately be adopted. July 25, 2013 The Bond-Review Committee holds its third public meeting. Mayor Isbell proposes a charter amendment to change Pasadena’s electoral map and plan to include at-large districts. Mayor Isbell responds to objections over Latino vote dilution that 70% of Pasadena’s Latinos are illegal residents. July 29, 2013 The Bond-Review turned Charter-Amendment Committee holds its fourth and final meeting, closed to the public. The Committee votes 10 to 1 against the proposed charter amendment to change Pasadena’s electoral map and plan. August 13, 2013 Mayor Isbell introduces a bond proposal to the City Council. The Council votes 5 to 3 to pass the proposal on its first reading. Council members from District C (Harrison), D (Van Houte), and E (Wheeler), all in North Pasadena, vote against the proposal. Council member Ybarra (District A) is absent. August 15, 2013 Mayor Isbell issues a memo to the City Council, retracting the bond 111 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 112 of 113 proposal and proposing a charter amendment to change Pasadena’s electoral map and plan. August 16, 2013 Larry Peacock, a member of the Bond-Review turned CharterAmendment Committee emails the Mayor to make clear that the Committee opposes any redistricting proposal. August 20, 2013 The City Council passes on first reading Ordinance 2013–126 calling for a charter-amendment election to change Pasadena to a 6–2 map and plan. The vote is 5 to 4. Council members from District A (Ybarra), C (Harrison), D (Van Houte), and E (Wheeler) oppose the measure. On the same vote, the Council passes the ordinance on its second reading. Fall 2013 Richard Scott, Director of Community Relations, uses City resources, City employees, and City time to campaign for Proposition 1 to change Pasadena to a 6–2 electoral map and plan. October 3, 2013 Mayor Isbell and other Pasadena officials hold an event billed as a homeowners’ association Board of Directors appreciation social. The event is used to solicit funds and campaign support for Proposition 1. Council members from Districts A (Ybarra), C (Harrison), D (Van Houte), and E (Wheeler) are not invited. The other Council members are all present. One, Steve Cote (District G) warns of Pasadena “turn[ing] blue,” which one Latina attendee, Homeowner’s Association President Bianca Gracia, interprets to be a racial reference. September 4, 2013 Director of Community Relations Scott directs a subordinate to use the Mayor’s campaign list for the Proposition 1 campaign but “pull out the Hispanic names.” November 4, 2013 Mayor Isbell states in an interview that he proposed the charter amendment when he did “because the Justice Department can no longer tell Pasadena what to do.” November 5, 2013 Proposition 1 passes by 3,292 votes to 3,213. February 18, 2014 City Council adopts a rule limiting debate time to three minutes per member per issue. The vote is 5 to 4. Council members from District A (Ybarra), C (Harrison), D (Van Houte), and E (Wheeler) oppose the measure. March 4, 2014 The City Council holds a public “Redistricting Workshop.” Mayor 112 Case 4:14-cv-03241 Document 151 Filed in TXSD on 01/06/17 Page 113 of 113 Isbell presents maps and plans prepared and reviewed by the law firm of Bickerstaff, Heath, Delgado and Acosta, LLP. The 6–2 map and plan ultimately enacted is among the proposals. March 18, 2014 The City Council holds a public hearing on redistricting. Council members and the public repeatedly object to the legality of the proposed 6–2 map and plan. The Mayor brings a firearm to the meeting which drops on the floor. April 1, 2014 The City Council holds a public hearing on redistricting. Council member Van Houte (District D) exceeds her three-minute debate limit and is removed by armed officers. Three other Council members (Ybarra from District A, Harrison from District C, and Wheeler from District E) depart in protest. The remaining Council members pass the 6–2 map and plan in their absence. April 15, 2014 The City Council passes on second reading the 6–2 map and plan. The vote is 5 to 4. Council members from District A (Ybarra), C (Harrison), D (Van Houte), and E (Wheeler) oppose the measure. November 12, 2014 The plaintiffs file suit challenging the dilution of Latino votes under the mixed 6–2 map and plan. May 2015 Pasadena holds elections for City Council under the 6–2 map and plan. Latino-preferred candidates are elected in single-member Districts A, C, and D, but not in District B, a Latino-majority district. Pat Van Houte, an incumbent Anglo Council member and Latinopreferred candidate, wins at-large Place G. Oscar Del Toro, a nonincumbent Latino-preferred candidate with a Spanish surname, loses with 39% of the vote in at-large Place H. 113