State of Texas v. United States of America , 887 F. Supp. 2d 133 ( 2012 )


Menu:
  •                                UNITED STATES DISTRICT COURT
    FOR THE DISTRICT OF COLUMBIA
    )
    STATE OF TEXAS,                              )
    )
    Plaintiff,                          )
    )
    v.                 )      Civil Action No. 11-1303
    )         (TBG-RMC-BAH)
    UNITED STATES OF AMERICA,                    )
    and ERIC H. HOLDER, in his                   )
    official capacity as Attorney General        )
    of the United States                         )
    )
    Defendants, and                     )
    )
    Wendy Davis, et. al.,                        )
    )
    Intervenor-Defendants.              )
    )
    MEMORANDUM OPINION
    Before: GRIFFITH, Circuit Judge, COLLYER and HOWELL, District Judges.
    Opinion for the Court filed by Circuit Judge GRIFFITH, in which District Judge HOWELL
    joins and District Judge COLLYER joins all except section III.A.3. Separate opinion for the Court
    with respect to retrogression in Congressional District 25 filed by District Judge HOWELL, in
    which District Judge COLLYER joins.
    Dissenting opinion with respect to retrogression in Congressional District 25 filed by
    Circuit Judge GRIFFITH.
    Appendix filed by District Judges COLLYER and HOWELL, in which Circuit Judge
    GRIFFITH joins.
    Opinion for the Court by GRIFFITH, Circuit Judge:
    Table of Contents
    I. Background.............................................................................................................................. 3
    II. Principles of Section 5 Analysis ............................................................................................. 5
    A. Retrogression ..................................................................................................................... 5
    1.    Texas’s Burden of Proof............................................................................................. 7
    2.    Election Analysis Methodologies ............................................................................... 8
    a. Types of Elections ....................................................................................................... 8
    b. Election Analysis Sample Sets ................................................................................. 11
    3.    Statewide Retrogression Analysis ........................................................................... 13
    4.    Coalition and Crossover Districts ........................................................................... 18
    a. Section 5 Analysis ..................................................................................................... 18
    b. Standard of Proof ..................................................................................................... 22
    B. Discriminatory Intent ..................................................................................................... 25
    III. Congressional Plan .............................................................................................................. 27
    A. Retrogression in the Congressional Plan ...................................................................... 27
    1.    Congressional District 27 ......................................................................................... 29
    2.    Congressional District 23 ......................................................................................... 29
    3.    Retrogression with New Congressional Seats ........................................................ 34
    B. Discriminatory Intent in the Congressional Plan ........................................................ 38
    IV. State Senate Plan ................................................................................................................. 43
    A. Retrogression in the Senate Plan ................................................................................... 43
    B. Discriminatory Intent in the Senate Plan ..................................................................... 45
    V. State House Plan................................................................................................................... 51
    A. Retrogression in the State House Plan .......................................................................... 51
    1.    Alleged Retrogressive Districts................................................................................ 51
    a. State House District 33 ............................................................................................. 51
    b. State House District 35 ............................................................................................. 53
    c. State House District 41 ............................................................................................. 55
    d. State House District 117 ........................................................................................... 58
    e. State House District 149 ........................................................................................... 60
    f. State House Districts 26, 106, and 144 .................................................................... 66
    2.    Alleged New Ability Districts................................................................................... 67
    B. Discriminatory Intent in the State House Plan ............................................................ 70
    VI. Conclusion ............................................................................................................................ 72
    2
    The latest Census reports that since 2000 the population of Texas grew by over four
    million. This dramatic increase required the Texas legislature to create new voting districts for
    the four seats added to the State’s congressional delegation, U.S. CONST. art. I, § 2, cl. 3; id.
    amend. XIV, § 2, and draw new boundaries for the state and congressional voting districts to
    comply with the mandate of one-person, one-vote, see Georgia v. Ashcroft, 
    539 U.S. 461
    ,
    488 n.2 (2003).
    Because Texas is a covered jurisdiction under section 5 of the Voting Rights Act of 1965
    (VRA), 
    42 U.S.C. § 1973
    , the Attorney General of the United States or a three-judge panel of
    this Court must approve, or “preclear,” any redistricting plan before it can take effect. 
    Id.
    § 1973c(a). Texas chose not to seek administrative preclearance and instead seeks from this
    Court a declaratory judgment that its redistricting plans will neither have “the purpose nor will
    have the effect of denying or abridging the right to vote on account of race or color, or [language
    minority group].” Id. The United States opposes preclearance of the redistricting plans for
    Texas’s congressional delegation and the State House of Representatives, but has no quarrel with
    the plan for the Texas Senate. Seven Intervenors raise a variety of challenges that collectively
    encompass all three plans. We conclude that Texas has failed to show that any of the redistricting
    plans merits preclearance.1
    I.    Background
    On July 19, 2011, Texas filed a complaint in this Court seeking a declaratory judgment
    that its newly enacted redistricting plans for the U.S. House of Representatives (Plan C185 or
    1
    Texas sought declaratory judgment that the three plans comply with section 5 in counts two, three, and four
    of the complaint. In its first count, Texas also sought from this Court preclearance of its redistricting plan for the
    State Board of Education. No party objected to the plan, either in their written answers or during a conference call
    the Court held with the parties on September 21, 2011. With no opposition and satisfied that the State Board of
    Education plan complies with section 5, we granted preclearance for that plan on September 22, 2011. See Minute
    Entry Order, Sept. 22, 2011.
    3
    Congressional Plan), the Texas House of Representatives (Plan H283 or House Plan), and the
    Texas Senate (Plan S148 or Senate Plan) comply with section 5 of the VRA. This Court has been
    properly convened as a three-judge court, 
    28 U.S.C. § 2284
    ; 42 U.S.C. § 1973c(a), and we took
    jurisdiction under 42 U.S.C. § 1973c and 
    28 U.S.C. §§ 1346
    (a)(2), 2201. After the United States
    and several Intervenors2 filed answers, Texas moved for summary judgment for all three plans
    on September 14, 2011. We heard argument on the motion on November 2, 2011, and issued an
    order denying summary judgment on November 8, 2011. Our memorandum opinion followed on
    December 22, 2011.
    The same three redistricting plans have been challenged under section 2 of the VRA
    before a three-judge district court in the Western District of Texas. The State’s population
    growth and the addition of four seats to its congressional delegation make it impossible for Texas
    to conduct elections using the district boundaries last approved under section 5. Our denial of
    Texas’s motion for summary judgment required the district court in the section 2 litigation to
    draw interim maps for the State’s fast-approaching primaries and the ensuing general election.
    After the Supreme Court invalidated those maps, see Perry v. Perez, 
    132 S. Ct. 934
     (2012), the
    court issued a second set, which have not been challenged. See Feb. 28, 2012 Order, Perez v.
    Perry, No. 11-cv-360 (W.D. Tex. filed May 9, 2011), ECF No. 681 (Congressional Plan interim
    map); Feb. 28, 2012 Order, Perez, No. 11-cv-360, ECF No. 682 (House Plan interim map); Feb.
    2
    This Court has granted Defendant-Intervenor status to seven parties, each of whom challenges various aspects
    of some or all of Texas’s proposed plans in their capacities as individual voters, elected state representatives, or civil
    rights advocacy groups. The Davis Intervenors are Texas State senators and representatives from districts in the Fort
    Worth area. The Mexican American Legislative Caucus is a caucus in the Texas House of Representatives. The
    Gonzales Intervenors are a group of Hispanic and Black Texas voters. The Texas Legislative Black Caucus is
    composed of seventeen members of the Texas House of Representatives. The Texas Latino Redistricting Task Force
    is a group of Hispanic organizations focusing on redistricting and voter registration. The Texas State Conference of
    NAACP Branches and the League of United Latin American Citizens are civil rights and advocacy groups
    concerned with minority voting rights in Texas.
    4
    28, 2012 Order, Davis v. Perry, No. 5:11-cv-00788 (W.D. Tex. filed May 9, 2011), ECF No. 141
    (Senate Plan interim map).
    Meanwhile, after expedited discovery, this Court sat for trial January 17-26, 2012, with
    closing arguments on January 31, 2012.3 The voluminous trial record includes evidence taken in
    open court, party exhibits, expert reports, post-trial briefing, and designated portions of the
    transcript from the section 2 trial in Texas.4 After reviewing this record and carefully considering
    the arguments of all parties, we now deny Texas preclearance and enter judgment for the
    defendants.
    In the discussion that follows, we do not recount the extensive background of the Voting
    Rights Act or of this case. Much of that is contained in our opinion at summary judgment. In
    addition, we do not repeat many of the factual findings set out in the appendix to this opinion.
    Using the framework for applying section 5 described in our summary judgment opinion, we
    first address a series of legal issues that remain outstanding after trial about what section 5
    requires for preclearance. Then, we examine the Congressional, Senate, and House Plans in turn.
    II. Principles of Section 5 Analysis
    A. Retrogression
    Texas must show that its redistricting plans have neither the effect nor the purpose of
    abridging minority voting rights. 42 U.S.C. § 1973c(a). We will take up the “purpose” prong
    below in section B. The goal of the “effect” prong is “to insure that no voting-procedure changes
    3
    Given the parties’ unanimous desire to proceed quickly to trial but faced with scheduling constraints from the
    panel members’ previously scheduled proceedings, the Court adopted a trial schedule in which all three judges heard
    evidence during the first four days of trial and two judges heard evidence the last four days, with the third judge
    reviewing the evidentiary record and transcript from those days. All three judges were present for closing
    arguments. The Court divided trial time so that Texas and the United States and the Intervenors would have equal
    time for argument when all three judges were physically present. No party raised an objection to these arrangements.
    4
    The full record in this case runs many thousands of pages, including over a thousand exhibits introduced by
    the parties.
    5
    would be made that would lead to a retrogression in the position of racial minorities with respect
    to their effective exercise of the electoral franchise,” Beer v. United States, 
    425 U.S. 130
    , 141
    (1976), regardless of whether the change was intended to do so. “Effective exercise,” in turn, has
    long been understood to include not only the “ability of minority groups to participate in the
    political process,” but also the ability “to elect their choices to office.” 
    Id.
     (quoting H.R. REP.
    NO. 94-196, at 60 (1975)). In the most recent reauthorization of the VRA, Congress further
    reinforced the meaning of the effect prong by stating that minority voters’ “ability to elect” their
    candidates of choice is the appropriate measure of whether a proposed change will be
    retrogressive. See 42 U.S.C. § 1973c(b) (stating that section 5 blocks voting changes that
    diminish minority citizens’ “ability . . . to elect their preferred candidates of choice”), id.
    § 1973c(d) (explaining that the “purpose of subsection (b) . . . is to protect the ability of
    [minority] citizens to elect their preferred candidates of choice”).
    As we explained in our summary judgment opinion, ensuring that a proposed plan will
    not undo the gains minority voters have achieved in electoral power requires a multi-factored,
    functional analysis. Texas v. United States, 
    831 F. Supp. 2d 244
    , 262-64 (D.D.C. 2011). A
    single-factor inquiry, such as the test Texas proposed relying on racial and ethnic population
    statistics alone, is inconsistent with precedent and too limited to provide an accurate picture of
    the on-the-ground realities of voting power.5 Id.; see also, e.g., Ashcroft, 
    539 U.S. at 480
     (“The
    5
    Indeed, analysis of the full record developed at trial has made it more clear that the test Texas initially
    proposed is insufficient to measure whether minority voters have an ability to elect. Several districts in the proposed
    plans show that population statistics alone rarely gauge the strength of minority voting power with accuracy. For
    example, the discussion that follows shows that Congressional District 23 and House District 117 were selectively
    drawn to include areas with high minority populations but low voter turnout, while excluding high minority, high
    turnout areas. Such districts might pass a retrogression analysis under Texas’s population demographics test (40%
    Black Voting Age Population or 50% Hispanic Citizen Voting Age Population as sufficient to establish ability
    status), even though they were engineered to decrease minority voting power. The 65% presumption of ability status
    we employ, discussed further below, is less susceptible to such problems. Our threshold is significantly higher than
    Texas’s proposed 50% test, and where it is met a district is only presumptively an ability district, not conclusively
    6
    ability of minority voters to elect a candidate of their choice is important but often complex in
    practice to determine.”). We do not repeat here the rationale for our conclusion, but instead
    address the additional arguments raised at trial about the appropriate standard to determine
    retrogression.
    1. Texas’s Burden of Proof
    Texas bears the burden of proving by a preponderance of the evidence that its
    redistricting plans are not retrogressive.6 City of Pleasant Grove v. United States, 
    479 U.S. 462
    ,
    469 (1987). Texas does not deny that it bears this burden. Instead, relying on the Supreme
    Court’s observation that a state is entitled to select its “own method of complying with the
    Voting Rights Act,” Bartlett v. Strickland, 
    556 U.S. 1
    , 23 (2009) (plurality opinion), Texas
    claims that “the flexibility to choose one theory of effective representation over the other,”
    Ashcroft, 
    539 U.S. at 482
    , gives it significant latitude in how to prove its case. Tex. Post-Trial
    Br. 3.
    We agree that section 5 does not interfere with many of the policy judgments a state must
    make during redistricting, such as whether to retain an ability district — a district in which
    minority citizens have the ability to elect their preferred candidates — or create a new one
    elsewhere. Yet Texas takes this point too far, claiming that the prerogative to choose among
    methods of redistricting extends to the type of evidence we should use to measure retrogression.
    so. The 65% presumption may be rebutted by other factors, such as voter turnout, that indicate the district is not
    effective for minority voters.
    6
    Significantly, the State’s expert, Dr. John Alford, declined to offer an opinion on whether the enacted plans
    are retrogressive, even when this Court directly questioned him on the point. He testified that his analysis provided
    only the first steps in the more complicated inquiry this Court must undertake, refused to offer an opinion on the
    number of districts protected by section 5 in the existing and enacted plans, and stated he was not offering an answer
    to the question whether the enacted plans preserve the current degree of ability to elect. See Trial Tr. 63:21-67:10,
    94:21-96:25, Jan. 24, 2012 PM. The State’s failure to produce testimony showing the enacted plans are not
    retrogressive may well be sufficient for us to find that Texas has not met its burden of proof under section 5.
    Nevertheless, because we find that the trial record is sufficient to show that the enacted plans cannot be precleared,
    this failing is not the only ground for our conclusions.
    7
    For example, Texas argues that we must defer to its decision to use the results of statewide
    elections to measure compliance with section 5. Id. at 5. We disagree. Ashcroft holds that states
    may choose between “theor[ies] of effective representation,” 
    539 U.S. at 482
     (emphasis added),
    but gauging effectiveness is a legal judgment that we must make. Texas is entitled to advocate its
    preferred methods of measuring minority voting strength, and we address those arguments
    below, but we need not defer to a state’s legal theory on how best to measure minority voters’
    ability to elect. That is a measure at the heart of the preclearance analysis that section 5 has left
    to the Attorney General or the judiciary.
    2. Election Analysis Methodologies
    The parties have submitted reports and testimony from fourteen experts in fields such as
    redistricting, election analysis, voting rights law, and the history of voting discrimination in
    Texas. Although we do not find the analysis of any one expert sufficient to guide our
    retrogression inquiry, we rely most heavily on the reports and testimony of Dr. Lisa Handley,
    expert for the United States; Dr. Richard Engstrom, expert for the Texas Latino Redistricting
    Task Force (TLRTF); and Dr. Stephen Ansolabehere, expert for the Gonzales Intervenors. We
    find their methodologies sound and their conclusions helpful to our analysis of the State’s
    redistricting. To explain our use of these experts we address two areas of disagreement between
    the parties about the merits of the various approaches the experts use: which type of elections to
    examine and the appropriate sample sets to use.
    a. Types of Elections
    Endogenous analysis examines the results of elections held within a district to determine
    how often minority-preferred candidates succeed.7 See, e.g., Defs.’ Ex. 326, Dr. Lisa Handley, A
    7
    All parties have agreed throughout this litigation that minority voters in Texas vote overwhelmingly
    Democratic, and thus there is generally no dispute about the identity of minority-preferred candidates in a given
    8
    Section 5 Voting Rights Analysis of the Proposed Texas State House Plan 3 [hereinafter Handley
    House Rep.]. Because endogenous analysis is based on actual election results within a single
    district, it is necessarily retrospective. It can only be used to determine whether a district in the
    existing, or benchmark, plan has an ability to elect. It cannot be used to assess whether a
    proposed district does as well, because a proposed district has not yet conducted any district-
    wide elections.
    Exogenous election analysis examines how minority-preferred candidates fared in a
    particular district in statewide or national elections. See, e.g., Pl.’s Ex. 175, Direct Written Test.
    of Dr. John Alford 5-6 [hereinafter Alford Rep.]. Take the 2008 presidential election as an
    example. In a state where minority voters almost always prefer Democratic candidates,
    exogenous election analysis suggests that minority voters lack an ability to elect in a benchmark
    district carried by John McCain over Barack Obama. Because exogenous analysis considers
    results from elections that occur across all districts in a state, such analysis allows comparison
    between benchmark and proposed districts. Precinct-level data from statewide or national
    elections can show if the minority-preferred candidate won the benchmark district, and by
    assembling, or “reconstituting,” the precinct-level returns into a district’s proposed new shape,
    exogenous election analysis can indicate whether the minority-preferred candidate would have
    won in the proposed district as well.
    Texas urges us to consider exogenous election analysis alone, see Tex. Post-Trial Br. 4-5,
    but we conclude that endogenous results are often more probative of ability to elect. As Dr.
    district. See, e.g., Trial Tr. 12:8-14, Jan. 17, 2012 AM (State’s opening statement, noting that “in virtually all of the
    elections in fact, all of the elections you’re going to hear about during this trial” the Hispanic-preferred candidate
    was the Democrat). In light of the parties’ agreement on this point, as a general matter we do not address the racially
    polarized voting data that makes this point. In the few districts in which there is a dispute over who is the candidate
    of choice of minority voters, discussed further below, we credit Dr. Handley’s assessment, which is based on her
    analysis of racial bloc voting in the districts.
    9
    Engstrom explained, exogenous elections are “not a good basis for predicting the specific
    number of elections in many new districts that will result in Hispanic preferred candidates
    winning,” partly because there are significant contextual differences between exogenous and
    endogenous elections. Defs.’ Ex. 747, Rebuttal Report of Dr. Richard Engstrom 6 [hereinafter
    Engstrom Reb. Rep.]. Likewise, Dr. Handley concluded that “the most essential piece of
    information” when determining benchmark ability districts “is whether minority voters have
    been successful at electing their preferred candidates to the legislative office at issue in the
    district.” Defs.’ Ex. 794, Rebuttal Report of Dr. Lisa Handley to Supplement Expert Report of
    Dr. John Alford 3 [hereinafter Handley Reb. Rep.]. Candidates in endogenous elections live in a
    particular district and focus their campaigns on local voters. Candidates in statewide elections are
    likely to make an appeal with a less direct connection to voters in that district. Nationwide
    contests are even more attenuated. Local connections and direct campaigning, then, may allow a
    minority-preferred candidate to win an endogenous election in a district the minority-preferred
    candidate for statewide office could not carry. We agree with Dr. Engstrom and Dr. Handley.
    Given the numerous and difficult-to-quantify factors that go into determining ability to elect, the
    best evidence is whether and how often minority voters have actually elected their candidate of
    choice to the position at issue, not the indirect proxy offered by exogenous analysis.
    Texas argues that endogenous analysis is an “impracticable” tool because it is available
    only for benchmark plans and does not provide the “common unit of measurement” available
    with exogenous results. Tex. Post-Trial Br. 4. As we have stated, we agree that endogenous
    elections are not well suited to prospective analysis, but when predicting the impact of
    redistricting changes on minority voters’ ability to elect, more information is better than less. We
    should not discount the powerful evidence of minority voting power that endogenous elections
    10
    provide in favor of a single tool that may be a less accurate gauge. When endogenous and
    exogenous analyses yield different results, we will give special attention to other relevant
    characteristics of the voting district.
    Texas argues that endogenous analysis may overvalue minority voting power and
    undervalue the advantage of incumbency in districts where the minority-preferred candidate has
    been repeatedly reelected. See id. at 5. We disagree with the premise that an incumbent’s
    advantage does “not bear on the ability-to-elect inquiry.” Id. The advantage incumbents enjoy
    during reelection campaigns is a factor that minority voters, like any other voters, often use to
    help elect their preferred candidate. Ability to elect is not less real simply because subsequent
    elections are easier to win than the first. Texas raises the more specific objection that endogenous
    results may be misleading in a district in which ability status is closely contested if a long-term
    incumbent plans to retire. Id. Yet as our analysis below bears out, our finding that endogenous
    elections are particularly probative evidence does not mean that a high endogenous score
    automatically implies ability status. Careful consideration of all factors matters, especially in
    close cases.
    We thus see no reason to exclude all endogenous election data from our analysis, nor to
    weigh exogenous data more heavily. Both types of data provide information about whether
    minority voters are or will be able to participate in the political process.
    b. Election Analysis Sample Sets
    The experts also vary widely in which elections they used for their sample sets. All use a
    similar methodology for their exogenous analysis. Starting with the boundaries in the benchmark
    plan, they count the number of times the minority-preferred candidate carried the district.
    Reconfiguring the districts by regrouping precincts as called for in the enacted plan, their
    11
    analyses then look to see how many times the minority-preferred candidate would have carried
    that district. Outcomes are determined by inputs, of course, and whether the analysis shows an
    ability to elect turns on variations in the sample set such as the number of elections chosen, the
    length of time they span, whether the sample is weighted toward more recent contests, and the
    offices at stake. For example, Texas’s expert, Dr. Alford, relies on reconstituted election results
    from a set of ten statewide elections weighted toward more recent years provided by the Texas
    Office of the Attorney General (the OAG 10). See Alford Rep. 9 tbl.2.8 Texas argues that we
    should give greatest weight to these exogenous results because they used a larger data set and
    relied more heavily on recent elections than did any other expert in the case.9 Tex. Post-Trial
    Br. 5.
    Although we agree that a larger data set generally improves accuracy, we are not
    persuaded that the OAG 10 is the best indicator of minority voting strength. A preference for
    recent elections may in fact distort the results. Dr. Handley, the expert for the United States,
    cautions against giving more weight to some years than others. To do so, she warns, would allow
    atypical election years to skew the picture of long-term minority voting power. See Handley Reb.
    Rep. 4 n.6. This caution is especially appropriate here because three of the OAG 10 elections are
    from the 2010 election cycle. As the evidence in this case shows, 2010 was an unusual year with
    8
    The OAG 10 includes one 2002 contest; two contests each from 2004, 2006, and 2008; and three contests
    from 2010. Dr. Alford’s analysis includes results using all ten of these contests, and also using only the five most
    recent elections on this list. See Alford Rep. 8-9.
    9
    Texas’s reliance on the OAG 10 exogenous analysis is a litigation position; the record is clear that this
    functional election analysis played little to no role in the map-drawing process itself. The OAG did not identify
    which districts were protected in the benchmark plans or even how many benchmark ability districts existed. In fact,
    the only evidence that analysis was performed regarding these critical facts was testimony from the primary House
    mapdrawer, Gerardo Interiano, that he made an effort to identify Hispanic ability districts in the benchmark. Trial
    Tr. 25:5-26:10, Jan. 17, 2012 PM. Both Interiano and the other main mapdrawer, Ryan Downton, testified that they
    did not look at the OAG 10 analysis of the benchmark and enacted districts until their work was essentially
    complete. See id. at 57:17-25, Jan. 18, 2012 AM; Trial Tr. 14:51-52, Perez, No. 11-cv-360, Sept. 12, 2011. And
    there is no evidence that the legislators and mapdrawers made any modifications to the proposed district lines when
    they did consult the OAG 10 analysis late in the process.
    12
    low Democratic turnout in which Republicans won several seats that had long been held by
    Democrats. See, e.g., Defs.’ Ex. 776, Seliger Dep. 15:1-7, Sept. 1, 2011, Perez, No.
    11-cv-360 [hereinafter Seliger San Antonio Dep.] It is too soon to tell if 2010 was an aberration
    or marked the start of a lasting change in Texas politics.
    Our concerns with the OAG 10 extend to the other sample sets used by the parties’
    experts. Dr. Engstrom’s exogenous election sample also places greater weight on recent years,
    considering elections from only 2006-2010. See Defs.’ Ex. 726, Supplemental Expert Report of
    Dr. Richard Engstrom 2 [hereinafter Engstrom Suppl. Rep.]; Defs.’ Ex. 799, Dr. Richard
    Engstrom Analysis: Retrogression in State’s Adopted House Plan [hereinafter Engstrom Chart].
    And all the experts in this case use relatively small sample sets. Dr. Handley, for example, uses
    only five elections from 2002-2010, and Dr. Engstrom uses just seven general elections. Handley
    House Rep. 3-4; Engstrom Suppl. Rep. 2; Engstrom Chart. Where there are so many elections
    from which to choose — the record contains analysis using races ranging from governor to
    railroad commissioner — it is hard to assess the merits of any one expert’s data when the sample
    sets are small and often do not overlap. In short, we are uncomfortable relying exclusively on the
    exogenous analysis of any single expert. Our solution is to consider the exogenous results from
    all three of these sources — the OAG 10, Dr. Handley, and Dr. Engstrom — as well as the
    analysis for the Congressional Plan conducted by Dr. Stephen Ansolabehere, the expert for the
    Gonzales Intervenors, finding all to be probative but none dispositive.
    3. Statewide Retrogression Analysis
    As the Supreme Court has made clear, our analysis of minority voting power “must
    encompass the entire statewide plan as a whole.” Ashcroft, 
    539 U.S. at 479
    . Section 5 is not
    concerned with the location of particular ability districts, but rather with whether the enacted
    13
    plan, in its entirety, preserves minority voters’ ability to elect. In other words, section 5 allows a
    state to dismantle an ability district as long as it offsets that loss by drawing a new ability district
    elsewhere.
    But Texas asks us to expand this principle to a point that is inconsistent with section 5.
    Texas’s expert submitted two reports to the Court, one at summary judgment and another at trial.
    His first report counted any district in which the number of registered Hispanic voters exceeded
    50% or the Black Voting Age Population (BVAP) exceeded 40% as an ability district, without
    giving attention to actual election performance. See Texas, 831 F. Supp. 2d at 263 n.23. After we
    rejected this single-factor test, Dr. Alford changed tack in his trial report, which uses what he
    calls a “statewide functional analysis.” See Alford Rep. 7. Rather than determine if particular
    districts are ability districts, Dr. Alford’s latest approach examines changes in the degree of
    minority voting power across the entire plan. Using the benchmark ability districts the United
    States listed, Dr. Alford counted every instance in which a minority-preferred candidate carried
    the district in an exogenous election. He then counted how many times the minority-preferred
    candidate would have carried the district in the enacted plan. If the total number of “wins” in the
    enacted plan meets or exceeds the number in the benchmark, Dr. Alford concludes that the plan
    is not retrogressive. See id. at 7-12. Dr. Alford contrasts his statewide approach to what he calls
    the “binary” approach of every other expert in the case. Those experts examine each district
    individually, using exogenous results as one factor when determining if a district is an ability
    district. See id. at 12-13. They then compare the number of ability districts in the benchmark
    map with the number in the proposed plan. Dr. Alford’s method counts election victories across
    all districts and does not label a district as “ability” or not. Texas argues this approach is superior
    14
    to the “blunt technique” of the binary method because it “captures the degree of minority voting
    strength across all relevant districts.” Tex. Post-Trial Br. 6.
    Perhaps, but this approach is a variation on the type of retrogression analysis that
    Congress rejected when it amended the VRA in 2006. In Georgia v. Ashcroft, 
    539 U.S. 461
    (2003), the Supreme Court concluded that courts “should not focus solely on the comparative
    ability of a minority group to elect a candidate of its choice,” but instead should consider the
    “totality of the circumstances” regarding minority participation in the electoral process. 
    Id. at 479-80
    . Specifically, the Court concluded that states could draw maps containing a combination
    of two different types of districts to satisfy section 5: traditional majority-minority districts, and
    “influence districts,” which are not ability districts, but rather those in which minority voters play
    a “substantial, if not decisive, role in the electoral process.” 
    Id. at 480-83
    .
    Congress rejected this holding in 2006 when it reauthorized section 5, making it clear that
    retrogression is not concerned with the degree of influence minority voters exert, but with their
    ability to elect their preferred candidates. See 42 U.S.C. § 1973c(b) (stating that voting changes
    must not diminish minority citizens’ “ability . . . to elect their preferred candidates of choice”),
    id. § 1973c(d) (defining subsection (b)’s purpose as protecting “the ability of [minority] citizens
    to elect their preferred candidates of choice”). The House Report explained that the 2006
    amendments were a response to Georgia v. Ashcroft, which allowed “the minority community’s
    own choice of preferred candidates to be trumped by political deals struck by State legislators
    purporting to give ‘influence’ to the minority community while removing that community’s
    ability to elect candidates.” H.R. REP. NO. 109-478, at 69 (2006). Congress decided that
    “[p]ermitting these trade-offs is inconsistent with the original and current purpose of Section 5.”
    Id.; see also id. at 68-72; S. REP. NO. 109-295, at 18-20 (2006) (stating that the amendments
    15
    “clarify that [section 5] protects the ability of minority voters ‘to elect their preferred candidates
    of choice,’” id. at 19). Congress does not view “ability to elect” in degrees; states may not add up
    districts in which minority voters have “partial” ability to elect to satisfy section 5. Instead,
    Congress views ability status as an on-off switch: minority voters either have an ability to elect
    in a district or they do not.
    Endorsing Dr. Alford’s analysis would be a return to the approach Congress rejected in
    2006. Consider, for example, a benchmark map with three districts. In two of the districts,
    minority voters elect their preferred candidates in six out of ten elections in a sample set, but in
    the third, they fail to win a single election. In all three districts in the enacted plan, minority-
    preferred candidates win in four out of the ten elections. A traditional binary approach would
    likely conclude that the benchmark map has two ability districts (where minority voters can elect
    their candidate of choice more often than not), and the enacted plan has no ability districts, just
    three influence districts. Such a plan would be clearly retrogressive under the current version of
    section 5. Yet Dr. Alford’s approach would show no retrogression because the total number of
    minority electoral victories remains the same (6 + 6 + 0 = 12 in the benchmark; 4 + 4 + 4 = 12 in
    the enacted).
    Texas argues that Dr. Alford’s approach yields better policy results, but such
    determinations belong to Congress, not the courts. In any event, the “benefits” Texas touts are
    illusory. Texas argues that the binary approach “ignores gradations in minority abilities to elect
    and gives States no credit for improving electoral performance in districts that stay above or
    below the ability-to-elect cutoff.” Tex. Post-Trial Br. 6.10 In other words, Texas seeks credit for
    10
    This observation is accurate, but we also note that the binary approach runs both ways: under a retrogression
    analysis, a State may not claim credit for strengthening an ability district, but neither is it penalized for reducing
    minority voting power in districts that are trending toward minority ability status but have not yet achieved it, as we
    discuss below with respect to HDs 26, 106, and 144.
    16
    strengthening an already-performing district from, say, six out of ten victories to ten out of ten.
    Yet giving credit in a scenario like this would allow Texas to use those four “additional”
    victories to offset a four-election decrease elsewhere. Such an approach would create a legal tool
    to dismantle ability districts as long as the state increases the effectiveness of others. In short, it
    would give states credit for packing minority voters into districts. A starker example would be a
    plan in which six benchmark districts perform for minority voters nine out of ten times, but
    perform ten out of ten times in the enacted plan. Statewide functional analysis would allow a
    state to use this six-election “increase” in minority effectiveness to weaken or even destroy
    ability districts in other parts of the state.
    We also find it significant that Dr. Alford can point to no other advocates of his approach
    within the well-populated field of voting rights and redistricting. Statewide functional analysis is
    not only foreclosed by the 2006 amendments, but it lies outside accepted academic norms among
    redistricting experts. See, e.g., Engstrom Reb. Rep. 2-6 (critiquing Dr. Alford’s approach and
    noting he was “not aware of any analysis, prior to this one by Dr. Alford, by any expert that
    completely ignores the results of endogenous elections in a benchmark plan in a retrogression
    analysis,” id. at 2); Handley Reb. Rep. 2-6 (critiquing Dr. Alford’s approach).
    Moreover, statewide functional analysis would be much more difficult to administer than
    the already fact-intensive binary approach because courts would need to make even more precise
    findings than whether a district is or is not an ability district. Courts would need to determine, for
    example, the difference between districts with effectiveness levels of 60% and 70%. Dr. Alford
    claims he can make these fine distinctions based on a district’s electoral performance in the
    limited set of elections that he chose. Yet as the multitude of experts in this case demonstrates,
    there is no agreed-upon method to choose how many elections are necessary to demonstrate
    17
    voting strength, much less which elections and over how long a period of time. We lack
    confidence that we, or any court, would be able to make findings at the level of precision Dr.
    Alford’s approach requires.
    Finally, we reject Texas’s argument that refusing to accept statewide functional analysis
    would increase the “substantial federalism costs” of preclearance by further limiting state
    flexibility, at the risk of rendering section 5 unconstitutional.11 See Tex. Post-Trial Br. 7 (quoting
    Reno v. Bossier Parish Sch. Bd. (Bossier II), 
    528 U.S. 320
     (2000)) (internal quotation marks
    omitted). The constitutional avoidance canon is no aid to Texas because we are not faced with
    two competing yet permissible interpretations of section 5. See United States v. X-Citement
    Video, Inc., 
    513 U.S. 64
    , 69 (1994) (describing the interpretative presumption “that a statute is to
    be construed where fairly possible so as to avoid substantial constitutional questions” (emphasis
    added)). As we have just discussed, retrogression analysis under section 5 as amended limits our
    analysis to ability to elect and does not permit us to weigh degrees of effectiveness. We cannot
    adopt an interpretation at odds with the statutory text to avoid possible constitutional concerns.
    4. Coalition and Crossover Districts
    a. Section 5 Analysis
    In a crossover district, a minority group “is large enough to elect the candidate of its
    choice with help from voters who are members of the majority and who cross over to support the
    minority’s preferred candidate.” Bartlett, 
    556 U.S. at 13
    . In a coalition district, two or more
    minority groups work together to elect their preferred candidate. 
    Id.
     We held at summary
    11
    Because Texas has not raised the argument, we have no opportunity in this case to consider whether the
    federalism costs of preclearance, when weighed against current conditions, call into question the constitutionality of
    section 5’s remedial scheme. Cf. Nw. Austin Mun. Util. Dist. No. One v. Holder, 
    557 U.S. 193
    , 202-05 (2009)
    (noting the Court’s serious concerns that “current needs” may no longer justify the burdens preclearance imposes on
    covered jurisdictions). The constitutionality of section 5 was neither briefed nor argued to us, and we express no
    opinion on this significant point. In fact, our Circuit has recently held that section 5 is constitutional. See Shelby
    Cnty. v. Holder, 
    679 F.3d 848
     (D.C. Cir. 2012).
    18
    judgment that because existing “coalition and crossover districts provide minority groups the
    ability to elect a preferred candidate, they must be recognized as ability districts in a Section 5
    analysis of a benchmark plan.” Texas, 831 F. Supp. 2d at 267-68. Texas asks us to reconsider our
    ruling in light of Bartlett v. Strickland, 
    556 U.S. 1
     (2009) (plurality opinion), and the Supreme
    Court’s recent decision in Perry v. Perez, 
    132 S. Ct. 934
     (2012). Having considered the parties’
    arguments, we reaffirm our conclusion that coalition and crossover districts are protected under
    section 5.
    The Supreme Court has never directly addressed whether section 5 protects coalition or
    crossover districts. A close reading of Georgia v. Ashcroft, however, suggests that it does. The
    Court described districts with “coalitions of voters who together will help to achieve the electoral
    aspirations of the minority group,” 
    539 U.S. at 481
    , concluding that such districts count as
    “effective representation” for purposes of section 5, just like “safe majority-minority districts.”
    
    Id. at 480-82
     (“Section 5 gives States the flexibility to choose one theory of effective
    representation over the other.” 
    Id. at 482
    .).12 The Court’s statements in Georgia v. Ashcroft are
    reinforced by the House Report accompanying the 2006 amendments, which spoke of coalition
    districts as a type of ability district: “Voting changes that leave a minority group less able to elect
    a preferred candidate of choice, either directly or when coalesced with other voters, cannot be
    precleared under Section 5.” H.R. REP. NO. 109-478, at 71 (emphasis added).13
    12
    Although the 2006 amendments rejected the portion of Georgia v. Ashcroft that directed courts to consider
    factors other than ability to elect in their retrogression analyses, this passage is from the opinion’s earlier section
    describing ability to elect.
    13
    As we noted at summary judgment, Senator Kyl wrote separately a week after the passage of the
    amendments “to explain why [he] believe[d] that Congress cannot require that state or local governments create or
    retain influence or coalition districts,” S. REP. NO. 109-295, at 22 (additional views of Senator Kyl), but those
    individual views were filed a week after the VRA had passed both houses of Congress, were not considered by
    Congress prior to the vote, and were neither adopted nor affirmed by Congress in its findings accompanying the
    2006 amendments. See Texas, 831 F. Supp. 2d at 267 n.30.
    19
    In addition, the Court’s jurisprudence under section 2, a distinct yet related provision of
    the VRA mandating equal opportunity for minority voters to participate in the electoral process,
    supports protecting coalition and crossover districts under section 5. The Court has long
    acknowledged the existence of coalition and crossover districts, recognizing at times that they
    can provide the means for minority voters to elect their candidates of choice. See Johnson v. De
    Grandy, 
    512 U.S. 997
    , 1020 (1994) (describing “communities in which minority citizens are able
    to form coalitions with voters from other racial and ethnic groups, having no need to be a
    majority within a single district in order to elect candidates of their choice” (emphasis added));
    Voinovich v. Quilter, 
    507 U.S. 146
    , 154 (1993) (describing a district in which a minority group
    was not large enough to elect its preferred candidate operating alone but could do so if it
    “attract[ed] sufficient cross-over votes from white voters”).14 In fact, the Court has suggested
    that such districts will become more common over time, replacing majority-minority districts as
    waning racial polarization makes it easier for minority voters to elect their preferred candidates
    even when they do not make up the majority of a district’s voters. See De Grandy, 
    512 U.S. at 1019-20
    . In other words, “ability” may look different now than it did when the VRA was first
    enacted. Our responsibility to protect the rights secured by section 5 calls that we be sensitive to
    these new, but real, forms of minority voting power.
    Texas argues that the Court’s decision in Bartlett precludes recognizing coalition and
    crossover districts under section 5. See Tex. Post-Trial Br. 8. But the Bartlett Court only
    concluded that section 2 does not compel states to draw new crossover districts under section 2,
    not that states can disregard the existence of established crossover and coalition districts in a
    14
    In lower court section 2 cases, courts have also frequently referred to coalition and crossover districts using
    the same adjectives used to describe traditional majority-minority districts, such as “performing,” “effective,” and
    “ability.” See, e.g., Ariz. Minority Coal. for Fair Redistricting v. Ariz. Indep. Redistricting Comm’n, 
    366 F. Supp. 2d 887
    , 904 (D. Ariz. 2005) (describing this trend).
    20
    section 5 inquiry.15 See Texas, 831 F. Supp. 2d at 267-68. Significantly, Bartlett noted that it did
    not reach the question of whether states could choose to draw crossover districts “as a matter of
    legislative choice or discretion,” and cited Georgia v. Ashcroft to show that drawing such
    districts may be the most effective way to strengthen minority voting power. Bartlett, 
    556 U.S. at 23
    . Far from revealing skepticism or hostility toward coalition districts, this language suggests
    that such districts can increase minority voters’ electoral ability, even while holding that states
    are not required to draw districts maximizing this potential.
    Nor do the Bartlett Court’s concerns under section 2 speak to our task under section 5.
    Part of the Court’s analysis rested on the difficulties of predicting whether a potential coalition
    would provide minorities with an opportunity to elect. 
    Id. at 17
    . Section 5, by contrast, asks
    whether an existing coalition has achieved an ability to elect. Section 5 does not call on us to
    guess the future, but to determine whether there is past evidence of a demonstrated ability to
    elect. And while section 2 does not demand granting “special protection to a minority group’s
    right to form political coalitions” or “impose on those who draw election districts a duty to give
    minority voters the most potential, or the best potential, to elect a candidate by attracting
    crossover voters,” 
    id. at 15
    , section 5 mandates that we ensure that “the gains thus far achieved in
    minority political participation [are] not destroyed,” Beer, 
    425 U.S. at 141
     (quoting S. REP. NO.
    94-295, at 19 (1975)). To be sure, forcing a state to create crossover districts would reach beyond
    section 2’s equality mandate, but nothing in Bartlett suggests that courts can turn a blind eye
    towards a district in which minority voters have already turned electoral opportunity into ability
    to elect.
    15
    Bartlett’s holding was limited to crossover districts. It did not analyze coalition districts. See Bartlett, 
    556 U.S. at 13-14
    .
    21
    And nothing in Perez extends the reasoning in Bartlett to section 5. Perez held only that
    the district court had no basis to draw a new coalition district under section 2, without addressing
    the separate question before us: whether preexisting coalition or crossover districts merit
    protection under section 5. See Perez, 
    132 S. Ct. at 944
    . Thus, although section 2 does not
    require states to draw new crossover districts, we read section 5’s ban on retrogression to extend
    protection to districts in which minority voters have demonstrated an ability to elect their
    preferred candidates via either assembling a coalition or attracting sufficient crossover votes, or
    both.
    b. Standard of Proof
    As we stated in our summary judgment opinion, proving the existence of coalition and
    crossover districts “require[s] more exacting evidence than would be needed to prove the
    existence of a majority-minority district.” Texas, 831 F. Supp. 2d at 268. The discussion that
    follows explains the test we have applied.
    At the outset, the minority group or groups must vote cohesively in coalition and
    crossover districts, just as they must in protected majority-minority districts. See Growe v.
    Emison, 
    507 U.S. 25
    , 41 (1993) (noting that proving political cohesion across an “agglomerated
    political bloc” — i.e, a coalition — “is all the more essential” than the need to prove cohesion
    within a single minority group).16 If minority groups split their vote between opposing candidates
    in the general election, there is by definition no candidate of choice, and the district is not
    protected under section 5.
    16
    Texas suggests that the test for proving cohesion across a coalition requires proof that the coalition votes
    together in primaries, not just general elections. The TLRTF joins Texas’s position in its post-trial submissions. See
    TLRTF Response to the Ct.’s Order of Mar. 6, 2012, at 9, ECF No. 219 (relying on Democratic primary results in
    Congressional District 25 as support for a conclusion that the district is not protected under section 5). As explained
    in our discussions of Congressional District 25 and House District 149 below, we reject this argument.
    22
    While the first inquiry considers whether minority voters have a candidate of choice, the
    next inquiry is grounded in a different part of section 5: do minority voters have the “ability to
    elect” their preferred candidate? See 42 U.S.C. § 1973c. In other words, are the groups large
    enough, motivated enough, or influential enough to elect their candidate of choice — and have
    they in fact done so? This question is in many respects similar to that for majority-minority
    ability districts. There is no single, clearly defined metric to determine when a minority group
    has an ability to elect, so we use a multi-factored approach to determine when a coalition or
    crossover district achieves that ability. See Growe, 
    507 U.S. at 41
     (pointing to anecdotal
    evidence, statistical evidence of minority political cohesion, and racial bloc voting as some of the
    factors relevant to prove the existence of a coalition district under section 2); Texas, 831 F. Supp.
    2d at 268 (“[T]here must be discrete data, by way of election returns, to confirm the existence of
    a voting coalition’s electoral power.”).
    A coalition district is protected under section 5 if there is sufficient evidence to find that
    minorities vote cohesively and have the ability to elect their preferred candidates. The same two
    inquiries apply to a crossover district, but the ability-to-elect analysis is more complicated.
    Although election returns are necessary to show that minority voters in a crossover district have
    a track record of success — results are the coin of the realm — it is not enough that they simply
    go along with the electoral decisions of some of the district’s Anglo voters.17 We must also be
    satisfied that it is the minority voters themselves who have the ability to elect their preferred
    candidate.
    17
    The same concern exists in majority-minority ability districts. A minority group that has low election day
    turnout despite comprising a little over half of the district’s voting age population may find itself consistently on the
    winning side in the district while providing relatively few votes and little influence. Nevertheless, courts have
    generally presumed that success electing the minority-preferred candidate in a majority-minority district is sufficient
    to find ability status. That such a presumption is rebuttable illustrates that we are not requiring a different kind of
    proof for coalition and crossover districts, only more exacting evidence.
    23
    The test to establish this ability must be rigorous enough to avoid the scenario Texas
    describes: that section 5 will be interpreted to protect any district that elects a Democrat, no
    matter how small its minority population. In other words, that minority voters are needed to win
    an election does not in itself prove that they have an ability to elect. As an extreme example of
    this concern, consider a district with a 90% Anglo and 10% minority population. If the Anglo
    vote splits evenly between Democrats and Republicans and minorities vote overwhelmingly
    Democratic, then the Democratic candidate will win with approximately 55% of the vote, and the
    minority vote will properly be viewed as essential to victory every time. Yet this would not be a
    district in which the minority group has an ability to elect; the Anglos do. Such a district would
    merely be a Democratic district that happens to contain a minority group. If we were to hold
    otherwise, then every district that consistently elects a Democrat with the minority vote
    providing the margin of victory, no matter how small, would qualify for protection under section
    5. This would stretch the scope of section 5 too far. A protected crossover district is not created
    each time Anglos and minorities vote together to elect a candidate.
    With respect to both coalition and crossover districts, we require “more exacting
    evidence” to prove that minority voters have an ability to elect than we do for majority-minority
    ability districts. Texas, 831 F. Supp. 2d at 268. Doing so ensures that we stay within the
    boundaries of section 5 and protect only those districts in which minority voters have
    demonstrated their effectiveness. Yet where that standard is met — where minority voters
    themselves “pull, haul, and trade” to elect their preferred candidates, De Grandy, 
    512 U.S. at
    1020 — then the district is one in which minority voters have an ability to elect, and section 5’s
    24
    safeguards apply.18
    B. Discriminatory Intent
    In Reno v. Bossier Parish School Board (Bossier II), 
    528 U.S. 320
     (2000), the Supreme
    Court considered whether section 5 barred a plan that “would have no retrogressive effect” but
    “nonetheless . . . was enacted for a discriminatory ‘purpose.’” 
    Id. at 325
    . The Court held that it
    did not, concluding that the purpose prong extended only to intent to retrogress, not to all
    intentional discrimination. Thus, section 5, the Court wrote, would catch only an “incompetent
    retrogressor,” but offered no recourse against a mapdrawer who intended to discriminate against
    minority voters using methods that did not create retrogression. 
    Id. at 332
    . In direct response, the
    2006 amendments to section 5 clarified that the term “purpose” must be read more broadly and
    includes “any discriminatory purpose.” 42 U.S.C. § 1973c(c); see also H.R. REP. NO. 109-478, at
    93 (stating that Congress “rejects the Supreme Court’s holding in Reno v. Bossier Parish”). As a
    result, we may not preclear any redistricting plan enacted with discriminatory intent.
    Texas argues that it should not be required to prove that it lacked any discriminatory
    purpose. Saddling a state with that burden, so the argument goes, adds too much to the serious
    federalism costs already imposed by preclearance and could “exceed Congress’ enforcement
    authority under the Fifteenth Amendment and violate the Tenth Amendment.” Tex. Post-Trial
    Br. 17-18. The only way to avoid this problem, Texas claims, is to shift the burden of proof for
    discriminatory intent from Texas onto the United States and the Intervenors. Id. at 18. We
    acknowledge the substantial federalism costs of section 5, see Nw. Austin Mun. Util. Dist. No.
    One v. Holder, 
    557 U.S. 193
    , 203-04 (2009) (stating that the preclearance remedy implicates
    18
    As described further in our discussions of Congressional District 25 below, although the Court agrees on the
    general standard outlined above, we disagree on the appropriate test to determine when minority voters possess
    sufficient voting power to have established their ability to elect.
    25
    serious federalism concerns), and recognize the difficulty of proving a negative. Yet it is settled
    law that Texas bears the burden of proving lack of discriminatory intent. See, e.g., Pleasant
    Grove, 
    479 U.S. at 469
     (“The burden of proving absence of discriminatory purpose and effect is
    on [the covered jurisdiction].”); City of Rome v. United States, 
    446 U.S. 156
    , 183 n.18 (1980)
    (“Under § 5, the city bears the burden of proving lack of discriminatory purpose and effect.”);
    Beer, 
    425 U.S. at 140-41
    ; Georgia v. United States, 
    411 U.S. 526
    , 538 (1973); South Carolina v.
    Katzenbach, 
    383 U.S. 301
    , 335 (1966). Texas has pointed to no evidence that Congress intended
    to modify this established understanding.
    Moreover, Texas’s burden is not insurmountable.19 There is no question, as the
    Supreme Court has previously stated, that “assessing a jurisdiction’s motivation in enacting
    voting changes is a complex task requiring a ‘sensitive inquiry into such circumstantial and
    direct evidence as may be available.’” Reno v. Bossier Parish Sch. Bd. (Bossier I), 
    520 U.S. 471
    ,
    488 (1997) (quoting Vill. of Arlington Heights v. Metro. Hous. Dev. Corp., 
    429 U.S. 252
    , 266
    (1977)). And as Texas rightly argues, this task is all the more difficult because disparate impact
    alone is insufficient to establish discriminatory purpose, see Bush v. Vera, 
    517 U.S. 952
    , 968
    (1996) (plurality opinion) (“If district lines merely correlate with race because they are drawn on
    the basis of political affiliation, which correlates with race, there is no racial classification to
    justify . . . .”). But we have clear direction how to conduct this “complex task” from Village of
    Arlington Heights v. Metropolitan Housing Development Corp., 
    429 U.S. 252
     (1977). See
    Bossier I, 
    520 U.S. at 488
     (“In conducting [a section 5 purpose] inquiry, courts should look
    19
    While Texas ultimately bears the burden of proving nondiscrimination, it may shift that burden to the
    defendants by making out a prima facie case for nondiscrimination. See Bossier Parish Sch. Bd. v. Reno, 
    907 F. Supp. 434
    , 446 (D.D.C. 1995), vacated on other grounds, 
    520 U.S. 471
     (1997) (noting that in section 5 cases
    “something like a burden shifting must occur in this, as in every other, civil case,” and that once “[a jurisdiction]
    makes out its prima facie case, it is entitled to preclearance unless its prima facie case is rebutted”). After the
    defendants respond to the prima facie case, the issue becomes whether Texas’s “evidence is more persuasive than
    the evidence proffered against it.” 
    Id.
    26
    to . . . Arlington Heights for guidance.”); see also H.R. REP. NO. 109-478, at 68 (“[T]he factors
    set out in [Arlington Heights] provide an adequate framework for determining whether voting
    changes submitted for preclearance were motivated by a discriminatory purpose . . . .”). There,
    the Court set forth a framework for analyzing “whether invidious discriminatory purpose was a
    motivating factor” in a government body’s decisionmaking. Arlington Heights, 
    429 U.S. at 266
    ;
    see also Bossier I, 
    520 U.S. at 488-89
     (collecting cases in which courts have applied Arlington
    Heights in the section 5 context). We follow this well-worn path and base our inquiry upon the
    five Arlington Heights factors: (1) discriminatory impact, (2) historical background, (3) sequence
    of events leading up to the decision, (4) procedural or substantive deviations from the normal
    decisionmaking process, and (5) contemporaneous viewpoints expressed by the decisionmakers.
    Arlington Heights, 
    429 U.S. at 266-68
    . Texas can carry its burden by showing that these
    factors — the longstanding yardstick for determining discriminatory intent — do not, taken
    together, show discriminatory purpose.
    III. Congressional Plan
    We now turn to the merits of the three plans before us, considering in turn whether
    Texas has carried its burden to prove the absence of discriminatory purpose and effect in the
    Congressional, Senate, and House Plans.
    A. Retrogression in the Congressional Plan
    There are thirty-six districts in the enacted Congressional Plan. Certain Intervenors argue
    that the enacted plan has one fewer ability district than the benchmark because three ability
    districts — Congressional Districts (CDs) 23, 25, and 27 — are lost and only two ability
    districts — CDs 34 and 35 — are added. There is no dispute that these two new districts are
    Hispanic ability districts. Texas agrees that CD 27 is a lost ability district, but disputes that
    27
    benchmark CDs 23 and 25 are ability districts. Under Texas’s theory, the Congressional Plan
    results in a net increase of one Hispanic ability district.
    The United States and certain Intervenors argue that the enacted Congressional Plan
    retrogresses by failing to draw an additional Hispanic ability district. They assert that CDs 23
    and 27, but not CD 25, were Hispanic ability districts in the benchmark whose loss in the enacted
    plan is offset by the gain of CDs 34 and 35. Nevertheless, in light of the growth in the State’s
    Hispanic population, they argue that failing to draw one of the four new congressional districts as
    a Hispanic ability district increases the degree of Hispanic disenfranchisement from the
    benchmark level and thus violates section 5.
    In addition to these arguments about Hispanic ability districts, some of the Intervenors
    argue that the Congressional Plan is retrogressive with respect to Black voters as well. All parties
    agree that CDs 9, 18, and 30 are ability districts for Black voters in both the benchmark and
    enacted congressional maps. Some of the Intervenors allege that the enacted plan “packed” these
    districts with Black voters from neighboring jurisdictions that were not performing for minority
    voters. But because section 5’s effect prong does not prohibit reductions in minority voting
    power in nonability districts, we find no retrogression in Black ability districts in the
    Congressional Plan.
    We do, however, conclude that the enacted Congressional Plan is retrogressive and thus
    cannot be precleared under section 5. Although we differ among ourselves whether benchmark
    CD 25 was an ability district, this disagreement does not affect our overall conclusion. At the
    outset, we discuss the two disputed ability districts upon which we agree, then explain the
    majority’s conclusion that Texas was required to draw a new ability district under section 5. We
    set out our separate views on CD 25 at the end of the opinion.
    28
    1. Congressional District 27
    Benchmark CD 27 includes the cities of Corpus Christi and Brownsville in southeastern
    Texas. With a Hispanic Citizen Voting Age Population (HCVAP) of 63.8%, Pl.’s Ex. 11, at 9,
    and, until 2010, a twenty-seven year history of representation by a Hispanic Democrat,
    benchmark CD 27 is a clear Hispanic ability district. Although an Anglo Republican won the
    seat with a 775 vote margin in 2010, Pl.’s Ex. 32, at 13, no party argues that this anomalous
    result is reason to doubt the district’s status as an ability district. Indeed, Texas’s own expert
    conceded that the district had “performed” from the time of its creation for close to thirty years
    until the 2010 election, Defs.’ Ex. 581, Trial Tr. 1870:16-1871:4, Sept. 14, 2011, Perez, No.
    11-cv-360, and Kel Seliger, chairman of the Texas Senate Select Committee on Redistricting,
    testified that benchmark CD 27 is clearly protected by the VRA and that he felt the legislature
    needed to draw another district to compensate for its loss, Seliger San Antonio Dep. 25:22-26:13;
    see also Trial Tr. 17:19-18:11, Jan. 24, 2012 AM.
    The enacted plan pivots CD 27 roughly 180 degrees such that the old northern boundary
    of the district is now the new southern boundary, with new CD 34 filling in much of CD 27’s old
    geography. The result is that enacted CD 27 is a majority-Anglo district: HCVAP drops to only
    41.1%. Pl.’s Ex.12, at 9. All parties agree that these significant geographic and demographic
    shifts mean that CD 27 will no longer perform for minority voters. We agree.
    2. Congressional District 23
    West Texas’s CD 23 has a complicated history under the VRA. In 2006, the Supreme
    Court held that CD 23, as then constituted, violated section 2. See LULAC v. Perry, 
    548 U.S. 399
    , 425-42 (2006). In response, the U.S. District Court for the Eastern District of Texas redrew
    its boundaries in 2006 to be an “opportunity district,” or one in which Hispanic voters would
    29
    have an opportunity to elect their preferred candidates, as required by section 2. See Defs.’ Ex.
    575, Trial Tr. 300:13-18, Sept. 7, 2011, Perez, No. 11-cv-360. We now find that the Hispanic
    voters in CD 23 turned that opportunity into a demonstrated ability to elect, but that the 2010
    redistricting took that ability away.
    Benchmark CD 23 has an HCVAP of 58.4%. Pl.’s Ex. 11, at 9. During the most recent
    redistricting, the mapdrawers in the Texas legislature acknowledged that CD 23 was a protected
    district under the VRA. See, e.g., Seliger San Antonio Dep. 13:19-15:11, 30:6-15, 31:6-16
    (testimony of Chairman Seliger describing his belief during the redistricting process that CD 23
    was a protected Hispanic district); Defs.’ Ex. 978 (email from congressional mapdrawer Doug
    Davis to National Republican Congressional Committee staffer noting VRA concerns when
    drawing CD 23). CD 23 elected the minority-preferred candidate in two out of the three
    endogenous elections since its boundaries were redrawn in 2006. Defs.’ Ex. 327, Dr. Lisa
    Handley, A Section 5 Voting Rights Analysis of the Proposed Texas Congressional Plan 5
    [hereinafter Handley Cong. Rep.]. The one narrow loss was in 2010, a year that Chairman
    Seliger described as “a bit of an aberration because of things like the Tea Party influence,”
    further noting that he “didn’t know if [that election] was reliable.” Seliger San Antonio Dep.
    15:5-7; see also Trial Tr. 11:15-21, Jan. 24, 2012 AM.20
    Texas counters that none of the experts found that benchmark CD 23 clearly performs as
    an ability district and points to the weak showing of minority voters in exogenous elections: only
    three out of ten victories in the OAG 10 and two out of five victories in Dr. Handley’s election
    20
    Texas argues that one of the two endogenous victories, the 2006 election, should be discounted because it
    did not occur on general election day. See Tex. Post-Trial Br. 16. VRA litigation left no time for a primary that year,
    and instead all eight candidates competed in a special election held the same day as Texas’s general election. See
    Trial Tr. 66:21-68:9, Jan. 26, 2012 AM. Texas is correct that the Republican candidate won the plurality of votes in
    the special election, but we find this result unremarkable because six of the eight special election candidates were
    Democrats. When the runoff election was held five weeks later, Hispanic-preferred candidate Ciro Rodriguez won a
    decisive victory. Pl.’s Ex. 17, at 368. We see no reason to discount Rep. Rodriguez’s victory.
    30
    set. Alford Rep. 23 tbl.4b; Handley Cong. Rep. 5. But these numbers do not tell the full story.
    Every expert save Dr. Alford concluded that benchmark CD 23 is an ability district despite
    marginal exogenous performance. Dr. Handley concluded that endogenous results are more
    probative than exogenous for this district, see Handley Cong. Rep. 5-6, and, as we have already
    discussed, we agree that this assessment is generally accurate. Dr. Ansolabehere’s analysis
    shows that minority-preferred candidates won the district “more often than not.” Defs.’ Ex. 724,
    Expert Witness Report of Dr. Stephen Ansolabehere 36-37 [hereinafter Ansolabehere Rep.]. And
    the TLRTF argues that a larger election sample set is necessary to make an informed judgment.
    When four additional racially contested contests are added to the OAG 10, the district’s
    exogenous success rises to seven out of fourteen. See Trial Tr. 111:14-113:4, Jan. 18, 2012 AM;
    Defs.’ Exs. 390, 647. These election results, combined with the endogenous elections discussed
    above, the fact that CD 23 was drawn to be an opportunity district, and the contemporary views
    of redistricting officials, are enough for us to find that benchmark CD 23 lived up to its potential
    as drawn in 2006 and became an ability district.
    But enacted CD 23 is not. Even though the district’s demographics remain relatively
    unchanged — HCVAP actually increased 0.1% from the benchmark to the enacted plan, Pl.’s
    Ex. 12, at 9 — this fact is inconclusive. Instead, we must look to other factors, including
    exogenous elections, testimony, and other evidence about changes made in the district.
    Enacted CD 23’s exogenous election results are significantly worse than those in
    benchmark CD 23. In the OAG 10, the number of victories decreases from three of ten to one. In
    Dr. Handley’s sample the number decreases from two of five to none. Alford Rep. 23 tbl.4b;
    Handley Cong. Rep. 7; see also Ansolabehere Rep. 37 (concluding that the enacted plan “lowers
    the electoral performance of minority-preferred candidates in the District to the point that it is
    31
    likely no longer a minority opportunity seat”). Minority voter turnout in enacted CD 23 declines.
    While Hispanic voters accounted for an average of 39% of total votes cast in benchmark CD 23
    over the past decade, they made up only 36.5% in enacted CD 23.21 Defs.’ Ex. 365, at 5-12; see
    also, e.g., Defs.’ Ex. 575, Trial Tr. 450:19-454:11, Sept. 7, 2011, Perez, No. 11-cv-360
    (testimony of Dr. Henry Flores, noting that Hispanic voter turnout was higher in areas moved out
    of the district than in areas that were moved in; turnout in some excluded areas was consistently
    over 30%, while turnout in areas that replaced them was only 25-30%). The changes were
    enough to “nudge” a district that was an ability district, but barely so, to a nonperforming
    district. See Ansolabehere Rep. 37 (noting that “in a competitive district such as this one,”
    seemingly small changes “made a huge difference”). Even Texas’s expert testified that CD 23 “is
    probably less likely to perform than it was, and so I certainly wouldn’t count and don’t [and]
    haven’t counted the 23rd as an effective minority district in the newly adopted plan.” Defs.’ Ex.
    581, Trial Tr. 1839:2-7, Sept. 14, 2011, Perez, No. 11-cv-360. Thus, CD 23 is an ability district
    in the benchmark, but would be no longer in the enacted plan.
    Texas claims that the enacted district has remained functionally identical to the
    benchmark, but these claims are undermined by the mapdrawers’ own admissions that they tried
    to make the district more Republican — and consequently, less dependable for minority-
    preferred candidates — without changing the district’s Hispanic population levels. The
    mapdrawers consciously replaced many of the district’s active Hispanic voters with low-turnout
    Hispanic voters in an effort to strengthen the voting power of CD 23’s Anglo citizens. In other
    words, they sought to reduce Hispanic voters’ ability to elect without making it look like
    anything in CD 23 had changed. See, e.g., Defs.’ Ex. 304 (email from Eric Opiela, counsel to
    21
    Judges Collyer and Howell do not depend on voter turnout data to conclude that CD 23 is a lost ability
    district.
    32
    Texas House Speaker Joe Strauss, to mapdrawer Gerardo Interiano in November 2010 urging
    Interiano to find a metric to “help pull the district’s Total Hispanic Pop[ulation] and Hispanic
    CVAPs up to majority status, but leave the Spanish Surname [Registered Voter] and [turnout
    numbers] the lowest,” which would be “especially valuable in shoring up [CD 23 incumbent]
    Canseco”); 
    id.
     (email from Interiano responding that he would “gladly help with this”); Defs.’
    Ex. 739, at 40 (email indicating that Opiela provided sample maps to Interiano as late as June 11,
    2011, that would “improve CD 23’s [H]ispanic performance while maintaining it as a
    Republican district”). We also received an abundance of evidence that Texas, in fact, followed
    this course by using various techniques to maintain the semblance of Hispanic voting power in
    the district while decreasing its effectiveness. See, e.g., Defs.’ Ex. 436 (evidence showing that
    over 600,000 persons were moved into and out of the district to redress overpopulation of only
    149,000); Defs.’ Ex. 903, at 1 (email noting that a draft map of CD 23 was “over 59% HCVAP,
    but still at 1/10 [exogenous election performance],” and commenting that there must be an
    HCVAP level high enough that low election results would not raise trouble under section 5);
    Defs.’ Ex. 978 (email commenting that a draft map of CD 23 “looks nice politically,” but still
    raises “concern[s] about the Voting Rights Act”); Trial Tr. 106:18-108:3, Jan. 18, 2012 AM
    (testimony of Ryan Downton that he drew the district’s lines precinct-by-precinct based on
    election results to keep Hispanic population numbers high while maximizing Republican
    performance); Id. at 12:2-16, Jan. 24, 2012 AM (testimony of Kel Seliger that CD 23 was drawn
    by considering “voting patterns and ethnicity” to see what could be done “to change the
    district”). Texas’s protestations that the district has remained functionally identical are weakened
    first by the mapdrawers’ admissions that they tried to reduce the effectiveness of the Hispanic
    33
    vote and then, more powerfully, by evidence that they did. We conclude that CD 23 is a lost
    ability district.
    3. Retrogression with New Congressional Seats22
    Texas’s population grew by approximately 4.3 million in the past decade, an increase of
    20.6%. Approximately 89% of this growth was from non-Anglo minorities: Hispanics comprise
    65% of the increase, Blacks 13.4%, and Asian-Americans 10.1%. See U.S. Req. for Judicial
    Notice ¶¶ 8, 20, 22, 24 (citing 2000 and 2010 Census data).23 As a result of this increase, the
    Texas delegation in the U.S. House of Representatives grew from 32 to 36 members, the largest
    growth ever in a jurisdiction fully covered by section 5. See Texas, 831 F. Supp. 2d at 257. The
    United States and various Intervenors argue that Texas was required to draw at least one of these
    new districts as an ability district. See, e.g., U.S. Post-Trial Br. 14-15. We agree.
    As already discussed, section 5’s prohibition on retrogression means that “the entire
    [enacted] statewide plan as a whole,” Ashcroft, 
    539 U.S. at 470
    , cannot “increase the degree of
    discrimination against [minority voters],”24 City of Lockhart v. United States, 
    460 U.S. 125
    , 134
    (1983). Abrams v. Johnson, 
    521 U.S. 74
     (1997), tells us how to measure the degree of
    discrimination when the number of districts remains the same or increases by one: there is no
    retrogression as long as the number of ability districts remains the same. 
    Id. at 97-98
    . At
    summary judgment we concluded that our case was similar to Abrams because “Texas’
    22
    Having found retrogression in the Congressional Plan because CD 25 was an ability district that was
    eliminated and not replaced, Judge Collyer does not reach the further question of retrogression based on proportional
    representation arising from multiple new congressional seats and a sizeable growth in minority population.
    23
    Likewise, minorities comprise 80.4% of the increase in Texas’s voting age population between 2000 and
    2010. U.S. Req. for Judicial Notice ¶ 19 (citing 2000 and 2010 Census data). We agree with the United States that
    U.S. Census data is an appropriate subject of judicial notice. See 
    id.
     at 2 (citing Hollinger v. Home State Mut. Ins.
    Co., 
    654 F.3d 564
    , 572-73 (5th Cir. 2011); City of Port Arthur v. United States, 
    517 F. Supp. 987
    , 993 n.5 (D.D.C.
    1981)).
    24
    The Supreme Court has also described our task as determining that the enacted plan “is no more dilutive than
    what it replaces.” Bossier II, 
    528 U.S. at 335
    .
    34
    percentage gain in congressional seats (12.5%) is similar to Georgia’s percentage gain in Abrams
    (10%).” Texas, 831 F. Supp. 2d at 269. Yet we also noted that “Abrams does not control.
    Although Abrams is clear that the VRA does not require there to be a new minority ability
    district for every new congressional seat, it does not hold that a state’s failure to draw new
    minority districts can never be retrogressive.” Id. Upon further examination and after weighing
    the arguments presented at trial, we have concluded that Texas’s failure to draw a new minority
    district does in fact make the enacted plan retrogressive under the specific facts of this case.
    Abrams spoke only to the case of a state that gained a single seat, 
    521 U.S. at 97-98
    , not to the
    case of a state that gains multiple seats.25
    Neither section 5’s text nor existing case law tells us how to measure the “degree of
    discrimination” in these circumstances. But guidance is available in the Supreme Court’s section
    2 cases. Even though section 5 is not ameliorative and has different purposes than section 2,
    some tools used in section 2 analysis reveal insights into the underlying principles of the VRA,
    see, e.g., Texas, 831 F. Supp. 2d at 261-62 & 262 n.21, which are especially helpful as we find
    ourselves in a setting no section 5 cases have yet considered.
    In the section 2 context, the Court has looked to the relationship between a minority
    group’s share of the CVAP statewide and the number of opportunity districts to help determine
    whether new opportunity districts must be created. See LULAC, 
    548 U.S. at 438
     (“Looking
    statewide, there are 32 congressional districts. The five reasonably compact Latino opportunity
    districts amount to roughly 16% of the total [number of districts], while Latinos make up 22% of
    25
    We agree with the United States that the holding of Abrams cannot be read to govern all cases in which a
    state gains seats in a district map. At the extreme, consider a state with a 100-member legislature and 30 ability
    districts in the benchmark map. If the state redrew its legislature to double the number of districts to 200, but created
    no new ability districts, it would be difficult to conclude that the new plan was not dilutive and had not increased the
    degree of discrimination against minority voters merely because it contained the same number of ability districts.
    35
    Texas’ citizen voting-age population. . . . Latinos are, therefore, two districts shy of proportional
    representation.”); De Grandy, 
    512 U.S. at
    1014 n.11 (examining “the number of majority-
    minority voting districts [compared] to minority members’ share of the relevant population”).
    We agree with the United States that this “representation gap” between the number of districts
    proportional representation would yield and the number of districts the legislature has actually
    created is a strong indicator of the “degree of discrimination.” U.S. Post-Trial Br. 15. When the
    representation gap grows, the degree of discrimination increases.
    This analysis squares with the outcomes of previous section 5 cases. Where the number
    of districts remains the same, the representation gap does not increase. Likewise, the
    representation gap in Abrams was unchanged between plans. There, Blacks constituted 27% of
    Georgia’s voting age population and had the ability to elect in only one of ten districts in the
    benchmark plan. See 
    521 U.S. at 103
     (Breyer, J., dissenting). That put the representation gap at
    two districts (27% of 10 is 2.7, which, when rounded up, is two more than one).26 In the enacted
    plan the representation gap remained the same (27% of 11 is 3.0, which is also two more than
    one). There was no increase in the degree of discrimination, and the plan did not retrogress.
    By contrast, the representation gap in Texas has increased. The Black and Hispanic
    communities currently make up 39.3% of Texas’s CVAP. Joint Stipulations of Fact ¶ 38. Thus, if
    districts were allocated proportionally, there would be 13 minority districts out of the 32 in the
    benchmark (39.3% of 32 is 12.6). Yet minorities have only 10 seats in the benchmark, so the
    representation gap is three districts. In the enacted plan, proportional representation would yield
    26
    We note that we are rounding 2.7 up to 3. We do so following the Court’s example in LULAC, in which it
    noted that “‘rough proportionality’ must allow for some deviations.” 
    548 U.S. at 438
     (quoting De Grandy, 
    512 U.S. at 1023
    ).
    36
    14 ability districts (39.3% of 36 is 14.1), but there are still only 10 ability districts.27 Thus, the
    representation gap in the enacted plan is four districts. Because this gap increases by one district,
    we cannot preclear the enacted plan.28
    We emphasize what our analysis does not do. It does not entitle minorities to proportional
    representation. It does not require a state to create new ability districts in proportion to increases
    in a minority group’s population.29 We require only that a state not “undo[] or defeat[] the rights
    recently won” by minorities, Beer, 
    425 U.S. at 140
     (quoting H.R. REP. NO. 91-397, at 8 (1969))
    (internal quotation marks omitted), by increasing the “degree of discrimination,” Lockhart, 
    460 U.S. at 134
    , which requires assessing the “number of majority-minority voting districts to
    minority members’ share of the relevant population,” De Grandy, 
    512 U.S. at
    1014 n.11.
    27
    Our calculations use the combined Black and Hispanic share of the CVAP (39.3%), the metric advanced by
    the United States and various Intervenors. See also De Grandy, 
    512 U.S. at
    1014 n.11 (“‘Proportionality’ as the term
    is used here links the number of majority-minority voting districts to minority members’ share of the relevant
    population.”). Nevertheless, we note that our method also yields one additional congressional seat if the Black and
    Hispanic representation gaps are calculated separately. Hispanics comprise 26.4% of Texas’s CVAP, Joint
    Stipulations of Fact ¶ 38, and the “Hispanic” representation gap increases by one in the enacted plan (Hispanics have
    seven ability districts in both plans, but 26.4% of 32 is 8.4 and 26.4% of 36 is 9.5). By contrast, Blacks comprise
    12.9% of Texas’s CVAP, 
    id.,
     and the “Black” representation gap does not change between plans. Blacks have three
    ability districts in both plans; 12.9% of 32 is 4.1, and 12.9% of 36 is 4.6. Following the “rough proportionality”
    principle, this increase of 0.5 in the representation gap does not require the State to draw a new district, just as we
    require the State to draw only one additional ability district above, even though there is a 1.5 increase in the
    representation gap.
    Similarly, this representation gap would exist even if CD 25 were counted as an ability district in the
    benchmark. In that case, the benchmark representation gap would be two districts (the difference between 13 and 11
    districts) and the enacted representation gap would be three districts (the difference between 14 and 11 districts).
    28
    We note that this requirement would likely be subject to the caveat that a state is only required to draw a new
    district if possible, i.e., if it can draw a new ability district without violating other principles such as one-person,
    one-vote or the demands of section 2. Yet the facts that minority population growth was largely concentrated in
    three areas in Texas and that the parties submitted several alternate plans drawing a new Hispanic ability district
    suggest that this will not be an issue here. In any event, the infeasibility of drawing a new district was not argued or
    briefed in any depth during this litigation.
    29
    Under our logic, if Texas had experienced the same population growth but had not gained additional
    congressional seats (because, for example, other states experienced equivalent or greater growth), it would have
    been required to draw only 10 ability districts. It is the growth in the number of districts that triggers our analysis,
    not growth in the population.
    37
    Because the Texas legislature purposes to increase this representation gap, we cannot preclear its
    Congressional Plan.
    B. Discriminatory Intent in the Congressional Plan
    Although we need not reach the issue of discriminatory intent because we conclude that
    the Congressional Plan will have a retrogressive effect, we do so here because, as we have just
    discussed, we do not all agree on the appropriate rationale for finding retrogression. But because
    we agree that the plan was enacted with discriminatory purpose, we reach this issue as an
    alternative, unanimous basis to deny preclearance for the Congressional Plan. If true, the
    allegations of the United States and the Intervenors that Texas drew the Congressional Plan with
    discriminatory purpose provide grounds to deny preclearance. Texas argues that intent to
    discriminate against minority voters played no role in the plan and that its decisions were
    motivated solely by partisan politics. See, e.g., Tex. Post-Trial Br. 26 (“Texas adopted the
    Congressional Plan with the lawful aim of protecting incumbents.”).
    There is no direct evidence that the enacted plan was motivated by discriminatory
    purpose; no emails, letters, or testimony about conversations between those members involved in
    congressional redistricting disclose such an intent. Cf. Diaz v. Kraft Foods Global, Inc., 
    653 F.3d 582
    , 587 (7th Cir. 2011) (“Direct evidence is something close to an explicit admission . . . that a
    particular decision was motivated by discrimination; this type of evidence is rare, but it ‘uniquely
    reveals’ the . . . intent to discriminate.” (quoting Rudin v. Lincoln Land Cmty. Coll., 
    420 F.3d 712
    , 720 (7th Cir. 2005))). Thus, we must assess the circumstances surrounding the drawing of
    the new maps. Our analysis follows the Supreme Court’s decision in Arlington Heights, which,
    as discussed in more detail above, identifies five “subjects of proper inquiry in determining
    whether racially discriminatory intent existed”: (1) discriminatory impact, (2) historical
    38
    background, (3) sequence of events leading up to the decision, (4) procedural or substantive
    deviations from the normal decisionmaking process, and (5) contemporaneous viewpoints
    expressed by the decisionmakers. Arlington Heights, 
    429 U.S. at 266-68
    .
    As we have already noted, CDs 9, 18, and 30 are the only Black ability districts in the
    benchmark and enacted plans. CD 9 is located south of Houston and incorporates parts of Harris
    and Fort Bend Counties, CD 18 is located within Houston, and CD 30 is within Dallas. The
    Texas legislature proposed substantial changes to these districts even though the 2010 Census
    data shows the population in each was already close to the ideal size.30 We have already
    determined that these changes are not retrogressive, but they raise serious concerns about what
    motivated the Congressional Plan.
    Congressman Al Green, who represents CD 9, testified that “substantial surgery” was
    done to his district that could not have happened by accident. The Medical Center, Astrodome,
    rail line, and Houston Baptist University — the “economic engines” of the district — were all
    removed in the enacted plan. Trial Tr. 124:6-20, Jan. 20, 2012 AM; see also Defs.’ Ex. 721, Pre-
    Filed Test. of Congressman Alexander Green 3-4. The enacted plan also removed from CD 9 the
    area where Representative Green had established his district office. Trial Tr. 124:16, Jan. 20,
    2012 AM. Likewise, Congresswoman Sheila Jackson Lee, who represents CD 18, testified that
    the plan removed from her district key economic generators as well as her district office. Id. at
    13:13-14:5, Jan. 23, 2012 PM. Congresswoman Eddie Bernice Johnson of CD 30 also testified
    that the plan removed the American Center (home of the Dallas Mavericks), the arts district, her
    district office, and her home from CD 30. Id. at 79:20-81:16, Jan. 18, 2012 PM. The mapdrawers
    30
    According to the 2010 Census, Texas’s population was 25,145,561. If this population were divided equally
    between the State’s thirty-six congressional districts, each district would have 698,488 individuals. Pl.’s Ex. 12, at 2.
    Benchmark CD 9 has a surplus of 35,508 people, or 5.05% of the district’s population. CD 18’s surplus is 22,503
    (3.22%), and CD 30’s is 7,891 (1.14%). Defs.’ Ex. 347, at 28-29.
    39
    also removed the district office, the Alamo, and the Convention Center (named after the
    incumbent’s father), from CD 20, a Hispanic ability district. Mem. Opp. Summ. J. Ex. 16, Decl.
    of Charles A. Gonzalez ¶¶ 3-9, 11, ECF No. 77.
    No such surgery was performed on the districts of Anglo incumbents. In fact, every
    Anglo member of Congress retained his or her district office. Trial Tr. 14:12-15, Jan. 23, 2012
    PM. Anglo district boundaries were redrawn to include particular country clubs and, in one case,
    the school belonging to the incumbent’s grandchildren. See Mem. Opp. Summ. J. Exs. 11, 18-19,
    ECF No. 77. And Texas never challenged evidence that only minority districts lost their
    economic centers by showing, for example, that the same types of changes had been made in
    Anglo districts.
    The United States and the Intervenors convincingly argue — and Texas does not
    dispute — that removing district offices from minority ability districts but not from Anglo
    districts has a disparate impact on the minority districts. See U.S. Post-Trial Br. 26. District
    offices help “provide[] a meaningful connection between a member and the people represented.”
    Defs.’ Ex. 721, Pre-Filed Test. of Congressman Alexander Green 4. Their locations are often
    well known to constituents, often placed to be easily accessible by freeway and public
    transportation, and serve as a way for members of Congress to communicate with and provide
    services to their constituents. See id. We are likewise troubled by the unchallenged evidence that
    the legislature removed the economic guts from the Black ability districts. Texas does not
    dispute that part of a member of Congress’s job is to “bring economic generators that will benefit
    that community,” id. Removing those economic generators harms the district. Id. at 3-4; U.S.
    Post-Trial Br. 26.
    40
    The only explanation Texas offers for this pattern is “coincidence.”31 Trial Tr. 95:5-19,
    Jan. 25, 2012 PM. But if this was coincidence, it was a striking one indeed. It is difficult to
    believe that pure chance would lead to such results. The State also argues that it “attempted to
    accommodate unsolicited requests from a bipartisan group of lawmakers,” and that “[w]ithout
    hearing from the members, the mapdrawers did not know where district offices were located.”
    Tex. Post-Trial Br. 29. But we find this hard to believe as well. We are confident that the
    mapdrawers can not only draw maps but read them, and the locations of these district offices
    were not secret. The improbability of these events alone could well qualify as a “clear pattern,
    unexplainable on grounds other than race,” Arlington Heights, 
    429 U.S. at 266
    , and lead us to
    infer a discriminatory purpose behind the Congressional Plan.
    When taken with the remaining Arlington Heights factors, Texas’s explanation becomes
    weaker still. First, the historical background gives us grounds for concern. In the last four
    decades, Texas has found itself in court every redistricting cycle, and each time it has lost. See,
    e.g., LULAC, 
    548 U.S. 399
    ; Vera, 
    517 U.S. 952
    ; Upham v. Seamon, 
    456 U.S. 37
     (1982); White v.
    Weiser, 
    412 U.S. 783
     (1973); White v. Regester, 
    412 U.S. 755
     (1973); Terrazas v. Slagle, 
    789 F. Supp. 828
     (W.D. Tex. 1992), aff’d sub nom., Richards v. Terrazas, 
    505 U.S. 1214
     (mem.). While
    a losing streak alone does not control our decision, Texas’s history of failures to comply with the
    VRA is one of the circumstantial factors that Arlington Heights instructs us to consider.
    Next, the sequence of events leading to the passage of the Congressional Plan also
    supports an inference of discriminatory purpose. Black and Hispanic members of Congress
    testified at trial that they were excluded completely from the process of drafting new maps, while
    the preferences of Anglo members were frequently solicited and honored. See, e.g., Mem. Opp.
    31
    Unlike in its arguments about retrogression, Texas never argued that the removal of district offices and
    economic generators was the product of political animus.
    41
    Summ. J. Exs. 18-19; Defs.’ Ex. 370, at 1, ECF No. 77. The Texas House and Senate
    redistricting committees released a joint congressional redistricting proposal for the public to
    view only after the start of a special legislative session, and each provided only seventy-two
    hours’ notice before the sole public hearing on the proposed plan in each committee. See, e.g.,
    Defs.’ Ex. 320, Decl. of Theodore S. Arrington 57-59; Defs.’ Ex. 366. Minority members of the
    Texas legislature also raised concerns regarding their exclusion from the drafting process and
    their inability to influence the plan via amendments. See, e.g., Defs.’ Ex. 370, at 1.
    Lastly, procedural and substantive departures from the normal decisionmaking process
    raise flags. Citing failure to release a redistricting proposal during the regular session, the limited
    time for review, and the failure to provide counsel with the necessary election data to evaluate
    VRA compliance, the Senate redistricting committee’s outside counsel described the proceedings
    as “quite different from what we’ve seen in the past.” 
    Id. at 2
    .
    Texas argues that, “[a]t worst, the evidence shows that [it] was guilty of blithe
    indifference to the wants to certain [minority] Congressmen.” Tex. Post-Trial Br. 29. But we do
    not find this explanation credible. Although we have already concluded that the Congressional
    Plan cannot be precleared under section 5’s effect prong, we are also persuaded by the totality of
    the evidence that the plan was enacted with discriminatory intent. Texas did not adequately
    engage with the evidence raised by the other parties on this point, and under Arlington Heights
    we find sufficient evidence to conclude that the Congressional Plan was motivated, at least in
    part, by discriminatory intent.32 Therefore, we deny Texas declaratory judgment with respect to
    the Congressional Plan on this ground as well.
    32
    The parties have provided more evidence of discriminatory intent than we have space, or need, to address
    here. Our silence on other arguments the parties raised, such as potential discriminatory intent in the selective
    drawing of CD 23 and failure to include a Hispanic ability district in the Dallas-Fort Worth metroplex, reflects only
    this, and not our views on the merits of these additional claims.
    42
    IV. State Senate Plan
    Next we consider Texas’s request to preclear its State Senate Plan. The United States has
    not objected to this plan, but the Davis Intervenors, the Texas State Conference of NAACP
    Branches, the League of United Latin America Citizens, and the Texas Legislative Black Caucus
    argue that the Senate Plan retrogresses and was enacted with discriminatory intent. Their
    arguments concern a single district, Senate District (SD) 10, which they contend is a coalition
    district in the benchmark plan, and which all parties agree is not an ability district in the enacted
    plan. These Intervenors also argue that discriminatory purpose motivated the legislature’s
    decision to break up SD 10. We conclude that benchmark SD 10 is not a coalition district, and
    thus that the Senate Plan is not retrogressive. Nevertheless, we deny preclearance because Texas
    failed to carry its burden to show that it acted without discriminatory purpose in the face of
    largely unrebutted defense evidence and clear on-the-ground evidence of “cracking” minority
    communities of interest in SD 10. Thus, we conclude that the Texas legislature redrew the
    boundaries for SD 10 with discriminatory intent.
    A. Retrogression in the Senate Plan
    Benchmark SD 10 is located entirely within Tarrant County, which includes Fort
    Worth. When the Texas legislature last drew the district in 2001, the population was 56.6%
    Anglo, 16.7% Black, and 22.9% Hispanic. Defs.’ Ex. 126, 2001 State of Texas Submission for
    State Senate Preclearance app. I (Aug. 15, 2001). Urging the Department of Justice to preclear
    the 2001 State Senate Plan, Texas justified SD 10’s configuration by arguing that “[t]he voting
    strength of these minority communities in the future will depend on the cohesion within and
    between Black and Hispanic voters and the ability of such voters to form coalitions with other
    43
    racial or ethnic groups in support of their preferred candidates.” 
    Id. at 18
    . In other words, Texas
    argued that SD 10 had the potential to become a coalition district.
    The Department of Justice precleared the 2001 map, and, over the past decade, the
    minority population in SD 10 has continued to grow. According to the 2010 Census, 47.6% of
    the population in SD 10 was Anglo, 19.2% Black, and 28.9% Hispanic. Defs.’ Ex. 151, at 5.
    Minorities made up a smaller portion of the 2010 CVAP, however: 62.7% were Anglo, 18.3%
    Black, and 15.1% Hispanic. Pl.’s Ex. 15, at 8. Republicans have won almost every election in SD
    10 in the past ten years, including the district’s endogenous State Senate elections from 2000-
    2008. No Democratic candidate running in a statewide or other exogenous election has ever won
    a majority of the vote in SD 10. See Alford Rep. 30.
    The only Democrat to win an election in SD 10 is the district’s current senator, Wendy
    Davis, who was elected to a four-year term in 2008. Davis’s path to the State Senate began when
    Democratic candidate Terri Moore lost the 2006 election for Tarrant County District Attorney,
    yet received nearly half of the vote in SD 10. See Trial Tr. 30:10-25, 31:1-17, Jan. 18, 2012 PM.
    In light of these results, Democratic elected officials and community leaders in Tarrant County
    were of the view that if the Black and Hispanic communities “came together as a coalition to
    vote . . . they could win Senate District 10.” 
    Id. at 30:15-16
    . These and other leaders within the
    district’s minority communities recruited Fort Worth City Council member Wendy Davis to run
    for State Senate. 
    Id. at 32:3-25, 33:1-17
    ; see also 
    id. at 16:1-5
    , Jan. 20, 2012 AM (Senator Davis,
    testifying, “I was approached by leaders in our minority community in large part because of the
    work I’d done as a City Council person and asked if I would consider running for the Texas State
    Senate.”). Senator Davis ran unopposed in the 2008 Democratic primary, see Pl.’s Ex. 135, at 3,
    44
    then won the general election with 49.9% of the vote, beating the incumbent by 2.4% —
    approximately 7,100 out of 288,000 votes cast.33 Pl.’s Ex. 31, at 14.
    According to Texas’s expert, Davis received 99.6% of the Black vote, 85.3% of the
    Hispanic vote, and 25.8% of the Anglo vote. Trial Tr. 32:24-25, 33:1-16, Jan. 25, 2012 AM.
    Although this is strong evidence that the minority communities in SD 10 voted cohesively in the
    2008 election, the argument that SD 10 is a coalition district runs into trouble when looking at
    evidence that the district’s minority communities have been effective in electing their preferred
    candidates.
    At summary judgment, we noted that “evidence that a coalition had historical success in
    electing its candidates of choice would demonstrate that the minority voters in that district had,
    and would continue to have, an ability to elect their preferred candidates.” Texas, 831 F. Supp.
    2d at 268. The case that SD 10 is an ability district turns on a single, razor-thin election victory,
    which is not “historical success.” Indeed, SD 10’s decade-long history of electing Republicans
    shows just the opposite. There is no doubt that the minority community came together to elect a
    preferred candidate in 2008, but a single victory is not the more exacting evidence needed for a
    coalition district. If it were, any single victory built upon the support of minority voters would
    create a claim for ability status.
    B. Discriminatory Intent in the Senate Plan
    There is no direct evidence that the Texas legislature acted with a racially discriminatory
    purpose in its reconfiguration of SD 10, and so we must look to circumstantial evidence. Once
    again, we look to the Arlington Heights factors to determine whether Texas has met its burden of
    disproving discriminatory intent.
    33
    Richard Cross, a libertarian candidate, received 2.6% of the vote (7,591 votes). Pl.’s Ex. 31, at 14.
    45
    Considering first the impact of the redistricting — “whether it ‘bears more heavily on one
    race than another,’” Arlington Heights, 
    429 U.S. at 266
     (quoting Washington v. Davis, 
    426 U.S. 229
    , 242 (1976)), there is little question that dismantling SD 10 had a disparate impact on racial
    minority groups in the district. Even Dr. Alford agreed that the enacted plan “diminishes the
    voting strengths of Blacks and Latinos in [SD 10],” Trial Tr. 39:14, Jan. 25, 2012 AM. In a letter
    he sent to the Department of Justice objecting to the enacted Senate Plan, Texas State Senator
    Rodney Ellis explained in detail how the new boundaries eliminate the ability of minority
    citizens to elect their preferred candidates by submerging their votes within neighboring and
    predominantly Anglo districts:
    The demolition of District 10 was achieved by cracking the African American and
    Hispanic voters into three other districts that share few, if any, common interests with the
    existing District’s minority coalition. The African American community in Fort Worth is
    “exported” into rural District 22 — an Anglo-controlled District that stretches over 120
    miles south to Falls [County]. The Hispanic Ft. Worth North Side community is placed in
    Anglo suburban District 12, based in Denton County, while the growing South side
    Hispanic population remains in the reconfigured majority Anglo District 10.
    Defs.’ Ex. 375, at 3. We find that Senator Ellis’s testimony is well supported by the record. See
    also Defs.’ Ex. 134, Expert Witness Report of Dr. Allan J. Lichtman ¶ 12 [hereinafter Lichtman
    Rep.] (“The state legislature, in dismantling benchmark SD 10 cracked the politically cohesive
    and geographically concentrated Latino and African American communities and placed members
    of those communities in districts in which they have no opportunity to elect candidates of their
    choice or participate effectively in the political process.”).
    Texas does not deny this disparate impact, but responds that its decision to “crack” SD 10
    is best explained by partisan, not racial, goals. Tex. Post-Trial Br. 25. While this is a potentially
    plausible rationale, Arlington Heights instructs that “[d]etermining whether invidious
    discriminatory purpose was a motivating factor demands a sensitive inquiry into such
    46
    circumstantial and direct evidence of intent as may be available,” and so we must “look to the
    other evidence.” 
    429 U.S. at 266
    .
    These other factors do not support Texas’s case. The second factor is Texas’s history of
    discrimination, and as we discussed in our analysis of the Congressional Plan above, history is
    not on Texas’s side. The third considers the “specific sequence of events leading up to the
    challenged decision.” 
    Id. at 267
    . The Senate’s principal mapdrawer and staff director of the
    Senate Redistricting Committee, Doug Davis (no relation to Senator Davis), began discussing
    draft maps of new Senate districts prior to the February 2011 release of official Census data by
    using projected population increases. Defs.’ Ex. 127, at 38-39. Once the 2011 general legislative
    session started in January, these maps were kept in an anteroom off the Senate floor, where many
    Republican members were taken individually by Chairman Seliger and Doug Davis to review the
    draft plans and provide input. See, e.g., Trial Tr. 39:15-25, Jan. 20, 2012 AM; Defs.’ Ex. 809,
    Dep. of Senator Judith Zaffirini 29:22-25, 30:1-19, Jan. 6, 2012. Senator Davis was consistently
    rebuffed when she asked to see the plans for SD 10, even as another senator told her that the
    proposed plan was “shredding” her district. Trial Tr. 38:2-8, 40:11-14, Jan. 20, 2012 AM.
    Senator Judith Zaffirini’s uncontroverted testimony shows that this scenario was not unique to
    Senator Davis, but reflected a larger pattern: every senator who represented an ability district
    was excluded from this informal map-drawing process and was not allowed into the anteroom to
    preview the maps. See Defs.’ Ex. 809, Dep. of Senator Judith Zaffirini 30:1-3. Indeed, none of
    the senators representing ability districts were shown their districts until forty-eight hours before
    the map was introduced in the Senate. See Defs.’ Ex. 129.
    Texas offered conflicting testimony in response. Doug Davis testified that “we were not
    printing maps and giving them to members,” Trial Tr. 172:10-11, Jan. 17, 2012 PM, suggesting
    47
    that at least part of this informal process that gave Republican senators opportunities to provide
    input into the plans did not occur. But Chairman Seliger, Davis’s boss, testified that he provided
    paper maps to at least three senators during this period, all of them Anglo. Trial Tr. 68:1-3, Jan.
    24, 2012 AM. In any case, it is clear that senators who represented minority districts were left
    out of the process.34
    Our skepticism about the legislative process that created enacted SD 10 is further fueled
    by an email sent between staff members on the eve of the Senate Redistricting Committee’s
    markup of the proposed map. The ostensible purpose of the markup was to consider amendments
    to the proposed plan, but the email suggests a very different dynamic at work. David Hanna, a
    lawyer for the Texas Legislative Council, a nonpartisan agency that provides bill drafting and
    legislative research to the Texas legislature, sent an email to Doug Davis and Senate
    Parliamentarian Katrina Davis (Doug Davis’s wife). Hanna’s email responded to an earlier
    message Texas did not produce, but which concerned “precook[ing]” the committee report, i.e.,
    writing the report before the hearing had been held. Trial Tr. 71:23-25, 72:1-7, Jan. 24, 2012
    AM. With a subject line titled, “pre-doing committee report,” Hanna’s email read:
    No bueno. RedAppl [the redistricting software Texas used] time stamps everything when
    it assigns a plan. Doing [the Committee Report on] Thursday [May 12] would create [a]
    paper trail that some amendments were not going to be considered at all. Don’t think this
    is a good idea for preclearance. Best approach is to do it afterwards and we’ll go as fast
    as possible.
    Defs.’ Ex. 359. Although the chairman of the redistricting committee, Kel Seliger, denied
    knowing of any advance decision to refuse to consider amendments, he acknowledged what is
    apparent from the email: the boundaries of the new Senate districts would be a fait accompli by
    34
    We also note that Texas did not refute testimony indicating that the field hearings held prior to the start of
    the 2011 legislative session were “perfunctory,” Trial Tr. 94:20-21, Jan. 20, 2012 AM, and “a sham,” with low
    attendance, low participation, and little invited testimony or prepared materials. Defs.’ Ex. 809, Dep. of Senator
    Judith Zaffirini 7:11-21.
    48
    the time of the markup and the committee did not intend to consider any amendments to the plan.
    Trial Tr. 71:3-25, 72:1-16, Jan. 24, 2012 AM. We agree with Chairman Seliger that, at a
    minimum, this email shows that a plan was in place, at least at the staff level, such that no new
    proposals or amendments to the district map would be entertained at the markup.
    Arlington Heights instructs that “departures from the normal procedural sequence also
    might afford evidence that improper purposes are playing a role.” 
    429 U.S. at 267
    . This factor
    focuses on comparing past redistricting cycles to the present one for anomalous behavior. The
    State held no field hearings after Census data was released and proposed plans were drawn,
    unlike the hearings that were held after such data was available in the past. Defs.’ Ex. 134, at 13.
    Additionally, Senator Zaffirini testified that she, a senator of a minority district, “had never had
    less input into the drawing of any [redistricting] map” in over thirty years of redistricting
    experience,” Defs.’ Ex. 370, at 1, and that the 2010 redistricting process was the “least
    collaborative and most exclusive” she had ever experienced. Lichtman Rep. app. 7, Decl. of
    Senator Judith Zaffirini ¶ 3. We find this unchallenged testimony sufficient to conclude that the
    2010 redistricting process was markedly different from previous years.
    Finally, Arlington Heights states that “the legislative or administrative history may be
    highly relevant especially where there are contemporary statements by members of the
    decisionmaking body.” 
    429 U.S. at 268
    . Aside from the “No Bueno” email described above, we
    have no evidence of contemporary statements by the majority members or their staff “concerning
    the purpose of the official action,” 
    id.
     But that email indicates, at a minimum, that redistricting
    committee staff feared their actions might create the appearance of impropriety under section 5.
    We do, however, have a statement published in the Senate journal from the eleven senators
    representing majority-minority districts and Senator Davis. They alleged that the fact they were
    49
    shut out from the map-drawing process until just forty-eight hours before the map was
    introduced in the Senate showed that the Senate Plan had a “racially discriminatory purpose.”
    Defs.’ Ex. 129, at 3. Other senators also wrote directly to Chairman Seliger to express their
    “disappointment in the process used to develop the Senate redistricting plan” and the
    “exclu[sion] [of] elected representatives of minority citizens” from that process. Defs.’ Ex. 132,
    at 1. Although statements from the senators aggrieved by the process do not necessarily show
    that it was racially discriminatory, instead of merely partisan, they do indicate that the majority
    was aware during redistricting that several members were upset by the irregular process, yet
    chose not to address their concerns.
    We conclude that Texas has not shown that the Senate Plan was enacted without
    discriminatory intent. Senator Davis and other Intervenors provided credible circumstantial
    evidence of the type called for by the Supreme Court in Arlington Heights, which, as a whole,
    indicates that an improper motive may have played a role in the map-drawing process. Rather
    than directly rebut this evidence, Texas asserts only that the legislature’s motivations were
    wholly partisan, untainted by considerations of race. We agree that a plan that impacts minority
    citizens more harshly than majority citizens is not necessarily at odds with section 5. But under
    the VRA and Arlington Heights, it is not enough for Texas to offer a plausible, nonracial
    explanation that is not grounded in the record. It must, at a minimum, respond to evidence that
    shows racial and ethnic motivation, which it has failed to do. See Arlington Heights, 
    429 U.S. at 266
     (“Absent a [clear pattern of discrimination] . . . the Court must look to other [circumstantial]
    evidence.”). Here, Texas has made no real attempt to engage with the Arlington Heights factors,
    even though it concedes that the Senate Plan has a disparate impact on minority voters in SD 10.
    We find it telling that the legislature deviated from typical redistricting procedures and excluded
    50
    minority voices from the process even as minority senators protested that section 5 was being run
    roughshod. One would expect a state that is as experienced with VRA litigation as Texas to have
    ensured that its redistricting process was beyond reproach. That Texas did not, and now fails to
    respond sufficiently to the parties’ evidence of discriminatory intent, compels us to conclude that
    the Senate Plan was enacted with discriminatory purpose as to SD 10.
    V. State House Plan
    A. Retrogression in the State House Plan
    The United States and the Intervenors argue that the enacted House Plan retrogresses
    minority voting power by eliminating eight ability districts (House Districts (HDs) 26, 33, 35,
    41, 106, 117, 144, and 149) without creating any others. Texas acknowledges retrogression in
    HD 33, but argues the House Plan works no abridgement of minority voting rights in any of the
    other districts. Texas maintains that the loss of HD 33 is offset by the plan’s provision for at least
    one and as many as three new ability districts. We conclude that the enacted plan will have the
    effect of abridging minority voting rights in four ability districts — HDs 33, 35, 117, and 149 —
    and that Texas did not create any new ability districts to offset those losses. Consequently, we
    conclude that the enacted plan cannot be precleared. We first analyze each of the eight alleged
    ability districts before turning to the three alleged offset districts.
    1. Alleged Retrogressive Districts
    a. State House District 33
    Nueces County in southeastern Texas includes three State House districts in the
    benchmark plan. HDs 33 and 34 are entirely within the county; HD 32 partially so. Benchmark
    HD 33 comprises the core of Corpus Christi. HD 34 includes the western part of the county, and
    HD 32 covers much of the eastern portion and extends into other counties immediately north of
    51
    Nueces County. The population of Nueces County grew at a slower rate than that of the rest of
    the State, so it was only entitled to 2.03 districts in the new map. Because the Texas Constitution
    mandates that any reapportionment of State House districts observe county lines where
    possible,35 the House mapdrawers drew only two districts in Nueces County, choosing to
    eliminate Hispanic-majority HD 33. See Trial Tr. 146:21-147:10, Jan. 17, 2012 AM.
    With an HCVAP of 60.4%, Pl.’s Ex. 13, at 13, and success electing the Hispanic
    candidate of choice in four out of the past five endogenous elections (with only a narrow victory
    by a Hispanic Republican in 2010 breaking this streak), Engstrom Suppl. Rep. 6 & n.5, there is
    no question that benchmark HD 33 was a Hispanic ability district. Even Texas concedes that if
    we accept, as we have, the binary analysis instead of Dr. Alford’s statewide functional approach,
    benchmark HD 33 would be an ability district. Tex. Post-Trial Br. 13.
    There is similarly little question that HD 33 is not an ability district in the enacted plan.
    The benchmark district’s population was redistributed to neighboring districts, and the new HD
    33 was transplanted to two predominantly Anglo counties near Dallas. The new HCVAP is only
    8.5%, Pl.’s Ex. 14, at 13, and no expert’s reconstituted election analysis shows any electoral
    victories for minority-preferred candidates. See, e.g., Alford Rep. 11 tbl.3b. At trial, Dr. Alford
    conceded that enacted HD 33 is not an ability district. Trial Tr. 99:16-18, Jan. 24, 2012 PM. The
    State also concedes that the binary approach supports this conclusion. Tex. Post-Trial Br. 13. We
    thus conclude that HD 33 is a lost ability district.
    35
    Under the County Line Rule, TEX. CONST. art. III, § 26, a district must be drawn to mirror a county’s
    boundary lines if that county has sufficient population for a voting district. When the population of more than one
    county is needed to make up a single voting district, the Rule requires that contiguous counties be joined to form that
    district. Likewise, when the population of a county requires more than one voting district, the districts must be
    contained within the county lines and any excess population must be joined wholly with population from a
    neighboring county to form a district.
    52
    b. State House District 35
    The parties who address this district agree that enacted HD 35 in south Texas is not an
    ability district. They disagree whether it is an ability district in the benchmark plan. The United
    States argues that benchmark HD 35 is an ability district because, just as in HD 33, the minority-
    preferred candidate won four out of the last five endogenous elections, and the fifth was a close
    election where a Hispanic Republican won the seat from the incumbent Hispanic Democrat.36
    U.S. Post-Trial Br. 5; see also Handley House Rep. 5. This track record of success is evidence
    that benchmark HD 35 is an ability district. Texas counters that the exogenous analysis tells a
    different story. The OAG 10 indicates that the district performs for minority voters only half the
    time. See Alford Rep. 11 tbl.3b. The other experts’ analyses place its success rate even lower: the
    district performed for minority voters in just two of Dr. Handley’s five elections, and two of Dr.
    Engstrom’s seven. See Handley Rep. 5; Engstrom Chart.
    Texas also argues that enacted HD 35 will perform much the same as benchmark HD 35.
    The district’s HCVAP drops only slightly, from 54.6% in the benchmark to 52.5% in the enacted
    plan, Pl.’s Exs. 13, at 13; 14, at 13, and the exogenous analyses show only minor changes
    between the plans. The analyses of Dr. Handley and the OAG 10 show a one election drop in
    effectiveness. Handley House Rep. 9; Alford Rep. 11 tbl.3b. Dr. Engstrom’s analysis, which
    weights recent elections more heavily, shows a one election increase. See Engstrom Chart. To
    Texas, all this suggests that there is no meaningful change in the district’s performance, and
    because all agree that enacted HD 35 is not an ability district, benchmark HD 35 must not be an
    ability district either.
    36
    We agree with Dr. Handley that Representative Jose Aliseda, who won in 2010 with only 22.3% of the
    Hispanic vote, is not the Hispanic candidate of choice. See Handley House Rep. app. D, at 34.
    53
    While true that only minor changes were made between benchmark and enacted HD 35,
    we think the best reading of the record is that the benchmark district is one in which minorities
    usually, although not always, elect their preferred candidate. Hispanic voters constitute the
    majority of the district, albeit barely, and they have been successful in electing their preferred
    candidate in endogenous elections held between 2002 and 2008. We find this to be persuasive
    evidence that Hispanic voters have attained an ability to elect their preferred candidates in HD
    35. Texas does not argue that endogenous results are misleading in this district, but instead
    repeats its general position that we should consider only exogenous election results. Tex. Post-
    Trial Br. 12. We have already rejected this argument. Exogenous analysis uses statewide and
    national elections to help determine political trends within a district. But by considering district-
    wide election results, endogenous analysis provides a more direct answer to the question posed
    by section 5: have minority voters shown an ability to elect their preferred candidates in that
    district? Because the exogenous results do not cut entirely against ability status — here, Texas’s
    own exogenous analysis shows a 50% benchmark success rate — and there is nothing in the
    record that calls into question the probative value of this district’s endogenous track record, we
    are confident that endogenous results accurately describe minority voting ability in the
    benchmark.
    As to enacted HD 35, Texas has not presented any evidence that HD 35 remains an
    ability district or that it tried to preserve the district’s ability status, and its argument based on the
    small changes in exogenous election performance is insufficient to counter the evidence we do
    have supporting the conclusion of the United States’s expert that the district loses ability status.
    When a district is close to the ability-to-elect line, even minor changes can be significant. The
    low exogenous election results for the enacted district combined with HCVAP changes that push
    54
    the district even closer to the majority line (52.5%) are not enough to show that the district will
    continue to perform for minority voters. We must conclude that the evidence Texas offers is not
    persuasive to meet its burden to show that the changes made to HD 35 will not have a
    retrogressive effect on minority voters.37
    c. State House District 41
    All parties agree that benchmark HD 41 in south Texas’s Hidalgo County is a minority
    ability district. Texas argues that the district remains so in the enacted plan, and we agree.
    The HCVAP in enacted HD 41 is 72.1%. Pl.’s Ex. 14, at 14. Although a decrease from
    77.5% in the benchmark, Pl.’s Ex. 13, at 14, that percentage remains well above the 65%
    threshold we laid down in our summary judgment opinion as a presumption of ability status.38
    See Texas, 831 F. Supp. 2d at 263 & n22. We agree that such a high Hispanic population density
    creates a strong presumption that HD 41 remains an ability district. Significantly, none of the
    experts thought that HD 41 lost ability status, a conclusion that both the OAG 10 and Dr.
    Engstrom’s analysis confirm. Alford Rep. 11 tbl.3b; Engstrom Chart.
    The United States takes issue with the value of a bright-line test, especially in a district
    like HD 41 where the uncontested record shows that voters have faced serious and pervasive
    socioeconomic barriers that depress voter turnout. U.S. Post-Trial Br. 8-9. The United States also
    argues that we know very little about enacted HD 41, and what we do know — its high Hispanic
    37
    This district presents a close and very difficult case. Presented with more or different evidence, we might
    conclude that the seemingly minor changes made to the district do not alter its ability status. Nevertheless, Congress
    has allocated the burden to prove lack of discriminatory effect to the State. On the record before us, we conclude
    that Texas has not done so.
    38
    Texas argues that our summary judgment opinion set out a 60%, not 65%, bright-line test. Tex. Post-Trial
    Br. 7 & n.5, 11. We find this argument puzzling given that our previous opinion stated that “a minority voting
    majority of sixty-five percent (or more) essentially guarantees that . . . a cohesive minority group will be able to
    elect its candidate of choice.” Texas, 831 F. Supp. 2d at 263. Texas argues that most of the authorities we cited used
    a 60% voting age population bright line, but we cited these (nonbinding) cases only as examples of the ways other
    courts have approached this issue.
    55
    population — is not enough for Texas to meet its burden to show no retrogression. The
    background for Texas’s approach to redrawing HD 41 centers on the decision of Representative
    Aaron Peña, the five-term incumbent in neighboring HD 40, to switch party affiliations from
    Democrat to Republican following the 2010 election. One of the mapdrawers’ goals during
    redistricting was to protect Rep. Peña’s chances of reelection. Trial Tr. 163:4-165:13, Jan. 17,
    2012 AM. They decided that the best way to do this was to have Peña, in effect, switch districts
    with HD 41’s incumbent, and then cut out of the district some strong Democratic areas “to
    increase the Republican performance of [enacted HD 41].” Id. at 168:1-3. The result is an oddly
    shaped district full of sharp corners that has earned the nickname “Transformer,” both here and
    in the section 2 litigation. See id. at 42:4-5, Jan. 23, 2012 PM. Enacted HD 41 splits apart
    seventeen of the forty-two voter tabulation districts (VTDs)39 in the district, Defs.’ Ex. 800, at
    35, in an effort to bolster Republican voting strength. Trial Tr. 165:17-168:17, Jan. 17, 2012 AM.
    Dr. Handley was unable to draw a conclusion whether enacted HD 41 remains an ability
    district because of these splits. Handley House Rep. 1 n.1. Election performance data is only
    available at the VTD level and not at the more precise level of a city block. Reconstituted
    election analysis uses the political averages for an entire VTD to assess how a portion of a VTD
    will perform. See, e.g., Trial Tr. 74:25-78:21, Jan. 24, 2012 PM; id. at 11:7-13, 50:19-23, 74:10-
    75:13, Jan. 26, 2012 AM. The higher the number of VTD splits in a new district, the more
    inconclusive these predictions become. Here, where over 31% of the district’s population lives in
    split-VTD areas and where the mapdrawers’ stated goal was to peel off from the district strong
    Democratic areas — suggesting that the general concerns about skewed exogenous results from
    political variance within split VTDs may be especially strong in this district — Dr. Handley
    39
    In Texas, VTDs are roughly equivalent to precincts elsewhere.
    56
    concluded that the results of her exogenous analysis were not reliable forecasters of the district’s
    future voting strength. See Handley House Rep. 9-10.
    We are not deaf to the concerns the United States raises, and we are skeptical of the
    State’s claim that high HCVAP is sufficient to prove continued ability status in light of the
    uncontested testimony that HD 41 was engineered to transform a reliable Democratic district into
    one that would elect a Republican instead. Nevertheless, we do not think the record calls into
    question enacted HD 41’s status as an ability district. Dr. Handley’s concerns would give us
    more pause were minority voting power less established, but we agree with the other experts that
    the shortcomings of reconstituted election analysis for enacted HD 41 are not enough to keep us
    from concluding the district does not retrogress. This is not a case in which the Hispanic
    population is close to the majority line, or even close to the supermajority 65% line we set out in
    our summary judgment opinion. Enacted HD 41 still has an HCVAP of 72.1%. We are hard
    pressed to find that minority voters lack an ability to elect in a district in which they comprise
    such a high percentage of the voting public. We need not decide whether the United States is
    correct that, in a rare case, 65% HCVAP may not be enough to ensure ability to elect, because in
    this case, 72.1% is.
    Lastly, if Texas succeeded in its goal to create a Republican district, Rep. Peña’s success
    would require support from a sizable portion of the district’s Hispanic community. This suggests
    either that Rep. Peña would be the Hispanic candidate of choice, or that Hispanic voting
    cohesion would have broken, perhaps to the point where there would no longer be one Hispanic-
    preferred candidate. If the former, Rep. Peña’s victory would not be a mark against Hispanic
    ability to elect. If the latter, finding retrogression would cause us to discount the preferences of
    57
    the district’s Hispanic Republican voters, which would put us at odds with section 5’s mandate.
    We conclude that HD 41 remains an ability district in the enacted plan.
    d. State House District 117
    As it does with regards to HD 41, Texas argues that the 63.8% HCVAP of southwestern
    San Antonio’s enacted HD 117, Pl.’s Ex. 14, at 16, satisfies our bright-line test for ability to
    elect. Yet as we have said, Texas misreads our summary judgment opinion. A minority voting
    population of 65% or higher, not 60%, is necessary to “essentially guarantee” ability to elect.
    Texas, 831 F. Supp. 2d at 263. We thus use the multi-factored analysis to assess the status of this
    district without starting from a presumption of ability status.40
    Benchmark HD 117’s protected status has not been seriously challenged, and we have
    been presented with no evidence indicating that the district does not perform for minority voters.
    Dr. Handley’s endogenous data shows the minority-preferred candidate won four of the five past
    elections, and only lost the fifth by a narrow margin. Handley House Rep. 9. The exogenous data
    shows an ability district, too. See Alford Rep. 11 tbl.3b (five out of ten elections); Handley
    House Rep. 5 (three out of five elections); Engstrom Suppl. Rep. 6 (four out of seven elections).
    As for enacted HD 117, Texas points out that the district’s boundaries remain essentially
    unchanged and notes that the district has been trending Republican in recent years. Considering
    only the five most recent elections in the OAG 10, exogenous results are the same for benchmark
    and enacted HD 117: minority-preferred candidates won only two out of five. See Tex. Post-Trial
    Br. 11-12; Alford Rep. 11 tbl.3. The United States, by contrast, argues that enacted HD 117 was
    40
    In its post-trial brief, Texas argues for the first time that enacted HD 117 satisfies the bright-line test as a
    coalition district because the Hispanic and Black communities comprise 68.4% of the district’s voting age
    population. Tex. Post-Trial Br. 11. We reject this new argument, especially because Texas has presented no
    evidence, such as election analysis or evidence of voting cohesiveness between the minority communities, to support
    a conclusion that HD 117 is a coalition district.
    58
    purposely engineered to appear unchanged from the benchmark, but that the proposed
    boundaries actually decrease minority voter power. U.S. Post-Trial Br. 9-10.
    We conclude that enacted HD 117 is no longer an ability district. Texas may be correct
    that minority voting power is beginning to weaken in the benchmark district, but it has not yet
    dropped below the ability-to-elect threshold. The exogenous data shows that changes made
    during redistricting, not shifting political trends, are responsible for enacted HD 117’s loss of
    ability status. The exogenous election analyses of all experts, including Texas’s, show that
    minority effectiveness decreases from the benchmark level, and all conclude that minority-
    preferred candidates carry HD 117 less than half the time. Alford Rep. 11 tbl.3 (two out of ten
    elections); Handley House Rep. 11 (one out of five elections); Engstrom Suppl. Rep. 8-9 (three
    out of seven elections).
    The high Hispanic population in enacted HD 117 — HCVAP increases five percentage
    points from the benchmark to 63.8%, Pl.’s Exs. 13, at 16; 14, at 15 — could suggest that enacted
    HD 117 remains an ability district despite its meager exogenous results. Yet HCVAP numbers
    do not tell the full story. The district’s Spanish Surname Voter Registration (SSVR)41 level is
    significantly lower at just 50.1%. Pl.’s Ex. 14, at 27. The record shows that the mapdrawers
    purposely drew HD 117 to keep the number of active Hispanic voters low so that the district
    would only appear to maintain its Hispanic voting strength, and that they succeeded.
    The primary mapdrawer for the House Plan, Gerardo Interiano, testified that a “ground
    rule[]” for drawing HD 117 was to keep the SSVR level just above 50%. Trial Tr. 106:25-108:1,
    Jan. 25, 2012 PM. The mapdrawers accomplished this goal by placing in the new district areas
    with high Hispanic populations but lower voter turnout, while excluding from the district high-
    41
    SSVR is a metric used to approximate the number of registered Hispanic voters in a given geographic area.
    The list is compiled by comparing state voter registration records against a Census list of common Spanish
    surnames.
    59
    Hispanic, high-turnout areas. For example, the heavily Hispanic communities of Somerset and
    Whispering Winds, part of benchmark HD 118, are both very poor and have low voter turnout.
    See id. at 9:7-13:7, Jan. 24, 2012 PM; Defs.’ Ex. 363, Garza Dep. 40:8-42:25, Oct. 19, 2011.
    They were moved to HD 117 despite repeated requests from HD 118’s representative, Joe Farias,
    to keep the communities within his district. Trial Tr. 7:11-14, 14:2-15:3, Jan. 24, 2012 PM. Rep.
    Farias’s offer to “trade” an area in HD 118 with similar Hispanic population numbers in
    exchange for keeping Somerset and Whispering Winds in his district was rejected, and according
    to his unchallenged testimony, the only plausible reason for this refusal was that Hispanic voters
    in the region he offered to trade have much higher turnout rates than the voters in Somerset and
    Whispering Winds. Id. at 14:19-17:23. Similarly, Interiano testified that Somerset was moved to
    HD 117 as a way to keep the district “above 50 percent [SSVR] and maintain [our] other goals in
    the district” — strengthening Representative John Garza’s chances at reelection. Id. at 107:7-11,
    Jan. 25, 2012 PM.
    These incidents illustrate Texas’s overall approach in HD 117: Texas tried to draw a
    district that would look Hispanic, but perform for Anglos. According to the experts, that was the
    result achieved. We conclude that HD 117 is a lost ability district.42
    e. State House District 149
    HD 149 in Houston-area Harris County is an alleged coalition district composed of
    Asian-American, Black, and Hispanic voters. The 2010 Census shows that Harris County was
    42
    Our conclusion that HD 117 is retrogressive may seem inconsistent with our conclusion regarding HD 41,
    given that HD 117’s HCVAP is only 1.2 percentage points below the 65% bright line. Yet there are significant
    differences between the two districts. First, HD 41’s HCVAP is eight points higher than that of HD 117, and eight
    points represents a significant difference in electoral power. Second, unlike HD 41, where no expert was willing to
    conclude that the district lost ability status, both Dr. Handley and Dr. Engstrom conclude HD 117 did. Handley
    House Rep. 11; Engstrom Suppl. Rep. 8-9. Finally, our concerns that finding HD 41 retrogressive would discount
    the choice of Hispanic Republicans is not an issue here. The evidence for HD 41 showed that the mapdrawers
    excluded Republican portions of the map; here it shows they excluded high-turnout portions. Selecting among
    Hispanic voters based on their political preferences may not raise a red flag under section 5, but selecting based on
    minority voters’ history of turnout, regardless of political preference, does.
    60
    entitled to twenty-four districts, not its current twenty-five, so HD 149 was selected for
    elimination. The legislature chose to draw the home of HD 149’s representative, Hubert Vo, into
    HD 137 so that Rep. Vo would be forced to run against Scott Hochberg, HD 137’s
    representative, in the next election. Defs.’ Ex. 352, Test. of Rogene Calvert, Trial Tr. 422:14-22,
    Perez, No 11-cv-360. Representatives Vo and Hochberg are the only Democrats in the county’s
    delegation. Benchmark HD 149’s population was redistributed to neighboring districts and
    enacted HD 149 was transplanted to an entirely different county in a different part of the state.
    The new district’s demographics shift dramatically from minority- to majority-Anglo. Pl.’s Exs.
    13, at 17 (benchmark Anglo CVAP of 37.6%); 14, at 17 (enacted Anglo CVAP of 77.4%). There
    is, unsurprisingly, no dispute that enacted HD 149 is not an ability district. Our only task is to
    determine whether benchmark HD 149 is a coalition district protected under section 5. As
    discussed above, we have concluded that section 5 protects coalition districts when there is clear
    evidence both of cohesion among the coalition’s members and demonstrated electoral success.
    Here, we conclude that this standard has been met.
    Rep. Vo is the minority candidate of choice and has won the last four endogenous
    elections in the district. Handley House Rep. 7 tbl.3. With such strong results, we would likely
    conclude that HD 149 is an ability district were there a single minority group in the district. But
    as we have already discussed, we must ask more when analyzing a claim that a coalition has
    created an ability district. There are four reasons why we conclude this endogenous success is
    persuasive evidence of the coalition’s demonstrated ability to elect.43
    First, population demographics give HD 149 the potential to perform as a coalition
    district. The district’s combined Asian-American, Black, and Hispanic CVAP is 61.3%. Pl.’s Ex.
    43
    Our conclusion is consistent with Dr. Handley’s assessment of the district. See Handley House Rep. 3, 7, 13.
    61
    13, at 17. This fact has limited value in assessing minority voting power without information
    such as voter turnout and cohesion statistics, but it does indicate that if the minority groups in the
    district came together, they would likely be able to elect their preferred candidate, potentially
    without any help from Anglo crossover voters.
    Second, the record shows that all three minority groups in the district vote cohesively.
    Texas has not contested that the district’s minority communities vote cohesively in general
    elections. And although the parties did not provide racially polarized voting analysis or a
    breakdown of election returns for Rep. Vo’s races, the Texas OAG’s analysis shows that
    Hispanic and Black voters in HD 149 uniformly prefer the same candidates in general elections
    and that their preferences consistently diverge from those of the district’s Anglo voters. See Pl.’s
    Ex. 26, at 3557-60. We have no statistical evidence of Asian-American voting patterns in the
    record, but the testimony at trial, discussed in more detail below, reports broad, cohesive support
    for Rep. Vo among all three minority communities and especially within the Asian-American
    community.
    Third, uncontroverted anecdotal evidence shows that a tripartite coalition of the Asian-
    American, Black, and Hispanic communities consistently elects its candidate of choice. HD
    137’s Rep. Hochberg testified to the strength of the coalition, concluding that “[p]olitically all
    three of [the minority] communities form a coalition, and the Asian community is the glue
    holding things together.” Defs.’ Ex. 738, Pre-Filed Direct Test. of Representative Scott
    Hochberg 13:12-13. Rogene Calvert, an associate of the Texas Asian American Redistricting
    Initiative, testified that Rep. Vo defeated a twenty-two year incumbent in 2004 on the strength of
    the district’s tri-ethnic coalition. Defs.’ Ex. 736, Pre-Filed Direct Test. of Rogene Calvert 11:3-
    16. The Asian-American community “really rallied behind Mr. Vo when he announced his
    62
    candidacy” and “took a lot of pride in Vo’s candidacy,” to the point that many Asian-Americans
    came out to support him who had “never participated in elections.” Id. at 11:8-11. Furthermore,
    he “wouldn’t have had a chance of success if he hadn’t received support from the other minority
    communities in District 149,” including endorsements from both Black and Hispanic political
    groups, and the Asian-American, Hispanic, and Black communities “all worked together to elect
    Mr. Vo.” Id. at 11:11-23; see also Defs.’ Ex. 352, Trial Tr. 420:14-17, Test. of Rogene Calvert,
    Perez, No. 11-cv-360, (Calvert, testifying that she has “seen Asian-Americans elected to office
    and other candidates of our choice due to the fact that we can coalesce with other groups to elect
    those people”); Defs.’ Ex. 353, Trial Tr. 425:18-24, Test. of Sarah Winkler, Perez, No.
    11-cv-360 (local school board member testifying that it is necessary to gain the support of all
    three minority groups to win office within HD 149). We find this testimony credible, and Texas
    has made no effort to dispute this evidence that the coalition is effective in local and district-wide
    elections.44
    Finally, the coalition has a track record of success, electing Rep. Vo in 2004 and in every
    election since. The tri-ethnic coalition has also had sustained success electing other local
    officials, such as school board and Houston City Council members. Defs.’ Ex. 736, Pre-Filed
    Direct Test. of Rogene Calvert 12:11-13:7. Although Texas points out that the district only
    performs in one of Dr. Handley’s five exogenous elections, Tex. Post-Trial Br. 13; see also
    Handley House Rep. 7 tbl.3, we do not find this persuasive. Texas’s expert did not provide
    general election exogenous analysis for this district; the only expert to do so is Dr. Handley, and
    44
    Although the Court agrees that this testimony is sufficient to conclude that the district is protected under
    section 5, it differs in its views of the strength of the evidence. Judge Griffith concludes that the testimony of Rep.
    Hochberg and Calvert shows that the Asian-American community leads the coalition and that the Black and
    Hispanic communities play a consistently supportive and vital role in its success. Judges Collyer and Howell need
    not reach the issue of leadership because they conclude that a tri-partite arrangement of equals is sufficient for
    protection under section 5.
    63
    she concluded that the endogenous results were more important for understanding voting patterns
    in the district. See Handley House Rep. 13 & n.20. Especially when combined with evidence that
    the coalition has success electing other local officials, we agree with Dr. Handley that
    endogenous elections, which speak to the ability of a particular voting community to coalesce
    around candidates for local office, are the best evidence of this coalition district’s success.
    Unlike the facts of SD 10, here we have evidence of both concerted efforts among a coalition to
    elect its preferred candidates and a pattern of success extending across multiple election cycles.
    Texas’s primary objection to this approach is to argue that the minority groups in HD 149
    do not vote cohesively in primaries and only come together to agree on a second- or third-best
    candidate in time for the general election. In Texas’s view, this does not prove an effective
    coalition district. Tex. Post-Trial Br. 9-10, 12-13; see also Alford Rep. 19-21 (explaining his
    analysis showing that Asian-American, Black, and Hispanic voters in HD 149 do not vote
    cohesively at the primary level). We agree that evidence of shared voting preferences at the
    primary level would be powerful evidence of a working coalition, but it is not needed to prove
    cohesion. In the first place, there is little support for Texas’s focus on primary elections. Texas
    cites LULAC for this point, but LULAC, a section 2 case, only talks about primaries as a method
    to determine one minority group’s candidate of choice; it says nothing about the need for two
    groups in a putative coalition to vote cohesively in a primary. See 
    548 U.S. at 444
    . More
    importantly, it does not hold that evidence of cohesion in a primary is necessary to identify a
    candidate of choice. 
    Id.
     (stating that without a contested primary there was “no obvious
    benchmark” to determine the minority-preferred candidate, and that the district court could draw
    multiple reasonable inferences from this lack of primary-level evidence). The same is true here,
    where there has been no contested endogenous Democratic primary since 2004, when Rep. Vo
    64
    first won his seat. Texas also cites two district court cases that rely on primary cohesion,
    Rodriguez v. Pataki, 
    308 F. Supp. 2d 346
    , 421 (S.D.N.Y. 2004); and Session v. Perry, 
    298 F. Supp. 2d 451
    , 478 (E.D. Tex. 2004), but these cases represent the minority view. Most courts to
    address this issue have expressed no preference about the election level at which voting cohesion
    must be shown. See, e.g., Lewis v. Alamance Cnty., 
    99 F.3d 600
    , 615 (4th Cir. 1996); LULAC,
    Council No. 4434 v. Clements, 
    999 F.2d 831
    , 886 (5th Cir. 1993) (en banc); Bridgeport Coal. for
    Fair Representation v. City of Bridgeport, 
    26 F.3d 271
    , 276 (2d Cir. 1994), vacated and
    remanded on other grounds, 
    512 U.S. 1283
    .
    We agree with the majority view. Courts regularly consider general election data to
    demonstrate voter cohesion in traditional majority-minority districts, without any indication that
    such a showing is insufficient without evidence of voter cohesion in the primary as well. See,
    e.g., Thornburg v. Gingles, 
    478 U.S. 30
    , 58-59 (1986); Old Person v. Cooney, 
    230 F.3d 1113
    ,
    1121 (9th Cir. 2000). Additionally, requiring cohesion in the primary election distorts the role of
    the primary. Although minority groups sometimes coalesce around a candidate at that point in
    time, minority voters, like any other voters, use the primary to help develop their preferences.
    We refuse to penalize minority voters for acting like other groups in a political party who do not
    coalesce around a candidate until the race is on for the general election. See Alamance Cnty., 
    99 F.3d at 614-16
     (“We reject the proposition that success of a minority-preferred candidate in a
    general election is entitled to less weight when a candidate with far greater minority support was
    defeated in the primary. . . . [S]uch a view is grounded in the belief that minority voters
    essentially take their marbles and go home whenever the candidate whom they prefer most in the
    primary does not prevail, a belief about minority voters that we do not share.” (citation and
    internal quotation marks omitted)). “Pull, haul, and trade” describes the task of minority and
    65
    majority voters alike, and candidates may be minority “candidates of choice” even if they do not
    “represent perfection to every minority voter.” De Grandy, 
    512 U.S. at 1020
    .
    We are persuaded the record establishes that benchmark HD 149 is a coalition district
    protected under section 5. The Asian-American, Black, and Hispanic voters in the district work
    together to support their preferred candidates, and they have a multi-year record of success.
    Benchmark HD 149 is a protected ability district, and Texas’s decision to dismantle it without
    offsetting the loss elsewhere is retrogressive.
    f. State House Districts 26, 106, and 144
    Various Intervenors have argued that HDs 26, 106, and 144 are also lost ability districts.
    We disagree. For two of the districts, HDs 26 and 106, the only evidence presented shows that
    neither is a majority-minority district and both are currently represented by Anglo Republicans.
    See Pl.’s Ex. 13, at 13, 16. Other than scant assertions about one endogenous election in which
    the Anglo Republican candidate won by a narrow margin and reputed exogenous success since
    2008, see Texas Legislative Black Caucus Post-Trial Br. 3-7, the parties have offered no election
    performance data or reconstituted election analysis. We cannot make findings of minority voting
    ability based on this thin record. At best, the evidence may show that the districts are beginning
    to favor minority-preferred candidates, but section 5’s effect prong protects only existing, not
    emerging, ability districts. See Texas, 831 F. Supp. 2d at 264-65.
    Similarly, HD 144 is not a majority-minority district and is represented by an Anglo
    Republican. Pl.’s Ex. 13, at 17. Both Dr. Handley’s and Dr. Engstrom’s exogenous election
    analyses show no victories for minority-preferred candidates in this district. See Handley House
    Rep. 5; Engstrom Chart. We find that benchmark HD 144 is not an ability district.
    66
    2. Alleged New Ability Districts
    Texas argues that the legislature created as many as three new ability districts, which
    offset the loss of any that might have been eliminated in the enacted plan. But the enacted plan
    does not draw any new ability districts. It only strengthens minority voting power in some
    districts that have already achieved the ability to elect. As we have already discussed,
    strengthening ability districts cannot salvage a retrogressive plan. A state may not offset the
    elimination of an ability district by “packing” additional minority voters into a district that
    already performs. What the State calls offsets are actually existing ability districts, and they do
    not compensate for the loss of others.
    During the course of this litigation, Texas has offered three different explanations for
    how the enacted plan creates new Hispanic ability districts. At summary judgment, Texas
    identified HD 148 in the Houston area as a new ability district. Mot. Summ. J. ¶ 11. At trial,
    Texas’s chief witness for the House Plan testified that he believed strengthening the SSVR
    percentages in HD 148 and Tarrant County’s HD 90 compensated for the loss of HD 33. Trial
    Tr. 11:24-12:6, Jan. 17, 2012 PM. And at closing arguments and in post-trial briefing, Texas
    appears to abandon these claims, shifting instead to the altogether new argument that enacted HD
    74 in western Texas is a new ability district. Tex. Post-Trial Br. 13-14.
    Texas’s decision no longer to rely on HDs 90 and 148 was sound. Although an initial
    examination of the demographic data shows that both districts are more strongly Hispanic in the
    enacted plan,45 all the experts’ election analyses show that both are already ability districts. Both
    achieved a perfect score under Dr. Handley’s endogenous election analysis, Handley House Rep.
    45
    The HCVAP in proposed HD 90 increases from 47.9% to 49.7%, and SSVR from 47.2% to 50.1%. Pl.’s
    Exs. 13, at 20; 14, at 20. HCVAP in proposed HD 148 increases from 42.1% to 51.4%, and SSVR from 40.0% to
    50.0%. Pl.’s Exs. 13, at 21; 14, at 21.
    67
    5 tbl.1, and no expert’s exogenous analysis shows any change between the performance of the
    benchmark and enacted districts. See Alford Rep. app. B; Handley House Rep. 5 tbl.1, 11 tbl.3;
    Engstrom Chart. Increasing the size of their minority populations had no impact on these districts
    for purposes of section 5’s effect prong.
    Whether enacted HD 74 is a new ability district is a closer call, but we conclude it is
    not.46 With an HCVAP of 69.4%, Pl.’s Ex. 14, at 15, all parties agree that enacted HD 74 is an
    ability district; the question is whether benchmark HD 74 is as well. Yet the rest of the evidence
    shows that, as with HDs 90 and 148, the district’s demographic changes only strengthen an
    already-performing minority district.
    Benchmark HD 74 is majority-Hispanic, with an HCVAP of 59.7%. Pl.’s Ex. 13, at 14.
    Representative Pete Gallego, the Hispanic candidate of choice, has represented the district since
    1990. Although Texas now argues that benchmark HD 74 is not an ability district, key players
    during redistricting believed it was. See Trial Tr. 25:5-22, Jan. 17, 2012 PM (Interiano, testifying
    that he identified HD 74 as a protected district at the outset of the redistricting process); Defs.’
    Exs. 214, 215, 820 (memoranda from Texas Legislative Council attorney David Hanna
    identifying benchmark HD 74 as a protected district). With a majority-Hispanic population,
    twenty-two years of success electing the minority-preferred candidate, and apparently little doubt
    by anyone that the district was protected until late in the litigation process, it seems clear that HD
    74 does not need the new boundaries of the enacted plan to perform for minority voters.
    In response, Texas points to the exogenous election analyses that paint a weaker picture
    of minority success. See Alford Rep. 11 tbl.3b (reporting minority victories in four of the OAG
    10 elections); Handley House Rep. 5 tbl.1 (one out of five victories). But see Engstrom Chart
    46
    We note that even if we agreed with Texas that enacted HD 74 is a new ability district, the enacted plan
    would still be retrogressive because the creation of one new ability district cannot offset the loss of several others.
    68
    (four out of seven victories). Texas argues that the endogenous results reflect only the fact that
    Rep. Gallego has held office in HD 74 for over two decades. According to Texas, that is
    insufficient evidence that HD 74 is an ability district. Tex. Post-Trial Br. 13-14.
    We are not persuaded. As discussed above, endogenous elections are the best indication
    of ability to elect. What a minority community actually does in a specific district on election day
    is more powerful evidence than reconstituted statewide results of its ability — or lack thereof —
    to elect a preferred candidate. To be sure, Rep. Gallego’s success is almost certainly attributable,
    in part, to the considerable advantages of incumbency. But Texas asks us to discount a long
    history of endogenous success without providing evidence that incumbency is the predominant
    reason the minority community is able to elect Rep. Gallego. We decline to speculate with Texas
    that this rationale, instead of, for example, a large Hispanic community of interest with a
    mobilized voter base, accounts for the district’s long history of electing the Hispanic-preferred
    candidate. Texas has failed to show that the minority community has reelected Rep. Gallego
    multiple times despite lacking an ability to elect.
    Moreover, we reject the premise that incumbency advantage is a mark against ability to
    elect. The minority community need not elect a different candidate in successive terms to prove
    continuing ability to elect. As we emphasized in our summary judgment opinion, analyzing
    ability to elect includes considering all relevant factors. Texas, 831 F. Supp. 2d at 260.
    Incumbency can be a tool that a minority community, like any other group of voters, uses to
    enhance its electoral power. We are sensitive to Texas’s concern that incumbency advantage
    could, at times, give a “false positive” for ability status, but we conclude that the best solution is
    to consider the record as a whole, not to exclude probative evidence. We are persuaded that what
    has happened on the ground in HD 74 for over two decades — the consistent reelection of Rep.
    69
    Gallego — reflects the reality of established minority voting power. We thus conclude that HD
    74 is an ability district in the benchmark plan, and that Texas’s attempt to add Hispanic voters to
    the district cannot be used to offset the loss of ability districts elsewhere.47
    B. Discriminatory Intent in the State House Plan
    Because of the retrogressive effect of the State House Plan on minority voters, we do not
    reach whether the Plan was drawn with discriminatory purpose. But we note record evidence that
    causes concern. First, the process for drawing the House Plan showed little attention to, training
    on, or concern for the VRA. See, e.g., Trial Tr. 61:1-66:23, Jan. 20, 2012 PM. And despite the
    dramatic population growth in the State’s Hispanic population that was concentrated primarily in
    three geographic areas, Texas failed to create any new minority ability districts among 150
    relatively small House districts.
    These concerns are exacerbated by the evidence we received about the process that led to
    enacted HD 117. As detailed above, the mapdrawers modified HD 117 so that it would elect the
    Anglo-preferred candidate yet would look like a Hispanic ability district on paper. They
    accomplished this by switching high-turnout for low-turnout Hispanic voters, hoping to keep the
    SSVR level just high enough to pass muster under the VRA while changing the district into one
    that performed for Anglo voters. This testimony is concerning because it shows a deliberate,
    47
    In a footnote in its post-trial briefing, Texas advances — for the first time — HD 101 as another potential
    offset district. Tex. Post-Trial Br. 14 n.7. It argues that if this Court finds coalition districts are protected under
    section 5, as we have, then enacted HD 101 is a new coalition district because the combined Black, Hispanic, and
    Asian-American CVAP is 55.5% and the district is located in Democratic-leaning Tarrant County (and so,
    presumably, the minority community will have help from crossover Anglo voters). We stated at summary judgment
    that the lack of election returns to show that two or more distinct minority communities will coalesce around a
    preferred candidate makes it “extremely difficult to confirm that minority voters would indeed have the ability to
    elect” in a prospective coalition district. Texas, 831 F. Supp. 2d at 268. Accordingly, we will not conclude, without
    evidence, that the minority groups in this new district will coalesce around the same candidates and turn out in
    sufficient numbers to elect them.
    70
    race-conscious method to manipulate not simply the Democratic vote but, more specifically, the
    Hispanic vote.
    Finally, the incredible testimony of the lead House mapdrawer reinforces evidence
    suggesting mapdrawers cracked VTDs along racial lines to dilute minority voting power. Texas
    made Interiano’s testimony the cornerstone of its case on purpose in the House Plan. Trial Tr.
    45:22-25, Jan. 17, 2012 AM (“[O]ur [discriminatory purpose] case rests largely on the credibility
    of one person. His name is Gerardo Interiano.”). Interiano spent close to a thousand hours — the
    equivalent of six months of full-time work — training on the computer program Texas used for
    redistricting, id. at 131:3-5, yet testified that he did not know about the program’s help function,
    id. at 85:18-25, Jan. 25, 2012 PM, or of its capability to display racial data at the census block
    level, id. at 93:13-19, Jan. 17, 2012 PM. As unequivocally demonstrated at trial, this information
    was readily apparent to even a casual user, let alone one as experienced as Interiano. See id. at
    93:1-15; id. at 88:5-89:17, Jan. 25, 2012 PM. The implausibility of Interiano’s professed
    ignorance of these functions suggests that Texas had something to hide in the way it used racial
    data to draw district lines. The data about which Interiano claimed ignorance could have allowed
    him to split voting precincts along racial (but not political) lines in precisely the manner the
    United States and the Intervenors allege occurred.
    This and other record evidence may support a finding of discriminatory purpose in
    enacting the State House Plan. Although we need not reach this issue, at minimum, the full
    record strongly suggests that the retrogressive effect we have found may not have been
    accidental.
    71
    VI. Conclusion
    We conclude that Texas has not met its burden to show that the U.S. Congressional and
    State House Plans will not have a retrogressive effect, and that the U.S. Congressional and State
    Senate Plans were not enacted with discriminatory purpose. Accordingly, we deny Texas
    declaratory relief. Texas has failed to carry its burden that Plans C185, S148, and H283 do not
    have the purpose or effect of denying or abridging the right to vote on account of race, color, or
    membership in a language minority group under section 5 of the Voting Rights Act.
    Date: August 28, 2012
    /s/
    THOMAS B. GRIFFITH
    United States Circuit Judge
    /s/
    ROSEMARY M. COLLYER
    United States District Judge
    /s/
    BERYL A. HOWELL
    United States District Judge
    72
    Separate opinion for the Court with respect to retrogression in Congressional District 25 by
    HOWELL, District Judge:
    The enacted Congressional Plan abridges the ability of minorities to elect their candidates
    of choice, and thus cannot be precleared under Section 5 of the VRA. As explained below, CD
    25 was among the districts that provided minorities the ability to elect their candidates of choice
    in the Benchmark Plan, and is therefore protected under the VRA. The elimination of this
    district, without a corresponding offset, was retrogressive.
    All parties agree that CD 25 in the enacted plan is not an ability district. They disagree,
    however, whether Benchmark CD 25 is a protected crossover district. Texas, the United States,1
    and one defendant-intervenor claim that it is not; the remaining intervenors argue that it is. As
    discussed above, we reaffirm the conclusion reached in our summary judgment opinion that
    crossover districts are protected under Section 5, and that proving their existence requires “more
    exacting evidence than would be needed to prove the existence of a majority-minority district,”
    with “discrete data, by way of election returns, to confirm the existence of a voting coalition’s
    electoral power.” Texas v. United States, 
    831 F. Supp. 2d 244
    , 268 (D.D.C. 2011). We conclude
    that the record before the Court demonstrates that minority voters are politically cohesive, have a
    demonstrated history of electoral success, and effectively exert their political power within the
    coalition that elects minority preferred candidates in CD 25. The district is therefore a protected
    ability district in the Benchmark that was lost in the enacted plan.
    In the Benchmark Plan, CD 25 draws a majority of its population from South Austin in
    Travis County, but also includes seven counties southeast of Austin. The total population in the
    1
    The United States does not dispute that minority voters have the ability to elect their preferred candidate in CD 25,
    but explained that, in its view, Section 5 does not apply because racially polarized voting is not present due to the
    number of Anglo crossover votes. While, in dissent, our colleague correctly notes the government does “not argue
    that benchmark CD 25 is a protected district,” CD 25 Dissent, at 1 n.1, the government’s underlying rationale for
    this position is based upon a restricted view of the protection provided by Section 5, which, as discussed infra, we
    reject.
    district is 49.8% Anglo, 38.8% Hispanic, and 8.7% Black. Pl.’s Ex. 11. Anglos constitute
    63.1% of the CVAP in CD 25 while Hispanics make up 25.3% and Blacks 9.1%. 
    Id.
     If Anglos
    voted cohesively in CD 25, they could elect their preferred candidate in every election, and the
    district would be beyond the ambit of Section 5. The Anglo vote in CD 25 is split, however; as
    many as half of Anglo voters cross over to vote with Hispanics and Blacks to elect Democratic
    candidates (a much greater crossover percentage than the statewide average of approximately
    25%). See Defs.’ Ex. 578 (Trial Tr. 1120-21, Perez v. Perry, Sept. 10, 2011). In contrast, the
    Hispanic and Black voters of CD 25 overwhelmingly vote cohesively for the Democratic
    candidates. Defs.’ Ex. 724 (Ansolabehere Rebuttal Report to the Supplemental Report of Dr.
    John Alford, Attach. 3) (“Ansolabehere Reb. Rep.”). For instance, in the 2008 and 2010
    congressional elections, 100% of Black voters cast ballots for Congressman Lloyd Doggett, as
    did over 80% of Hispanic voters. 
    Id.
     Anglo support for Congressman Doggett, however, has
    varied considerably. In 2008, he received 53% of the Anglo vote, but Anglo support dropped to
    37% in the 2010 election. 
    Id.
     The dominant political force in CD 25 is thus described by some
    as a “tri-ethnic coalition” composed of almost all the district’s Black and Hispanic voters, and up
    to half, but as little as 37%, of Anglo crossover voters. See, e.g., Trial Tr. 84-86, Jan. 19, 2012
    PM (Dukes); Ansolabehere Reb. Rep., Attach. 3.
    To determine if a crossover district is protected under Section 5, the Court must assess
    whether minority voters (1) vote cohesively and (2) successfully elect their preferred candidate
    by effectively exerting their political power within the voting coalition.2
    2
    Although our dissenting colleague characterizes this test as “novel” and “divorced from Supreme Court
    precedent,” CD 25 Dissent, at 1, the test we outline above is no more novel than the application of any precedent to
    a unique set of facts, and fully comports with this panel’s conclusion at summary judgment that crossover and
    coalition districts are protected by Section 5 of the VRA, Texas, 831 F. Supp. 2d at 266-68, as well as our reading of
    the Supreme Court precedent regarding protected districts outlined in the Majority Opinion. Majority Op., at 18-25.
    Nevertheless, the dissent argues that the test we delineate “sweeps too wide because it provides no way to
    distinguish between unprotected influence districts, where minority voters play a substantial role, and protected
    2
    The parties do not dispute that minorities in CD 25 combine with some Anglo voters to
    form a “tri-ethnic coalition,” that this coalition votes cohesively in general elections, and that the
    coalition has had considerable success in electing minorities’ candidates of choice.3 For
    example, despite the fact that Anglos comprise 63.1% of the citizen voting age population, “the
    candidate preferred by Blacks and Hispanics [in CD 25] has won every Congressional election
    this decade.” Ansolabehere Reb. Rep., at 5; Pl.’s Ex. 11. Indeed, Texas’s own expert agreed
    that Benchmark CD 25 is a district in which minorities have the ability to elect the candidates of
    their choice in general elections.4 Trial Tr. 21-22, Jan. 25, 2012 AM (Alford); see also Alford
    Dep. 181-82, Jan. 22, 2012.
    crossover districts, in which they have an ability to elect.” CD 25 Dissent, at 2. The “unprotected influence
    districts” referenced by the dissent, however, are districts “where minority voters may not be able to elect a
    candidate of choice but can play a substantial, if not decisive, role in the electoral process.” Georgia v. Ashcroft,
    
    539 U.S. 461
    , 482 (2003). In other words, minority voters in influence districts fall short of exercising sufficient
    power to be a protected district. By contrast, our inquiry under the test we apply focuses on whether minority voters
    are able to “successfully elect their preferred candidate by exerting their political power.” As we make clear, it is
    not enough that minorities exert political power. They must also be successful in electing the candidates of their
    choice. See LaRoque v. Holder, 
    650 F.3d 777
    , 793-94 (D.C. Cir. 2011) (“Essentially overruling Georgia v.
    Ashcroft, Congress added subsections (b) and (d) to section 5, which make clear that the section 5 inquiry should
    focus on whether the proposed change ‘has the purpose of or will have the effect of diminishing the ability of any
    citizens of the United States on account of race or color . . . to elect their preferred candidates of choice.’ 42 U.S.C.
    § 1973c(b)”). As discussed below, and our dissenting colleague concedes, CD 25 Dissent, at 3-4 (stating that “. . .
    there is evidence that a coalition of Black, Hispanic, and some Anglo voters consistently elects minority-preferred
    candidates in CD25 . . . .”), it is undisputed that the tri-ethnic coalition elected Congressman Doggett, the minority
    candidate of choice, in each of the past three elections.
    3
    Given that Hispanic and Black voters in Benchmark CD 25 prefer the same candidate of choice in the general
    election, the Court considers these voters together for purposes of assessing minority voting power. See Majority
    Op., at 5-6 (stating that “[t]he goal of the ‘effect’ prong [of section 5] is ‘to insure that no voting-procedure changes
    would be made that would lead to a retrogression in the position of racial minorities with respect to their effective
    exercise of the electoral franchise,” Beer v. United States, 
    425 U.S. 130
    , 141 (1976) . . . .”) (emphasis added). While
    the dissent queries whether such aggregation is permissible under Section 5, CD 25 Dissent, at 2 n.2, this Court has
    already answered this question by confirming that coalitions formed by minority voters, who have united to elect
    their preferred candidate in a district, are protected under Section 5. Majority Op., at 18-22.
    4
    Although our dissenting colleague faults us for citing Dr. Alford’s expert opinion regarding CD 25 because we
    reject his methodology, Texas bears the burden of proof and its only expert credibly opined, in disagreement with
    his own client, the State of Texas, that Benchmark CD 25 is a district in which minority voters are able to elect the
    candidates of their choice. The dissent states that because “Dr. Alford uses a metric . . . we have emphatically
    rejected[,] [t]here is no reason that his assessment should be legally conclusive for this district, yet no other.” CD 25
    Dissent, at 5 n.5. There are two inaccuracies in this statement. First, Dr. Alford’s conclusion regarding Benchmark
    CD 25 was not based on his rejected metric, but upon the undisputed fact that the minority candidate of choice has
    won all endogenous elections in the district. Trial Tr. 21, Jan. 25, 2012 AM (Dr. Alford responding “yes” to the
    3
    Given that there is no dispute that CD 25’s tri-ethnic coalition votes cohesively and has
    had considerable and proven success in general elections, the only remaining question before the
    Court is whether minorities in CD 25 exert their political power effectively in the tri-ethnic
    coalition, or are rather just “hangers-on” to the choices of Anglo voters.
    A. Minority Groups Effectively Exert Political Power Within the Tri-Ethnic Coalition
    that Elects Minority Preferred Candidates in CD 25
    Texas argues that minorities succeed in CD 25 because Anglos do not vote as a racial
    bloc and some Anglos happen to vote for Democratic candidates, who are preferred by
    minorities. See Texas Post-Trial Brief, ECF No. 201, at 16 (stating that “[t]he demographics for
    the district show why” CD 25 is not a protected district and arguing that it performs for
    minorities because it is a “reliable Democratic district”). Texas’s expert, while agreeing that
    minorities in Benchmark CD 25 have the ability to elect the candidates of their choice, similarly
    asserts that the sole cause of the minority groups’ undisputed success is the partisan makeup of
    the Anglo population in the district. See Pl.’s Ex. 175, at 26-27 (Alford Pre-Filed Direct
    Testimony) (stating that “the key factor at work is partisanship”); Defs.’ Ex. 319, Alford Report,
    at 2 (“Because these ‘tri-ethnic’ coalitions are driven by partisanship, they cannot be easily
    disentangled from partisanship . . . .”). According to this argument, minority voters in CD 25 are
    subject to the whims of Anglo Democrats and have no effective voice in the electoral process.
    This argument is untenable for two reasons. First, the fact that a number of Anglo voters
    share the same political party as minority voters does not remove those minority voters from the
    protections of the VRA. The statute makes clear that this Court must focus on whether
    question: “You would agree . . . that on the benchmark plan Congressional District 25 was a district in which Blacks
    and Hispanics were able to elect the candidates of their choice in general elections, correct?”). In fact, Dr. Alford
    testified that he did not even include CD 25 in his statewide functional analysis. Id. 21-22. Second, we do not deem
    Dr. Alford’s statement on CD 25 to be “legally conclusive,” which is why we discuss the law and evidence
    pertaining to CD 25 at length.
    4
    minorities are able to elect the candidate of their choice, no matter the political party that may
    benefit. Second, as detailed below, the record does not support Texas’s argument concerning the
    political dynamics in CD 25. Both factual and expert testimony establish that Anglos do not
    control the election outcomes in CD 25 and that power is shared equally among Hispanics,
    Blacks, and Anglos in this district, giving minority voters the ability to elect their preferred
    candidates.5
    The record demonstrates that no single group in CD 25’s tri-ethnic coalition is
    sufficiently numerous to elect its candidate alone, but together the coalition consistently wins
    general elections in the district. Contrary to the assertion that Anglo Democrats control the
    district, evidence shows that candidates supported by the minority groups within the tri-ethnic
    coalition are the ones who win. Trial Tr. 104, Jan. 19, 2012 PM (Dukes). For example, Texas
    State House Representative Dawnna Dukes testified that candidates are not able to bypass
    minority voters, and candidates who only obtain endorsements from Anglo groups in the tri-
    ethnic coalition do not win elections. Id. at 106 (Rep. Dukes testifying that “ . . . in general
    elections in Travis County [] if you do not win the Hispanic and African-American boxes that are
    largely located in the central portion of Travis County, then you are not going to win an election
    5
    Writing in dissent, our colleague argues that to draw “the line between protected crossover districts and non-
    protected districts that simply vote Democratic,” a minority group “must lead” a crossover coalition and that “an
    equal voice” in a district’s electoral decisions is not enough. CD 25 Dissent, at 1, 3. For this reason, the dissent is
    critical of the testimony that “could support a conclusion that Anglos do not control CD 25, but [] doesn’t tell us
    anything more.” Id. at 5. This new “leadership” test sets down a hurdle for which we find no basis in the law or
    precedent and, consequently, to which we do not subscribe. Section 5 of the VRA protects “the ability of [minority
    voters] to elect their preferred candidates of choice.” 42 U.S.C. § 1973c(d). This text charges the Court, quite
    simply, with assessing whether minority voters are able effectively to elect their preferred candidates. The Supreme
    Court has never stated that minorities must “lead” a voting coalition, but rather that when minorities “pull, haul, and
    trade” to elect their preferred candidate, the district is one in which minority voters have an ability to elect and
    section 5’s safeguards apply. See Johnson v. De Grandy, 
    512 U.S. 997
    , 1020 (1994); see also Georgia v. Ashcroft,
    
    539 U.S. 461
    , 481 (2003). The Supreme Court’s oft-used description belies an interpretation of Section 5 that
    would require minority voters to “lead,” with the implication that they must eschew any “trade” or compromise in
    power sharing, even though such trading and compromise are a necessary part of the process in a political coalition.
    We decline to adopt a new “leadership test,” as outlined in the dissent, when the text of the statute and Supreme
    Court construction of the law provide no basis for the assertion that minorities are only able to elect their candidate
    of choice if they are “leaders,” as opposed to equal participants in the process of political coalition building.
    5
    in Travis County without the progressive Anglo Black and Hispanic communities. I may not
    have an Excel spreadsheet, but I can tell you I know my county.”).
    Representative Dukes provided specific examples of elections to support her analysis of
    minority groups’ voting power in CD 25. She recalled the 2008 election for Travis County Tax
    Assessor, in which the African-American supported by the coalition successfully defeated, with
    74% of the vote, an Anglo male “progressive Democrat.” Id. at 112. Before the Court is similar
    testimony from David Escamilla, the Travis County Attorney, regarding the power of minority
    voters in the tri-ethnic coalition. See Defs.’ Ex. 735 (Pre-Filed Direct Testimony of David
    Escamilla). Mr. Escamilla not only echoed Representative Dukes’ testimony that Anglos do not
    control the election outcomes in the tri-ethnic coalition, but also provided the example of a race
    in 2008 in which an Anglo Assistant County Attorney lost a race for a county judgeship despite
    having “the lion’s share of endorsements from the local Democratic clubs” because he was
    “unable to gain significant support from the Hispanic or African American community.” Id. at
    9-10.
    The evidence presented to the Court regarding the power of minority voters in the tri-
    ethnic coalition is persuasive, particularly because it is corroborated by the expert analysis
    performed by Dr. Stephen Ansolabehere.6 To assess the relative power of the groups comprising
    the tri-ethnic coalition, Dr. Ansolabehere examined each groups’ success in Democratic primary
    elections in Travis County. In the 43 Travis County primaries he analyzed, the Anglo preferred
    6
    As discussed infra, Dr. Ansolabehere’s analysis could be more comprehensive. His findings are nonetheless
    probative of the voting dynamics within CD 25. Our dissenting colleague believes that some of Dr. Ansolabehere’s
    statistics result in “discrepancies,” CD 25 Dissent, at 7 n.8, but our colleague’s deconstruction of Dr.
    Ansolabehere’s data has not been corroborated by any statistical expert. Regardless, even taking into account any
    alleged “discrepancies,” it is undisputed that Dr. Ansolabehere’s data indicates that Congressman Doggett enjoys
    virtually unanimous support from Black voters and overwhelming support from Hispanic voters.
    6
    candidate won only once without support from the Hispanic and Black communities.7
    Ansolabehere Reb. Rep., Attach. 6. On the other hand, minority preferred candidates won
    twelve elections without the support of Anglo voters. Id. While Texas – as well as the TLRTF –
    argues that Anglo Democrats control the tri-ethnic coalition and drown out minority voters, these
    election results belie that conclusion. To the contrary, Dr. Ansolabehere concludes that
    “[l]ooking across the different patterns of group coalitions reveals that no one group dominates
    the primary process. Power is shared very equally and in such a way that the racial groups
    succeed in nominating their preferred candidates 75 percent of the time.” Id. at 23. By way of
    example, in the 43 primaries Dr. Ansolabehere analyzed, Anglo voters backed the winner in 31
    primaries; Hispanic voters backed the winner in 32 primaries; and Black voters backed the
    winner in 31 primaries. Id., Attach. 6.8
    These statistics support the testimony presented to the Court that the tri-ethnic coalition
    consistently elects candidates in the Democratic primary that appeal to the minority voters in the
    tri-ethnic coalition. Mr. Escamilla effectively described the political cohesion and cooperation
    within the tri-ethnic coalition, which “consistently produces broad agreement to support
    individual candidates and slates of candidates. The high frequency of agreement on candidates
    among the organizations within the Coalition also stems from the fact that many individuals are
    members of more than one of the organizations. This overlap in membership promotes
    7
    These primary election results are cited only to assess the relative power among groups comprising a voting
    coalition, not to assess whether a voting coalition exists. See Majority Op., at 64-66 (noting that groups comprising
    a voting coalition need not vote cohesively at the primary level).
    8
    A second expert also noted that voter turnout in primary elections is generally low in Travis County and CD 25,
    which effectively amplifies the preference of minority voters in Democratic primaries. He explained: “[I]n the low-
    turnout Travis County and CD 25 primaries, minority voters vote almost exclusively in the Democratic election,
    while the Anglo majority in Travis County, and elsewhere in CD 25, splits its vote in the March partisan balloting.
    That means minority voters, especially in Travis County, combining with the minority of Anglos who remain in the
    Democratic primary, are very effective in determining the nominee of their party.” Murray Suppl. Rep., ECF No.
    218, Ex. 1, at 1; see also id. at 5 (“Fewer and fewer Anglos vote in Democratic primaries in the 25th District.”).
    7
    agreement on common slates of political candidates.” Defs.’ Ex. 735, at 7. Indeed, the evidence
    demonstrates that Anglos do not dominate the tri-ethnic coalition that successfully elects
    candidates in CD 25. Rather, the record shows that the views and preferences of minority voters
    in the tri-ethnic coalition are not only necessary but, more importantly for Section 5 analysis,
    regularly prevail in the coalition’s selection of candidates. In our view, as noted, the tri-ethnic
    crossover coalition at work in Benchmark CD 25 reflects equal power-sharing among the
    members of the coalition rather than domination by Anglo voters.
    In addition to the anecdotal and expert evidence of the dynamics within the tri-ethnic
    coalition, there is no greater evidence of the power of minority voters in CD 25 than the
    reelection of Congressman Doggett in 2010. In 2008, 53% of Anglo voters supported
    Congressman Doggett’s successful reelection campaign. In 2010, however, Congressman
    Doggett won reelection despite receiving only 37% of the Anglo vote because 100% of Blacks
    and 86% of Hispanics voted for him. Ansolabehere Reb. Rep., Attach. 3.9 Thus,
    notwithstanding the fact that a large majority of Anglos voted against the minority preferred
    candidate, minority voters effectively exerted their political power (with the aid of a number of
    crossover Anglo voters) to elect the candidate of their choice.
    Texas argues that minority success is solely due to the partisan makeup of the district, but
    the 2010 election alone refutes this conclusion. Indeed, despite Texas’s burden of proof, Texas
    supplies no evidence aside from demographic statistics to support its argument that minority
    9
    Most of the experts agree that such endogenous election results are the most probative evidence of whether a
    minority group or minority coalition has the ability to elect the candidate of choice. See, e.g., Defs.’ Ex. 794, at 3
    (Handley Rebuttal Report) (“[T]he most essential piece of information in determining if a Benchmark district is a
    district that provides minority voters with the ability to elect their candidates of choice is whether minority voters
    have been successful at electing their preferred candidates to the legislative office at issue in the district.”); Defs.’
    Ex. 327, at 4 (Handley Congressional Report); Defs.’ Ex. 724 (Ansolabehere Oct. 21, 2011 Report, at 31). While
    the retrogression expert proffered by Texas disagrees with reliance on endogenous election analysis, as noted
    previously, even he agreed that Benchmark CD 25 is a district in which minorities have the ability to elect the
    candidates of their choice. Trial Tr. 21-22, Jan. 25, 2012 AM (Alford).
    8
    voters do not have power in the tri-ethnic coalition nor does it supply evidence to undercut the
    intervenors’ argument that they do. The intervenors argue that minority voters’ repeated
    electoral success as well as the unrebutted factual and expert testimony regarding equal power-
    sharing among the groups comprising the tri-ethnic coalition is sufficient to establish that
    Benchmark CD 25 is a minority ability district. We agree.
    B. Evidence Discrediting Minority Voting Power is Unpersuasive
    Despite the success of minority voters in electing the candidates of their choice in CD 25,
    unrebutted testimony of elected officials from within this district, and expert evidence
    corroborating the political power of minority voters within the tri-ethnic coalition, two
    arguments are asserted for support of the position that minority voters do not exert political
    power in Benchmark CD 25 and that this district is therefore not eligible for protection as an
    ability district. The United States also takes the position that Benchmark CD 25 is not a
    protected district, but does so on the belief that Section 5 does not apply to CD 25 because Anglo
    voting in the district is not characterized by racial polarization. Each of these arguments merits
    consideration.
    First, the TLRTF discounts the expert evidence presented by Dr. Ansolabehere
    demonstrating the electoral success of minority voters and their power within the tri-ethnic
    coalition because this evidence relies on information from Travis County, as opposed to all of the
    counties that comprise CD 25 in the Benchmark plan. TLRTF’s Resp. to the Ct.’s Order of Mar.
    6, 2012, ECF No. 219, at 7.10 We acknowledge that, as with other experts in this case, the
    analysis by Dr. Ansolabehere does not cover all possible useful data. Nonetheless, the Court
    finds evidence of the tri-ethnic coalition’s performance in Travis County probative of its analysis
    10
    Our dissenting colleague also cites this focus on Travis County as a weakness in the expert analysis and
    testimony. CD 25 Dissent, at 5-6.
    9
    of whether minorities in CD 25 have the ability to elect the candidates of their choice. As an
    initial matter, there is only one endogenous election within CD 25 as a whole: the election for a
    representative to the U.S. Congress, which Congressman Doggett has won since the district’s
    initial formation. Thus, in order to measure the effectiveness and power of minorities in the tri-
    ethnic coalition that elects Congressman Doggett, one must necessarily look to the performance
    of the coalition in other political subdivisions, such as in Travis County. The portion of CD 25
    that encompasses Travis County not only comprises a significant majority of the population of
    CD 25 (59.7%), but also contains a large majority of the district’s minority population (66%).
    See Pl.’s Ex. 11. The voting dynamics in the district’s most populous county have a significant
    impact on the voting dynamics in the rest of the district. The minority population’s other
    successes in Travis County are therefore significant in assessing its power and influence within
    the crossover coalition. 11
    Second, like Texas, the TLRTF contends that Benchmark CD 25 is not an ability district
    because Anglo voters dominate the electoral outcomes in CD 25. As the prior discussion
    reveals, this argument is factually wrong. It is also based upon faux data. In support of its view
    of the relative voting power of minority versus Anglo voters, the TLRTF cites two different sets
    of data supplied by the OAG: racially polarized voter turnout estimates and exogenous election
    results in statewide Democratic primaries. Prior to discussing the reliability of these datasets,
    11
    No party, including Texas, presented any evidence regarding the tri-ethnic coalition’s performance in the six
    smaller counties wholly contained in Benchmark CD 25. Our colleague states that we are “mistaken” on this point,
    and writes that “[w]e received evidence indicating that the tri-ethnic coalition was ineffective in these counties in
    2010.” CD 25 Dissent, at 6 n.6. The exhibit to which he cites, however, is a 206-page table of election results,
    which indicates that the Democratic candidate lost in the elections he references. See id.; Pl.’s Ex. 34. This table
    and the election results our colleague discovered on the Internet, see CD 25 Dissent, at 6 n.6, indicate that the other
    six counties wholly contained in Benchmark CD 25 vote overwhelmingly Republican. Id. (concluding, based on an
    analysis of aggregate data, that “the tri-ethnic coalition prevailed in only three of one hundred and twenty elections
    held in these counties in 2010”). This confirms that at least the majority of voters in these counties are not part of
    the tri-ethnic coalition, and thus do not affect the voting dynamics within the tri-ethnic coalition, which is the focus
    of our inquiry. It further indicates that the tri-ethnic coalition is able to prevail in endogenous elections in
    Benchmark CD 25 despite the fact that most of the voters in these six smaller counties do not vote for the minority
    preferred candidate.
    10
    the Court briefly reviews the peculiar manner in which the TLRTF first raised its arguments to
    the Court.
    Over three weeks after trial and following submission of the parties’ proposed findings of
    fact and post-trial briefs, the TLRTF argued for the first time in an “advisory” that the Court
    should not count Benchmark CD 25 as a protected district because Anglo voters “dominate the
    Democratic primary.”12 Advisory of Def. Intervenor TLRTF, ECF No. 210, at 3; see also
    TLRTF Resp. to Gonzales Intervenors’ Brief Regarding CD 25, ECF No. 223, at 2 n.3 (TLRTF
    concedes that prior to filing its advisory, this intervenor had “never previously ‘suggest[ed] to
    the Court that CD 25 was not a minority ability district[.]’”). As support for its blanket statement
    that Anglos “dominate” Democratic primaries in CD 25, the TLRTF cited tables of exogenous
    election results from statewide primary and general elections in four years (2002, 2006, 2008 and
    2010). Advisory of Def. Intervenor TLRTF, ECF No. 210, at 3 n.10 (citing Defs.’ Exs. 437,
    439-41). Since the tables of election results did not identify the minority candidates of choice,
    the exhibits did not corroborate TLRTF’s statement, prompting the Court to issue a Minute
    Order directing the TLRTF to provide a “fuller explication of its reasoning for and the evidence
    behind its conclusion.” Minute Order dated Mar. 6, 2012.
    In response to the Court’s Order, the TLRTF argued for the first time that Anglos often
    cast a majority of votes in primary elections. 13 TLRTF’s Resp. to the Ct.’s Order of Mar. 6,
    12
    According to the other intervenors, when the U.S. District Court in the Western District of Texas initially adopted
    a congressional map that preserved CD 25, the TLRTF “defended that map in Texas’s appeal to the Supreme Court,
    never suggesting to the Court that CD 25 was not a minority-ability district . . . .” Resp. of Certain Def. Intervenors
    to TLRTF’s Briefing Relating to CD 25, ECF No. 221, at 3.
    13
    The TLRTF also urges the Court to look to a second OAG dataset of exogenous election results in CD 25 for
    statewide Democratic primaries, which the TLRTF interprets as showing that Hispanic candidates of choice only
    prevail in three out of nine primary elections. TLRTF’s Resp. to the Ct.’s Order of Mar. 6, 2012, ECF No. 219 at
    12-13; TLRTF’s Resp. to Gonzales Intervenors’ Brief Regarding CD 25, ECF No. 223 at 13 (“Latino candidates
    won only three out of the nine Democratic Primary elections”) (emphasis in original). Other intervenors dispute the
    TLRTF’s interpretation of this data and argue that the data shows that minority preferred candidates prevailed in
    11
    2012, ECF No. 219, at 13. The Court agrees with the remaining intervenors, however, that the
    data presented by the TLRTF to support this argument is not persuasive. 14
    According to the TLRTF and our dissenting colleague, the OAG turnout estimates for
    certain selected elections in four elections cycles between 2002 and 2010 indicate that Anglo
    voters cast the majority of votes in both the Democratic primary and general elections in CD
    25.15 Id.; see also TLRTF Resp. to Gonzales Intervenors’ Brief Regarding CD 25, ECF No. 223,
    at 1-2 (arguing that “Anglo voter preferences drive the outcome of both the Primary and General
    Election”). These turnout estimates, however, were appropriately criticized by Dr. Alford
    because they are unreliable on their face. Dr. Alford explained:
    . . . if you’ll take a quick look at the last two columns [of the data] I think you’ll
    agree with me there’s very little reason to put any faith in this particular analysis.
    I don’t put any faith in the analysis. I’ve not relied on the analysis. Precisely
    what you’re about to talk about here, because of a variety of technical things, we
    don’t need to discuss. I mean, look at the general election in 2004. This model
    estimated that the turnout was 26 percent. The actual turnout in the election was
    40.8 percent. The error in this model is enormous, and it’s increased when we try
    to estimate the increase in categories. I don’t rely on this.
    “six of eight primaries.” Resp. of Certain Def. Intervenors to TLRTF’s Briefing Relating to CD 25, ECF No. 221, at
    5. Resolving this dispute, which the TLRTF raised for the first time in a post-trial brief, is unnecessary since the
    Court finds that exogenous primary evidence is not probative to assess whether a voting coalition exists or to
    measure the effectiveness of minority voters in CD 25. In any event, exogenous primary election results would not
    rebut the testimonial and expert evidence demonstrating that minority voters in CD 25 have fulfilled their
    “obligation to pull, haul, and trade to find common political ground” and achieve electoral success. Georgia v.
    Ashcroft, 
    539 U.S. at 481
     (quoting De Grandy, 
    512 U.S. at 1020
    ).
    14
    The timing and weak evidentiary basis for TLRTF’s belated “advisory” suggest that tactical considerations were
    at play. See generally Resp. of Certain Intervenors to TLRTF’s Briefing Relating to CD 25, ECF No. 221, at 3
    (implying that the “Task Force has now created that argument in an attempt to justify the deal it cut with Texas” in
    the interim map-drawing process in the U.S. District Court for the Western District of Texas). Indeed, the TLRTF
    uses this Court as a forum to contend that a new Latino-majority district in CD 35 in Central Texas may properly
    encompass portions of Travis County, a matter that is not at issue before this Court. Advisory of Def. Intervenor
    TLRTF, ECF No. 210, at 3; TLRTF’s Resp. to the Ct.’s Order of Mar. 6, 2012, ECF No. 219, at 2. Whether CD 25
    is a protected district only has a bearing on the key issue of the retrogressive impact of the enacted plan, and not the
    location or boundaries of any offsetting new ability district.
    15
    The TLRTF specifically referenced turnout in the 2002 Democratic primary race for Governor; the 2006
    Democratic primary for Lt. Governor; the 2008 Democratic primary race for U.S. Senator; and the 2010 Democratic
    primary race for Lt. Governor. TLRTF’s Resp. to the Ct.’s Order of Mar. 6, 2012, ECF No. 219, at 13.
    12
    Trial Tr. 86-87, Jan. 24, 2012 PM (Alford). In other words, the predictions of voter turnout
    produced using the methodology employed by the OAG relied on assumptions that differed
    drastically from what occurred in real life. Dr. Alford’s comments pertained to the OAG turnout
    data for elections for the State House, but there is no indication that the OAG’s methodology
    differed when compiling data for congressional elections or that Texas’s expert’s views of the
    unreliability of this data would be any different for congressional elections. Indeed, examination
    of the turnout predictions for congressional elections reveals that the error rate Dr. Alford
    referenced is equally as large, if not greater, than in the State House portion of this dataset.16
    For example, the statistical model used to produce the OAG turnout data projected that
    turnout in CD 25 in the 2010 general election was 74.6%, when the actual turnout was 31%.
    Pl.’s Ex. 24, at 577. In the 2008 general election, the OAG projected turnout was 100%, but the
    actual turnout was 48.6%. 
    Id.
     In 2006, the OAG projected turnout was 65.7% when the actual
    turnout was 28.1%. 
    Id.
     In short, the projected congressional turnout data contains enormous
    error rates, similar to that found in the turnout estimates for the State House, and is subject to the
    same criticism of unreliability.17 Our dissenting colleague states that he relies on “different data”
    16
    The dissent states that Dr. Alford rejected the turnout data because he found significant discrepancies in the “‘last
    two columns [of the data],’” CD 25 Dissent, at 9 n.11, while we find discrepancies in other parts, or columns, of this
    data, an observation that leads our colleague to conclude that “Dr. Alford’s concern is not the same as” ours. This is
    simply wrong. Dr. Alford pointed out blatant error rates in certain columns of the dataset merely as examples of the
    significant issues with relying on this data, and made clear that there were a “variety” of issues with this data. Trial
    Tr. 86-87, Jan. 24, 2012 PM (Alford) (“I’ve not relied on the analysis . . . because of a variety of technical things,
    we don’t need to discuss.”) (emphasis supplied). Thus, when our dissenting colleague “use[s] that same metric
    [namely, the difference between the real and projected voter turnout] —the only ground Dr. Alford gave as support
    for his critique — to test the data,” CD 25 Dissent, at 9 n.11, he may be missing the remaining “variety of technical
    things” referenced by Dr. Alford that were not fully developed in the record by the experts and that make this data
    insufficiently reliable for even Texas’ own expert to rely upon.
    17
    Notably, at no point did Texas cite to the OAG turnout data in response to the intervenors’ argument that CD 25
    was a protected ability district in the Benchmark Plan. The only time any party cited to this data in regards to CD 25
    was in the post-trial submission filed by the TLRTF in response to a show cause order issued by the Court. See
    Minute Order dated Mar. 6, 2012; TLRTF’s Resp. to the Ct.’s Order of Mar. 6, 2012, ECF No. 219. Even then, the
    TLRTF cited to turnout data for primary elections. If the OAG’s turnout data were reliable and minority voter
    turnout was 10% in CD 25, CD 25 Dissent, at 9, the Court’s inquiry into whether CD 25 is a protected district in the
    Benchmark plan would meet a swift end. See Majority Op., at 24 (noting that minority voters who make up 10% of
    13
    from the OAG than the one we discuss, but it appears that all of the OAG turnout projections
    were produced using the same methodology.18 Dr. Alford stated that error rate in the “model”
    used by the OAG led him to completely disregard that data. Trial Tr. 86-87, Jan. 24, 2012 PM
    (Alford). Indeed, none of the experts in this case appeared to rely on OAG turnout projections,
    in any of its forms.19 The dissent expends much energy attempting to demonstrate that the OAG
    turnout data could be accurate, and even “extrapolat[es] [] missing turnout data” from Dr.
    Ansolabehere’s analysis as an alternative, but, quite frankly, our colleague’s own calculations of
    turnout data are defective.20 While reliable and accurate data regarding voter turnout by racial
    group in Benchmark CD 25 would be highly probative, the Court is not presented with such
    evidence and will not endeavor to create its own.21 Consequently, we do not rely on such data
    a population and provide a margin of victory to Anglo Democratic voters could not be protected under Section 5). If
    the OAG turnout data had sufficient reliability to have any probative value, Texas’s failure to rely on this data would
    be inexplicable. Rather, the explanation lies in the fact that this data is simply so unreliable as to be irrelevant.
    18
    Our colleague states in dissent that unlike the OAG dataset criticized by Dr. Alford, the OAG dataset upon which
    the dissent relies appears to “accurately predict[] the overall turnout in a given election.” CD 25 Dissent, at 9 n.11.
    This may be so, but the projections relevant to the Court’s analysis are those pertaining to the number of votes cast
    by minority groups. There is no testimony, expert or otherwise, in the record that the data on which the dissent
    relies is not as flawed as the turnout numbers rejected by Dr. Alford. Texas’s failure to cite to this data again
    indicates to the Court that it has little probative value.
    19
    Our dissenting colleague fails to acknowledge that none of the experts in this case appears to rely on OAG
    turnout projections. He also dismisses, without explanation, the fact that Texas – the party that compiled and
    calculated the turnout projections – never relied upon this data.
    20
    For example, our colleague acknowledges that his calculation that minorities cast only 19% of the vote for
    Congressman Doggett is “imperfect” because “relative turnout among minority groups . . . could have changed
    between elections.” CD 25 Dissent, at 8 n.10. Overall turnout changed dramatically between the 2008 and 2010
    elections. In 2008, 291,296 voters voted in the election for the U.S. Congress in Benchmark CD 25. Pl.’s Ex. 31, at
    10. In 2010, voter turnout dropped over 33%, by more than 100,000 votes, to 189,247. Pl.’s Ex. 32, at 13. The
    dissent’s assumption that turnout among racial groups remained constant as a percentage of the overall voter turnout
    is unsupported speculation. Given the number of variables affecting voter turnout generally and the complexity of
    predicting turnout on a demographic basis, none of the experts in this case – including the one who compiled the
    data used by our colleague to compute his 19% figure –apparently viewed such predictions as sufficiently reliable to
    offer an opinion.
    21
    We disagree with our dissenting colleague’s assertion that turnout data is the only way to provide a “context” for
    the expert testimony before the Court. CD 25 Dissent, at 9 n.11. In the absence of reliable turnout projections,
    unrebutted witness testimony and endogenous election results are sufficient to corroborate expert analysis and to
    provide a “context” for such evidence. See Texas, 831 F. Supp. 2d at 268 (recognizing crossover and coalition
    districts may be ability districts based upon “discrete data, by way of election returns, to confirm the existence of a
    14
    and give neither the OAG turnout projections nor our colleague’s extrapolated calculations
    weight or further consideration. In its place, we rely on the most probative evidence presented to
    the Court: testimony from elected officials, endogenous election results, and expert analysis.
    C. The United States’ Position that CD 25 is Not Protected because of the Absence of
    Racially Polarized Voting Among Anglo Voters is Incorrect
    The United States does not argue that CD 25 is a protected district in the Benchmark Plan
    and has remained generally silent as Texas and the intervenors argue over the district’s status
    under Section 5. At closing argument, in response to a direct question posed by the Court, the
    United States clarified its position that while minority voters have an ability to elect the
    candidate of their choice in CD 25, it believes that CD 25 is not protected by Section 5 because
    Anglo voting in the district is not racially polarized. Trial Tr. 82-85, Jan. 31, 2012 AM.22
    The Court is presented with four distinct positions regarding racial polarization in CD 25:
    The United States and certain intervenors posit that Anglo voters in CD 25 are not racially
    polarized, but that finding compels the government to conclude that no Section 5 protection
    applies,23 while the intervenors reach the opposite conclusion. According to these intervenors,
    the lack of racially polarized voting among Anglos is irrelevant to the Court’s Section 5 analysis.
    See, e.g., Trial Tr. 34-35, Jan. 31, 2012 PM (Gonzales intervenors’ counsel acknowledging that
    voting coalition's electoral power,” citing as example “evidence that a coalition had historical success in electing its
    candidates of choice”).
    22
    Counsel for the Department of Justice explained as follows: “Congressional District 25 does perform . . . the
    issues there . . . has to do with polarized voting . . . we aren’t finding the polarized voting in that area . . . . [T]here is
    polarized voting in most of Texas . . . . The question is regarding this area around Travis County area, and the
    success where it’s not crossover anymore if 52 percent of the . . . Anglo voters are voting the same as the [B]lack
    voters and the same as the Hispanic voters or whatever the percent may be depending upon which election contest
    you’re looking at. So at a certain level again, is the protections that flow do deal with the realities of polarized
    voting and whether or not there exists polarized voting.” Trial Tr. 84-85, Jan. 31, 2012 AM.
    23
    The Department does not indicate the threshold number of Anglo cross-over votes that would remove from the
    protection of Section 5 an otherwise minority ability district. In any event, rather than focus on the single
    demographic statistic of the number of Anglo cross-over votes, a functional test must be applied to assess whether
    minority voters effectively exercise power to elect their candidate of choice in a cross-over coalition.
    15
    voters in Travis County are “colorblind,” but stating that “[t]he only way the Voting Rights Act
    could not protect that district . . . is if somehow, there had been a bailout. . . . In fact, Section 5
    covers all of Texas as a matter of law, and so Travis County, CD 25, is covered, just like every
    other jurisdiction in Texas.”). By contrast to those two positions, the TLRTF argues that voting
    is polarized in all elections. TLRTF Resp. to Gonzales Intervenors’ Brief Regarding CD 25,
    ECF No. 223, at 10. Finally, Texas appears to take the position that Anglo voting is not racially
    polarized in general elections, but asserts that CD 25 should not be protected because racial
    polarization is present among minority voters in Democratic primary elections. Trial Tr. 131-35,
    Jan. 19, 2012 PM. The Court need not determine which of these views is ultimately correct
    because, regardless, CD 25 is a protected district under Section 5.
    The Court agrees with the Gonzales intervenors that Section 5 covers all of Texas as a
    matter of law and Travis County, including CD 25, is covered, just like every other jurisdiction
    in Texas. The position of the United States would have the anomalous consequence that once
    minorities successfully elect their candidates of choice in a cross-over district, Section 5 would
    no longer apply. 24 That is not the law. Such an ability district remains protected by the VRA
    and, if it is eliminated as an ability district, it must be offset, which CD 25 is not. This loss of
    Benchmark CD 25 as an ability district, without the creation of any new ability district, renders
    the enacted Congressional plan retrogressive.
    24
    The United States’ position is that the protections of Section 5 need not apply to CD 25 because of the presence of
    crossover Anglo voters. The presence of these crossover Anglo voters does not sufficiently protect minority voters
    in CD 25, but, in fact, may create a ripe target for actors in other parts of the state to retrogress minority voting
    power in CD 25 by fracturing the district and submerging its pieces in areas where race-based voting remains
    prevalent. Indeed, the enacted plan divides Travis County into five different districts and as a result, “[t]his is the
    only county in which the population exceeds the number required to constitute a [congressional] district, but the
    county is not the seat of any single district.” Ansolabehere Report on Minority and White Representation and
    Voting Patterns in the Texas Congressional District Plan C185 at 47, Perez v. Perry, Aug. 8, 2011, ECF No. 123-1.
    16
    GRIFFITH, Circuit Judge, dissenting with respect to retrogression in Congressional District 25:
    I, too, reaffirm our decision at summary judgment that crossover districts are protected
    under section 5, and I agree that enacted CD 25 is not an ability district. But I cannot join in my
    colleagues’ proposed test for the existence of a crossover district, which is divorced from
    Supreme Court precedent and dangerously broad. I first explain why the test to find a protected
    crossover district is more demanding than that my colleagues employ. Then I show that even
    under their standard, the record does not contain the “more exacting evidence” needed to show
    that benchmark CD 25 is a crossover district.1 Texas v. United States, 
    831 F. Supp. 2d 244
    , 268
    (D.D.C. 2011).
    As my colleagues’ analysis shows, Blacks and Hispanics vote cohesively in CD 25, and
    their support is necessary to victory. But as we have already agreed, these factors alone are not
    enough to show that minority voters can effectively exercise their electoral power to elect their
    preferred candidates. My colleagues and I agree that section 5 does not protect every district in
    which “Anglos and minorities vote together to elect a candidate,” or “that elects a Democratic
    candidate no matter how small its minority population.” Majority Op. at 25, 24. We disagree
    over where section 5 draws the line between protected crossover districts and nonprotected
    districts that simply vote Democratic.
    My colleagues hold that a district is protected when minority voters “effectively exert[]
    their political power within the voting coalition.” Under this novel rephrasing of “ability to
    elect,” they establish a false dichotomy, testing “whether minorities in CD 25 exert their political
    1
    As my colleagues note, Texas and the Texas Latino Redistricting Task Force argue that benchmark CD 25 is
    not an ability district under section 5. In addition, the United States and its expert, Dr. Handley, do not argue that
    benchmark CD 25 is a protected district. Indeed, the United States explicitly noted in its post-trial briefing that, in its
    view, retrogression in the Congressional Plan is based on the failure to add an additional ability district, “not on
    a . . . determination that benchmark [CD] 25 provides minority voters with the ability to elect preferred candidates of
    choice.” U.S. Post-Trial Br. 16 n.9.
    power effectively in the tri-ethnic coalition, or are rather just ‘hangers-on’ to the choices of
    Anglo voters.” CD 25 Majority Op. at 4. Although my colleagues do not provide a full definition
    of what it means for minority voters to “effectively exert[] their political power,” it appears that
    they view anything more than “hanging on” as sufficient to prove that the district is protected.
    But this overbroad result runs headlong into the 2006 amendments to the VRA. As we noted at
    summary judgment, Texas, 831 F. Supp. 2d at 251, Congress amended the VRA to make clear
    that section 5’s retrogression prong did not include “influence districts” — ones in which
    minorities play a “substantial, if not decisive, role in the electoral process,” Georgia v. Ashcroft,
    
    539 U.S. 461
    , 482 (2003); see also LaRoque v. Holder, 
    650 F.3d 777
    , 793-94 (D.C. Cir. 2011).
    The majority’s “effectively exert” test sweeps too wide because it provides no way to distinguish
    between unprotected influence districts, where minority voters play a substantial role, and
    protected crossover districts, in which they have an ability to elect.
    The line between influence and protected crossover districts2 is admittedly difficult to
    draw. But Supreme Court precedent — which my colleagues do not cite as support for their
    “effectively exert” test3 — helps us at least to sketch its location, and CD 25 falls on the
    unprotected side. Whenever the Court has examined crossover districts, it has spoken of Anglo
    voters providing supplemental support to minority voters. See Bartlett v. Strickland, 
    556 U.S. 1
    ,
    2
    CD 25 is, in fact, a combination of a potential coalition district (because Blacks and Hispanics band together)
    and a potential crossover district (because that joint minority group combines with Anglos to elect its candidate of
    choice). Even if there were only one minority group in the district, however, my analysis would yield the same
    result. If CD 25 were 35% Hispanic, for example, I would still conclude based on this record that it was not an
    ability district. For that reason, I do not assess the possible impact that a multi-ethnic coalition within a crossover
    district might have on the ability-to-elect inquiry. My colleagues, who hold the district is protected, do not address
    this issue either. Rather, they treat the Black and Hispanic communities as a single minority group for purposes of
    their crossover district analysis, with no explanation why such aggregation is permissible under section 5.
    3
    My colleagues note that the VRA “charges the Court, quite simply, with assessing whether minority voters
    are able effectively to elect their preferred candidates.” CD 25 Majority Op. at 5 n.5. But the majority’s “effectively
    exert” test, just like the statute’s “ability to elect” language, is not self-defining. As we noted above, the Supreme
    Court has never directly addressed the test to determine ability to elect in the context of crossover districts. Majority
    Op. at 19-20. Nevertheless, to the extent the Court has spoken to the issue in previous cases, we must look to those
    precedents for guidance.
    2
    13 (2009) (plurality opinion) (defining a crossover district as one in which the minority group
    can “elect the candidate of its choice with help from voters who are members of the majority and
    who cross over to support the minority’s preferred candidate” (emphasis added)); Voinovich v.
    Quilter, 
    507 U.S. 146
    , 158 (1993) (describing a crossover district as one in which minority
    voters can elect “their candidate of choice with the assistance of crossover votes from the white
    majority” (emphasis added)). The Court’s language reflects its assumption that minority voters
    take the leadership role in a crossover district, with Anglo voters providing necessary — but
    ultimately secondary — support. Likewise, the Court’s use of vivid, active phrases to describe
    the part minority voters play in a crossover district suggests a leading role. The Court has stated
    that minority voters must “attract[] sufficient cross-over votes from white voters,” Voinovich,
    507 U.S. at 154 (emphasis added), and “pull, haul, and trade” to elect their preferred candidates,
    Johnson v. De Grandy, 
    512 U.S. 997
    , 1020 (1994) (emphases added). This is the line we must
    draw: the minority group must lead in order to have the ability to elect. The leadership needed to
    prove ability can be demonstrated in a variety of ways, such as by consistently casting the
    majority of votes for the winning candidate in most elections, coordinating get-out-the-vote
    drives, or recruiting the lion’s share of candidates. It makes little difference how that leadership
    is asserted. What is crucial is that minority voters do more than provide the margin of victory or
    have simply an equal voice in a district’s electoral decisions.4
    Such a showing is especially important in a district like CD 25, where minority voters
    comprise only 35% of the Citizen Voting Age Population (CVAP). Although there is evidence
    4
    My colleagues argue that leadership requires that minority voters “eschew any ‘trade’ or compromise in
    power sharing,” CD 25 Majority Op. at 5 n.3, but leaders can (and good leaders often do) trade at times without
    relinquishing their position at the head of a coalition. Trading alone, however, is not enough; minority voters must
    also “pull” and “haul.” The Supreme Court case my colleagues and I both cite for this point reads “pull, haul, and
    trade,” De Grandy, 
    512 U.S. at 1020
     (emphasis added) — and both “pull” and “haul” imply taking the lead. My
    colleagues’ critique isolates “trade” both from its immediate context and from the balance of Supreme Court
    precedent, which supports a test that requires minority voters to take a more active role in a coalition than simply
    being “effective.”
    3
    that a coalition of Black, Hispanic, and some Anglo voters consistently elects minority-preferred
    candidates in CD 25, there is none showing that minority voters lead the effort. For example, no
    testimony was presented that they play a critical role in recruiting the candidates who run in CD
    25 (in contrast to Senator Davis’s uncontroverted testimony about SD 10), are instrumental in the
    coalition’s efforts to get out the vote (in contrast to the Asian-American community in HD 149),
    or that minorities consistently make up the majority of a winning candidate’s votes. The record
    shows only that the minority-preferred candidate wins consistently in CD 25, but that fact alone
    tells us little (and perhaps nothing) about who is responsible for engineering these wins.
    Because there is no evidence that minorities lead in CD 25, I would stop my analysis here
    and find that it is not protected. But even assuming my colleagues’ test is correct, and that a
    district in which power is shared equally would satisfy such a test, there is insufficient evidence
    in the record to support a finding of equal electoral power. First, I address the anecdotal evidence
    regarding CD 25; then I turn to the expert and statistical evidence about the district’s voter
    turnout and electoral results.
    My colleagues place much weight on the anecdotal testimony of one of Travis County’s
    State House representatives, Dawnna Dukes, and Travis County Attorney David Escamilla
    regarding the tri-ethnic coalition in Travis County. I am not confident their testimony can bear
    this weight. First, evidence about Travis County voting patterns does not adequately describe
    minority voting power within CD 25 because, as discussed in more detail below, less than half of
    Travis County is in CD 25, and approximately 40% of CD 25 lies outside Travis County. More
    importantly, this testimony — at best — only indicates that minority votes are needed to win, not
    that minority voters have an equal role in the coalition. The testimony boils down to this: to win
    local elections in Travis County, a candidate must have the support of Black and Hispanic voters.
    4
    For instance, Rep. Dukes testified that candidates running in Travis County cannot win without
    support from “the progressive Anglo, Black, and Hispanic communities.” Trial Tr. 106:10-18,
    Jan. 19, 2012 PM. Escamilla testified about a candidate who lost the primary because he was
    “unable to gain significant support from the Hispanic or African American community.” Defs.’
    Ex. 735, Pre-Filed Direct Test. of David Escamilla 9-10. My colleagues are surely right that this
    testimony could support a conclusion that Anglos do not control CD 25, but it doesn’t tell us
    anything more. The testimony of Dukes and Escamilla simply doesn’t address the critical issue:
    do minority voters in Travis County play some role beyond providing votes necessary to win?
    Minority voters may have veto power in Travis County, but the same is true whenever a minority
    group, however small, consistently provides the margin of victory.
    Neither do the expert analysis nor statistical data show that CD 25 is a protected district.5
    My colleagues place much weight on the Travis County primary election results analyzed by Dr.
    Ansolabehere. But as noted above, Travis County is not CD 25. Travis County contains only
    59.7% of CD 25’s population, even though it has a larger portion (66%) of the district’s minority
    population. See Pl.’s Ex. 11, at 7. The remaining 40.3% of CD 25 — which votes Republican —
    5
    As a preliminary matter, my colleagues place much faith in Dr. Alford’s statement that benchmark CD 25 is a
    district in which minorities have an ability to elect. But as we have already explained at length, Majority Op. at 14-
    19, Dr. Alford uses a metric that determines ability to elect by degree — a metric we have emphatically rejected.
    There is no reason that his assessment should be legally conclusive for this district, yet no other. My colleagues
    respond that the burden of proof is on Texas, and “its only expert credibly opined . . . that Benchmark CD 25 is an
    ability district.” CD 25 Majority Op. at 3 n.4. But Texas did not concede benchmark CD 25’s ability status. See, e.g.,
    Pl.’s Mem. Concerning Congressional District 25, at 1 (“[U]nless all Democratic districts are ipso facto ability
    districts, no minority group in benchmark CD 25 had the ability to elect candidates of their choice.”). That Texas’s
    expert uses the word “ability” in his assessment of benchmark CD 25 does not mean that he was offering a legal
    opinion on its protected status, properly defined, contrary to the State’s position. As we agreed in the opinion, Dr.
    Alford has a different view of an “ability district” than that called for in section 5. In fact, he stated that if “the 25 th
    District is a protected district, then it is hard to see how any other majority Democratic district, assuming it had at
    least one eligible minority resident, would not also be a protected district.” Pl.’s Ex. 175, Pre-Filed Direct Test. of
    Dr. John Alford 28-29. And we did not rely on Dr. Alford’s similar concession that SD 10, under his metric, was an
    ability district. Trial Tr. 39:5-21, Jan. 25, 2012 AM.
    5
    does not figure into Dr. Ansolabehere’s calculations at all.6 Even more troubling, over half of
    Travis County lies outside CD 25, but is nonetheless included in the analysis. 
    Id.
     Dr.
    Ansolabehere’s data set is thus both over- and under-inclusive in the extreme. This is particularly
    problematic because we have been provided with no explanation why it is appropriate to draw
    conclusions about CD 25 from voting data drawn from only a subset of the relevant population
    together with voters from a different district entirely.7 This is not the type of “more exacting
    evidence” necessary to prove a crossover district. We must consider the district’s ability status,
    not Travis County’s.
    Even assuming Travis County can stand in for CD 25, this primary data still does not
    show that minority voters themselves have an ability to elect in the district. Dr. Ansolabehere’s
    report shows that the Anglo-preferred candidate won only one primary without support from the
    Hispanic and Black communities in the contests he analyzed, but that the Hispanic- and Black-
    preferred candidates won twelve elections without Anglo support. And the vast majority of the
    time — in 31 out of 43 primaries — the prevailing candidate had support from the Black,
    6
    In contrast, the smaller counties of Caldwell, Colorado, Fayette, Gonzales, Hays, and Lavaca are all wholly
    contained within CD 25 and are not addressed in either the anecdotal testimony or Dr. Ansolabehere’s analysis. My
    colleagues state that “[n]o party, including Texas, presented any evidence regarding the tri-ethnic coalition’s
    performance in the six smaller counties wholly contained in Benchmark CD 25.” CD 25 Majority Op. at 10 n.11.
    They are mistaken. We received evidence indicating that the tri-ethnic coalition was ineffective in these counties. In
    2010, Republican candidates won (and the tri-ethnic coalition lost) all eighteen elections held within Gonzales and
    Lavaca Counties. Pl.’s Ex. 34, at 49-52, 189-92. The tri-ethnic coalition fared little better in Hays County, where
    Democrats won only one of twenty-two elections, and Caldwell County, where Democrats won only two of
    eighteen. Id. at 165-68. According to the Texas Secretary of State, the tri-ethnic coalition lost all twenty-three
    elections in Colorado County in 2010 and all nineteen elections in Fayette. See Historical Election Results, TEX.
    SEC’Y OF STATE, http://www.sos.state.tx.us/elections/ historical/index.shtml (last visited Aug. 16, 2012) (official
    website of the Texas Secretary of State listing past election results). Thus, the tri-ethnic coalition prevailed in only
    three of one hundred and twenty elections held in these counties in 2010. My colleagues would have us disregard
    this data because “the majority of voters in these counties are not part of the tri-ethnic coalition,” CD 25 Majority
    Op. at 10 n.11. But nowhere else do we examine only a subset of a district to determine the district’s ability status.
    To determine voting dynamics in CD 25, we must examine CD 25.
    7
    My colleagues concede that Dr. Ansolabehere does not “cover all possible useful data,” but they argue that
    Travis County data is useful nonetheless because “one must necessarily look to the performance of the coalition in
    other subdivisions, such as in Travis County.” CD 25 Majority Op. at 9-10. But it is a far jump from useful to
    conclusive. My colleagues give no indication, for example, why such an analysis would not include even a passing
    glance at the six “other subdivisions” that are wholly contained within CD 25.
    6
    Hispanic, and Anglo communities. More importantly, the conclusion he drew from this evidence
    is only that “[p]ower is shared very equally” in Travis County, Defs.’ Ex. 724, Expert Witness
    Report of Dr. Stephen Ansolabehere 105-06 [hereinafter Ansolabehere Rep.], not that minorities
    lead the way. At best, this shows that all members of the coalition play a vital role at the primary
    level in Travis County. Even taking this conclusion as true, evidence that power is shared equally
    does not show that minority voters are at the helm, and thus that they themselves have an ability
    to elect in CD 25.
    The final piece of evidence my colleagues marshal — and the only one that concerns CD
    25 as a whole — is also from Dr. Ansolabehere. His report considers the breakdown of votes by
    racial and ethnic group for Representative Lloyd Doggett, the minority candidate of choice in
    CD 25. Even taking Dr. Ansolabehere’s calculations as accurate,8 this evidence is still
    insufficient to conclude that CD 25 is a crossover district. Dr. Ansolabehere calculates that
    Doggett won in 2008 with 53% of the Anglo vote, 83% of the Hispanic vote, and 100% of the
    Black vote in CD 25. Ansolabehere Rep. attach. 3. In 2010, he calculates that Doggett won with
    37% of the Anglo vote, 86% of the Hispanic vote, and the entire Black vote. Id. At first blush,
    this seems persuasive. With the support of a little more than one-third to one-half of the Anglo
    vote, Rep. Doggett’s victories seem attributable to a minority community doing the heavy lifting.
    But Dr. Ansolabehere’s analysis begs the question, because it tells us nothing about voter
    turnout. Without that crucial element, there is no way to put his analysis into context. See Texas,
    8
    The regression analysis Dr. Ansolabahere provides is also not without its flaws. Dr. Ansolabehere uses VAP,
    not CVAP, in his calculations. Ansolabehere Rep. attach. 3. The HCVAP of CD 25 is 25.3%; its HVAP 34%. We
    are left to guess if this significant difference between citizen and noncitizen minority population, a highly relevant
    factor in light of citizen voting requirements, would change his conclusions. Moreover, there are unexplained
    discrepancies in Dr. Ansolabehere’s data. He states that Hispanics comprised 83% of Doggett’s coalition in 2008,
    id., but his retrogression calculation appears to indicate that figure was 93%, id. attach. 4. And he calculates Black
    support for Doggett in 2008 at 111%. Id. attach. 3, an overestimation by (at least) 11%. These problems are
    additional reasons why I am hesitant to find that this evidence supports finding that CD 25 is an ability district. At a
    minimum, it is not “more exacting evidence.”
    7
    831 F. Supp. 2d at 263 (“However, when there is no supermajority in a district, a Section 5
    analysis must go beyond mere population data to include factors such as minority voter
    registration, minority voter turnout, election history, and minority/majority voting behaviors.”
    (emphasis added)). In other words, Dr. Ansolabehere does not answer the question, “83% and
    86% of how many Hispanic voters, and 100% of how many Black voters?” That minorities voted
    overwhelmingly for Rep. Doggett tells us very little about their role in the coalition. This data
    could support a story in which minorities lead the way to victory, but it could also tell a story in
    which minority voters have an equal voice to Anglos, or even one where Anglo voters take the
    lead in CD 25. Incomplete data from which we might infer ability status is not the type of “more
    exacting evidence” necessary to find a protected district.
    Even taking Dr. Ansolabehere’s data at face value — a limb on which I am extremely
    loathe to perch for the reasons stated above — and using it to try to extrapolate the missing
    turnout data9 would indicate that Anglos cast an average of 81% of all votes in CD 25 in 2008
    and 2010, and thus that minorities cast only 19%.10 Absent any indication that minorities play a
    9
    To be clear, I do not think that we should engage in this type of endeavor. In my view, the fact that the
    experts in this case did not provide sufficient information to show ability to elect should be the end of the inquiry. I
    set out this analysis only because my colleagues do not share my view that Dr. Ansolabehere’s data, as presented, is
    insufficient.
    10
    Assuming that Dr. Ansolabehere is correct, I calculate turnout in the following manner. In 2008, Rep.
    Doggett received 65.82% of the vote. Pl.’s Ex. 31, at 10. In 2010, he garnered 52.82%. Pl.’s Ex. 32, at 13. This
    change, according to Dr. Ansolabehere’s data, was due almost exclusively to the decrease in Anglo support for Rep.
    Doggett from 53% to 37% (the only other change was an increase in Rep. Doggett’s Hispanic vote share from 83%
    to 86%, which is negligible and within the standard margin of error). Thus, a 16% change in Anglo preferences
    (53% - 37%) triggered a 13% change in votes for Rep. Doggett’s vote share (65.82% - 52.82%). This implies that
    Anglos comprised 81% of the total number of votes cast in 2008 and 2010 (13% / 16% = 81.25%). While this
    analysis is imperfect — relative turnout among minority groups (as opposed to overall turnout, which my colleagues
    cite, CD 25 Majority Op. at 13 n.19) could have changed between elections — it is the best we can accomplish with
    the limited data provided by Dr. Ansolabehere. As discussed above, I conclude that Dr. Ansolabehere failed to
    provide any evidence regarding turnout data, and so would prefer to stop my analysis there. I engage in this
    calculation only because my colleagues find Dr. Ansolabehere’s analysis to be persuasive.
    8
    leadership role in the coalition, I cannot conclude that a district where (according to the most
    favorable reading of expert testimony) minorities cast only 19% of the votes can be protected.
    This reading of Dr. Ansolabehere’s data is also largely consistent with the OAG voter
    turnout statistics my colleagues discard.11 In the 2008 election, the OAG analysis indicates that
    minorities comprised approximately 18% of voters, almost exactly the 19% composition Dr.
    Ansolabehere appears to predict. Pl.’s Ex. 24, at 579. In 2010, the OAG data indicates voter
    turnout of 10%, which is lower than Dr. Ansolabehere’s apparent, average prediction of 19%, but
    11
    My colleagues reject this data based on a single comment by Dr. Alford, Texas’s expert, during oral
    argument. CD 25 Majority Op. at 12. But Dr. Alford addressed the OAG’s racially polarized voting analysis
    concerning the House, not the Congress. Trial Tr. 86:12-87:7, Jan. 24, 2012 PM. And he compared a different subset
    of that data than my colleagues do. Compare id. at 86:19-20 (“[I]f you’ll take a quick look at the last two columns
    [of the data] . . . .” (emphasis added)), with CD 25 Majority Op. at 13 (discussing the difference between “estimated
    turnout % in district” and “actual turnout % in district,” Pl.’s Ex. 24, at 576, which are the third to last and the last
    columns.). In other words, Dr. Alford’s concern is not the same as my colleagues’. Moreover, Alford and my
    colleagues raise concerns about different data than I examine here. He critiqued the total estimated turnout
    calculated “as a percent of VAP,” Trial Tr. 85:23-87:7, Jan. 24, 2012 PM (discussing Defs.’ Ex. 6, at 358), i.e., what
    percentage of eligible voters in a minority group voted in an election.
    Neither Alford nor my colleagues assess the OAG’s calculations concerning “distribution of votes in [a]
    contest,” Pl.’s Ex. 24, at 579, i.e., what percentage of votes in any given election was cast by each minority group.
    Unlike the data my colleagues examine (and on which I do not rely in any way), this analysis accurately predicts the
    actual overall turnout in a given election, including every election my colleagues identify as problematic. Compare
    Pl.’s Ex. 24, at 587-88 (predicting 190,223 votes in the 2010 general election when 173,309 were actually cast,
    resulting in an error rate of 9.8%), with CD 25 Majority Op. at 13 (calculating an error rate in 2010 of 43.6%); Pl.’s
    Ex. 24, at 587 (predicting that 300,273 votes were cast in the 2008 general election when 282,161 votes were
    actually cast, resulting in an error rate in 2008 of 6.4%), with CD 25 Majority Op. at 13 (calculating an error rate of
    51.4%); Pl.’s Ex. 24, at 587 (predicting that 172,695 votes were cast in the 2006 general election when 159,507 were
    actually cast, resulting in an error rate of 8.2%), with CD 25 Majority Op. at 13 (calculating an error rate of 28.1%).
    Additionally, this data predicts the “number of votes cast by [each] minority group,” see Pl.’s Ex. 35, at 585-89,
    despite my colleagues’ apparent statements to the contrary. CD 25 Majority Op. at 13 n.18.
    My colleagues state that “there is no testimony, expert or otherwise, in the record that the data on which the
    dissent relies is not as flawed as the turnout numbers rejected by Dr. Alford.” Id. But my colleagues would discard
    the State House OAG data (and, by extension the Congressional OAG data) based on the metric Dr. Alford
    described in his testimony: the gap between the predicted turnout and “real life.” Id. at 12. I use that same metric —
    the only ground Dr. Alford gave as support for his critique — to test the data. My colleagues also note that “Texas’s
    failure to cite to this data again indicates to the Court that it has little probative value.” Id. at 13. While I am
    skeptical that our assessment of evidence contained in the record should be influenced by whether a particular party
    chose to cite it, I note that the United States cited the OAG turnout data favorably. See U.S. Proposed Findings of
    Fact ¶¶ 24, 54, 58, 163, 197.
    Finally, it bears reemphasizing that even if my colleagues’ concerns were serious enough to warrant
    discarding this data entirely, the proper consequence should be to conclude that Dr. Ansolabehere’s analysis is not
    the “more demanding evidence” necessary to prove a coalition district. As I have explained above, turnout data is
    the only way to provide context for the data on which my colleagues rely so heavily. My attempt to provide that
    missing data should not detract from the more important fact: Dr. Ansolabehere’s report does not include this
    essential information at all.
    9
    not absurdly so. Id. The rest of the OAG data indicates that minorities cast closer to 10% of the
    vote in CD 25, Pl.’s Ex. 24, at 579-80. This data indicates that CD 25 looks much like the
    hypothetical district we described before in which the Anglo voters that made up 90% of the
    district split their vote evenly and minority voters comprise just 10% of the votes, providing the
    margin of victory. We agreed that such a district would not be protected. Majority Op. at 24-25.
    Even assuming that minorities cast 19% of the vote, as Dr. Ansolabehere’s data appears to
    indicate, this would be enough only to show influence, not that benchmark CD 25 is a district in
    which minority voters themselves have an ability to elect.12
    In sum, we heard testimony and received expert reports that minorities are essential to
    victory in Travis County, but that is not enough to find that CD 25 is a protected crossover
    district. To protect CD 25, we must find that minorities themselves have an ability to elect in CD
    25 — that they lead the coalition there. It is not enough that they provide the margin of victory in
    a competitive Democratic district. Most of the evidence concerns Travis County alone. No
    evidence includes turnout data, in Travis County or in the district as a whole. At best, the
    evidence shows that minorities cast no more than 20% of the votes in CD 25, and possibly
    significantly less. If this is the “more exacting” evidence we require to prove the existence of a
    coalition district, it is hard to see what Democratic district in Texas would not be so protected.
    Respectfully, I dissent from my colleagues’ assessment that benchmark CD 25 is an ability
    district.
    12
    Even assuming that the 2010 election alone did rise to that high level of proof (which I do not believe it
    does), we have previously stated in the context of SD 10 that a single election does not indicate a proven history of
    ability to elect.
    10
    APPENDIX TO THE MEMORANDUM OPINION
    FINDINGS OF FACT & CONCLUSIONS OF LAW REGARDING DISPUTED ABILITY
    DISTRICTS
    COLLYER & HOWELL, District Judges:
    I.   THE CONGRESSIONAL PLAN
    A. Congressional Redistricting Plan, C185
    1. In 2006, a three-judge district court adopted a redistricting plan for the Texas
    congressional delegation. See LULAC v. Perry, 
    457 F. Supp. 2d 716
    , 716–18 (E.D. Tex.
    2006) (per curiam). That plan, known as C100, is the Benchmark Plan for the purposes
    of this case.
    2. The 2010 Census showed that the population of Texas increased by 4,293,741, from
    20,851,820 in 2000 to 25,145,561 in 2010. Pl.’s Ex. 75. This growth represented a
    20.6% increase in the State’s overall population, with 89.2% of the increase attributable
    to growth in the minority populations. Mot. for Judicial Notice, ECF No. 180, ¶¶ 8, 18.
    Hispanics comprise 65% of the increase and Blacks comprise 13.4%. Id. at ¶¶ 20, 22.
    3. As a result of this population growth, Texas was entitled to four new seats in the U.S.
    House of Representatives, increasing the State’s number of representatives from 32 to 36
    members. This increase required the State to reallocate congressional districts, and
    necessitated the drawing of new district maps to govern congressional elections in 2012.
    B. The Legislative Process
    a. 2010 Field Hearings
    4. Anticipating that the State’s population growth would result in additional congressional
    districts, in 2010, prior to the start of the 2011 legislative session, the Texas House
    Committee on Redistricting, the Texas House Judiciary Committee, and the Texas Senate
    Select Committee on Redistricting jointly or separately held approximately 19 field
    hearings around the State regarding the redistricting process for the State Legislature and
    congressional plans. Defs.’ Ex. 320, at 58-60 (Rep. of Dr. Arrington); Trial Tr. 86, Jan.
    17, 2012 AM (Rep. Todd Hunter); Pl.’s Ex. 39; Pl.’s Ex. 42. The purpose of the hearings
    was to receive input before the formal redistricting process began in 2011. Trial Tr. 54,
    Jan. 17, 2012 AM (Rep. Hunter); Trial Tr. 145, Jan. 17, 2012 PM (Doug Davis).
    5. At the time of these hearings, the official 2010 Census data had not yet been released, nor
    had any of the State legislative committees participating in the hearings furnished for
    public comment any proposed Congressional redistricting plans. Trial Tr. 115-16, Jan.
    17, 2012 AM (Rep. Hunter). Testimony was presented at the hearings regarding the need
    to retain minority communities of interest, recognize minority population growth in the
    Dallas-Fort Worth metroplex with a new minority ability district, and maintain those
    congressional districts where minority voters had been able to elect their candidates of
    choice. See, e.g., Trial Tr. 10-11, Jan. 23, 2012 PM (Congresswoman Jackson Lee).
    Nevertheless, these hearings were of limited utility since no plans were available for the
    witnesses to review or to offer specific comment on. Furthermore, the sponsoring
    legislative committees prepared no written reports summarizing the information presented
    at the hearings to facilitate communication of any concerns or recommendations raised to
    members of the legislature who were not present. Trial Tr. 115, Jan. 17, 2012 AM (Rep.
    Hunter).1
    6. Testimony at trial made clear that minority elected representatives from Texas viewed the
    2010 field hearings as a “sham” or “just for show.” Trial Tr. 91, Jan. 19, 2012 AM (Rep.
    Dawnna Dukes) (testifying that the 2010 field hearings were “just a circus to show that a
    hearing had been held around the state, but it was not of substance because there was
    absolutely nothing before the committee for individuals to testify on, for or against.”);
    Defs.’ Ex. 809, at 4 (Senator Judith Zaffirini, Hispanic representative for SD 21,
    describing the 2010 field hearings as “a sham” with “low attendance, [] low participation,
    [] lack of invited testimony, [and] the lack of prepared materials for [members of the
    Senate Redistricting Committee].”); see also Trial Tr. 94, Jan. 20, 2012 AM (Sen.
    Rodney Ellis testifying that the 2010 field hearings were “perfunctory”).
    7. In addition, the only Black member of the House Redistricting Committee,
    Representative Marc Veasey, testified that some field hearings, specifically in the Dallas-
    Fort Worth metroplex, were held in locations inconvenient for minority voters that did
    not have public transport, which limited their participation. Trial Tr. 8-12, Jan. 18, 2012
    PM (Rep. Veasey). Representative Veasey offered to help find locations convenient for
    minority voters, but ultimately locations were picked without regard to the concerns of
    minority members of the redistricting committee. Id. at 12 (“But when it came to, you
    know, trying to make sure that you know, southeast Ft. Worth and the city of Ft. Worth,
    which, like I said, is the third largest concentration of African-Americans in the state --
    trying to find a place to do hearings there, that no one came to consult with me or any
    other minority members of the committee. They just decided they were going to have
    this field hearing in Arlington, which just, you know, still to this day makes no sense at
    all.”).
    8. No evidence was presented that the 2010 field hearings addressed the topics of the number
    of districts that provided minority citizens the ability to elect the candidates of their choice
    or the minimal number of minority ability districts required for compliance with the
    Voting Rights Act (“VRA”) under any new congressional plan. See generally Pl.’s Ex.
    50 (Texas Legislative Council Redistricting Guidance, dated August 2011, stating that
    courts generally compared the number of minority districts in the benchmark plan and in
    the enacted plan).
    1
    The only way for legislators to review the information presented during these hearings was to obtain from the
    Committee Clerks any material submitted during the field hearings, or by viewing the hearings by webcast. Trial Tr.
    114-15, Jan. 17, 2012 AM (Rep. Hunter).
    2
    b. 2011 Regular Texas Legislative Session
    9. After convening in January 2011, the Texas Legislature faced the task of enacting
    redistricting maps for the State House of Representatives (“State House”), State Senate,
    and U.S. House of Representatives in response to the population growth in the state. Trial
    Tr. 59-60, Jan. 18, 2012 AM (Downton). The regular session of the Texas Legislature ran
    from January 11 through May 30, 2011. Joint Stipulation, ECF No. 177, ¶¶ 3-4. No
    congressional redistricting plan was publicly released by the Redistricting Committees of
    either the State House or State Senate, nor were any hearings held concerning a
    congressional plan during the regular session.2 Defs.’ Ex. 509, at 29-32; Joint Stipulation,
    ¶¶ 4-5.
    10. During the Legislature’s regular session, only informal discussions were held concerning
    the congressional redistricting plan. Interested advocacy groups, including the Mexican
    American Legal Defense and Educational Fund (“MALDEF”) and Mexican American
    Legislative Caucus (“MALC”), proposed congressional maps to the House Redistricting
    Committee. Trial Tr. 60-61, Jan. 18, 2012 AM (Ryan Downton). Members of Texas’s
    congressional delegation also submitted proposals and attempted to meet with State
    legislators to discuss proposed plans.
    c. 2011 Special Legislative Session
    11. The Legislature’s failure to enact a new congressional plan during the regular session
    prompted Governor Rick Perry, on May 31, 2011, to order the State Legislature to sit in
    Special Session to address, among other things, legislation relating to congressional
    redistricting. Joint Stipulation, ¶ 5. On the first day of this special session, on May 31,
    2011, Chairman Kel Seliger, chairman of the Senate Redistricting Committee, and
    Chairman Burt Solomons, chairman of House Redistricting Committee, publicly released
    C125, which was the first congressional redistricting plan proposed publicly by the
    leadership of the State Legislature. Defs.’ Ex. 366.
    12. Hispanic and Black members of the State House were not included in the map-drawing
    process for C125. State Representative Marc Veasey, a Black member of the House
    Redistricting Committee, testified that no minority state representative had any input into
    the proposed congressional redistricting map before it was made public. Defs.’ Ex. 335
    (Veasey Dep. at 25-27, Aug. 19, 2011); see also Trial Tr. 91-93, Jan. 19, 2012 PM (Rep.
    Dukes) (testifying that she first saw the proposed congressional map on Friday, June 9,
    2011 well after its release).3
    13. In the late afternoon of May 31, 2011, the State House and Senate noticed public hearings
    on C125. Less than 48 hours later, at 9:00 a.m. on June 2, 2011, the House Redistricting
    2
    The focus of the legislative redistricting efforts during this session was on the State House and Senate plans,
    which, if not enacted during the regular session, would have been determined by the Texas Legislative Redistricting
    Board. Trial Tr. 59-60, Jan. 18, 2012 AM (Downton).
    3
    Representative Dukes did not explain why she first saw the map on June 9 when the map was publicly released on
    May 31, 2011.
    3
    Committee held its only public hearing on the proposed plan at the State Capitol in
    Austin. The following day, on Friday, June 3, 2011, the Senate Redistricting Committee
    held its only hearing on C125, also in the State Capitol in Austin. Defs.’ Ex. 320, at 59
    (Rep. of Dr. Arrington); Defs.’ Ex. 366 (Congressional Redistricting Timeline); Defs.’
    Ex. 509, at 39.
    14. At the June 3, 2011 Senate Redistricting Committee hearing, minority members of the
    Committee complained of being excluded from the congressional redistricting process.
    Defs.’ Ex. 370 at 1. Specifically, Senator Judith Zaffirini, a Hispanic Senator
    representing SD 21, and Senator Royce West, a Black Senator representing SD 23,
    complained that the process was too rushed and stated that neither they nor the public had
    adequate time to study the proposed map or meaningfully participate. Id. Senator West
    stated: “For the purposes of the record, I did not have any input into the map 125. I
    never saw map 125 before you published it.” Id. Similarly, Senator Zaffirini told
    Chairman Seliger: “I’ve been on every redistricting committee since my election in 1986
    and I must say that I have never had less input into the drawing of any map until this
    session.” Id.
    15. Experts retained by the Senate Redistricting Committee from Baylor University’s School
    of Law and the University of Texas Law School, Professors David Guinn, Mike
    Morrison, and Robert Heath (“Senate Redistricting Committee Outside Experts”),
    echoed concerns about the lack of opportunity for public scrutiny of C125 in comparison
    to redistricting processes in previous years. Trial Tr. 73, 81, Jan. 24, 2012 AM,
    (Chairman Seliger); Defs.’ Ex. 370, at 2. These outside experts indicated that they did not
    have an opportunity to review the proposed congressional redistricting plan before it was
    presented in the Committee hearing. Defs.’ Ex. 370, at 2; Defs.’ Ex. 568, at 1. Professor
    Morrison testified that “this process has been quite different from what we’ve seen in the
    past. . . [n]obody has had the opportunity to study it the way it has been done in the past.”
    Id. He explained further that this procedure differed from the one followed in 2003 when
    the committee’s staff “went all over the state . . . spent sixteen hours in one place, twenty
    in another. We sat down . . . we visited. We hired experts to do retrogression analysis.”
    Id. In fact, evidence presented at trial shows, for example, that prior to passage of the
    congressional redistricting plan in 2003, the Redistricting Committees held seven
    public hearings, and the committee substitute bill was the focus of six of those hearings.
    Defs.’ Ex. 300.
    16. At the June 3, 2011 hearing, the Senate Redistricting Committee Outside Experts
    cautioned Members about the care required for compliance with the VRA, testifying that
    they “furnished the committee an advisement to take [the DOJ 2011 Guidelines] and read
    them all very carefully.” Defs.’ Ex. 370. Indeed, Chairman Seliger testified that the sole
    responsibility of these outside counsel was “to vet the maps as we drew them and to
    inform me or anyone else on the committee whether they were legal or not.” Trial Tr. 81,
    Jan. 24, 2012 AM (Chairman Seliger). In his pre-filed written direct testimony,
    Chairman Seliger claimed that he relied on these experts to “inform me if the
    demographics, performance, or any other attribute of a proposed district would raise
    concerns under the Voting Rights Act.” Pl.’s Ex. 162, ¶ 4 (Seliger Pre-filed Direct
    4
    Testimony). To the contrary, these experts testified before the Senate Redistricting
    Committee that they did not “provide[] verbal or written guidance or []opinion to the
    committee regarding whether [the proposed Congressional plans were] in compliance
    with Section 5” because they were not asked to do so. Defs.’ Ex. 370, at 3.
    17. On June 6, 2011, the Monday immediately following the Friday hearing, the full Senate
    considered the proposed congressional redistricting plan, C185 (the “Congressional
    Plan”). On the floor of the Senate, Senator Zaffirini asked Chairman Seliger if “any
    minority Members [were] involved in developing” the redistricting maps under
    consideration. Chairman Seliger bluntly responded, “[n]ot that I recall.” Devaney Decl.,
    ECF No. 77, Ex. 9 (Texas State Senate Journal, June 6, 2011, at A-12). Chairman Seliger
    also admitted during the floor debate that the Senate Redistricting Committee Outside
    Experts he hired had not seen the Congressional Plan until it was released in committee
    and that these outside experts had not evaluated the plan for compliance with the VRA.
    Defs.’ Ex. 568 at 1. Nevertheless, the Senate passed the proposed Congressional Plan in
    Senate Bill 4 (“SB4”) on June 6, 2011 by a party-line vote of 18-12. Joint Stipulation, ¶¶
    16-17, 19.
    18. Following passage of SB4, the State House leadership gave notice that the House
    Redistricting Committee would meet to consider the Senate Bill at 9:00 a.m. on June 9,
    2011. On June 9, 2011, the House Redistricting Committee met to consider the proposed
    Congressional Plan, and passed it out of Committee without taking any public comments.
    Defs.’ Ex. 320, at 59 (Rep. of Dr. Arrington); Defs.’ Ex. 366. Representative Dukes, who
    is not on the House Redistricting Committee, testified that she first saw the proposed
    Congressional Plan on June 9, 2011. Trial Tr. 91-93, Jan. 19, 2012 PM (Rep. Dukes).
    That same day, the State House passed a Calendar Rule requiring any amendments to the
    proposed map to be filed “prior to Monday.” Id. This effectively gave any representative
    two days to prepare and submit proposed alterations to the congressional map.
    Representative Dukes testified that she worked through the weekend on an amendment
    and proposed a new map, but Chairman Solomons tabled her amendment and it was
    never considered. Trial Tr. 93-94, Jan. 19, 2012 AM (Rep. Dukes). State Representative
    Dukes further testified that every Democratic proposal to amend C185 was tabled. Id.
    19. On June 15, 2011, the State House passed the proposed Congressional Plan by a vote of
    93-47-3,4 after incorporating minor amendments. Joint Stipulation, ¶¶ 16-17. All
    Democratic members of the State House voted against passage of SB4. Texas State
    House Journal, June 15, 2011, at 421. The State Senate concurred with the State House
    amendments to the proposed Congressional Plan on June 20, 2011, and SB4 was reported
    as enrolled on June 20, 2011. SB4 was then signed by the State Senate on June 22, 2011,
    and the State House on June 24, 2011. Joint Stipulation, ¶ 16. On June 24, 2011, SB4
    containing the proposed congressional map, C185, was transmitted to Governor Rick
    Perry, who signed it into law three weeks later, on July 18, 2011. Id.
    20. The legislative process under which the Congressional Plan was made public, considered
    4
    Three Representatives voted “present.”
    5
    and enacted was rapid. The timing of the two public hearings in the House and Senate
    Redistricting Committees within the short span of 48 and 72 hours, respectively, after
    first public release of the Congressional Plan severely circumscribed the opportunity for
    meaningful public scrutiny and comment, including by minority citizens and their elected
    officials. Defs.’ Ex. 320, at 58-60 (Rep. of Dr. Arrington); Trial Tr. 16, Jan. 18, 2012
    PM (Rep. Veasey). Outreach by Representative Todd Hunter, Chair of the House
    Judiciary Committee, to the congressional delegation during a 2010 visit to Washington,
    D.C., and by Congresswoman Sheila Jackson Lee and Congressman Gene Green to
    Chairman Solomons in 2011, appear to have been “meet and greet” sessions with
    minimal to no substantive discussion about the changes planned by the Texas legislators
    to the districts represented by minority Members of Congress. Pl.’s Ex. 162, ¶ 12
    (Chairman Seliger Pre-Filed Direct Testimony); Pl.’s Ex. 148, ¶ 8 (Chairman Solomons
    Pre-Filed Direct Testimony) (stating that the meetings with Congressman Green and
    Congresswoman Jackson Lee were “more of a ‘meet and greet,’ neither of the
    congresspersons provided me with any details requesting specific changes to their
    districts”).
    C. Mapdrawers’ View of the Redistricting Process
    21. Ryan Downton, the general counsel to the House Committee on Redistricting under
    Chairman Burt Solomons, was the principal drafter of the Congressional Plan. Trial Tr.
    44-45, Jan. 18, 2012 AM (Downton); Trial Tr. 47, Jan. 17, 2012 PM (Interiano). Mr.
    Downton was primarily responsible for “zeroing-out”5 districts to make them conform to
    the required population size and for allocating Texas’s four new congressional districts.
    Gerardo Interiano, counsel to Speaker of the State House Joe Straus, also testified that he
    periodically helped Mr. Downton with the congressional map to zero-out population
    deviations. Trial Tr. 105, Jan. 17, 2012 PM (Interiano); Trial Tr. 44-45, Jan. 18, 2012
    AM (Downton); Trial Tr. 47, Jan. 17, 2012 PM (Interiano).
    22. Upon release of the 2010 Census data on February 17, 2011, Mr. Downton testified that
    he learned that “there were three areas where the population growth per region
    significantly outpaced growth in the rest of the state. Those three regions, the first being
    north central Texas around the Dallas-Fort Worth metroplex. The second being the
    suburban areas around Harris County in kind of Southeast Texas, and the third being the
    I-35 corridor running from San Antonio north through Austin.” Trial Tr. 61-62, Jan. 18,
    2012 AM (Downton).6 Mr. Downton believed that of the four new congressional seats
    allotted to Texas, “one had to go in each of those regions and in the fourth one [Texas]
    had some flexibility.” Id. at 62.
    5
    Congressional districts must be drawn within one person of the ideal district size. Congressional districts therefore
    must be “zeroed out” by the mapdrawer, meaning that the district must deviate from the required population by at
    most one person. Trial Tr. 91-92, Jan. 18, 2012 AM (Downton); Trial Tr. 71-72, Jan. 25, 2012 PM (Interiano).
    Based on the 2010 Census, the ideal population for each of the 36 congressional districts in Texas is 698,488. Joint
    Stipulations, ¶ 15.
    6
    Mr. Downton later clarified that although “Dallas County itself lost population relative to the rest of the State[,]
    Tarrant County on the west and Colin and Denton counties on the north gained population.” Trial Tr. 62, Jan. 18,
    2012 AM (Downton).
    6
    23. Prior to assignment of map-drawing responsibilities, Mr. Downton was aware of the
    VRA and actively sought to educate himself on its requirements. Trial Tr. 45, Jan. 18,
    2012 AM (Downton). To this end, Mr. Downton consulted with the Texas Legislative
    Council7 (“TLC”), including a lawyer named David Hanna. See id. at 50. Mr.
    Downton testified that he viewed compliance with the VRA on par in importance with
    getting enough votes to get the map passed, but this testimony is not credible. Defs.’
    Ex. 778A (Downton Dep. 62, Aug. 12, 2011).
    24. Mr. Downton testified that during the map-drawing process he identified districts
    protected by the VRA in the Benchmark Plan “based on Census level. If they were
    above 50%, then they were Hispanic majority districts.” Trial Tr. 63, Jan. 18, 2012
    AM (Downton). The specific demographic statistics that Mr. Downton relied upon
    were Hispanic Citizen Voting Age Population (“HCVAP”) and Spanish Surname Voter
    Registration (“SSVR”). Defs.’ Ex. 778A (Downton Dep. 22, Aug. 12, 2011); Trial Tr.
    67, Jan. 18, 2012 AM (Downton). If these statistics were above the 50% mark, he
    believed the district was protected under the VRA. Defs.’ Ex. 577 (Trial Tr. 966, Perez
    v. Perry, civil action no. SA:11-360 (W.D. Tex. Sept. 9, 2011)). Based on Census data
    alone, Mr. Downton identified seven districts (CDs 15, 16, 20, 23, 27, 28 and 29) as
    protected districts in the Benchmark Plan that provided Hispanic citizens the ability to
    elect their candidates of choice.8 Trial Tr. 63-65, Jan. 18, 2012 AM (Downton)
    25. The Office of the Attorney General (the “OAG”) performed a racially polarized voting
    analysis of the Benchmark and enacted districts. Pl.’s Ex. 26, 27.
    26. The OAG also performed reconstituted election analyses that estimated what
    percentage of a specific racial or language-minority group voted for certain candidates
    in chosen primary and general elections. Pl.’s Ex. 27. These analyses were based on
    ten general elections (the “OAG 10”) selected by Todd Giberson, an employee in the
    OAG’s Legal Technical Support Division, because they were “racially contested
    elections,” i.e., ones that involved minority candidates running against each other or a
    minority candidate running against a non-minority candidate. Giberson Dep. 16, 20-21,
    Oct. 18, 2011.
    27. Mr. Interiano, who assisted in drawing the Congressional Plan, confirmed that any
    initial understanding of protected districts in the Benchmark Plan was made solely by
    looking at the demographic population statistics of the district. Trial Tr. 26-27, 47-48,
    Jan. 17, 2012 PM (Interiano). He testified that the mapdrawers did not look at election
    result analyses for the Benchmark Plan to help identify protected districts until they had
    already submitted draft redistricting plans to the OAG. Id. Mr. Downton was also clear
    that he did not factor the State’s reconstituted election analysis into his determination of
    7
    The TLC is an agency within the legislative branch of the Texas State government that provides nonpartisan,
    technical support and services to each member of the Legislature. Archer Dep. 8-9, Oct. 12, 2011.
    8
    By contrast, Texas argued at summary judgment that any district with a Black voting age population of 40% or
    more is an ability district. Pl.’s Mem. in Supp. of Mot. for Summ. J., ECF. No. 41, at 30.
    7
    whether a district was a Hispanic majority district and therefore a protected district in
    the Benchmark Plan. Defs.’ Ex. 778A (Downton Dep. 22-23, Aug. 12, 2011). In his
    view, political performance was not particularly relevant. Id. at 24. If a district met the
    mapdrawers’ own standard of over 50% in HCVAP and SSVR, he classified the district
    as an ability district regardless of whether it elected the minority candidate of choice 3
    out of 10 times, or 1 out of 10 times. Id. at 24-25; Defs.’ Ex. 577 (Trial Tr. 966, Perez
    v. Perry, Sept. 9, 2011 (Downton)). Mr. Interiano also testified that demographic
    information, including Hispanic Voting Age Population (“HVAP”), HCVAP and SSVR,
    must be considered to determine if a district is a Hispanic “opportunity” district. Defs.’
    Ex. 579 (Trial Tr. 1451, Perez v. Perry, Sept. 12, 2011 (Interiano)). Mr. Hanna advised
    the mapdrawers, however, that even if a district were over a 51% threshold based upon
    demographic data, it might not perform for the minority population. See Defs.’ Ex. 305;
    Defs.’ Ex. 312, at 5 (when editing Texas’s informal preclearance submission to the
    DOJ, Mr. Hanna commented that demographic benchmarks were “phony”).
    28. Mr. Downton ignored Mr. Hanna’s advice about identifying minority ability districts.
    Relying solely on demographic statistics to identify a minority population’s ability to
    elect, Mr. Downton testified that when drawing the Congressional Plan, he tried to keep
    the demographic numbers of protected districts “at their benchmark levels.” Trial Tr.
    65-66, Jan. 18, 2012 AM (Downton). The Congressional Plan was legally compliant
    with the VRA, in his opinion, because seven districts in South and Central Texas have
    over 50% HCVAP. Defs.’ Ex. 577 (Trial Tr. 950, Perez v. Perry, Sept. 9, 2011
    (Downton)). Mr. Downton asserted, however, that based on the reconstituted election
    analysis conducted after the Congressional Plan was submitted to the OAG, in his view,
    the Congressional Plan actually increases the number of districts that provide Hispanics
    the ability to elect their candidate of choice. Trial Tr. 67-68, Jan. 18, 2012 AM
    (Downton).
    29. Messrs. Downton and Interiano both testified that they did not look at reconstituted
    election analyses or performance prior to completing the Congressional Plan, even
    though they both received legal advice that, for VRA compliance, reliance solely on
    demographic data is insufficient to measure the number of protected districts in the
    benchmark or the enacted plan. Trial Tr. 1451-52, Perez v. Perry, Sept. 12, 2011
    (Interiano); Trial Tr. 57, Jan. 18, 2012 AM (Downton).
    30. While the mapdrawers’ reliance solely on demographic data to assess VRA compliance
    was erroneous, their superiors were negligent of their responsibilities under the VRA.
    The Chairmen of the Redistricting Committees testified that they relied on the
    mapdrawers to ensure that the map was “legal,” but made little independent effort to
    ensure that minority districts were protected. Chairman Solomons did not utilize the
    Senate Redistricting Committee Outside Experts hired to evaluate whether the
    Congressional Plan complied with the VRA. Neither he nor Chairman Seliger ever asked
    for the specific number of minority ability districts required, at a minimum, to ensure that
    the congressional map complied with the VRA. Trial Tr. 11, Jan. 24, 2012 AM
    (Chairman Seliger); Trial Tr. 65-67, Jan. 20, 2012 PM (Chairman Solomons testifying
    that he did not know or identify the number of protected districts in the Benchmark Plan
    8
    because that determination was made by his staff).
    D. Congressional Districts at Issue
    a. Congressional District 23
    31. In the Benchmark Plan, CD 23 is based in West Texas and incorporates Brewster,
    Crockett, Culberson, Dimmit, Edwards, Hudspeth, Jeff Davis, Kinney, Maverick,
    Medina, Pecos, Presido, Reeves, Terrel, Uvalde, Val Verde, and Zavala counties, as well
    as portions of Bexar, El Paso, and Sutton counties. Pl.’s Ex. 11, at 5-6. In terms of
    metropolitan areas, CD 23 in the Benchmark Plan includes the cities of Del Rio and
    Eagle Pass, as well as areas of Bexar County that fall outside San Antonio’s city limits.
    This district was drawn in 2006, following the Supreme Court’s ruling in LULAC v.
    Perry, by the U.S. District Court for the Eastern District of Texas in order to remedy the
    State’s violation of Section 2 of the VRA and provide Hispanics the opportunity to elect
    candidates of their choice. Defs.’ Ex. 826, at 5 (Rep. of Dr. Flores); Defs.’ Ex. 575
    (Trial Tr. 300, Perez v. Perry, Sept. 7, 2011 (Flores)).
    32. Based on demographic statistics, Hispanics are a clear majority of the population in the
    Benchmark CD 23 and endogenous election results indicate that the district often elected
    a Hispanic candidate of choice, even if not every time. See infra ¶ 35.
    33. The only expert proffered by Texas on the issue of retrogression disagrees with Texas
    and concludes that CD 23 is no longer an ability district. Defs.’ Ex. 581 (Trial Tr. 1839,
    Perez v. Perry, Sept. 14, 2011) (Dr. Alford testifying: “I don’t think that the 23rd is any
    more likely to perform that it was. I think it is probably less likely to perform than it was,
    and so I certainly wouldn’t count and don’t – in all of this discussion, I haven’t counted
    the 23rd as an effective minority district in the newly adopted plan, but it does remain a
    majority district.”) (emphases added).
    34. CD 23 in the Congressional Plan is no longer an ability district.
    i. Demographic and Election Result Data for Benchmark Congressional District 23
    35. Texas has identified CD 23 as a Hispanic ability district in the Benchmark. See Pl.’s
    Mem. Supp. Mot. Summ. J., ECF No. 41, at 6. In the Benchmark Plan, CD 23 has an
    overall Hispanic population of 66.4%, an HCVAP of 58.4%, and an SSVR of 52.6%.
    Pl.’s Ex. 11, at 10. According to the OAG’s election analysis, Hispanic citizens in
    Benchmark CD 23 elected their candidate of choice in three out of ten elections. Defs.’
    Ex. 390. The Texas Latino Redistricting Task Force (“TLRTF”) argues that the OAG 10
    does not accurately reflect the ability of Hispanics to perform in the district. If four
    additional racially contested elections are examined, the Hispanic candidate of choice
    wins in 7 out of 14 elections. Trial Tr. 111-13, Jan. 18, 2012 AM (Downton); Defs.’ Ex.
    647. Moreover, Dr. Richard Engstrom, an expert offered by TLRTF, emphasizes that
    from 2006 to 2010, the candidate of choice of Hispanics won two of three endogenous
    elections in Benchmark CD 23. Defs.’ Ex. 575 (Trial Tr. 513-14, Perez v. Perry, Sept. 7,
    9
    2011) (Engstrom).
    36. Mr. Interiano testified that prior to redrawing CD 23, he never made a determination as to
    whether CD 23 was a protected district in the Benchmark Plan. Trial Tr. 49, Jan. 17,
    2012 PM (Interiano). Chairman Seliger, however, testified that in the Benchmark Plan
    CD 23 is a Hispanic “opportunity” district and was drawn to be a Hispanic “opportunity”
    district by the court. Defs.’ Ex. 776 (Seliger Dep. 13, Sept. 1, 2011). The State’s expert
    witness, Dr. Alford, similarly testified that since the creation of CD 23 in 2006, it elected
    the Hispanic-preferred candidate in 2006 and 2008. Defs.’ Ex. 964 (Alford Dep. 121,
    Sept. 2, 2011).
    37. CD 23 is currently represented by Congressman Francisco Canseco, a Hispanic
    Republican. Defs.’ Ex. 406, at 7. Congressman Canseco was first elected to office in the
    2010 election, in which he defeated incumbent Ciro D. Rodriguez, a Hispanic Democrat,
    by a vote of 74,853 to 67,348, or 49.39% to 44.44%. Pl.’s Ex. 32, at 13. Voting in the
    2010 election was racially polarized, with 84.7% of Hispanics voting for Mr. Rodriguez.
    Defs.’ Ex. 728, at 25 (Rep. of Dr. Engstrom). While Hispanics overwhelmingly
    supported Mr. Rodriguez, he received only 18.1% of votes cast by non-Hispanics. Id.
    38. The evidence presented demonstrates that Congressman Canseco won a close election for
    CD 23 in 2010. With regards to this election, and others during 2010, Chairman Seliger
    testified that “the 2010 election was a bit of an aberration because of things like the Tea
    Party influence and I didn’t know if it was reliable.” Defs.’ Ex. 776 (Seliger Dep. 15,
    Sept. 1, 2011); Trial Tr. 11, Jan. 24, 2012 AM (Chairman Seliger).
    39. Although Chairman Seliger acknowledged that the 2010 election may not be “reliable,”
    he expressed his belief that Congressman Canseco was the preferred candidate of
    Hispanics in CD 23. Defs.’ Ex. 776 (Seliger Dep. 15, Sept. 1, 2011); Trial Tr. 11, Jan.
    24, 2012 AM (Chairman Seliger). He conceded that his belief is not based upon any
    analysis to determine whether Congressman Canseco was in fact the Hispanic candidate
    of choice. Defs.’ Ex. 776 (Seliger Dep. 31, Sept. 1, 2011). Furthermore, despite his
    stated belief that Congressman Canseco was the Hispanic candidate of choice in a
    Hispanic district, Chairman Seliger testified that he wanted to change CD 23 to make it
    safer for Congressman Canseco. Id. at 14; Trial Tr. 11, Jan. 24, 2012 AM (Chairman
    Seliger). Indeed, he testified that it was possible that Congressman Canseco would lose
    in 2012 if CD 23 were not reconfigured in some way. Defs.’ Ex. 776 (Seliger Dep. 15,
    Sept. 1, 2011); Trial Tr. 11-12, Jan. 24, 2012 AM (Chairman Seliger). Notwithstanding
    his desire to improve Congressman Canseco’s electoral performance, Chairman Seliger
    testified that he stressed to staff that CD 23 needed to remain a Hispanic district. Defs.’
    Ex. 776 (Seliger Dep. 13, 15, 30, 37, Sept. 1, 2011). He believed that the Legislature was
    legally required to build a district to elect the Hispanic candidate of choice in CD 23. Id.
    at 31; Trial Tr. 14-16, Jan. 24, 2012 AM (Chairman Seliger). Chairman Seliger further
    testified that if he had understood that Congressman Canseco was not the Hispanic
    preferred candidate, and he was taking steps to make CD 23 safer for Congressman
    Canseco, that would have created a concern in his mind regarding compliance with the
    VRA. Trial Tr. 11-16, Jan. 24, 2012 AM (Chairman Seliger).
    10
    ii.       Plan to Protect Congressman Canseco
    40. The Senate Redistricting Committee staff attempted to draw a district safe for
    Congressman Canseco’s reelection but found this to be a difficult challenge. Chairman
    Seliger stated: “in order to keep it as an opportunity district we just couldn’t piece it
    together where it served Congressman Canseco; and we wanted to if we could. And
    then [the House] came up with their design and we thought it was good.” Defs.’ Ex.
    776 (Seliger Dep. 14, Sept. 1, 2011).
    41. The mapdrawers in the State House, Messrs. Downton and Interiano, testified that
    there were “two goals” with regard to CD 23 when drawing the enacted map: “to
    maintain or strengthen the Hispanic nature of 23 and also strengthen the [R]epublican
    nature of 23.” Trial Tr. 80, Jan. 18, 2012 AM (Mr. Downton); Trial Tr. 47, Jan. 17,
    2012 PM (Interiano); Defs.’ Ex. 579 (Trial Tr. 1454-55, Perez v. Perry, Sept. 12, 2011
    (Interiano)); Defs.’ Ex. 779A (Interiano Dep. 102, Aug. 2, 2011). Mr. Interiano
    acknowledged, however, that he never conducted any analysis to determine if
    Congressman Canseco is the Hispanic preferred candidate in Benchmark CD 23.
    Defs.’ Ex. 579 (Trial Tr. 1456, Perez v. Perry, Sept. 12, 2011 (Interiano)); Defs.’ Ex.
    779A (Interiano Dep. 86-87, Aug. 2, 2011); Trial Tr. 49, Jan. 17, 2012 PM (Interiano).
    Mr. Downton conceded that he knew when he was drawing CD 23 that Congressman
    Canseco was not the Hispanic candidate of choice. Defs.’ Ex. 577 (Trial Tr. 966,
    Perez v. Perry, Sept. 9, 2011 (Downton)); Defs.’ Ex. 778A (Downton Dep. 90, Aug.
    12, 2011). He nonetheless drew CD 23 to “giv[e] Mr. C[a]nseco his best chance to be
    re-elected while maintaining and increasing the . . . total . . . Hispanic voting age,
    Hispanic citizen voting age, and Spanish surname voter registration.” Trial Tr. 105-
    107, Jan. 18, 2012 AM (Downton).
    42. The mapdrawers were aware that because Congressman Canseco was not the minority
    candidate of choice, increasing CD 23’s performance for Congressman Canseco would
    be problematic. For example, on April 13, 2011, a staffer at the National Republican
    Congressional Committee, Lee Padilla, requested in an email that Doug Davis,
    Director for the Senate Select Committee on Redistricting, “check on the latest
    Canseco version.” Defs.’ Ex. 978. Mr. Davis responded that “[i]t looks nice
    politically. We’re still concerned about the Voting Rights Act.” Mr. Davis continued
    that “[w]e’re going to have to put our best legal minds on the 23rd.” Id.
    43.    During the map-drawing process, legislative staffers understood that drawing a map to
    protect Congressman Canseco while maintaining the benchmark demographic
    statistics would require careful uses of demographic statistics. As early in the
    redistricting process as November 2010, Eric Opiela9 sent an email to Mr. Interiano,
    explaining that “certain data would be useful in identifying a nudge factor by which
    one can analyze which census blocks, when added to a particular district, especially
    50-plus-1 majority-minority districts, help pull the districts total Hispanic pop[ulation]
    9
    Mr. Interiano testified that at the time of this email, Eric Opiela was his colleague doing political work for Speaker
    Straus. Trial Tr. 54, Jan. 17, 2012 PM (Interiano).
    11
    and the Hispanic CVAP up to majority status, but leave the Spanish surnamed
    registered voters and turnout the lowest. This is especially valuable in shoring up
    Canseco and Farenthold.” Defs.’ Ex. 304; Trial Tr. 52-53, Jan. 17, 2012 PM
    (Interiano). According to Mr. Interiano, the import of this November 2010 email was
    to use demographic data, such as HVAP, HCVAP and SSVR, to draw a district that
    featured lower turnout of Spanish surname voters, while leaving the HCVAP at the
    benchmark level. Trial Tr. 53, Jan. 17, 2012 PM (Interiano).
    44. Mr. Opiela was not an outsider to the redistricting process and played a role in the
    manner in which districts were drawn. Mr. Downton testified that he communicated
    with Mr. Opiela during the drawing of the Congressional Plan and understood that the
    latter was “speaking on behalf of the Republican Congressmen from Texas with the
    exception of Representative Barton.” Trial Tr. 104, Jan. 18, 2012 AM (Downton);
    Trial Tr. 56, Jan. 17, 2012 PM (Interiano). Indeed, Mr. Downton acknowledged that
    he incorporated some of Mr. Opiela’s ideas into the Congressional Plan. Trial Tr. 104,
    Jan. 18, 2012 AM (Downton). Mr. Opiela also gave pointers to Mr. Interiano during
    the redistricting process. In particular, after Mr. Opiela informed Mr. Interiano in the
    November 2010 email that data available at the block level could be used to lower a
    district’s turnout of voters with Spanish surnames while raising its total Hispanic
    population, Messrs. Interiano and Opiela requested SSVR data at the block level from
    the TLC. Defs.’ Ex. 820; Defs.’ Ex. 980.
    iii.   Alterations to Congressional District 23 in the Congressional Plan
    45. The 2010 Census indicated that CD 23 was overpopulated by about 149,000 people.
    Defs.’ Ex. 575 (Trial Tr. 450, Perez v. Perry, Sept. 7, 2011 (Flores)); Defs.’ Ex. 436. Mr.
    Downton testified that CD 23 was “a very sensitive district” throughout the redistricting
    process because “[i]t was previously a court drawn district. We wanted to make sure we
    maintained the SSVR and HCVAP level of District 23.” Trial Tr. 78, Jan. 18, 2012 AM
    (Downton). As noted above, Mr. Downton also wanted to improve the district’s
    performance for Congressman Canseco. Id. at 105.
    46. Mr. Downton testified that while drawing CD 23 in the Congressional Plan he shaded
    precincts by election results and moved precincts in and out of CD 23 based on their
    election performance. Trial Tr. 107-08, Jan. 18, 2012 AM (Downton). In choosing
    between two precincts with similar SSVR, Mr. Downton testified that he would select the
    precinct with the greater percentage of Republican votes. Id. at 109. He did not,
    however, have any data showing which voters in a precinct were both Hispanic and
    Republican. Id. at 108. Mr. Downton sought to protect Congressman Canseco’s
    reelection prospects by including in CD 23 those precincts that voted for Senator John
    McCain in the 2008 Presidential election, even though he recognized the possibility that
    these precincts voted for Senator McCain because Anglo voters preferred Senator
    McCain and turned out at higher rates than Hispanic voters. Id. at 109-10; Defs.’ Ex. 577
    (Trial Tr. 956, Perez v. Perry, Sept. 9, 2011 (Downton)); Defs.’ Ex. 778A (Downton Dep.
    76-77, Aug. 12, 2011). Mr. Downton testified, however, that he “never looked at turnout
    data for any map.” Trial Tr. 89, Jan. 18, 2012 AM (Downton).
    12
    47. To address the overpopulation in CD 23 of approximately 149,000 people, mapdrawers
    moved over 600,000 residents in and out of the district. Defs.’ Ex. 575 (Trial Tr. 450,
    Perez v. Perry, Sept. 7, 2011 (Flores)); Defs.’ Ex. 436. The Congressional Plan adds
    approximately 33,000 people from traditionally Anglo counties along Benchmark CD
    23’s northern border. Id. at 448; Defs.’ Ex. 430, at 1. Chairman Seliger testified that he
    did not know why some of these counties were added to CD 23 because it was done by
    his counterparts in the State House, but stated that no study was done in these counties to
    determine if the Republican primary voters would support a Hispanic candidate. Defs.’
    Ex. 776 (Seliger Dep. 31, 36, Sept. 1, 2011); Trial Tr. 15, Jan. 24, 2012 AM (Seliger).
    48. Instead of adding population from Anglo counties in the northern part of CD 23 – north
    of the Pecos river – Chairman Seliger testified that the excess population in CD 23 could
    have been addressed by simply moving CD 23 down toward the border with Mexico,
    without extending the district northward. Defs.’ Ex. 776 (Seliger Dep. 38, Sept. 1,
    2011); Trial Tr. 20-21, Jan. 24, 2012 AM (Seliger). Chairman Seliger acknowledged that
    if CD 23 were pulled down closer to the border, Hispanic voters would “determine[] the
    outcome” of the election in CD 23. Defs.’ Ex. 776 (Seliger Dep. 38, Sept. 1, 2011); Trial
    Tr. 20-21, Jan. 24, 2012 AM (Seliger). Mr. Downton similarly testified that because CD
    23 lies adjacent to the border with Mexico and New Mexico, it is mathematically possible
    to achieve the ideal population in CD 23 by removing precincts from the northern and
    western part of the district. Defs.’ Ex. 778A (Downton Dep. 85, Aug. 12, 2011).
    49. In addition to adding population from Anglo counties to the north of CD 23, over
    300,000 people in Bexar County were moved out of, and about 60,000 individuals in
    Bexar County were moved into CD 23. Defs.’ Ex. 575 (Trial Tr. 485, Perez v. Perry,
    Sept. 7, 2011 (Flores)); Defs.’ Ex. 436.
    50. At the same time that he attributed population shifts in CD 23 as furthering the goal of
    making the district safer for Congressman Canseco, Mr. Downton also testified that
    changes were made to CD 23 in Bexar County to accommodate requests by State
    Representative Jose Menendez and Congressman Charles Gonzales for CD 20 and CD
    35, a new congressional district in the Bexar County area. Trial Tr. 78-79, Jan. 18, 2012
    AM (Downton). These requests with respect to CD 23 in the San Antonio area,
    according to Mr. Downton, “dropped the HCVAP of [CD] 23 below the Court [drawn]
    level” and required “other changes to [CD] 23 in other parts of the map to try to bring it
    back up. So it was kind of a constant ripple between 20, 23, 35 and to a lesser extent 21,
    and it might be 15, and other districts coming into Bexar County to try to get all of that to
    work.” Id. In order to increase the HCVAP and SSVR of CD 23 to benchmark levels,
    Mr. Downton testified that he altered the boundary between CD 16 and CD 23 near El
    Paso County, and made changes to the “southern region” of CD 23. Id. at 81-83.
    Specifically, Mr. Downton testified that he split Maverick County at the southern end of
    enacted CD 23 and moved half of that County into enacted CD 28 in order to raise
    enacted CD 23’s HCVAP level. He did this, in part, because he did not want to split
    Webb County, given previous litigation regarding a split of Webb County in LULAC v.
    Perry. Id. at 83-84.
    13
    iv.     The Splitting of Maverick County
    51. In the Benchmark Plan, Maverick County, and its most populous city, Eagle Pass, are
    entirely contained in CD 23. Defs.’ Ex. 428, at 4. The Congressional Plan, however,
    moves half of Maverick County from CD 23, splitting the city of Eagle Pass between CD
    23 and CD 28. Id.; Defs.’ Ex. 340, at 1; Defs.’ Ex. 575 (Trial Tr. 447, Perez v. Perry,
    Sept. 7, 2011 (Flores)); Defs.’ Ex. 430, at 1.
    52. Maverick County is located along the Mexican border and is among the “poorest counties
    in the United States.” Trial Tr. 113, Jan. 18, 2012 PM (Saucedo). The County Judge for
    Maverick County, David Saucedo, testified that despite their relative poverty, “the
    citizens of Maverick County have been educated on the electoral process. They’re aware
    of the fact of the investments that are made in that district. They’re aware of the fact [of]
    the money that’s invested by candidates to run in that district. And Maverick – the people
    in Maverick County understand that you can actually have a larger margin come [from]
    one community like Maverick County than you would in all of the San Antonio portion
    that is represented by that congressman. So that is what has given a community, a mid-
    sized community like ours, more influence.” Id. at 118. Judge Saucedo further testified
    that the Maverick County community is united and “[w]hen we go out, . . . we vote for
    one candidate and we’ve finally seen some of that change come about. We’re fighting for
    four-year universities, we’re fighting for veterans clinics, things that don’t exist in
    Maverick County that actually exist in smaller communities outside Maverick County.”
    Id. at 115.
    53. The Congressional Plan splits Maverick County in half between enacted CD 23 and CD
    28. Defs.’ Ex. 428, at 4; Defs.’ Ex. 340, at 1. During his testimony at trial, Mr. Downton
    could not remember how he split Maverick County, but believed “a large part of it
    follows the road . . . it was essentially just cutting the county in half.” Trial Tr. 85, Jan.
    18, 2012 AM (Downton). Mr. Downton later conceded, however, that the split of
    Maverick County in the enacted plan does not follow just one road and also resulted in at
    least three precinct cuts. Id. at 114; Defs.’ Ex. 575 (Trial Tr. 449, Perez v. Perry, Sept.
    7, 2011 (Flores)).
    54. Mr. Downton indicated that he was not aware that he cut the city of Eagle Pass in half
    when he split Maverick County. Trial Tr. 86, Jan. 18, 2012 AM (Downton). In any
    event, he appeared to discount the impact of this decision, stating his belief that “there’s
    roughly a thousand people that live there. So it didn’t change the nature of either
    district.” Id. The City of Eagle Pass actually has a population of 26,248 and is 95.5 %
    Hispanic. Defs.’ Ex. 391, at 1012.
    55. Mr. Downton testified that he removed portions of Maverick County from CD 23 because
    Maverick County does not have a good record of voting Republican. Defs.’ Ex. 577
    (Trial Tr. 963, Perez v. Perry, Sept. 9, 2011 (Downton)); Defs.’ Ex. 778A (Downton
    Dep. 87-90, Aug. 12, 2011). In the 2010 general election, Ciro Rodriguez, the candidate
    of choice of Hispanics, won 80.29 % of the vote in Maverick County and Congressman
    14
    Canseco won only 15.64 % of the vote. Defs.’ Ex. 393, at 3. In the 2010 Republican
    Primary Election, Congressman Canseco received only 23.07 % of the vote in Maverick
    County. Id. at 4. Judge Saucedo testified that for the past ten years Maverick County has
    turned out about 12,000 to 14,000 voters for presidential elections, and 8,000 to 9,000
    voters in other elections, and they vote heavily for the Hispanic-preferred candidate.
    Defs.’ Ex. 576 (Trial Tr. 771, Perez v. Perry, Sept. 8, 2011 (Saucedo)); Defs.’ Ex. 576
    (Trial Tr. 681, Perez v. Perry, Sept. 8, 2011 (Korbel)). Splitting Maverick County,
    according to Judge Saucedo, could make the difference in an election. Defs.’ Ex. 576
    (Trial Tr. 771, Perez v. Perry, Sept. 8, 2011 (Saucedo)).
    v.     Hispanic Citizens’ Ability to Elect in Congressional District 23 in the
    Enacted Plan
    56. In the Congressional Plan, CD 23 is 67.8% Hispanic, with an HCVAP of 58.5% and an
    SSVR of 54.8%. Pl.’s Ex. 12, at 6, 11. The Congressional Plan slightly increases CD
    23’s HCVAP by 0.01% and its SSVR by 2.2% over the Benchmark. See Defs.’ Ex.
    575 (Trial Tr. 454-55, Perez v. Perry, Sept. 7, 2011 (Dr. Flores stating that “even
    though the number SSVR is higher I don’t consider it a Hispanic opportunity district at
    all. I think that a Hispanic candidate would find it very difficult to get elected in the
    new configuration.”)); Pl.’s Ex. 11; Pl.’s Ex. 12.
    57. The evidence demonstrates that mapdrawers sought to ensure that the overall
    performance of Hispanic candidates of choice would decrease. On May 28, 2011,
    Messrs. Downton, Davis, and Interiano had an email exchange regarding the Attorney
    OAG’s election analysis results for the Congressional Plan, in which Mr. Interiano
    asked, “Any guidance on your 23. Have you been able to make any of the changes that
    we all discussed?” Mr. Downton responded, “Have it over 59 % HCVAP, but still at
    1/10. There has to be some level of HCVAP where it doesn’t make a difference what the
    election results are. It is more Hispanic than the other two San Antonio based districts . .
    . . ” Defs.’ Ex. 903, at 1. In this email, mapdrawers referenced the OAG’s
    reconstituted election analysis, which indicated that candidates supported by Hispanics
    dropped from winning three out of ten elections in the Benchmark Plan, to one out of
    ten in enacted CD 23. Pl.’s Ex. 65; Defs.’ Ex. 390.
    58. Mr. Interiano conceded that enacted CD 23 does not perform as a minority ability
    district. Defs.’ Ex. 779A (Interiano Dep. 96-97, Aug. 2, 2011). David Hanna and
    Jeffrey Archer from the TLC expressed concern that CD 23 was not “really effective in
    the proposed map.” Defs.’ Ex. 288. The goal of changes to CD 23 was, in fact, to make
    the district safer for Congressman Canseco, who is not the Hispanic candidate of choice.
    59. Mr. Downton, however, expressed little concern about the performance of CD 23. In
    proceedings before the U.S. District Court for theWestern District of Texas, Mr.
    Downton testified that he did not consider political performance as particularly relevant.
    Defs.’ Ex. 577 (Trial Tr. 966, Perez v. Perry, Sept. 9, 2011 (Downton)). He would
    classify a district as a majority-minority district if it elected the minority candidate of
    choice three out of ten times or one out of ten times because he believes “that any
    15
    district where the Hispanic citizen voting age population exceeds 50 percent, it is, by
    definition, a Hispanic opportunity district.” Id.; Defs.’ Ex. 778A (Downton Dep. 24-25,
    Aug. 12, 2011). Notwithstanding Texas’s position before this Court that CD 23 is an
    ability district both in the Benchmark and in the enacted plan, Mr. Downton does not
    view it as such. He testified that he believed CD 23 “was not an ability to elect district .
    . . before or afterward. It was performing in three out of ten elections before, and one of
    ten afterward, so in neither case was it performing.” Trial Tr. 87, Jan. 18, 2012 AM
    (Downton). The Chairmen of the Redistricting Committees, however, testified
    otherwise. Chairman Seliger testified that no one had told him that CD 23 in the
    Congressional Plan was predicted to elect the Hispanic preferred candidate in only one
    out of ten elections. Defs.’ Ex. 776 (Seliger Dep. 24, Sept. 1, 2011); Trial Tr. 13-14,
    Jan. 24, 2012 AM (Seliger). Chairman Solomons similarly testified that he would
    consider it problematic if a new congressional plan were to reduce the number of wins
    by the minority candidate of choice by three or more in a VRA protected district. He
    stated that it would also be a problem if the number of wins went from three in the base
    plan down to one and would necessitate a change to the plan. Defs.’ Ex. 580 (Trial Tr.
    1605-07, Perez v. Perry, Sept. 13, 2011 (Solomons)). He further testified that if an
    election analysis reduces the number of wins for minority preferred candidates by one
    out of ten, it would get his attention and that it was his understanding that legislative
    council was using that as a basis of their analysis. Trial Tr. 89, Jan. 20, 2012 PM
    (Solomons).
    60. The ability of Hispanic voters to elect their candidate of choice is lost in enacted CD 23.
    b. Congressional District 25
    61. In the Benchmark Plan, CD 25 draws 59.7% of its population from south Austin in
    Travis County, and also incorporates counties southeast of Austin, including Caldwell,
    Colorado, Fayette, Gonzales, Hays, and Lavaca counties, as well as portions of
    Bastrop County. Pl.’s Ex. 11. The District as configured in the Benchmark Plan is
    38.8% Hispanic, 8.7% Black, and 49.8% Anglo. The citizen voting age population
    (“CVAP”) is 25.3% Hispanic, 9.1% Black, and 63.1% Anglo. Id. at 7, 9. The SSVR
    in the District is 20.4%. Id. at 10. As reflected by the above statistics, the combined
    minority citizen voting age population totals 34.4% and Anglos constitute a majority
    of voters in the district.
    62. CD 25 is currently represented by Congressman Lloyd Doggett. Defs.’ Ex. 802; Trial
    Tr. 115, Jan. 19, 2012 PM (Dukes). Congressman Doggett won the special election
    for CD 25 in December 2006, and was reelected in 2008 and 2010.
    63. Congressman Doggett is the candidate of choice of minority voters in CD 25. Trial Tr.
    101, Jan. 19, 2012 PM (Dukes). A tri-ethnic coalition of Black, Hispanic, and cross-
    over Anglo voters work together to elect Democratic candidates in the area that CD 25
    16
    encompasses.10 Id. at 85.
    64. The success of this tri-ethnic coalition depends on support of some Anglos for
    Democratic candidates. Trial Tr. 86, Jan. 19, 2012 PM (Dukes).
    i. Minority Election Performance
    65. Despite the fact that Anglos comprise 63.1% of the CVAP, “the candidate preferred by
    Blacks and Hispanics [in CD 25 in the Benchmark] has won every congressional
    election this decade.” Defs.’ Ex. 724,at 5 (Ansolabehere Reb. Rep Jan 16, 2012);
    Pl.’s Ex. 11.
    66. Elected officials from areas encompassed by CD 25 testified at trial about the
    effectiveness of the tri-ethnic coalition and the role of minority voters within that
    coalition. State Representative Dukes testified that candidates supported by the tri-
    ethnic coalition are the ones who win in Travis County. Trial Tr. 104, Jan. 19, 2012
    PM (Dukes). Candidates are not able to bypass minority voters, and those who only
    obtain endorsements from Anglo groups in the tri-ethnic coalition do not win elections
    in Travis County. Id. at 106 (“[I]n general elections in Travis County [] if you do not
    win the Hispanic and African-American boxes that are largely located in the central
    portion of Travis County, then you are not going to win an election in Travis County
    without the progressive Anglo[,] black and Hispanic communities. I may not have an
    Excel spreadsheet, but I can tell you I know my county.”). As an example,
    Representative Dukes testified that Nelda Wells Spears, an African-American
    supported by the coalition, successfully defeated an Anglo male “progressive
    Democrat” with 74% of the vote. Id. at 112.
    67. In addition to Representative Dukes, David Escamilla, the Travis County Attorney,
    provided unrebutted written testimony that political cohesion and cooperation in the
    tri-ethnic coalition “consistently produces broad agreement to support individual
    candidates and slates of candidates. The high frequency of agreement on candidates
    among the organizations within the Coalition also stems from the fact that many
    individuals are members of more than one of the organizations. This overlap in
    membership promotes agreement on common slates of political candidates.” Defs.’
    Ex. 735, at 7. He provided the example of the 2008 election, in which an Anglo
    Assistant County Attorney lost a race for a county judgeship despite having “the lion’s
    share of endorsements from the local Democratic clubs” because he was “unable to
    10
    Representative Dukes testified that this coalition includes “multiple democratic organizations. There is the
    Black Austin Democrats, the Tejano Democrats, the Mexican-American Democrats, the lesbian-gay, or
    Stonewall Democrats, there's the Central Austin Progressives, there's the University Democrats, the Northwest
    Democrats that help northeast, and the list goes on and on and on and on, coupled with labeled [sic]
    organizations, especially the Central Labor Counsel [sic] made up of 15 labor unions, the police association, fire
    fighters, all working together, and the candidates work very hard to get the endorsements because they will go out
    and work the community, through literature and create a slate. It is very rare, if you have that coalition’s support
    that you are not successful in winning.” Trial Tr. 85, Jan. 19, 2012 PM (Dukes).
    17
    gain significant support from the Hispanic or African American community.” Id. at 9-
    10.11
    ii.        Congressional District 25 in the Congressional Plan
    68. In the enacted plan, CD 25 is significantly altered. While CD 25 in the Benchmark
    extended southeast of Travis County, CD 25 in the enacted plan takes a smaller
    population from Travis County and extends north to Tarrant County. Compared to its
    Benchmark configuration, CD 25 in the enacted plan loses population from south
    Austin and five counties and gains eleven counties. CD 25 in the enacted plan no
    longer incorporates Bastrop, Caldwell, Colorado, Fayette, Gonzales, and Lavaca
    counties, and now includes Bosque, Burnet, Coryell, Hamilton, Hill, Johnson,
    Lampasas, and Somervell counties, as well as portions of Bell, Erath, and Tarrant
    counties. Compare Pl.’s Ex. 11 with Pl.’s Ex. 12.
    69. State House Representative Dawnna Dukes testified that the Congressional Plan “takes
    the historical African-American community that was forced by segregation into
    central Austin and moved it into a majority Republican district that runs west . . . .”
    Trial Tr. 129, Jan. 19, 2012 PM (Dukes).
    70. CD 25 in the Benchmark Plan was overpopulated by 115,893 voters, or by 16.59%,
    and needed to shed this excess population. Pl.’s Ex. 11. Compared to the Benchmark,
    enacted CD 25 retained only 126,507 of the District’s original voters, lost 489,434,
    and added 392,869 voting age persons. Defs.’ Ex. 724, tbl. C.2 (Ansolabehere Rep.
    Oct. 21, 2011). In sum, only 28% of the voters in enacted CD 25 are from the
    Benchmark district. Id. at 39.
    71. In the Congressional Plan, CD 25 is 70.3% Anglo, 17.3% Hispanic, and 8.3% Black.
    Pl.’s Ex. 12. The CVAP is 78.2% Anglo, 10.3% Hispanic, and 8.1% Black. Id. In
    short, the citizen voting age population of Hispanics was cut by more than half and of
    Blacks was reduced by half a percentage point, while the population of Anglos was
    increased by over fifteen percentage points. In addition, the Anglo population in the
    new areas added to CD 25 “shows high levels of racial cohesion and polarization” and
    “85 % of Whites in this new district vote for the same candidate.” Defs.’ Ex. 724, at
    35-36 (Ansolabehere Rep. Oct. 21, 2011). In contrast to the new areas added to CD
    25, the small area that remains from the old district votes 60% for the minority-
    preferred candidates. Id. at 39. According to Dr. Ansolabehere, CD 25 in the enacted
    plan no longer provides minorities living in the district the ability to elect their
    candidates of choice. Id.
    c. Congressional District 27
    11
    Mr. Escamilla stated that since 2002 minority candidates have prevailed in 34 county-wide elections in Travis
    County, 18 of whom were Black and 16 were Hispanic. Defs.’ Ex. 735, at 8. He did not provide the total number of
    countywide elections from which this information is draw, which undermines the usefulness of this evidence in
    evaluating the ability of minority voters in the district.
    18
    i.      Congressional District 27 in the Benchmark Plan
    72. In the Benchmark Plan, CD 27 is located in southeastern Texas, and includes the cities
    of Corpus Christi and Brownsville, the counties of Kenedy, Kleberg, Willacy, and
    Nueces, as well as portions of Cameron and San Patricio counties. Pl.’s Ex. 11; Defs.’
    Ex. 575 (Trial Tr. 458, Perez v. Perry, Sept. 7, 2011 (Flores)); Defs.’ Ex. 818. Based
    on 2010 demographic data, CD 27 in the Benchmark had a total Hispanic population
    of 73.2%, an HVAP of 69.2%, an HCVAP of 63.8%, and an SSVR of 61.1%. Pl.’s
    Ex. 11.
    73. CD 27 is currently represented by Congressman Blake Farenthold, an Anglo
    Republican. Congressman Farenthold has been representing CD 27 since 2010, when
    he defeated twenty-seven year incumbent Solomon Ortiz, a Hispanic Democrat. Pl.’s
    Ex. 32, at 13; Trial Tr. 16, Jan. 24, 2012 AM (Chairman Seliger). In the 2010
    election, Congressman Farenthold defeated Mr. Ortiz by only 775 votes and received
    51,001 votes, or 47.84 %, compared to Mr. Ortiz, who received 50,226 votes, or 47.11
    %. Pl.’s Ex. 32, at 13.
    74. Chairman Seliger recognized that Congressman Farenthold was not the Hispanic
    candidate of choice in CD 27. Defs.’ Ex. 776 (Seliger Dep. 20, Sept. 1, 2011); Trial
    Tr. 16-17, Jan. 24, 2012 AM (Seliger). Despite their inability to reelect Mr. Ortiz in
    2010, Hispanic citizens in Benchmark CD 27 elected their candidate of choice to the
    United States House of Representatives in 2004, 2006 and 2008. Defs.’ Ex. 327, at 5
    (Handley Congress Rep.).
    75. According to Texas’s expert, CD 27 had “performed” from the time of its creation for
    close to thirty years until the 2010 election. Defs.’ Ex. 581 (Trial Tr. 1870-71, Perez
    v. Perry, Sept. 14, 2011 (Alford)). Indeed, Chairman Seliger testified that CD 27 in
    the Benchmark Plan is “clearly an opportunity district.” Defs.’ Ex. 776 (Seliger Dep.
    25-26, Sept. 1, 2011); Trial Tr. 17-18, Jan. 24, 2012 AM (Seliger). Chairman
    Solomons similarly understood that CD 27 was protected under the VRA. Defs.’ Ex.
    777 (Solomons Dep. 153, Aug. 31, 2011).
    ii.     Congressional District 27 in the Congressional Plan
    76. According to 2010 Census data, CD 27 in the Benchmark Plan was overpopulated by
    about 43,000 people. Trial Tr. 99, Jan. 18, 2012 AM (Downton); Pl.’s Ex. 11. In the
    Congressional Plan, CD 27 is reconfigured and moved north, keeping the city of
    Corpus Christi and adding the cities of Victoria, Wharton, and Bay City, but
    eliminating Brownsville from the district. Mr. Downton acknowledged that CD 27 in
    the Benchmark Plan and CD 27 in the Congressional Plan are very different districts.
    Defs.’ Ex. 577 (Trial Tr. 971, Perez v. Perry, Sept. 9, 2011 (Downton)); Defs.’ Ex.
    778B (Downton Dep. 48, Aug. 31, 2011). The Congressional Plan removes the
    southern counties of Kenedy, Kleberg, Willacy, and Cameron from the district, and
    adds Aransas, Calhoun, Jackson, Lavaca, Matagorda, Refugio, Victoria, and Wharton
    counties, as well as parts of Bastrop, Caldwell, and Gonzales counties. Pl.’s Ex. 12.
    19
    77. Mr. Downton testified that he believed that CD 27 was a district protected by the VRA
    in the Benchmark but was no longer a majority Hispanic district in the Congressional
    Plan. Defs.’ Ex. 778A (Downton Dep. 32-33, Aug. 12, 2011); Defs.’ Ex. 778B
    (Downton Dep. 54, Aug. 31, 2011). Similarly, Texas’s expert, Dr. Alford, testified
    that CD 27 in the Congressional Plan “has flipped, in almost exactly the same way 23
    was flipped previously, so it is CD 27 this time that is flipped into being a majority . . .
    Anglo district.” Defs.’ Ex. 581 (Trial Tr. 1829-30, Perez v. Perry, Sept. 14, 2011
    (Alford)).
    78. In the Congressional Plan, CD 27 has a total Hispanic population of 49.5%, an HVAP
    of 45.1%, an HCVAP of 41.1%, and an SSVR of 36.8%. Defs.’ Ex. 859, at 2; Defs.’
    Ex. 881, at 1. When compared to the Benchmark Plan, the HVAP decreases by
    24.4%, SSVR decreases by 22.6%, and the HCVAP decreases by 22.7% in enacted
    CD 27.
    79. While enacted CD 27 no longer includes counties in South Texas, Nueces County
    remains in the district. Nueces County is thus no longer included in the South and
    West Texas configuration of Hispanic ability districts. Trial Tr. 103, Jan. 18, 2012
    AM (Downton). Mr. Downton testified that, Nueces County effectively is in a
    different district in the Congressional Plan than in the Benchmark Plan. Defs.’ Ex.
    778B (Downton Dep. 49, Aug. 31, 2011).
    80. Nueces County has a population of 340,223 and an HCVAP of 54.6%. Pl.’s Ex. 11;
    Defs.’ Exs. 883, 746B, 391. In the Benchmark Plan, Nueces County voters constitute
    over 50% of the total registered voters of CD 27, while in the Congressional Plan, they
    do not. Trial Tr. 119-20, Jan. 18, 2012 AM (Downton); Defs.’ Ex. 778B (Downton
    Dep. 54-55, Aug. 31, 2011).
    81. According to Mr. Interiano, a goal of the Congressional Plan was to allow Nueces
    County to anchor a congressional district. That said, Mr. Interiano testified that he did
    not know what portion of CD 27 voters were in Nueces County under the Benchmark
    Plan. Defs.’ Ex. 579 (Trial Tr. 1461-62, Perez v. Perry, Sept. 12, 2011 (Interiano));
    Defs.’ Ex. 779A (Interiano Dep. 112, Aug. 2, 2011).
    82. Mr. Downton conceded that because Benchmark CD 27 was overpopulated by only
    about 43,000 individuals, if it had simply been the State’s goal to maintain CD 27, he
    would have had to remove only a few precincts. Trial Tr. 119, Jan. 18, 2012 AM
    (Downton). Mr. Downton further testified that CD 27 was redrawn to give
    Congressman Farenthold a better chance of reelection. This could have been
    accomplished in the Congressional Plan by carving out a small portion of Nueces
    County containing the incumbent’s home and moving that portion into a northern
    district, leaving the bulk of Nueces County in a South Texas district. Defs.’ Ex. 778B
    (Downton Dep. 53-54, Aug. 31, 2011).
    20
    83. Chairman Seliger similarly testified that it is conceptually possible to take
    Congressman Farenthold’s neighborhood, which is located along Gulf Shore Drive in
    Corpus Christi, and pair it with counties to the north to make him a safer district,
    leaving the remainder of Nueces County in the district that runs south to Cameron
    County. Defs.’ Ex. 776 (Seliger Dep. 27-28, Sept. 1, 2011); Trial Tr. 19, Jan. 24,
    2012 AM (Chairman Seliger).
    84. Mr. Downton testified that the mapdrawers considered and rejected proposals to
    include Nueces County’s Hispanic population in the South Texas configuration of
    congressional districts. Trial Tr. 103-04, Jan. 18, 2012 AM (Downton). This decision
    was in large part a political choice. Id. at 104. According to Mr. Downton, he moved
    Nueces County north into CD 27 in part because “the Cameron County delegation in
    the House and the Senate had expressed a preference that they have a District
    anchored in Cameron County without Nueces so that their county would be the sole
    anchor point and could control the election.” The Cameron County delegation
    included State Senator Eddie Lucio, Jr., State Representative Eddie Lucio, III, and
    State Representative Renee Oliveira, who are all Democrats. Mr. Downton fulfilled
    these representatives’ requests by creating enacted CD 34, a new district that was
    intended to be an offset for the loss of CD 27.12 Trial Tr. 71, Jan. 18, 2012 AM
    (Downton); Trial Tr. 118, Jan. 18, 2012 AM (Downton); Defs.’ Ex. 575 (Trial Tr.
    485-86, Perez v. Perry, Sept. 7, 2011 (Flores)); Defs.’ Ex. 577 (Trial Tr. 971, Perez v.
    Perry, Sept. 9, 2011 (Downton)); Defs.’ Ex. 778A (Downton Dep. 31-32, 66, Aug. 12,
    2011).
    85. As configured in the Congressional Plan, CD 27 does not provide Hispanic citizens the
    ability to elect their candidates of choice.
    E.        Discriminatory Purpose in the Congressional Plan
    a. Disparate Impact on Minority Congresspersons
    i. Congressional District 9
    86. In the Benchmark Plan, CD 9 is located south of Houston and incorporates parts of
    Harris and Fort Bend counties. Pl.’s Ex. 11. This district provides Black and
    Hispanic citizens the ability to elect the candidate of their choice.
    12
    CD 34, one of four new districts created in the Congressional Plan, is located in southeast Texas, and includes
    De Witt, Goliad, Bee, Jim Wells, Kleberg, Kenedy, Willacy and Cameron counties, as well as portions of Hidalgo
    County, San Patricio County, and Gonzales County. Pl.’s Ex. 12, at 1. All parties agree that proposed CD 34
    provides Hispanic citizens living in the district the ability to elect candidates of their choice. Defs.’ Ex. 726, at 4.
    Specifically, CD 34 has an HVAP of 79%, an HCVAP of 71.7%, and an SSVR of 71.9%. Defs.’ Ex. 885; Pl.’s
    Ex. 12, at 9. Chairman Seliger testified that he created CD 34 because he felt he was required to create a Hispanic
    district in South Texas, particularly after the loss of CD 27. Defs.’ Ex. 776 (Seliger Dep. 25-26, Sept. 1, 2011);
    Trial Tr. 18, Jan. 24, 2012 AM (Chairman Seliger).
    21
    87. Based on 2010 demographic data, CD 9 is 36.7% Black and 42.4% Hispanic. The
    district has a BVAP of 36.3%, an HCVAP of 19.1%, and an SSVR of 16.2%. Pl.’s
    Ex. 11, at 4, 10-11. Congressman Al Green has represented CD 9 since 2005.
    88. In the Benchmark Plan, CD 9 has a surplus of 35,508 people, or 5.05 %. Defs.’ Ex.
    347, at 28. While this district was required to shed a small percentage of population,
    Congressman Green testified that his district had “substantial surgery” done to it. Trial
    Tr. 124-25, Jan. 20, 2012 AM (Congressman Green). Primarily Black communities,
    such as Hiram Clarke, were removed from his district, along with “economic
    engines,” such as the rail line, Houston Baptist University, the Medical Center, and the
    Astrodome. The removal of these areas substantially decreased the political power of
    the citizens in his district. Defs.’ Ex. 721, at 4; Trial Tr. 124-25, Jan. 20, 2012 AM
    (Congressman Green).
    89. In addition to removing key landmarks from his district, the Congressional Plan
    removes Congressman Green’s district office. Congressman Green testified that the
    “district office provides a meaningful connection between a member and the people
    represented. Our district office is in a location that is well-known to my constituents
    and has been in its present location since 2006; it has easy access to major freeways,
    mass transit, and many of the important centers of business activity within the Ninth
    Congressional District such as the Texas Medical Center, the VA hospital, and the
    Astrodome complex. Other similar properties in the area have been surgically
    removed; this couldn’t have been done by accident.” Defs.’ Ex. 721, at 4
    (Congressman Green Pre-filed Direct Testimony).
    ii.     Congressional District 18
    90. In the Benchmark Plan, CD 18 is located in Houston and within Harris County. Pl.’s
    Ex. 11. Based on 2010 demographic data, CD 18 in the Benchmark Plan is 43.5%
    Hispanic, 37.6% Black, and 15.8% Anglo. The district has a BVAP of 46.4%, an
    HCVAP of 22.3%, and an SSVR of 18.4%. Id. at 6, 9-10. Congresswoman Sheila
    Jackson Lee has represented CD 18 since 1995.
    91. Congresswoman Jackson Lee testified that during the 2011 redistricting process, she
    traveled to Texas to meet with the Chairmen of the House and Senate Redistricting
    Committees, and went to a public redistricting hearing to urge state lawmakers to
    respect communities of interest in CD 18. Trial Tr. 9-11, Jan. 23, 2012 PM
    (Congresswoman Jackson Lee). These requests, however, were unheeded and she was
    never contacted to discuss changes to CD 18. Id.
    92. Based on 2010 demographics, CD 18 in the Benchmark is over-populated by 22,503
    people, or 3.22%, which thus required only minor changes to reach the ideal
    population size. Defs.’ Ex. 347, at 29 (Murray Rep.). Nonetheless, in the
    Congressional Plan, the district’s key economic generators, as well as
    Congresswoman Jackson Lee’s district office are removed. Trial Tr. 13-14, Jan. 23,
    2012 PM (Congresswoman Jackson Lee). Congresswoman Jackson Lee testified that
    22
    her district office has been in the same location for a lengthy period of time, having
    been used by the previous two representatives of CD 18, including former
    Congresswoman Barbara Jordan. Consequently, constituents in the district know
    where the office is and go there to seek services. Id. In addition to removing her
    district office, the Congressional Plan also splits the historic Third Ward-MacGregor
    area, an important Houston community and home to many of Houston’s African-
    American leaders. This area has been in CD 18 since the district’s creation in 1972.
    Defs.’ Ex. 577 (Trial Tr. 1051, Perez v. Perry, Sept. 9, 2011 (Murray)); Trial Tr. 12-
    13, Jan. 23, 2012 PM (Congresswoman Jackson Lee).
    iii.     Congressional District 30
    93. In the Benchmark Plan, CD 30 is located in Dallas within Dallas County. Pl.’s Ex. 11.
    94. Based on 2010 demographic data, CD 30 is currently 42.4% Black, 39.7% Hispanic,
    and 16.7% Anglo. The district has a BVAP of 42.5%, an HCVAP of 19.8%, and an
    SSVR of 14.6%. Pl.’s Ex. 11, at 7, 9-10. Since 1992, Congresswoman Eddie Bernice
    Johnson has represented CD 30. Trial Tr. 67, 69, Jan. 18, 2012 PM (Congresswoman
    Johnson).
    95. Benchmark CD 30 has only 7,891 people over the ideal population, or 1.14%, and thus
    required only minor changes to reach the ideal population size. Pl.’s Ex. 11. Despite
    this fact, significant changes were made to CD 30, including the addition of a large
    prison, which artificially inflated the Black population in the enacted district. Trial Tr.
    81, Jan. 18, 2012 PM (Congresswoman Johnson); Defs.’ Ex. 579 (Trial Tr. 1276,
    Perez v. Perry, Sept. 12, 2011 (Congresswoman Johnson)).
    96. The Congressional Plan removes from the district Congresswoman Johnson’s district
    office and even her own home. Trial Tr. 79, Jan. 18, 2012 PM (Congresswoman
    Johnson). Congresswoman Johnson testified that the removal of her district office
    would be a significant loss to her community because her constituents are familiar
    with her office and it is easily accessible. Id. at 79-80.
    97. In addition to her district office and home, the Congressional Plan removes economic
    generators from the district, including areas that Congresswoman Johnson has worked
    to improve, such as the American Center, where the Dallas Mavericks play,
    transportation areas for the downtown park, and the arts district. Trial Tr. 81, Jan. 18,
    2012 PM (Congresswoman Johnson).
    iv.    Congressional District 20
    98. Hispanic Congressman Charlie Gonzalez represents CD 20. In the Congressional Plan,
    his district office is removed from CD 20. The enacted plan also removes key
    economic and cultural landmarks from Congressman Gonzalez’s district, including the
    Alamo and the Convention Center named after Congressman Gonzalez’s father.
    23
    Devaney Decl., Ex. 16 (Decl. of Congresman Charles Gonzales, ¶¶ 3-9, 11).
    v.        Comparative Treatment of Anglo Congresspersons
    99. While all three Black Congresspersons and Hispanic Congressman Gonzalez had their
    district offices removed from their re-configured districts, and Congresswoman
    Johnson even had her home removed, no Anglo Congressperson had his or her district
    office or home removed from his or her district as a result of the re-districting process.
    Trial Tr. 14, Jan. 18, 2012 PM (Congresswoman Jackson Lee); Trial Tr. 80, Jan. 18,
    2012 PM (Congresswoman Johnson).
    100. While minority Congresspersons had key landmarks in their districts removed, the
    mapdrawers accommodated requests from Anglo Congresspersons to include in their
    districts country clubs and grandchildren’s schools. Devaney Decl., Exs. 19-21. For
    example, mapdrawers ensured that Anglo Congresswoman Kay Granger, whose office
    originally had been drawn out of her district, had her office restored before adoption
    of the final plan. Devaney Decl., Ex. 5 ( Opiela Dep. 63, Aug. 22, 2011); id., Ex. 17.
    Anglo Congressman Kenny Marchant requested on May 31, 2011, that his district
    lines be drawn to cross a street to include his grandchildren’s school. Id., Ex. 18.
    Mapdrawers accommodated that request. Anglo Congressman Lamar Smith requested
    on June 8, 2011, that his district lines be drawn to include a precinct with the San
    Antonio Country Club. Id., Ex. 19. The Republican leadership also granted that
    request. Devaney Decl., Ex. 20.
    101. With regard to the removal of district offices, Mr. Interiano testified that the
    mapdrawers did not have the addresses of any congressional district offices when they
    were redrawing the congressional map. Mr. Interiano stated that it was just
    “coincidence” that minority Congresspersons had their district offices removed. Trial
    Tr. 95, Jan. 25, 2012 PM (Interiano). The Court finds that this testimony is not
    credible.
    b. Configuration of Congressional Districts in North Texas
    102. In the Benchmark Plan, nine congressional districts converge in the Dallas-Fort Worth
    metroplex. Defs.’ Ex. 818, at 1. Of these nine districts, CD 30 is the only minority
    ability district. Defs.’ Ex. 327, at 2 (Handley Congress Rep.).
    103. The Dallas-Fort Worth metroplex is located in north Texas, and is spread across Dallas
    and Tarrant counties. Between 2000 and 2010, the non-Anglo population in Dallas
    County grew by 28%, accounting for 100% of the population growth in the county.
    During the same period, the Anglo population declined by 20%. Defs.’ Ex. 734, at 5-6
    (Supp. Rep. of Rogelio Saenz). The non-Anglo population in Tarrant County grew 12
    times faster than the Anglo population, accounting for 89% of the population growth
    there. Id. According to 2010 census data, Blacks and Hispanics now account for a
    combined 55% of the voting age population in Dallas County and 37% of the voting
    24
    age population in Tarrant County. Mot. for Judicial Notice, ECF No. 180, ¶¶ 33, 37.
    104. Due to significant population growth in Dallas and Tarrant counties, the Congressional
    Plan allocates CD 33, one of the State’s newly apportioned congressional districts that
    will be Anglo controlled, to the area. Pl.’s Ex. 12. The Dallas-Fort Worth metroplex
    will thus have ten congressional districts converge in the area in the Congressional
    Plan, but CD 30 remains the only minority ability district among them. Id.
    105. Despite significant minority population growth, and the addition of another
    congressional district, the Congressional Plan does not reflect the minority growth in
    the Dallas-Fort Worth metroplex. Trial Tr. 13, Jan. 18, 2012 PM (Rep. Veasey); Defs.’
    Ex. 320, ¶¶ 148-52 (Arrington Rep.). The Congressional Plan divides the urban,
    minority population in the Dallas-Fort Worth metroplex among four Anglo-controlled
    congressional districts, CD 6, CD 12, CD 26, and new CD 33. Trial Tr. 16-17, Jan. 18,
    2012 PM (Rep. Veasey); Defs.’ Ex. 320, ¶¶ 148-152 (Arrington Rep.); Defs.’ Exs. 677-
    80, 683-84; Trial Tr. 75, Jan. 18, 2012 PM (Congresswoman Johnson testifying that the
    Congressional Plan has been configured to break “solid African-American and Latino
    growth up in one, two, three, four, five or six different districts”).
    106. To rebut claims that minorities in the Dallas-Fort Worth metroplex were either
    fractured into Anglo-dominated districts or packed into CD 30, Mr. Downton testified,
    that it was difficult to draw a Hispanic district in the Dallas-Fort Worth metroplex
    because “a significant part” of the population growth was either non-citizen, under 18,
    or “assimilated.” Trial Tr. 74, Jan. 18, 2012 AM (Downton).
    107. The Republican congressional delegation, through Congressman Lamar Smith,
    MALDEF, MALC and Representative Veasey all presented Mr. Downton with
    “concepts” for North Texas, but Mr. Downton stated that none of these groups
    provided him with a proposed map during the regular or special session. Trial Tr. 73,
    Jan. 18, 2012 AM (Downton).
    108. Mr. Downton’s testimony on this issue is not accurate. In early April 2011,
    Congressman Lamar Smith, on behalf of a majority of the Texas Republican
    congressional delegation, distributed a draft congressional map to Republican leaders
    of the Texas Legislature, as well as the Lieutenant Governor and Governor. Defs.’ Ex.
    394. Congressman Smith’s map created “one new Voting Rights Act district in the
    Dallas-Ft. Worth area,” which, inter alia, “reflects the population growth in Texas over
    the last decade.” Id.
    109. Representative Veasey testified that when he heard through the local paper that the
    Congressman Lamar Smith and the Republican congressional delegation had proposed
    a map with another minority congressional district in north Texas, in addition to CD
    30, he approached Chairman Solomons and asked to see it. Chairman Solomons
    responded that there was no such map. Representative Veasey testified that he
    subsequently learned that Chairman Solomons had, in fact, seen Congressman Smith’s
    proposed map. Trial Tr. 15-16, Jan. 18, 2012 PM (Rep. Veasey).
    25
    i.       Congressional Districts 6, 12, and 33
    110. CD 6 in the Benchmark Plan is anchored in the heavily Anglo counties of Ellis and
    Navarro and reaches into both Dallas County and Tarrant County to include heavily
    Hispanic neighborhoods in Dallas County and areas of Tarrant County with rapidly
    growing Hispanic and Black population areas. Trial Tr. 21, Jan. 18, 2012 PM (Rep.
    Veasey); Defs.’ Ex. 819 at 1. CD 6 in the Benchmark has a combined Black and
    Hispanic CVAP of 38.6%, with most of the minority population in Dallas County.
    Defs.’ Ex. 857, at 2.
    111. CD 12 in the Benchmark is based in northern Tarrant County, which is comprised of
    affluent Anglo communities. The district also incorporates southeast Fort Worth,
    which is a Black community. Southeast Fort Worth is situated south of Interstate 30
    and east of Interstate 35 and is made up of several inner-city, low-income communities
    that are predominantly minority.
    112. CD 33 is one of the State’s newly apportioned congressional districts. In the
    Congressional Plan, this district includes all of Parker County and parts of Wise
    County, both of which are predominantly comprised of Anglo, suburban areas. Anglos
    make up 85.3% of the population in Parker County and the portion of Wise County
    included in CD 33 is 78.7% Anglo. Pl.’s Ex. 12. In addition to those Anglo areas, CD
    33 cuts into Tarrant County to include Tarrant County’s fast-growing minority
    populations. Representative Veasey testified that enacted CD 33 “goes around
    southwest -- underneath southeast Ft. Worth in the unincorporated Tarrant County, and
    then moves into Arlington, into the heavily Anglo part of Arlington, and then picks up
    the fast minority growth area in southeast Tarrant County, Arlington -- southeast
    Arlington-Grand Prairie area.” Trial Tr. 23, Jan. 18, 2012 PM (Rep. Veasey).
    ii.       Congressional District 26
    113. CD 26 in the Benchmark Plan covers parts of Cooke, Dallas, Denton, and Tarrant
    counties. Pl.’s Ex. 11. Benchmark CD 26 is anchored in Denton County, and then
    reaches south into the center of Tarrant County in a long peninsula-like strip,
    incorporating 363,872 individuals. The population of benchmark CD 26 in Tarrant
    County is 45.5% Anglo, 24.4% Black, and 26.5% Hispanic. Pl.s’ Ex. 12.
    114. In the Congressional Plan, CD 26 covers parts of three counties: Dallas, Denton, and
    Tarrant, with most of Denton County within the district. Pl.’s Ex. 12, at 1, 7. The
    portion of Denton County in CD 26 is 67.1% Anglo. Id.; Trial Tr. 18, Jan. 18, 2012
    PM (Rep. Veasey). CD 26 also includes a small portion of Dallas County containing
    841 individuals, who are 43.9% Anglo. Pl.’s Ex. 12, at 7. In addition to these Denton
    County and Dallas County areas, CD 26 in the Congressional Plan runs down the
    center of Tarrant County in an exaggerated “lightning bolt” shape to capture 147,815
    individuals, 65.2% of whom are Hispanic. Id.; Defs.’ Ex. 75.
    26
    Benchmark CD 26                                         Enacted CD 26
    115. The “lightning bolt” into Tarrant County divides two significant minority communities
    of interest in Tarrant County – North Side and South Fort Worth – and moves these
    areas to CD 12, a district represented by Anglo Republican Congresswoman Kay
    Granger. North Side is an urban, low-income, majority Hispanic community in Fort
    Worth. South Fort Worth is another urban, low-income, majority Hispanic community
    in Fort Worth. Trial Tr. 19, Jan. 18, 2012 PM (Rep. Veasey); Trial Tr. 98-99, Jan. 18,
    2012 PM (Jiminez).
    116. The boundary between enacted CD 26 and enacted CD 12 in Tarrant County – the
    eastern boundary of the “lightning bolt” – divides minority communities according to
    race. Defs.’ Ex. 630; Pl.’s Ex. 133.
    117. The “lightning bolt” running through Tarrant County in the Congressional Plan
    contains 38 splits of voter tabulation districts (“VTD”).13 Defs.’ Ex. 875, at 10-11.
    The purpose behind the split VTDs was to move Hispanic populations into enacted CD
    26 and split the non-Hispanic population out of the district. Defs.’ Ex. 887 at 74-82,
    185-88. Mr. Downton testified that he drew the map to keep the Hispanic population
    together, even though he also testified that these Hispanic populations may not want to
    be submerged into Denton County. Defs.’ Ex. 778A (Downton Dep. 130-131, Aug.
    12, 2011).
    118. In an effort to explain the configuration of the “lightning bolt” in the Congressional
    Plan, Mr. Downton testified that the “lightning bolt” running from Denton County to
    Tarrant County went through “multiple iterations and changes based on concerns raised
    by various people throughout the process.” Trial Tr. 68, Jan. 18, 2012 AM (Downton).
    According to Mr. Downton, the “lightning bolt” went further down into Fort Worth
    because “concerns were raised that we had split the Hispanic population of the City of
    Fort Worth between a group in the . . . northern side of Fort Worth which we had put in
    13
    While not precisely the same, the parties agree that VTDs are essentially the same as voting precincts.
    27
    26 and a group down in the southern part of Fort Worth that we had put in 12. So to
    rectify that concern we reached down further . . . and . . . the two Fort Worth Hispanic
    communities which we were told shared a community of interest, we put them in 26.
    Initially when we did that, it looked a little cleaner coming down, but in doing that we
    had taken out downtown Fort Worth and the [Trinity Vision] project out of District 12.
    Congressman Granger expressed concern that she really needed those areas in her
    District. . . . Then we made an additional change over Representative Veasey’s request
    that primarily the black community we had put in 26 and he asked us to move that to
    12 and so we did that as well.” Id. at 68-69.
    119. The assertion that that the jagged edges in the “lightning bolt” were due to
    Congresswoman Granger’s request to keep the Trinity Vision Project in CD 12, is
    disputed by other evidence in the record. Specifically, Tarrant County Commissioner
    Roy Brooks stated in a memorandum, dated September 15, 2011, to the DOJ:
    “Frankly, you are not being told the truth . . . . I have no doubt that the State wants
    Trinity Vision to remain in CD 12. However, using the project to explain and excuse
    their racially discriminatory map is flatly dishonest.” Defs.’ Ex. 113. Mr. Brooks
    further stated that “[t]he contorted lines south of the Trinity Vision site reflect a careful
    effort to include Hispanic precincts and blocks in CD 26 while placing African
    American precincts and blocks in CD 12. Any difficulty in retaining Trinity Vision in
    CD12 was caused by the State placing a higher priority on separating black and
    Hispanic voters from each other.” Id.
    iii.        Packing of Minorities into Congressional District 30
    120. Evidence presented to the Court demonstrates that the Congressional Plan concentrates
    a large part of Dallas County’s minority population into enacted CD 30. Enacted CD
    30 has a BVAP of 45.6% and an HVAP of 40.3 %, which increases the combined
    minority voting age population from the Benchmark district by 4.8 %. Compare Defs.’
    Ex. 859, at 2 with Defs.’ Ex. 858, at 2.
    II. STATE SENATE PLAN
    121. There are 31 seats in the State Senate. Senators serve terms of four years.
    122. On July 24, 2001, following the failure of the Texas State Legislature to enact a
    redistricting plan for the State Senate, the Texas Legislative Redistricting Board
    adopted a plan to redistrict all 31 Senate seats. This plan was precleared by the DOJ on
    October 15, 2001. This is the Benchmark State Senate Plan (the “Benchmark Plan”)
    for the purposes of this case.
    123. Redistricting maps for the State House and State Senate must be passed during the
    general legislative session. Otherwise, the maps are drawn by a Legislative
    Redistricting Board that is designated by statute and consists of the State’s Lieutenant
    Governor, Speaker of the House, Attorney General, Comptroller, and Land
    Commissioner.
    28
    124. On May 17, 2011, the State Senate passed Senate Bill 31, containing a new redistricting
    plan for the State Senate based on the 2010 Census, and the Governor signed it into law
    on June 17, 2011 (the “State Senate Plan”). This is the Plan for which Texas is seeking
    preclearance.
    125. SD 10 is represented by Senator Wendy Davis in the Benchmark Plan and its
    configuration in the State Senate Plan is the only challenge to the State Senate Plan
    before the Court.
    A. State Senate Redistricting Process
    126. Doug Davis, the director for the Senate Select Committee on Redistricting, was the
    principle mapdrawer for the State Senate Plan. Trial Tr. 140, Jan. 17, 2012 PM (D.
    Davis).
    127. In September 2010, the Senate Redistricting Committee held seven field hearings across
    the State to “receive input from the public” on redistricting. Trial Tr. 145, Jan. 17,
    2012 PM (D. Davis).
    128. Redistricting hearings for the State Senate Plan were held in “population centers”
    around the State, but none was held in Tarrant County by the Senate Redistricting
    Committee. Trial Tr. 17, Jan. 18, 2012 AM (D. Davis).
    129. The State House Committee on Redistricting held the only hearing in Tarrant County in
    Arlington, which has the dubious distinction of being the largest city in the United
    States that lacks both public bus and rail service, Trial Tr. 8, Jan. 18, 2012 PM (Rep.
    Veasey), and therefore is not fully accessible without private transportation. As
    discussed, Rep. Veasey specifically asked that a public hearing be held in Fort Worth
    and offered to locate an appropriate site, but his request was ignored. Id. at 8-10.
    130. Senator Judith Zaffirini, a Hispanic who served on the Senate Redistricting Committee
    in 2011 and who had been through several past Senate redistricting cycles, said that the
    field hearings were “a sham” because of low attendance and participation, lack of
    invited testimony, and lack of prepared materials for members of the Redistricting
    Committee. Defs.’ Ex. 189 (Zaffirini Dep. 7-8, Jan. 6, 2012). Senator Rodney Ellis
    similarly testified that these hearings had very limited attendance and were “fairly
    perfunctory.” Trial Tr. 94-95, Jan. 20, 2012 AM. Both Senators’ testimony is credited
    by this Court.
    131. Chairman Seliger and Doug Davis met with Senator Davis in March 2011 and asked her
    what changes she would like made to SD 10. Trial Tr. 35, Jan. 20, 2012 AM (Sen.
    Davis). She told them that the urban cores of Fort Worth and Arlington were “very
    important” to the District and that she felt “it was important to keep the district wholly
    contained within Tarrant County.” Id.
    29
    132. During late April 2011, draft redistricting maps were available for viewing in a room
    adjacent to the State Senate floor, but only by invitation. Those senators who were
    invited to look would leave the floor of the State Senate with Chairman Seliger and Mr.
    Davis to review the draft proposals and provide comments on them. Trial Tr. 39, Jan.
    20, 2012 AM (Sen. Davis).
    133. Senator Ellis testified that senators who represented “minority districts” were left out of
    the redistricting process. Trial Tr. 95, Jan. 20, 2012 AM (Sen. Ellis). Senator Zaffirini
    testified that Anglo senators had access to view the plans for their districts and the
    overall State Senate Plan but minority senators did not. Defs.’ Ex. 809 (Zaffirini Dep.
    29-30, Jan. 6, 2012). She characterized the redistricting in 2011 as the “least
    collaborative and most exclusive” she had ever experienced. Defs.’ Ex. 134, Lichtman
    Rep. app. 7 (Decl. of Judith Zaffirini, ¶ 3).
    134. Throughout April and May, Senator Davis constantly asked Mr. Davis and Chairman
    Seliger to see the map for her district. Trial Tr. 38-39, 42, Jan. 20, 2012 AM (Sen.
    Davis). Senator Davis was not shown any drafts and it was not until May 10, 2011 that
    she saw her district for the first time. Id. at 42; Defs.’ Ex. 128. On that date, Senator
    Davis sent a letter to Chairman Seliger, expressing concern that minority voting rights
    were badly undermined by the State Senate Plan and that excluding her from the
    process, as the representative of several minority communities, contributed to this
    result. Defs.’ Ex. 128.
    135. Mr. Davis explained that he did not show Senator Davis how her district was re-drawn
    because “we were not printing maps and giving them to members.” Trial Tr. 172, Jan.
    17, 2012 PM (D. Davis); see also id. at 173. On the contrary, Chairman Seliger
    admitted that he provided maps to three other senators who represent majority-Anglo
    districts. Trial Tr. 67-68, Jan. 24, 2012 AM (Chairman Seliger); see also Defs.’ Ex.
    646 (Dep. of Sen. Jane Nelson 10-11, Jan. 6, 2012).
    136. Senator Davis appeared at the May 12, 2011 Senate Select Committee on Redistricting
    hearing to testify against the State Senate Plan. Trial Tr. 9-10, Jan. 18, 2012 AM (D.
    Davis). Because Senator Davis was not a member of the Committee, she could not
    propose amendments in Committee. Senator Zaffirini sponsored amendments on her
    behalf. Trial. Tr. 43-45, Jan. 20, 2012 AM (Sen. Davis). These amendments were
    designed to “strengthen[] the African American and Latino makeup of [SD 10].” Id. at
    45. Neither of these plans passed the committee vote. Senator Davis again offered
    amendments on the floor of the State Senate, but was defeated in a “party line” vote.
    Trial. Tr. 12, Jan. 18, 2012 AM (D. Davis).
    137. The formal Senate redistricting process began on May 12, 2011 and the bill was passed
    to the State House on May 18, 2011, six days later. Defs.’ Ex. 156. The State Senate
    passed the redistricting plan by a roll call vote of 29-2. Only Senators Davis and Ellis
    voted against it.
    30
    138. On Wednesday, May 11, 2011, the day before the public hearing on the State Senate
    Plan, David Hanna an attorney at the TLC, responded to an email from Karina Davis,
    the Senate Parliamentarian (and Doug Davis’s wife), with a copy to Mr. Davis. Ms.
    Davis had inquired about “pre-doing the committee report,” Defs.’ Ex. 359, but Mr.
    Hanna advised “No bueno [no good]. RedAppl time stamps everything when it assigns
    a plan. Doing it Thursday would create paper trail that some amendments were not
    going to be considered at all. Don’t think this is good idea for preclearance.” Id.
    139. Although Chairman Seliger testified that he never knew about the Davis-Hanna email,
    he did agree that he knew on May 11, 2011 that none of Senator Davis’s proposed
    amendments to the State Senate Plan would pass. Trial Tr. 70-71, Jan. 24, 2012 AM
    (Chairman Seliger).
    B. Senate District 10
    140. SD 10 in the Benchmark is a geographically compact district located entirely within
    Tarrant County that includes most of Fort Worth, Texas.
    141. The 2010 Census reported that SD 10’s population is 19.2% Black, 28.9% Hispanic,
    4.9% other minorities (i.e., approximately 53% minorities) and 47.6% Anglo. Defs.’
    Ex. 151, at 5. However, the 2010 Census showed an 18.3% Black Citizen Voting Age
    Population (“BCVAP”), 15.1% HCVAP, 62.7% White Citizen Voting Age Population
    (“WCVAP”), and a 2.6% Asian-American citizen voting age population in the district.
    Pl.’s Ex. 15, at 8.
    142. In 2006, an Anglo Democrat, Terri Moore, ran for District Attorney in Tarrant County
    but lost with close to 50% of the vote. Trial Tr. 30, Jan. 18, 2012 PM (Rep. Veasey).
    This election result caught the attention of State House Representative Marc Veasey,
    who studied the 2006 election results to see whether Black and Hispanic voters could
    elect a candidate of choice in SD 10. Representative Veasey concluded that “when
    African-American and Hispanic communities came together as a coalition to win, . . .
    they could win Senate District 10.” Id.
    143. Thereafter, a coalition of Black and Hispanic community leaders in SD 10 visited
    Wendy Davis, an Anglo with deep minority support who was then serving on the Fort
    Worth City Council, and asked her to run for the State Senate in SD 10 in 2008. Trial
    Tr. 16-19, Jan. 20, 2012 AM (Sen. Davis).
    144. During her campaign, Senator Davis “spent a great deal of time going into [Black and
    Hispanic] neighborhood meetings, knocking [on] doors in those communities and
    attending churches and speaking to church congregations in those communities.” Trial
    Tr. 21, Jan. 20, 2012 AM (Sen. Davis).
    145. According to the Chairman of the Texas State Democratic Party, Boyd Richie, “there
    was a concerted effort to build support from and mobilize African-American and
    Hispanic voters and to have them unite in their electoral support for Wendy Davis.”
    31
    Defs.’ Ex. 732, at 3 (Decl. of Boyd Richie). Senator Davis corroborated this testimony
    and is credited by this Court.
    146. Senator Davis had no primary opponent and ran against the Anglo Republican
    incumbent, Senator Kim Brimer, in the general election in 2008. Senator Davis
    testified that Senator Brimer was “incredibly well funded” and had “the endorsement of
    every mayor and the police and fire unions, and had mayors appearing in television
    commercials with him endorsing him.” Trial Tr. 67-68, Jan. 20, 2012 AM (Sen.
    Davis).
    147. Senator Davis won the election to the State Senate in 2008 by approximately 7,100 of
    the 288,000 votes cast in SD 10. Senator Davis received 147,832 votes (49.91%);
    former Senator Brimer received 140,737 votes (47.52%); and Libertarian Party
    candidate Richard Cross received 7,591 (2.56 %). Pl.’s Ex. 110, at 4.
    148. Dr. Alford, Texas’s expert witness, calculated that Senator Davis was elected with
    99.6% of the Black vote, 85.3% of the Hispanic vote, and 25.8% of the Anglo vote.
    Trial Tr. 32-33, Jan. 25, 2012 AM (Alford).
    149. The Court finds that the election of Senator Davis in 2008 demonstrated a successful
    three-way coalition of Blacks, Hispanics and some cross-over Anglos in SD 10. Since,
    however, there has been no occasion for Senator Davis to win reelection, and no
    evidence of the coalition’s success in other elections, the Court concludes that the
    ability of minorities to elect candidates of choice in SD 10 has insufficient history to
    afford it protection as a Benchmark ability district for purposes of redistricting in 2011.
    150. Nonetheless, the Court concludes that the record demonstrates purposeful
    discrimination in the re-drawing of SD 10.
    C. Dismantling of Senate District 10’s Minority Coalition
    151. SD 10 in the Benchmark is comprised of almost all the traditional and growing minority
    neighborhoods of Tarrant County in and around Fort Worth, including the historic
    Northside Hispanic area, the growing Southside Hispanic area, Defs.’ Exs. 138, 657;
    Trial Tr. 21-22, Jan. 18, 2012 AM (D. Davis), and the predominantly Black areas of
    Southeast Fort Worth, Forest Hill, and Everman. Defs.’ Exs. 136, 657; Trial Tr. 21-22,
    Jan. 18, 2012 AM (D. Davis).
    152. These areas were broken apart and placed into Anglo-controlled districts in the State
    Senate Plan, specifically enacted SDs 12 and 22. Defs.’ Ex. 141; Pl.’s Ex. 16 (showing
    that enacted SD 12 has a 61% Anglo population and a 75.5% WCVAP and enacted SD
    22 has a 61.3% Anglo population and a 72.3% WCVAP).
    153. The ideal district size for a senate district in the State Senate Plan is 811,147
    individuals. Trial Tr. 149, Jan. 17, 2012 PM (D. Davis). The 2010 Census showed
    Benchmark SD 10 to have a population of 834,265, which is 23,118 more than the
    32
    ideal number for a senate district in Texas. Defs.’ Ex. 151, at 5. The additional
    population in Benchmark SD 10, however, is well within the populations deviation
    accepted for redistricting in the State Senate Plan by the State and there is no evidence
    this “over-population” played any part in redrawing the district. See Pl.’s Ex. 35 at 32.
    154. The maps below show Benchmark SD 10 and enacted SD 10 and the surrounding
    districts, particularly SDs 12, 9, and 22.
    Benchmark SD 10                                       Enacted SD 10
    155. A closer examination of the area of Fort Worth (below) clarifies the cracking of the
    minority communities from SD 10 in the State Senate Plan. An excerpt of the map
    above is below depicting Fort Worth and its surrounding areas, which are enclosed
    within the loop shape. The loop, which represents Interstate 820, is intersected by
    Texas State Highway 30 running east to west and Interstate 35 running north to south,
    which create four quadrants within the loop.
    12
    22
    Benchmark SD 10
    33
    Enacted SD 10
    156. In the southeast quadrant lies a large Black community in Southeast Fort Worth,
    described as “the core urban community of Fort Worth,” Trial Tr. 42, Jan. 20, 2012
    AM (Sen. Davis). Southeast Fort Worth continues south of Interstate 820 where it
    remains a predominantly minority community. Defs.’ Exs. 657, 136. This area is
    moved from Benchmark SD 10 into enacted SD 22 in the State Senate Plan. Id.
    157. Within the northwest quadrant is the community known as the “north side Latino
    community,” Trial Tr. 28-29, Jan. 20, 2012 AM (Sen. Davis); Defs.’ Ex. 657, which is
    moved out of SD 10 into enacted SD 12 in the State Senate Plan. Id. at 42.
    158. Overall, an examination of enacted SD 10 shows that it is drawn in a bow-tie shape in
    order to exclude many of the urban minority communities in Tarrant County that are in
    Benchmark SD 10. Defs.’ Ex. 141.
    159. Senator Rodney Ellis explained in a letter to the DOJ: “The demolition of [Senate]
    District 10 was achieved by cracking the African American and Hispanic voters into
    three other districts that share few, if any, common interests with the existing District’s
    minority coalition. The African American community in Fort Worth is ‘exported’ into
    rural District 22 – an Anglo-controlled District that stretches over 120 miles south to
    Falls [County]. The Hispanic Ft. Worth North Side community is placed in Anglo
    suburban District 12, based in Denton County, while the growing South side Hispanic
    population remains in the reconfigured majority Anglo District 10.” Defs.’ Ex. 375, at
    3.
    160. Dr. Allan J. Lichtman, an expert witness for the Davis Intervenors, echoed in his report:
    “[The] [S]tate legislature, in dismantling benchmark SD 10[,] cracked the politically
    cohesive and geographically concentrated Latino and African American communities
    and placed members of those communities in districts in which they have no
    opportunity to elect candidates of their choice or participate effectively in the political
    process.” Defs.’ Ex. 134, ¶ 12 (“Lichtman Rep.”).
    161. Over 53,000 persons are moved from SD 10 into SD 12 in the State Senate Plan, of
    whom 89.5% are Hispanic or Black, even though Benchmark SD 12 is already over-
    populated by more than 200,000 people. Defs.’ Ex. 151, at 2. Likewise, 104,703
    persons are moved from SD 10 to SD 22, of which 78.2% are either Hispanic or Black.
    Id. at 3.
    162. Doug Davis admitted that he knew that the areas he cut out of SD 10 were minority
    neighborhoods. Trial Tr. 22, Jan. 18, 2012 PM (D. Davis). Chairman Seliger also
    admitted that he knew that many of the neighborhoods being moved out of SD 10,
    including Everman and Forest Hills, were minority neighborhoods. Trial Tr. 56-57,
    Jan. 24, 2012 AM (Chairman Seliger).
    34
    163. Mapdrawers attribute the changes to SD 10 to partisanship, stating that one of the goals
    in drawing the State Senate Plan was to increase the Republican voting strength in four
    districts. Trial Tr. 144, 161, Jan 17, 2012 PM (D. Davis). SD 10 was one of those
    districts. Id. at 160.
    164. SD 10 is a majority Anglo district in the State Senate Plan. The Anglo population is
    54.5% of the enacted district’s population, 6.9% higher than the Benchmark; the Black
    population is 14.6%, a 4.6% decrease from the Benchmark; and the Hispanic
    population is 25.9%, a 3% decrease from the Benchmark. Pl.’s Ex. 15, at 4; Pl.’s Ex.
    16, at 4. Furthermore, the WCVAP increases to 69.5% of the enacted district’s citizen
    voting age population, 6.8% higher from the Benchmark; the HCVAP in enacted SD 10
    is 13.6%, 1.5% lower than in the Benchmark; the BCVAP is 12.8%, 5.5% lower than
    in the Benchmark; and the Asian-American citizen voting population CVAP increases
    from .1% to 2.7%. Pl.’s Ex. 15, at 8; Pl.’s Ex. 16 at 9.
    165. In 2001, the State of Texas predicted that SD 10 could become a district where the
    minority community would grow to be able to elect a candidate of its choice. Trial Tr.
    36, Jan. 20, 2012 AM (Sen. Davis). In the State Senate election of 2008, SD 10
    exhibited the real-life potential of that prediction. In 2011, the State Senate cracked SD
    10 and removed most of its minority populations, spreading them into predominately
    Anglo districts and effectively dismantling the coalition that had elected Senator Davis.
    166. The dismantling of SD 10 will have a disparate and negative impact on minority groups
    in the District.
    I.   STATE HOUSE PLAN
    167. There are 150 Members (the “Members”) of the State House, who run for office every
    two years. There are currently 101 Republican Members and 49 Democratic Members.
    Trial Tr. 133, Jan. 17, 2012 AM (Interiano).
    168. The Texas State Constitution provides that its Legislature will meet every two years.
    The Legislature is sworn in on the second Tuesday of every odd-numbered year and
    meets for 140 days, unless a special session is called by the Governor. The committees
    within the Legislature are typically not appointed until February and, therefore,
    legislation is not usually considered until mid-February, when the general legislative
    session begins. Actual legislative consideration of bills and their passage primarily
    takes place between February and May of a legislative year. Trial Tr. 61-62, Jan. 17,
    2012 AM (Hunter). There are approximately 70 to 80 days that are available within a
    regular legislative session to pass a bill in the State House. Trial Tr. 70, Jan 20, 2012
    PM (Chairman Solomons).
    169. On November 28, 2001, a three-judge district court adopted a redistricting plan for the
    State House in Balderas v. Texas, No. 6:01-cv-158, 
    2001 WL 34104833
     (E.D. Tex.
    Nov. 28, 2001), based on the 2000 Census. That plan is the Benchmark Plan for
    purposes of this case.
    35
    170. When the State House failed to pass a redistricting plan for the State House in 2001, the
    Legislative Redistricting Board drew the map. Pl.’s Ex. 148, at 2 (Solomons’ Pre-filed
    Testimony). Avoidance of this default process was important to the State House in
    2011. 
    Id. 171
    . On February 17, 2011, the U.S. Census Bureau released redistricting data from the 2010
    Census to Texas. During the 82nd Legislature, which met between January and May
    2011, the State House took up redistricting based upon population numbers from the
    2010 U.S. Census. The Texas Legislature did not begin the actual map-drawing
    process until the U.S. Census Bureau released block-level population data. Pl.’s Ex.
    35, at 26.
    172. The redistricting plan for the State House contained in House Bill 150 (the “State House
    Plan”) is the redistricting plan for which Texas seeks preclearance.
    A. Map-Drawing Process
    173. Speaker of the State House Joe Straus appointed Representative Burt Solomons, who
    represents HD 65, to chair the State House Redistricting Committee. Straus Dep. 63-
    64, Oct. 24, 2011. Chairman Solomons had never before chaired a redistricting
    committee, Pl.’s Ex. 148 at 1 (Pre-filed Testimony of Chairman Burt Solomons), and
    had “no background in redistricting.” Trial Tr. 65, Jan. 20, 2012 PM (Chairman
    Solomons).
    174. Speaker Straus did not give Chairman Solomons any specific instructions other than to
    prepare a map that would be supported within the State House. Straus Dep. 63-64, Oct.
    24, 2011.
    175. Chairman Solomons spent no time learning anything about the VRA or Texas’s
    obligations thereunder, and was entirely reliant on legislative staff members (Ryan
    Downton, Gerardo Interiano, and the TLC) and the OAG throughout the redistricting
    process. Trial Tr. 65-66, Jan. 20, 2012 PM (Chairman Solomons).
    176. Mr. Interiano was the principal mapdrawer for the State House Plan. Trial Tr. 127-32,
    Jan. 17, 2011 AM (Interiano).
    177. Members provided Mr. Interiano with proposed maps for their districts and it was his
    responsibility to put the pieces together to create the 150-district map. Trial Tr. 132,
    Jan. 17, 2012 AM (Interiano). Mr. Interiano worked with Members on drawing and re-
    drawing their districts. Trial Tr. 105, Jan. 25, 2012 PM (Interiano).
    178. Mr. Interiano understood from both Speaker Straus and Chairman Solomons that one
    major goal behind the State House Plan was “to give members the opportunity to be
    reelected.” Trial Tr. 160, Jan. 17, 2012 AM. Additionally, he was tasked by Speaker
    Straus to pass a legal map; ensure that map-drawing was a member driven process; and
    36
    pair the least number of Members to run against each other in re-drawn or new
    districts. 
    Id. at 133-34
    .
    179. Mr. Downton’s role in the process was to assist Members in drawing maps and to help
    mediate disagreements. Trial Tr. 46, Jan 18, 2012 AM (Downton).
    180. Speaking on the floor of the State House, Chairman Solomons told the Members in
    early 2011 that he wanted the redistricting process to be a member-driven process,
    which meant that he wanted the Members to draw their own districts. Pl.’s Ex. 148 at
    3.
    181. As a result, redistricting was in large measure left to the State House Members without,
    as far as the record reveals, any instruction on or attention to the State’s obligations
    under the VRA or its history of discrimination in voting. The process promoted
    Members’ self-interest in reelection – with Republicans preferred by the Republican
    House majority – ahead of all other considerations for redistricting. See, e.g., infra ¶¶
    240, 249, 258.
    182. Chairman Solomons never addressed, or contemporaneously even knew, the number of
    minority districts protected by the VRA under the Benchmark Plan. Trial Tr. 61-62,
    Jan. 20, 2012 PM (Solomons). There is no evidence that any other legislator was any
    better informed. Speaker Straus relied on Chairman Solomons and staff to assure him
    that the State House Plan was compliant with the VRA. Straus Dep. 71, Oct. 24, 2011.
    Likewise, Chairman Seliger in the State Senate relied on assurances from Chairman
    Solomons that the State House Plan complied with the VRA. Trial Tr. 33-34, Jan. 24,
    2012 AM (Chairman Seliger).
    183. When issues concerning the VRA arose, Mr. Interiano and/or Mr. Downton would
    usually make any necessary decision, but they went to the political leadership on
    critical issues. Trial Tr. 99, Jan. 25, 2012 PM (Interiano) (“[I]n the vast majority of the
    process of drawing, the decision was made by the staff. . . . [W]hen it was a decision
    that we were not comfortable making [,] we would take [it] to [Chairman Solomons
    and Speaker Straus].”).
    B. Redistricting Principles
    184. It is a requirement of Texas law that a candidate live in the State House district from
    which s/he is running for election. Trial Tr. 166, Jan. 17, 2012 AM (Interiano). This
    requirement resulted in strangely-shaped new districts, such as HD 41 (known as the
    “Transformer” because of its abrupt angles) in which lines were carefully drawn to
    include the home of Representative Aaron Pena and to exclude the home of
    Representative Veronica Gonzales so that the two incumbents would not have to run
    against each other. 
    Id. 185
    . Based on the State’s population of 25,145,561 people in 2010, State House districts
    with perfectly equalized population would each contain 167,637 residents. Pl.’s Ex. 35
    37
    at 15. A ten percent total population deviation from 167,000 individuals per State
    House district was acceptable in redistricting. Trial Tr. 1473-74, Perez v. Perry, Sept.
    12, 2011 (Interiano); Trial Tr. 149, Jan. 17, 2012 AM (Interiano).
    186. The County Line Rule greatly shaped the State House Plan. It stems from Section 26,
    Article 3 of the Texas State Constitution, which provides that State House districts
    must be apportioned within counties “as nearly as may be” according “to the most
    recent United States Census.” TEX. CONST., art. III, § 26. The mapdrawers first divided
    the total population of the State into 150 districts and then assigned the appropriate
    number of districts to each county. They interpreted the County Line Rule to mean
    that as many whole districts as possible must be drawn within a county and that any
    surplus population must be wholly joined with other counties or with whole surpluses
    from other counties to form a district. Trial Tr. 143-45, Jan. 17, 2012 AM (Interiano);
    Pl.’s Ex. 9. Adherence to the County Line Rule was the explanation offered for the
    elimination of Hispanic ability districts. Trial Tr. 147, Jan. 17, 2012 PM (Interiano);
    see infra § E(a).
    187. Where all proposed State House districts in a county were projected to be wholly
    contained within the county lines, the affected State House Members drew their own
    maps because any changes did not affect the rest of the State House Plan. Those
    counties were “dropped-in” to the overall redistricting map. Trial Tr. 46-47, Jan. 18,
    2012 AM (Downton). El Paso, Dallas, Tarrant, Denton, Bexar, Travis, Nueces, and
    Harris Counties were treated as drop-in counties. Id. at 48. The process worked
    smoothly for some counties, but was more difficult in others.
    188. Mr. Interiano testified that he used HCVAP and SSVR data to determine that the State
    House Plan complied with the VRA. Trial Tr. 138-39, Jan. 17, 2012 AM (Interiano).
    189. Neither Mr. Interiano nor Mr. Downton even looked at the OAG election analyses until
    the work was basically done. Trial Tr. 1451-52, Perez v. Perry, Sept. 12, 2011
    (Interiano); Trial Tr. 57, Jan. 18, 2012 AM (Downton). There is no testimony that the
    OAG analyses prompted any changes in the State House Plan after Messrs. Interiano
    and Downton actually looked at them.
    190. The OAG did not identify or analyze districts in which minority voters had the ability to
    elect a candidate of choice. Giberson Dep. 20-21, Oct. 18, 2011. Using a test of 50.1%
    or more for an ability district, both Mr. Interiano and Mr. Hanna identified 29 Hispanic
    ability districts in the Benchmark Plan, and thought this number increased to 30 in the
    State House Plan because the SSVR of HDs 90 and 148 increased. See Trial Tr. 25-26,
    32, Jan. 17, 2012 PM (Interiano); id. at 181, Jan. 17, 2012 AM (Interiano); Pl.’s Ex. 6.
    191. Based upon the testimony at trial, the Court does not find credible or persuasive
    representations by counsel that election analysis was an important tool to determine
    whether proposed State House districts would assure minority voters the ability to
    elect. It is clear that Texas adopted the principle that districts with SSVR above 50.1%
    were Hispanic ability districts under the VRA. See, e.g., Pl.’s Ex. 6 (email between
    38
    Mr. Hanna and Mr. Interiano discussing the number of districts with an SSVR over
    50%); Trial Tr. 183, Jan. 17, 2012 AM (Interiano); Trial Tr. 66-68, Jan. 25, 2012 PM
    (Interiano).
    192. Although Mr. Interiano agreed that it would have been possible to draw another
    Hispanic ability14 district in the map, he did not do so because “this was a member-
    driven map [and] we were not going to be asking [members] to make changes unless
    we believed that it was required by the Voting Rights Act or any other law.” Trial Tr.
    35, Jan. 17, 2012 PM (Interiano).
    C. Data Available to Draw the Maps
    193. Mr. Interiano testified that he spent close to one thousand hours learning the RedAppl
    software program used for redistricting even before any census results were available.
    Trial Tr. 131, Jan. 17, 2012 AM (Interiano). He also attended several conferences and
    read major cases on redistricting. Id. at 130. David Hanna and Jeffery Archer of the
    TLC contributed legal advice when asked, id. at 134-35, but, according to the record,
    were frequently ignored.15
    194. RedAppl has a function that shades a map to indicate the percentage of ethnic
    (Hispanic) or racial (Black) voting age population in a certain voter tabulation district
    (“VTD”). As relevant here, RedAppl further disaggregates this data to allow a user to
    view the variations of voting age population or total population by race or ethnicity at
    the census block level through color shading. RedAppl also allows a user to view
    variances in SSVR16 between VTDs through color shading, but does not show
    variances in SSVR between census blocks within a particular VTD. Trial Tr. 70, Jan.
    14
    Mr. Interiano testified that another Hispanic opportunity district could have been drawn. Because he was
    discussing retrogression the Court finds that he was testifying to the possibility of an additional Hispanic ability
    district. Trial Tr. 35, Jan. 17, 2012 PM (Interiano) (“Q. And you agreed also with me that it was possible to have
    avoided retrogression in [the] house plan by creating a Latino Opportunity District elsewhere in the state, but you
    did not do that? A. That’s correct. And that was due to the fact that this being a member-driven map – in many
    circumstances, the delegation bought us a map where they had all agreed to it – we were not going to be asking them
    to make changes to it unless we believed that it was required by the Voting Rights Act or any other law. But at this
    point, we felt comfortable that the fact -- with the map, that it was a legal map.”). A minority opportunity district is
    meaningful under Section 2 of the VRA, which is concerned with whether minority “voters have less opportunity
    than other members of the electorate to . . . elect representatives of their choice.” 
    42 U.S.C. § 1973
    (b).
    15
    Mr. Hanna prepared various memoranda during the redistricting process that noted Section 5 problems with the
    developing map. See Pl.’s Ex. 3 (April 6, 2011); Pl.’s Ex. 4 (April 10, 2011); Pl.’s Ex. 5 (April 20, 2011). Mr.
    Interiano reviewed all of these memos, Trial Tr. 175-78, Jan. 17, 2012 AM (Interiano), although he identified no
    significant changes or actions he took as a result.
    16
    Mapdrawers for the State House used non-suspense data with respect to SSVR, instead of total voter registration
    data. Trial. Tr. 181-83, Jan. 17, 2012 AM (Interiano). According to the Texas Secretary of State’s website, “[a]
    suspense voter is a voter known to have an incorrect address or outdated address. The county has sent the voter a
    form to obtain a new current address, but no response has been received. The voter is however considered to be an
    active voter for voting purposes.” Texas Secretary of State’s Voter Registration Public Information Request Form,
    http//www.sos.state.tx.us/elections/forms/pi.pdf, (last visited Aug. 10, 2012). The Court takes judicial notice of this
    information and all SSVR numbers referenced are non-suspense numbers for 2010.
    39
    25, 2012 PM (Interiano). Similarly, election information, i.e. the percentage of
    population that voted for a certain candidate in a prior election, is only available at the
    level of a VTD. Trial Tr. 69-71, Jan. 19, 2012 AM (Korbel).
    195. Mr. Interiano testified that the State has no specific policy against splitting VTDs when
    drawing new electoral maps. Trial Tr. 63-64, Jan. 25, 2012 PM (Interiano). As a
    result, VTDs were split readily in districts like HD 41 in Hidalgo County. Splitting
    VTDs reduces minority voting, as confusion and lack of language skills causes some
    minority voters not to vote. Trial Tr. 61-62, Jan. 19, 2012 AM (Korbel) (“[I]t has a
    disproportionate impact because minority, a great number of minority voters don’t
    have transportation for example, don’t read . . . , aren’t able to read these notices in the
    newspaper about changes in polling place and it results in a great deal of confusion.”).
    196. Mr. Interiano testified that he used shading functions on RedAppl to indicate population
    by racial or ethnic group very early in the process, but he did not use this function to
    shade for Hispanic voting age population at the census block level; instead, he only
    examined Hispanic voting age population at the VTD level. Trial Tr. 86-87, Jan. 25,
    2012 PM (Interiano). He further testified that he did not know that the RedAppl
    software could even show shading for different racial populations at the census block
    level. Trial Tr. 94, Jan. 17, 2012 PM (Interiano) (“I never did it at the bloc [sic]. I did
    not know that it was even possible, as I testified in several of my depositions. I did not
    know that it was possible to do it, and because it was not possible I certainly never
    tried and never used bloc [sic] racial shading.”).
    197. This testimony is not credible and is not accepted by this Court. As demonstrated at
    trial, when racial/ethnic population shading is used in RedAppl, a drop-down menu
    gives the user the option to show variances in that demographic between census blocks.
    Trial Tr. 88-89, Jan. 25, 2012 PM (Interiano). After one thousand hours of training
    and experience, the Court is confident Mr. Interiano would be aware of this
    functionality in the software and saw the drop-down menu.
    198. Furthermore, it is clear Mr. Interiano knew that using census block data to identify the
    demographics of voters could advance the goal of maximizing Republican electoral
    strength by suppressing the minority vote. As previously discussed, in early December
    2010, Eric Opiela, counsel to Speaker Straus, suggested to Mr. Interiano that voting
    and population data might permit distinctions between minorities who turn out heavily
    to vote and those who do not; with such information, he suggested, districts could be
    drawn that would retain a large minority population but actually include a much
    smaller number of minority voters. Defs.’ Ex. 304. Mr. Interiano tried to obtain
    demographic information on the level of census blocks but learned that only data on
    “Spanish Surname VR/Total Hispanic Population” was available. Defs.’ Ex. 197.
    Although he obtained the data and sent it to Mr. Opiela, Mr. Interiano insisted that he
    never opened the files containing SSVR information on a census block level. Interiano
    Dep. 69, Jan. 10, 2012. Given the results in some districts, such as CD 23, where low-
    voting minorities were substituted for politically-active minorities, the Court does not
    credit this testimony.
    40
    199. In addition, the fact that HD 41, a Benchmark ability district, was crafted with 17 VTD
    splits in the State House Plan, Defs.’ Ex. 886, at 76-77, and the reasons for all of these
    splits was not explained further leads the Court to discredit this testimony. Trial Tr.
    168, Jan. 17, 2012 AM (Interiano).
    D. Hearings, Procedures, and Passage
    200. The first statewide proposal put forth by the Chairman of the State House Redistricting
    Committee, known as Plan H113, was released on Wednesday, April 13, 2011. Notice
    of a public hearing on the plan was provided on that day, and public hearings were held
    immediately, in Austin, Texas, on Friday, April 15 and Sunday, April 17. No hearings
    outside Austin were conducted.
    201. The House Rules provide for a three-to-five day rule for posting advance notice of
    hearings. The notice for the April 15th hearing on the H113 was posted on April 13,
    2011, which resulted in less than two days’ notice. Solomons Dep. 83-85, Aug. 31,
    2011. David Hanna advised that the hearing schedule was too tight, but his caution
    was ignored. Defs.’ Ex. 971, at 2.
    202. On Monday, April 18, 2011, it was announced on the floor of the House that the
    Redistricting Committee would meet on Tuesday, April 19. Defs.’ Ex. 509 at 18.
    When it met, the House Redistricting Committee passed a plan known as H153 out of
    Committee.
    203. State House Members were given until Monday, April 25, 2011 (the Monday after
    Easter weekend) to file any amendments to the bill. Trial Tr. 801, Perez v. Perry, Sept.
    8, 2011 (Turner).
    204. The rushed schedule severely hampered the ability of citizens to attend the two hearings
    on the bill and of legislators to prepare objections or proposed amendments. See Defs.’
    Ex. 738 at 19 (Rep. Hochberg Pre-filed Direct Testimony).
    205. On Thursday, April 28, 2011, the State House passed an engrossed version of the plan,
    H283, in House Bill 150 by a vote of 92-54-3.
    206. No changes were made to the State House Plan in the State Senate. Pl.’s Ex. 162, ¶ 5
    (Seliger Pre-filed Testimony). Thus, when the Texas Legislature passed House Bill
    150 on May 23, 2011, the State House Plan that was adopted was the one drawn by the
    State House.
    207. The Governor signed House Bill 150 into law on June 17, 2011.
    E. Alleged Lost Hispanic Ability Districts
    a. Nueces County & House District 33
    41
    208. The Benchmark contained three House districts in Nueces County: 32, 33, and 34. HD
    32 was the only district not fully contained in the county. HD 33 was dropped from
    Nueces County in the State House Plan. The demographics of the districts containing
    some part of Nueces County under the Benchmark Plan are as follows:
    Benchmark HD               HCVAP                 SSVR
    HD 32                      35.3%                 33.2%
    HD 33                      60.4%                 54.3%
    HD 34                      58.2%                 53.8%
    Pl.’s Ex. 13, at 13, 23.
    209. Benchmark HD 32 is represented by Representative Todd Hunter, an Anglo
    Republican. It is only partially in Nueces County and also includes Aransas, San
    Patricio and Calhoun counties. The majority of voters in HD 32 in general elections
    are Anglo. Defs.’ Ex. 737, at 11 (Abel Herrero Pre-filed Direct Testimony ).
    210. Benchmark HD 34 is represented by Representative Connie Scott, an Anglo
    Republican. It is made up of both urban and rural areas of Nueces County and is a
    majority Hispanic district. The Hispanic candidate of choice won four out of the past
    five endogenous elections in HD 34. Defs.’ Ex. 326, at 5 (Handley House Report). It
    is a district where Hispanic voters have the ability to elect a candidate of their choice in
    the Benchmark.
    211. Benchmark HD 33 is represented by Representative Raul Torres, a Hispanic
    Republican. It is made up of the historic core of Corpus Christi. Defs.’ Ex. 737, at 8.
    Hispanic voters are ordinarily the majority of voters in Benchmark HD 33. 
    Id. at 9-10
    .
    It is a district where Hispanic voters have the ability to elect in the Benchmark and it
    no longer provides Hispanic voters the ability to elect in the State House Plan.
    212. The voting demographics of Benchmark and enacted HD 33 are as follows:
    HD 33                 Pop.        VAP17       CVAP      HCVAP BCVAP          WCVAP        SSVR
    Benchmark           148,929      109,257     97, 255    60.4% 4.5 %          33.5%        55.3%
    Enacted             172,135      119,518     109,865    8.5 %    5.9 %       81.2%        6.5%
    Pl.’s Exs. 13, 14.
    213. Hispanic voters were successful in electing their candidate of choice in four of the past
    five endogenous elections in Benchmark HD 33. Handley House Rep., at 5. In the
    2010 election, Representative Solomon Ortiz, the Hispanic candidate of choice, won
    17
    VAP represents voting age population.
    42
    47.5% of the vote, but lost to Representative Raul Torres, the current representative of
    HD 33. Defs.’ Ex. 726 at 6, n.5 (Engstrom Supp. Rep.).
    214. Hispanic candidates of choice won at least 50% of the exogenous elections the experts
    in this case analyzed in Benchmark HD 33. Handley House Rep. at 5 (five out of five
    elections); Defs.’ Ex. 799 (“Engstrom Chart”) (four out of seven elections); Pl.’s Ex.
    175, at 11, tbl. 3b (“Alford Rep.”) (six out of ten elections).
    215. According to the 2010 American Community Survey 1-year estimates, Nueces County
    had a citizen voting-age population of 238,102 persons, including 91,467 Anglos
    (38.4%) and 133,084 Hispanic persons (55.9%). As a whole, however, Nueces County
    had an SSVR below 50%. Trial Tr. 9, Jan. 17, 2012 PM. It was allotted 2.03 districts
    based upon the County Line Rule in the State House Plan. Trial Tr. 147, Jan. 17, 2012
    AM (Interiano). The State calculated this number by dividing Nueces County’s
    population of 340,233 by the ideal district size (167,637). Pl.’s Ex. 35, at 19.
    216. Messrs. Hanna and Interiano decided that Nueces County “needed to have two districts
    and only two districts” in the State House Plan. Trial Tr. 147, Jan. 17, 2012 PM
    (Interiano). They informed Chairman Todd Hunter, who represents HD 32, and the
    rest of the Nueces County delegation of this fact. 
    Id.
     HD 33, a Hispanic ability district
    in the Benchmark, was chosen for elimination.
    217. Legislative staff drew one district that was a “Latino Democratic” district and one that
    “would likely be Republican and not Latino.” Defs.’ Ex. 742 (Hanna Dep. 46, Jan. 12,
    2012). The core of Benchmark HD 33 is moved into enacted HD 34. Defs.’ Ex. 737,
    at 13-14. Representative Raul Torres, the current incumbent in Benchmark HD 33, and
    Representative Connie Scott, the current incumbent in Benchmark HD 34, are the
    junior Republican members of the Nueces County delegation and are paired under the
    State House Plan in HD 34, while Representative Todd Hunter is drawn into enacted
    HD 32 in Nueces County. Trial Tr. 122, Jan. 17, 2012 AM (Rep. Hunter).
    218. Mr. Hanna recognized that drawing two districts in Nueces County may have to yield to
    the VRA. He wrote on April 20, 2011: “While there are two 50% SSVR plus districts
    within the county currently that may constitute performing Hispanic districts, they are
    both significantly underpopulated [sic] and the remaining people in Nueces County are
    predominantly Anglo. The county line rule likely requires two districts to be wholly
    contained within Nueces County with no surplus coming out; however this would have
    to yield to the federal Voting Rights Act if it can be shown retrogression could be
    avoided by splitting the county.” Pl.’s Ex. 5, at 1. He further advised that splitting the
    Hispanic population in half would only result in two districts with SSVRs under 50%
    which were unlikely to perform as “Hispanic districts of choice.” 
    Id.
     In keeping with
    the County Line Rule, it was not possible to draw two districts that had an SSVR of
    above 50% in Nueces County and therefore only one district was drawn with an SSVR
    of above 50%.18 Trial Tr. 147, Jan. 17, 2012 AM (Interiano).
    18
    The County Line Rule was broken in Henderson County to comply with the federal one-person one-vote
    requirement. Trial Tr. 85-86, Jan 20, 2012 PM (Chairman Solomons).
    43
    219. Mr. Hanna did not know if the loss of a district that “performed” for Hispanic voters in
    Nueces County was made up somewhere else in the State House Plan. Defs.’ Ex. 742
    (Hanna Dep. 46, Jan. 12, 2012). Mr. Interiano testified that he felt that the loss of HD
    33 would be accounted for through the increase in SSVR in enacted HDs 90 and 148.
    Under the Benchmark, neither of these districts had an SSVR above 50%, Trial Tr. 10,
    Jan. 17, 2012 PM (Interiano), although each was in fact an ability district. See infra §
    F.
    220. In the State House Plan, only HDs 32 and 34 remain in Nueces County, see Pl.’s Ex.
    88, with the following demographics:
    Enacted HD                 HCVAP                  SSVR
    HD 32                      44.2%                  36.6%
    HD 34                      64.6%                  60.1%
    Pl.’s Ex. 14, at 13, 23.
    221. The area that makes up Benchmark HD 33 is relocated to Dallas County in the State
    House Plan. Minority preferred candidates have no success in exogenous elections in
    enacted HD 33. Handley House Rep. at 9; Alford Rep at 11, tbl. 3b; Engstrom Chart.
    b. House District 35
    222. Benchmark HD 35 is located in southern Texas and contains Atascosa, Karnes, Goliad,
    Bee, Live Oak, and McMullen counties. Enacted HD 35 loses the counties of Karnes,
    Goliad, and Jim Wells. It gains San Patricio and Duvall counties.
    223. Benchmark HD 35 is a district where Hispanic voters have the ability to elect. It cannot
    be determined whether minority voters have the ability to elect in HD 35 in the State
    House Plan.
    224. The voting demographics for the Benchmark and enacted HD 35 are as follows:
    HD 35            Pop.      VAP      CVAP       HCVAP BCVAP WCVAP             SSVR
    Benchmark 151,882         113,190   107,735    54.6%       5.3 %   38.8 %    55.3%
    Enacted         172,482   127,314   121,925    52.5 %      4.2 %   42.0 %    53.4 %
    Pl.’s Exs. 13, 14.
    225. The minority candidate of choice won the last four of five endogenous elections in
    Benchmark HD 35. Handley House Rep., at 5. The current representative of
    Benchmark HD 35 is Jose Aliseda, a Hispanic Republican, who is not running for re-
    election. Representative Aliseda is a freshman representative who beat former
    Represenative Yvonne Gonzales, a Democrat, to win the seat in 2010. Trial Tr. 39, Jan
    44
    20, 2012 PM (Chairman Solomons). Representative Aliseda was not the candidate of
    choice of Hispanic voters. Handley House Rep., at 34.
    226. The exogenous election results for Benchmark HD 35 show that Hispanic candidates of
    choice may win approximately half or fewer of the elections analyzed by the experts in
    this case. See Handley House Rep. at 5 (two out of five elections); Alford Rep. at 11,
    tbl. 3 (five out of ten elections); Engstrom Chart (two out of seven elections). The
    exogenous election results of the experts vary with respect to their measurement of
    minority voting strength in enacted HD 35, making these results inconclusive in
    evaluating the change in minority voting power in enacted HD 35. Handley House
    Report at 5 (one out of five elections); Alford Rep. at 11, tbl. 3 (four out of ten
    elections); Engstrom Chart (three out of seven elections).
    227. Representative Aliseda wanted to draw a proposed HD 35 that was more Republican in
    the State House Plan. Trial Tr. 113, Jan. 17, 2012 PM (Aliseda). He himself proposed
    a large portion of the map for this district, but he did not receive every area he
    proposed. Id. He worked with Mr. Interiano on the map and relied on Mr. Interiano’s
    advice regarding the manner in which the district needed to be drawn. Id. at 111, 113,
    118. He understood that, due to the VRA, he needed to keep the Hispanic population
    in HD 35 at current percentages. Id. at 122.
    c. House District 117
    228. Both the Benchmark and enacted HD 117 are wholly located within Bexar County. HD
    117 shares its eastern border with HD 118 in both the Benchmark and State House
    Plans.
    229. HD 117 is a Hispanic ability district in the Benchmark, but Hispanic voters lose the
    ability to elect in this district in the State House Plan.
    230. The Hispanic candidate of choice won three out of five of the last endogenous elections
    in Benchmark HD 117. Defs.’ Ex. 326, at 5 (Handley House Rep.).
    231. HD 117 is currently represented by John Garza, a Hispanic Republican and a freshman
    representative in the State House who defeated Representative David Liebowitz, an
    Anglo Democrat, by a very close margin in 2010. Representative Garza is not the
    Hispanic candidate of choice. Handley House Rep. at 34. The Hispanic candidate of
    choice, Representative Liebowitz, won 48.1% of the vote in 2010. Id. at 34; Engstrom
    Supp. Rep. at 7.
    232. HD 118, which lost some area to HD 117 in the State House Plan, is currently
    represented by Representative Jose Farias, a Hispanic Democrat.
    233. The exogenous election results for Benchmark HD 117 show that Hispanic citizens are
    successful in electing their candidate of choice at least 50% of the time. See Alford
    Rep., at 11, tbl. 3b (five out of ten elections); Handley House Rep., at 5 (three out of
    45
    five elections); Engstrom Supp. Rep., at 6 (four out of seven elections). In enacted HD
    117, the exogenous results show that minority effectiveness in such elections will drop;
    in particular, Dr. Alford and Dr. Handley’s analyses both show decreases of at least
    30% in exogenous election results in the enacted district. Alford Rep., at 11, tbl. 3
    (two out of ten elections); Handley House Rep., at 11 (one out of five elections); see
    also Engstrom Supp. Rep., at 8-9 (three out of seven elections).
    234. The voting demographics for Benchmark and enacted HD 117 are as follows:
    HD 117              Pop.    VAP      CVAP       HCVAP       BCVAP       WCVAP       SSVR
    Benchmark       220,360    155,490   106,595   58.8%       6.1%        32.3%        50.8%
    Enacted         171,249    116,261   71,395    63.8%       4.6 %       29.4%        50.1%
    Pl.’s Exs. 13, 14.
    235. The Bexar County Delegation, made up of seven Democrats and three Republicans, met
    as a whole to decide how to redistrict the County. Trial Tr. 105, Jan. 25, 2012 PM
    (Interiano). Representative Mike Villarreal, a Hispanic Democrat, and Representative
    Ruth McClendon, a Black Democrat, led the process. Defs.’ Ex. 363 (Garza Dep. 25,
    Oct. 19, 2011). Mr. Interiano was present at meetings where the delegation discussed
    how to draw the new map. Id. at 28; Trial Tr. 105, Jan. 25, 2012 PM (Interiano).
    236. Mr. Interiano described the process for drawing the Bexar County map as one where all
    of the Members proposed their ideal district to Representative Villarreal, who put the
    districts together in a map that showed where requests overlapped. The Members then
    negotiated to determine who would receive specific parts of the map. Trial Tr. 107,
    Jan. 25, 2012 PM (Interiano). Nine out of ten Members from the Bexar County
    delegation approved the districts for the County. Trial Tr. 105, Jan. 25, 2012 PM
    (Interiano). Representative Farias did not agree.
    237. Mr. Interiano’s goal in drawing enacted HD 117 was to keep its SSVR above 50%.
    Trial Tr. 159, Jan. 17, 2012 AM (Interiano). Mr. Interiano told Representative Garza
    that he needed to keep his district above a 50% SSVR and “maintain [his] other goals
    in the district.” Trial Tr. 107, Jan. 25, 2012 PM (Interiano).
    238. Representative Garza said that he did not have much control and that the delegation
    agreed on the map for Bexar County on a consensus basis. Defs.’ Ex. 363 (Garza Dep.
    69, Oct. 19, 2011). He said that he wanted his district to move northward, where the
    area was more Anglo and more Republican. Id. at 30-31. And while he testified that
    he did not identify any specific areas that he wanted in his district, id. at 33, he then
    said that he wanted to keep Port San Antonio, the University of Texas at San Antonio,
    and Lackland Air Force Base in his district. Id. at 34-35.
    239. Representative Garza was aware that minority representation had to be maintained in
    HD 117. Defs.’ Ex. 363 (Garza Dep. 26, Oct. 19, 2011). He was advised by both the
    OAG and Representative Villarreal that he could not move his district northward and
    46
    that he had to continue to keep certain indicators of Hispanic voting strength the same
    as in the Benchmark district. Id. In his deposition, Representative Garza could not
    identify those specific indicators, id. at 51, although contemporaneously he told
    Representative Farias that he had to have an SSVR of 50.1% in his enacted district.
    Trial Tr. 14-15, Jan. 24, 2012 PM (Rep. Farias).
    240. The mapdrawers thus decided to maintain SSVR levels while minimizing the actual
    Hispanic vote so that Representative Garza, as a Republican, could be reelected. Part
    of the attention to this issue is revealed by a March 24, 2011 email from Representative
    Villarreal to Mr. Interiano containing a chart stating that “[o]f the 27 VTD’s won by
    Garza, 4 had a majority of SSRV.”19 Defs.’ Ex. 917, at 3.
    241. In order to accomplish this goal, the communities of Somerset and Whispering Winds
    were moved from Representative Farias’s district to Representative Garza’s district.
    Somerset is a rural, minority community with low Hispanic voter turnout. Defs.’ Ex.
    363 (Garza Dep. 39-40, Oct. 19, 2011). Whispering Winds is another largely, Hispanic
    community with low voter turnout. Trial Tr. 13, Jan. 24, 2012 PM (Farias).
    242. Both Somerset and Whispering Winds are very poor communities that have poor water
    quality and poor housing. Farias Dep. 25, Jan. 5, 2012. As a result, Representative
    Farias paid special attention to the needs of these communities and was working
    actively to improve both. Trial. Tr. 9-11, Jan. 24, 2012 PM (Farias). He was very
    concerned that he continue to represent both communities. Trial Tr. 7-8, Jan. 24, 2012
    PM (Rep. Farias).
    243. Despite Representative Farias’s strong objections, Whispering Winds and Somerset
    were drawn into HD 117, Representative Garza’s district. Trial Tr. 8, 23, Jan. 24, 2012
    PM (Rep. Farias).
    244. Representative Villarreal proposed taking Somerset and Whispering Winds out of
    Representative Farias’s district, Trial Tr. 24, Jan. 25, 2012 PM (Rep. Farias), at the
    behest of Speaker Straus to ensure Representative Garza’s reelection. Farias Dep. 26,
    Jan. 5, 2012.
    245. Mr. Interiano admitted that the “political numbers” of Representative Garza’s
    benchmark district meant that Representative Garza could not be reelected if Somerset
    were not included in his district in the State House Plan. In order to maintain
    “demographic [i.e. SSVR] and political numbers” for his reelection, Somerset was a
    necessary addition. Trial Tr. 160, Jan. 17, 2012 AM (Interiano).
    246. Because Representative Farias’s goal was to keep the communities of Somerset and
    Whispering Winds in HD 118, he asked Representative Garza to take South San
    Antonio Independent School District (“ISD”) from HD 118 in exchange for allowing
    Somerset and Whispering Winds to remain in his district. Trial Tr. 15-16, Jan. 25,
    2012 PM (Rep. Farias). Notably, South San Antonio ISD had a very high voter turnout
    19
    SSVR may also be referred to as SSRV (Spanish Surname Registered Voters).
    47
    as compared to Somerset and Whispering Winds. Trial Tr. 17, Jan. 24, 2012 PM.
    Representative Garza refused to make the trade. Id. at 16-18.
    247. Representative Farias also visited with Mr. Interiano and Speaker Straus,
    unsuccessfully, in his effort to keep Whispering Winds and Somerset in his district.
    Trial Tr. 14, Jan 24, 2012 PM (Rep. Farias).
    248. Representative Farias ultimately offered an amendment on the floor of the State House
    to allow Whispering Winds to stay in his district in exchange for moving the area
    around Lackland Air Force Base to Representative Garza’s district. Trial Tr. 17, Jan.
    24, 2012 PM (Rep. Farias); Defs.’ Ex. 323, at 36. Representative Garza said he would
    leave the Amendment to the will of the House. Trial Tr. 17, Jan. 24, 2012 PM (Rep.
    Farias).
    249. During the House discussion of Representative Farias’s amendment, Representative
    Aliseda expressed concerns regarding what it would do to the “Republican numbers”
    of Representative Garza’s new district. Defs.’ Ex. 323, at 36.
    250. Representative Farias was unsuccessful in passing his amendment and the communities
    of Somerset and Whispering Winds are in HD 117 in the State House Plan.
    d. House District 41
    251. Representative Veronica Gonzales, a Hispanic Democrat, currently represents
    Benchmark HD 41 in Hidalgo County. Representative Aaron Pena, a Hispanic
    Republican, currently represents Benchmark HD 40, immediately adjacent to HD 41 in
    Hidalgo County. Representative Pena is a five-term incumbent who switched political
    affiliation from Democrat to Republican at the end of 2010. Trial Tr. 163, Jan. 17,
    2012 AM (Interiano).
    252. Both Benchmark enacted HD 41 are a Hispanic ability districts.
    253. In Benchmark HD 41, the Hispanic candidate of choice was elected in five out of five
    of the past endogenous elections. Handley House Rep., at 4. Furthermore, in
    Benchmark HD 41, the minority candidate of choice wins the exogenous elections the
    experts analyzed over 50% of the time. Handley House Rep., at 5 (four out of five
    elections); Alford Rep., at 11 tbl. 3 (seven wins out of ten elections); Engstrom Chart
    (five out of seven elections).
    254. The demographics of Benchmark and enacted HD 41 are as follows:
    HD 41          Pop.      VAP        CVAP      HCVAP        BCVAP       WCVAP        SSVR
    Benchmark 185,892       125,055    86,940     77.5%       0.9%        20.2%        69.2%
    Enacted      160,238    111,689    79,770     72.1%       0.9%        25.3%        64.6%
    Pl.’s Exs. 13, 14.
    48
    255. According to the 2010 American Community Survey 1-year estimates, Hidalgo County,
    Texas had a citizen voting-age population of 363,615 persons, including 48,087 Anglos
    (13.2%), and 309,690 Hispanics (85.2%).
    256. Mr. Interiano drew all of the proposed State House districts in Hidalgo County. Trial
    Tr. 1426, Perez v. Perry, Sept. 12, 2011 (Interiano). He first drew HD 41, id. at 1476-
    77, with the objective of boosting Representative Pena’s chances for reelection. Trial
    Tr. 165, Jan. 17, 2012 AM (Interiano).
    257. Mr. Interiano knew when he drew enacted HD 41 that it would have to be a majority-
    Hispanic district because it was impossible to draw a district that was not majority-
    minority in Cameron or Hidalgo counties. Trial Tr. 42-43, Jan. 17, 2012 PM
    (Interiano).
    258. In the process of drawing the map for HD 41, Mr. Interiano split 17 VTDs. Defs.’ Ex.
    886, at 76-77. VTDs were cut in order to avoid pairing incumbents, allow
    Representative Pena’s house to be moved into enacted HD 41, and to cut out heavily
    Democratic areas “because [mapdrawers] wanted to increase the [sic] Republican
    performance of that district.” Trial Tr. 168, Jan. 17, 2012 AM (Interiano); Defs.’ Ex.
    785 (Pena Dep. 160-61, Oct. 19, 2011). Many splits, however, were not specifically
    explained. Over 31% of the population of enacted HD 41 was drawn into the district
    from split VTDs. Handley House Report, at 9.
    F. New Hispanic Ability Districts
    a. Alleged New Hispanic Ability Districts
    i. House Districts 90 & 148
    259. Mr. Interiano specifically testified that the loss of HD 33 was made up in part by the
    increased SSVR in enacted HDs 90 and 148. Trial Tr. 10, Jan. 17, 2012 PM
    (Interiano). Both of these districts are Hispanic ability districts in the Benchmark and
    remain so in the State House Plan.
    260. Representative Jessica Farrar, a Hispanic Democrat, represents Benchmark HD 148,
    located in Harris County. Representative Lon Burnam, an Anglo Democrat, represents
    Benchmark HD 90, located in Tarrant County.
    261. The voting demographics for Benchmark and enacted HD 148 are as follows:
    HD 148         Pop.      VAP        CVAP      HCVAP        BCVAP       WCVAP        SSVR
    Benchmark 140,946       109,647    79,785     42.1%       10.0%       45.4%         40.0%
    Enacted      175,324    126,854    86,715     51.4%       9.4%        37.0%         50.0%
    49
    Pl.’s Exs. 13, 14.
    262. In Benchmark HD 148, the Hispanic candidate of choice won five out of five of the past
    endogenous elections. Handley House Rep. at 5. The Hispanic candidate of choice
    also won all of the exogenous elections the experts analyzed in this case. Id. at 5;
    Engstrom Chart. These results do not change in the enacted State House Plan.
    Handley House Rep., at 11; Engstrom Chart.
    263. The voting demographics for Benchmark and enacted HD 90 are as follows:
    HD 90            Pop.        VAP         CVAP         HCVAP          BCVAP          WCVAP           SSVR
    Benchmark 141,349          97,594       62,045        47.9%         12.6%          37.2%           47.2%
    Enacted         159,428    105,582      67,570        49.7%         15.6%          32.5%           50.1%
    Pl.’s Exs. 13, 14.
    264. In Benchmark HD 90, the Hispanic candidate of choice won five out of five of the past
    endogenous elections. Handley House Rep. at 5. In Benchmark HD 90, the Hispanic
    candidate of choice also won all of the exogenous elections the experts analyzed. Id.;
    Engstrom Chart. These results do not change in the State House Plan. Handley House
    Rep., at 11; Engstrom Chart.
    265. Other than Mr. Interiano’s testimony at trial, there is no evidence that the decision to
    increase the SSVR in these two districts was intended to offset the loss of HD 33.
    Instead, it appears that HDs 90 and 148 were drawn with SSVRs at or above 50% in
    the State House Plan at the request of MALDEF. Trial. Tr. 10, Jan. 17, 2012 PM
    (Interiano). Luis Figueroa of MALDEF testified at a Redistricting Committee Hearing
    and requested this change.20 Id.
    266. Mr. Interiano never determined whether HDs 90 and 148 are Hispanic ability districts in
    the Benchmark. Defs.’ Ex. 779A (Interiano Dep. Vol. 1, 151-53, Aug. 2, 2011).
    However, he decided that both districts are ability districts in the State House Plan
    because their SSVR increased above 50%. Id. at 153.
    267. Mr. Interiano explained that he did not do an election analysis of HD 90 because “[i]t
    was going to perform ten out of ten, and it performed ten out of ten because it was a
    Democrat [sic] district, not because it was a district that was always electing the
    candidate of choice of the Latino community.” Trial. Tr. 14, Jan. 17, 2011 PM
    (Interiano). Similarly, Mr. Interiano did no analysis of HD 148 to determine whether it
    was or would become a Hispanic ability district. Id. at 32.
    20
    MALDEF wrote to Chairman Solomons on April 27, 2011, stating that it considered HDs 90 and 148 to be
    Benchmark ability districts so that raising their SSVR in the State House Plan would not create new ability districts.
    Defs.’ Ex. 649, at 2.
    50
    268. Mr. Interiano based some of his assessment of enacted HD 90’s effectiveness for
    minority voters on politics. Mr. Interiano believed Hispanic voters would be able to
    elect their preferred candidate in enacted HD 90 because the sitting representative, Lon
    Burnam, opposed the increase in SSVR in this district, while MALDEF supported it.
    Trial. Tr. 12-13, Jan. 17, 2012 PM (Interiano).
    ii.    House District 74
    269. Benchmark HD 74 encompasses the counties of Uvalde, Edwards, Val Verde, Terrell,
    Pecos, Brewster, Presidio, Jeff Davis, Ward, Reeves, Loving, Culberson, and
    Hudspeth. Enacted HD 74 combines Hudspeth, Culberson, Loving, Jeff Davis,
    Reeves, Presidio, Brewster, Pecos, Terrell, Val Verde, Kinney and Maverick counties.
    270. Benchmark HD 74 is a Hispanic ability district.
    271. Benchmark HD 74 is represented by Representative Pete Gallego, a Hispanic Democrat,
    who has represented this district since 1991. Representative Gallego is running for
    Congress in CD 23 in 2012 and he does not plan to run for reelection to the State
    House. Pl.’s Ex. 10. It is uncontested that he has been the candidate of choice of
    Hispanic voters.
    272. Mr. Interiano testified that he believed that HD 74 was a Hispanic opportunity district in
    the Benchmark. Trial Tr. 25, Jan. 17, 2012 PM (Interiano).
    273. The demographics for Benchmark and enacted HD 74 are as follows:
    HD 74               Pop.    VAP    CVAP HCVAP BCVAP WCVAP SSVR
    Benchmark          143,566 104,522 85,920 59.7% 1.8% 36.7% 58.1%
    Enacted            162,357 115,236 86,210 69.4%           1.5%       27.2%       69.6%
    Pl.’s Exs. 13, 14.
    274. The exogenous election results of the experts in this case for Benchmark HD 74 vary
    with respect to their measurement of minority voting strength. Alford Rep. at 11
    (reporting minority victories in four of ten elections); Handley House Rep. at 5 (one
    out of five elections); Engstrom Chart (four out of seven elections).
    iii.   Failure to Draw Additional Hispanic Ability Districts
    275. Mr. Interiano thought that that there did not need to be an additional Hispanic ability
    district in the State House Plan because it was a Member-driven map. As a result, if
    the Members did not add such districts, he would not ask them to make changes. He
    felt comfortable with this decision because he believes that the State House Plan is
    legal under the VRA. Trial Tr. 35, Jan. 17, 2012 PM (Interiano). However, Mr.
    51
    Interiano did admit that it was possible to draw another majority-Hispanic district,
    which he would classify as an “opportunity” district in Southern Texas in the Rio
    Grande Valley.
    276. In this area of the State, the State House Plan continues to maintain two State House
    districts in Cameron County and spills its excess population northward into a district
    shared with Willacy and Kennedy Counties. Pl.’s Ex. 14, at 1. It also maintains four
    districts in Hidalgo County and spills its excess population towards Starr County to
    form another district. Id.
    277. Mr. Interiano testified it was possible to use excess population from Cameron and
    Hidalgo counties to create a majority Hispanic district in the State House Plan that
    likely would have performed as a Hispanic “opportunity”21 district. Trial Tr. 42, Jan.
    17, 2012 PM (Interiano).
    278. If such a district were drawn, then a ripple effect would have caused a county line split
    in the northern portion of the map around Nueces County. The TLRTF submitted such
    a proposal. Trial Tr. 39-40, Jan. 17, 2012 PM (Interiano). Due to the violation of the
    County Line Rule that would result if the populations from the two districts were
    spilled towards each other, another majority-Hispanic district was not created in this
    area in the State House Plan. Id. at. 40.
    279. Chairman Solomons was aware that there was excess population in both Cameron and
    Hidalgo Counties, but testified that he never realized that a district could have been
    created using these populations because his staff did not advise him of that fact. Trial.
    Tr. 76-77, Jan. 20, 2012 PM (Chairman Solomons).
    280. Though population is available to draw a potential new Hispanic opportunity district,
    the State did not choose to do that nor does it argue that any other new potential
    Hispanic opportunity/ability district was drawn in the State House Plan
    281. There are no new Hispanic ability districts in the State House Plan.
    H. Lost Coalition District: House District 149
    282. Benchmark HD 149 is in Harris County, but is eliminated from the County in the State
    House Plan.
    283. HD 149 is a district where a coalition of Asian-American, Black, and Hispanic voters
    have the ability to elect and its elimination from Harris County in the State House Plan
    leads to the loss of a minority ability district.
    21
    In this instance, the Court cannot make a finding that Mr. Interiano was discussing a potential ability district,
    because this portion of his testimony is unclear.
    52
    284. Hubert Vo, a Vietnamese-American Democrat, is the representative of Benchmark HD
    149. Mr. Vo was elected in 2004 and is the only Vietnamese-American in the State
    House.
    285. The voting demographics of Benchmark and enacted HD 149 are as follows:
    HD 149          Pop.     VAP        CVAP        HCVAP       BCVAP       WCVAP         Asian
    CVAP
    Benchmark 169,836       123,771    90,245      19.0%        26.1%      37.6%        16.2%
    Enacted       164,376   116,361    98,445      12.9%        4.4%       77.4%        3.8%
    Pl.’s Exs. 13, 14.
    286. Benchmark HD 149 contains the community of Alief, a large Asian-American
    population in the Houston area. Defs.’ Ex. 736, at 5-6, 8 (Rogene Calvert Pre-filed
    Testimony ).
    287. In general, a review of election results from the OAG 10 shows that Hispanic and Black
    voters uniformly prefer different candidates from Anglo voters in HD 149 in general
    elections and that voting is racially polarized in this district. Pl.’s Ex. 26, at 3557-60.
    288. Asian-Americans, Hispanics, and Blacks in the area of HD 149 often work together to
    support candidates in local elections and Asian-American candidates have successfully
    been elected to school boards in Alief and to the Houston City Council. Defs.’ Ex.
    736, at 12.
    289. In particular, Hispanic, Black, and Asian-American communities came together to help
    to elect Representative Vo. Defs.’ Ex. 736, at 11. Mr. Vo received endorsements from
    the Tejano Democrats and the African Coalition for his candidacy in 2004. Id.
    Representative Hochberg, the Democratic representative from HD 137, testified that
    Hispanics, Blacks, and Asian-Americans form a coalition in HD 149, with the Asian-
    American community acting as the glue for this coalition. Defs.’ Ex. 738, at 13.
    Representative Vo has had a very diverse base of volunteers working on his
    campaigns, including Asian-American, Hispanic, and African-American volunteers.
    Id. at 13; Defs.’ Ex. 736, at 11.
    290. Mr. Vo would not have been successful in his bid for election if he had not received
    support from all of the minority communities in HD 149. Defs.’ Ex. 736, at 11. In
    particular, the Asian-American community has taken a great deal of pride in
    Representative Vo’s election and many Asian-Americans turned out to vote for him
    who had not before participated in elections. Id.
    291. Because Harris County went from 25 districts in the Benchmark to 24 in the State
    House Plan, one fewer district was available. HD 149 is eliminated from Harris
    County and Representative Scott Hochberg and Representative Vo are paired in
    53
    enacted HD 137 in the State House Plan. Enacted HD 137 contains only one VTD that
    Representative Vo previously represented. Defs.’ Ex. 738, at 20. Representative
    Hochberg understands that enacted HD 137 was drawn to give him a chance to win the
    district, not Representative Vo. Id. at 21.
    292. The decision to decrease the number of districts in Harris County was based upon 2010
    census data. Dividing Harris County’s population of 4,092,459 by the ideal district
    size (167,637) yielded 24.4126 districts for the County.22 Pl.’s Ex. 35, at 20; Trial Tr.
    148, Jan. 17, 2012 AM (Interiano).
    293. Members of the minority community regarded the decision to eliminate this minority
    ability district as detrimental to minority voting interests and strength. Leaders of the
    Texas Asian American Redistricting Initiative, MALDEF, and the NAACP in Houston
    sent Chairman Solomons a letter protesting the elimination of HD 149. The letter
    highlighted that the State House Plan would break up the community of interest in
    Alief. Defs.’ Ex. 632.
    294. Mr. Hanna of the TLC concluded that either 24 or 25 districts would be permissible in
    Harris County, but he thought the choice to draw 24 districts in Harris County was
    “absolutely defensible.” Defs.’ Ex. 742 (Hanna Dep. 106, Jan. 12, 2012).
    295. Initially, Chairman Solomons had stated on the floor of the State House that there
    would be 25 districts in the enacted Harris County map. Trial Tr. 43, Jan. 19, 2012 PM
    (Coleman); Defs.’ Ex. 738, at 15. Mr. Interiano told Representative Wayne Smith, an
    Anglo Republican, and Representative Senfronia Thompson, a Black Democratic from
    Harris County, to draw maps for Harris County that had both 24 and 25 districts. Trial
    Tr. 148, Jan. 17, 2012 AM (Interiano). It was generally understood that Representative
    Smith would lead the redistricting effort in Harris County. Defs.’ Ex. 738, at 15.
    296. Between March and April of 2011, Representative Smith and Representative Thompson
    worked with the whole Harris County delegation on a 25-member map. Trial Tr. 46-
    48, Jan. 19, 2012 PM (Coleman).
    297. Chairman Solomons later told Representative Smith, however, that the Harris County
    map would only have 24 districts. Representative Smith then drew a 24-district
    version of the Harris County map that merged Representative Vo and Representative
    Hochberg’s districts. Smith Dep. 22-23, 37, 38, Oct. 13, 2011. Mr. Interiano provided
    instruction to the delegation on which districts could be eliminated. He testified: “If
    the courts would have found or do find that coalition district[s] are protected by the
    Voting Rights Act, then we believed that the district that was going to most likely be
    protected by the Voting Rights Act was Scott Hochberg’s district. As a result, . . . we
    instructed the Harris County delegation . . . that the demographics of that district, that
    was the combination of Hochberg and Vo, needed to more closely assemble [sic] Mr.
    22
    Chairman Solomons testified that he decided that there would be 24 not 25 districts in Harris County in the State
    House Plan based upon the “advice of counsel.” Trial Tr. 1567, Perez v. Perry, Sept. 13, 2011 (Chairman
    Solomons).
    54
    Hochberg’s district rather than Mr. Vo’s.” Trial. Tr. 153, Jan. 17, 2012 AM
    (Interiano).
    298. Representative Smith sent the 24-district version of the map to Speaker pro tem of the
    State House, Beverly Woolley. Smith Dep. 34-35, Oct. 13, 2011. Speaker Woolley
    separately worked on a map that had 24 districts, which she presented to the State
    House Redistricting Committee. Trial Tr. 46-48, Jan. 19, 2012 PM (Coleman). All of
    the Republican members of the Harris County delegation signed off on Speaker
    Woolley’s map, which she then showed to Mr. Interiano. Woolley Dep. 17, Oct. 13,
    2011. Democratic members of the Harris County delegation objected to the decision to
    decrease the number of districts in Harris County. Trial Tr. 150, Jan. 17, 2012 AM
    (Interiano). Speaker Woolley’s map ultimately was the basis for the way that Harris
    County was drawn in the State House Plan. Trial Tr. 52, Jan. 19, 2012 PM (Coleman).
    299. Mr. Hanna told Mr. Interiano that he felt that a coalition district composed of three
    different minority groups would be novel. Trial Tr. 30-31, Jan. 17, 2012 PM
    (Interiano). Nonetheless, he thought that both Representative Hochberg’s and
    Representative Vo’s districts fell “into [a] potential coalition district situation.” Defs.’
    Ex. 742 (Hanna Dep. 39, Jan. 12, 2012). Indeed, Mr. Hanna advised Mr. Interiano via
    email on February 17, 2011 that cutting Harris County down to 24 seats would lead to
    two Republicans being paired because all of the Democratic seats constituted
    “minority” seats. Defs.’ Ex. 293.
    300. In Mr. Interiano’s estimation, neither Representative Vo’s district (HD 149) nor
    Representative Hochberg’s district (HD 137) could be classified as a coalition district
    within his understanding of the term. Interiano Dep. 46-47, Jan. 10, 2012. However,
    he realized there was a chance that Benchmark HD 137 might be protected because the
    district is majority-minority based upon the population of two minority groups whereas
    Benchmark HD 149 is majority-minority based upon the combination of three minority
    groups. Trial Tr. 30-31, Jan. 17, 2012 PM (Interiano).
    I. Other Disputed Districts
    a. House District 26
    301. Benchmark HD 26 is located in Fort Bend County and is represented by Charlie
    Howard, an Anglo Republican. Enacted HD 26 is also located in Fort Bend County
    and continues to share a border with Benchmark HD 149.
    302. The voting demographics of enacted and benchmark HD 26 are as follows:
    HD 26          Pop.       VAP        CVAP      HCVAP      BCVAP      Asian      WCVAP       SSVR
    CVAP
    Benchmark    180,72    133,838     108,535     10.5%      12.6%      22.2%      53.5%      9.7%
    9
    Enacted      160,09    117,247     85,950      11.6%      10.6%      19.6%      57.3%      10.3%
    1
    55
    Pl.’s Exs. 13, 14.
    303. No election analysis regarding this district was offered to the Court.
    b. House District 106
    304. Benchmark HD 106 is located in Dallas County and is represented by Representative
    Rodney Anderson, an Anglo Republican. Enacted HD 106 is relocated out of Dallas
    County.
    305. The voter demographics of Benchmark and enacted HD 106 are as follows:
    HD 106         Pop.      VAP      CVAP          HCVAP BCVAP WCVAP             SSVR
    Benchmark 159,716       110,146   81,165        29.0%   12.8%     52.0%      23.6%
    Enacted       161,947   110,568   74,515        8.8%    6.5%      80.1%      7.6%
    Pl.’s Exs. 13, 14.
    306. No election analysis was offered to the Court regarding this district.
    c. House District 144
    307. House District 144 is currently represented by an Anglo Republican and is not a
    minority-majority district in terms of citizen voting age population. Pl.’s Ex. 13.
    308. Minority preferred candidates do not win endogenous elections in this district. Handley
    House Report, at 5 (zero out of five elections); Engstrom Chart (zero out of five
    elections).
    56
    

Document Info

Docket Number: Civil Action No. 2011-1303

Citation Numbers: 887 F. Supp. 2d 133

Judges: Judge Rosemary M. Collyer

Filed Date: 8/28/2012

Precedential Status: Precedential

Modified Date: 8/31/2023

Authorities (40)

Arizona Minority Coalition for Fair Redistricting v. ... , 366 F. Supp. 2d 887 ( 2005 )

bridgeport-coalition-for-fair-representation-national-association-for-the , 26 F.3d 271 ( 1994 )

Janine Rudin v. Lincoln Land Community College , 420 F.3d 712 ( 2005 )

Hollinger v. Home State Mutual Insurance , 654 F.3d 564 ( 2011 )

league-of-united-latin-american-citizens-council-no-4434-and-jesse , 999 F.2d 831 ( 1993 )

ernestine-lewis-sylvester-j-lewis-and-minnetta-e-hill-v-alamance , 99 F.3d 600 ( 1996 )

LaRoque v. Holder , 650 F.3d 777 ( 2011 )

Johnson v. De Grandy , 114 S. Ct. 2647 ( 1994 )

White v. Weiser , 93 S. Ct. 2348 ( 1973 )

Diaz v. Kraft Foods Global, Inc. , 653 F.3d 582 ( 2011 )

Shelby County, Ala. v. Holder , 679 F.3d 848 ( 2012 )

earl-old-person-carol-juneau-bill-whitehead-herman-red-elk-ronald-williams , 230 F.3d 1113 ( 2000 )

Rodriguez v. Pataki , 308 F. Supp. 2d 346 ( 2004 )

City of Port Arthur, Tex. v. United States , 517 F. Supp. 987 ( 1981 )

Washington v. Davis , 96 S. Ct. 2040 ( 1976 )

Georgia v. United States , 93 S. Ct. 1702 ( 1973 )

White v. Regester , 93 S. Ct. 2332 ( 1973 )

City of Rome v. United States , 100 S. Ct. 1548 ( 1980 )

Reno v. Bossier Parish School Board , 120 S. Ct. 866 ( 2000 )

City of Lockhart v. United States , 103 S. Ct. 998 ( 1983 )

View All Authorities »