League of United Latin American Citizens 4552 v. Roscoe Independent School District

HIGGINBOTHAM, Circuit Judge: This case involves a challenge to the at-large election scheme used to elect trustees for the Roscoe Independent School District. The district court entered a final judgment declaring that the electoral structure did'not violate the Voting Rights Act of 1965, 42 U.S.C. § 1973. Plaintiff filed a timely notice of appeal. We have jurisdiction pursuant to 28 U.S.C. § 1291 and now AFFIRM. I. RISD is a small school district with limited financial resources. It spans 147 square miles in the northwest part of Nolan County, Texas. According to the 1990 census, RISD had approximately 1,786 residents with a racial makeup of 68.1% Anglo and 31.2% Mexican-American. The voting-age population of RISD as of the 1990 census was 1,277 with 73.6% Anglo and 25.5% Mexiean-American. Though RISD includes portions of four counties and substantial rural territory, 81% of the population in RISD lives in the town of Roscoe. During the 1994-95 school year, RISD had 438 students who were housed in two buildings, an elementary school and a combined middle school and high school. The Board of Trustees is responsible for the management and governance of RISD. Tex. Educ.Code Ann. § 11.051 (West 1996). The Board is comprised of seven members. Since 1995, voters residing in RISD have used a non-place, at-large election scheme to elect trustees. Under this structure, the candidates with the highest number of votes win according to the number of vacancies to be filled. There is no majority vote requirement. The trustees serve staggered three year terms such that two or three trustees are elected each year. Each voter is entitled to cast one vote for as many candidates as there are vacancies on the Board. For example, if two trustees are to be elected, then each voter is allowed to place one vote for two different candidates. However, a voter may single shot his vote and vote for fewer candidates than the number of positions to be filled. Cumulative voting is not allowed. There is one polling place in RISD. The election system used from 1978 to 1994 was identical to the one just described, except each candidate ran for a specific numbered position. RISD changed its voting scheme in response to pressure from minority groups complaining of dilution in their voting power. The Justice' Department approved RISD’s change in election structure and adoption of the current system. Appellant LULAC filed suit on June 22, 1994, seeking a declaratory judgment that the at-large election scheme used to elect trustees in RISD violated Section 2 of the Voting Rights Act of 1965, 42 U.S.C. § 1973, arid an injunction against the continued use of this electoral system. LULAC contended that RISD’s election system diluted Mexi-ean-American voting strength and denied these voters an equal opportunity to participate in the political process. After a bench trial, the district court held that LULAC had not proved a violation of the Voting Rights Act and entered a take nothing judgment. LULAC attacks the judgment on three bases. First, LULAC argues that the district court’s findings of fact and conclusions of law were insufficient under Fed.R.Civ.P. 52(a) and require the case to be remanded — 846 — for more detailed findings. Second, LULAC asserts that the district court’s findings with respect to the Gingles preconditions and the ultimate issue of vote dilution were clearly erroneous and merit reversal. Finally, LU-LAC claims that the district court improperly applied a strict scrutiny analysis. We reject each contention and affirm the district court’s judgment. II. We have often stressed the special need for detailed findings of fact in vote dilution cases. Rule 52(a) requires here that the record adequately reflects the factual and legal bases for the trial court’s decision. Westwego Citizens For Better Gov’t v. City of Westwego, 872 F.2d 1201, 1203 (5th Cir.1989) (Westwego I); Velasquez v. City of Abilene, 725 F.2d 1017, 1020 (5th Cir.1984); Cross v. Baxter, 604 F.2d 875, 879 (5th Cir.1979), vacated on other grounds, 704 F.2d 143 (5th Cir.1983). Our cases hold that the trial court has two primary obligations in making its findings in a vote dilution ease. First, the court must specifically state the evidence found credible and the reasons for its conclusions. Rollins v. Fort Bend Indep. Sch. Dist., 89 F.3d 1205, 1221 (5th Cir.1996); Westwego I, 872 F.2d at 1203-04. Second, the trial court must discuss all “substantial” evidence contrary to its decision. Velasquez, 725 F.2d 1017, 1020 (5th Cir.1984). However, “this Court does not require the district court to expressly mention all the evidence in its opinion.” Rollins, 89 F.3d at 1221. At the outset, it is important to note that LULAC had the burden of proof. It was required to prove by a preponderance of the evidence that all of the Gingles preconditions were satisfied and that based on the totality of the circumstances the at-large election system diluted the voting strength of Mexican-Americans in RISD. Overton v. City of Austin, 871 F.2d 529, 532 (5th Cir.1989). Any lack of evidence in the record regarding a violation of the Voting Rights Act of 1965 must be attributed to LULAC, not to the district court. LULAC’s primary quarrel with the district court’s findings of fact is that the court did not consider the testimony and exhibits of its expert witnesses regarding the Gingles preconditions and Zimmer factors. We disagree. The district court’s memorandum of opinion expressly considered the evidence the experts submitted and found it to be unimpressive. The trial court specifically found RISD’s expert’s testimony to be “much more persuasive than that of the plaintiffs witnesses.” Rec. Vol. 7 at 1786. Much of this trial was a familiar battle of experts. “The credibility determination of witnesses, including experts, is peculiarly within the province of the district court.” Orduna S.A. v. Zen-Noh Grain Corp., 913 F.2d 1149, 1154 (5th Cir.1990). Consequently, we give deference to the findings and credibility choices trial courts make with respect to expert testimony. I.U. Tech. Corp. v. Research-Cottrell, Inc., 641 F.2d 298, 305 n. 7 (5th Cir.1981). The district court found RISD's expert to be the more credible expert and that his testimony strongly rebutted that of opposing experts. As a result, the district court was not persuaded that LULAC’s experts offered substantial contrary evidence. We see no reason to dispute the district court’s evaluation of the experts or to remand this case because the district court did not discuss in its order the testimony of unpersuasive witnesses. LULAC asserts that our holding in Teague v. Attala County, 17 F.3d 796 (5th Cir.1994), requires a remand for additional findings of fact. LULAC’s reliance on Teague is misplaced. In Teague, we remanded for a more in-depth analysis of the plaintiffs’ statistical evidence and for determinations on the credibility of the trial witnesses. Teague, 17 F.3d at 798. No such concerns arise here. The main reason for remand in voting rights cases is for explanation of the district court’s treatment of statistical data. See, e.g., Houston v. Lafayette County, 56 F.3d 606, 611-13 (5th Cir.1995); Clark v. Calhoun County, 21 F.3d 92, 96 (5th Cir.1994) (Clark I); Teague, 17 F.3d at 798; Westwego Citizens For Better Government v. City of Westwego, 906 F.2d 1042, 1044 (5th Cir.1990) (Westwego II). LULAC’s able counsel was unable to generate substantial statistical evidence as compared to many of our vote dilu — 847 — tion cases.1 The reality is that the database was thin, given the manner of conducting the elections. The statistics LULAC gathered were from an exit poll of the City of Roscoe election and thus are entitled to less weight than if they had derived from a Board of Trustees election in RISD. Magnolia Bar Ass’n v. Lee, 994 F.2d 1143, 1149 (5th Cir.1993), cert. denied, 510 U.S. 994, 114 S.Ct. 555, 126 L.Ed.2d 456 (1993). Finally, unlike in Teague, the district court in this case made known its credibility determinations and the evidence upon which its conclusions of law were based. Though this Court has instructed trial courts “to thoroughly discuss the statistics offered by making specific references to the evidence”, Rollins, 89 F.3d at 1221, and the district court’s discussion of statistical data in this case was closemouthed at best, there are no grounds to remand given the evidence at trial. III. LULAC objects to virtually every factual and legal conclusion the trial court made. “We need not address all of [its] contentions, however, because failure to establish any single criterion of [Gingles] is fatal to [its] case.” Overton, 871 F.2d at 538. “Failure to establish any one of the Gingles factors precludes a section 2 violation, because ‘[t]hese circumstances are necessary preconditions for multimember districts to operate to impair minority voters’ ability to elect representatives of their choice.’ ” Magnolia Bar Ass’n, 994 F.2d at 1146 (quoting Thornburg v. Gingles, 478 U.S. 30, 50, 106 S.Ct. 2752, 2766, 92 L.Ed.2d 25 (1986)). We review the district court’s findings on the Gingles threshold requirements for clear error. Magnolia Bar Ass’n, 994 F.2d at 1147. Since the district court’s ruling with respect to Gingles III was not clearly erroneous, we affirm its entry of judgment against LULAC. The third Gingles precondition requires the plaintiff to show that the Anglo majority votes sufficiently as a bloc to enable it, absent special circumstances, usually to defeat the minority’s preferred candidate. Gingles, 478 U.S. at 51, 106 S.Ct. at 2766. By establishing this circumstance, “the minority group demonstrates that submergence in a white multimember district impedes its ability to elect its chosen representatives.” Id. A white majority bloc voting pattern is ordinarily established through evidence of racially polarized voting. Westwego I, 872 F.2d at 1207. Since RISD only has one polling place, LULAC could not use commonly employed statistical analyses like ecological regression analysis or extreme case analysis to prove the existence of racially polarized voting in the district. Instead, LULAC had to reach for other evidence in an attempt to satisfy the third Gingles requirement. In particular, LULAC noted that only one Mexican-American had ever been elected to the Board, that in the 1995 school board election the sole Mexican-American candidate garnered only 14% of the Anglo vote, and that an exit poll of the 1995 City of Roscoe election revealed that 78% of the Anglo voters surveyed did not cast any of their five votes for the sole Mexican-American candidate, who was victorious.2 In addition, LULAC highlighted the fact that since 1978 95.3% of the winning candidates in the school board elections have been Anglo and 4.7% have been Mexican-American. This was not, however, the only evidence bearing on the third Gingles precondition. RISD showed that a Mexican-American, Jose Villafranca, had been elected to the Boax’d in 1991 and re-elected in 1993. LU-LAC contends that Villafranca won in 1991 because two Anglo candidates split the white vote and that Villafranca was not the preferred candidate of Mexican-Americans. — 848 — The district court did not find either assertion to be true. In the 1991 election, Villafranea received 153 votes, while the two Anglo candidates received 143 votes and 60 votes, respectively. LULAC adduced no evidence at trial showing that Villafranea would have lost if there had only been one Anglo competitor. In fact, Villafranea defeated the only Anglo candidate to oppose him in the 1993 school board election. We do not think the district court’s finding that Villafranea’s success in 1991 was not due to “special circumstances” was clearly erroneous. We reach the same conclusion regarding the district court’s determination that Villaf-ranca was the preferred candidate of the Mexican-Ameriean community. Mexican-American residents in RISD testified to that effect. The trial court found LULAC’s witnesses who testified to the contrary to be unreliable. Since we give deference to the trial court’s credibility assessments, we cannot say its finding on this matter was clearly erroneous. As for the 1993 school board election, Vil-lafi*anea received a majority of the votes in defeating his Anglo and Mexican-American competitors.3 Villafranea had also twice won election to the Roscoe City Council. Moreover, a Mexican-Ameriean was elected to the Roscoe City Council in 1995, which according to LULAC’s expert would not have occurred but for the Anglo cross-over vote. Though minority electoral success alone cannot act to defeat a vote dilution claim, Gingles, 478 U.S. at 75, 106 S.Ct. at 2779, we have consistently recognized that such success and racially polarized voting are the two most probative factors in evaluating the merits of such an allegation. Clark v. Calhoun Co., 88 F.3d 1393, 1397 (5th Cir.1996) (Clark II). Given the degree of minority success in this case and the failure of LULAC to produce sufficient evidence showing that Anglo and Mexican-American voters in RISD vote along strict racial lines, we are not left with the definite and firm conviction that the district court made a mistake in finding that LULAC failed to meet the third Gingles precondition. Because “ ‘the district court’s account of the evidence is plausible in light of the record viewed in its entirety,’ its findings will not be reversed.” Magnolia Bar Ass’n, 994 F.2d at 1147 (quoting Anderson v. City of Bessemer City, 470 U.S. 564, 573, 105 S.Ct. 1504, 1511, 84 L.Ed.2d 518 (1985)). IV. LULAC urges that the district court improperly evaluated its case under a strict scrutiny analysis. LULAC bases its argument on the last paragraph of the trial court’s order. Applying a strict scrutiny test, this court finds that the plaintiff has failed to prove that under the totality of the circumstances, its members have less opportunity than other members of the electorate to participate in the political process and to elect serious, preferred candidates of the Hispanic community to the board of trustees of RISD. Rec. Vol. 7 at 1788. The use of the strict scrutiny language is puzzling, but the district court did not apply strict scrutiny to the voting claim. It used the legal analysis announced in Gingles and explicated in our precedent. This stranger to the case wandered in but made no impression. We are pointed to no prejudice and have found none. V. RISD appeals the denial of its motion for attorneys’ fees. A prevailing defendant in a Voting Rights Act case is entitled to an award of attorneys’ fees if the plaintiffs claim was frivolous, unreasonable, or groundless. 42 U.S.C. § 1973i(e); see Christiansburg Garment Co. v. EEOC, 434 U.S. 412, 422, 98 S.Ct. 694, 700-701, 54 L.Ed.2d 648 (1978). This is not such a case. Nor do we find that the district court abused its discretion in taxing $6,965.95 in costs against LULAC. These costs are for exemplification — 849 — and copies necessarily obtained for use in the case and for expenses incident to taking depositions. They are high but are within the purview of 28 U.S.C. § 1920 and within the discretion of the district court. VI. The judgment of the district court is AFFIRMED. Footnotes [1] The only statistical evidence LULAC presented at trial was the results of an exit poll for the City of Roscoe elections. This paucity of statistical data is dramatically less than that adduced in the vote dilution cases we have remanded for more detailed findings. See, e.g., Houston, 56 F.3d at 609 (noting plaintiff’s evidence from fourteen elections); Teague, 17 F.3d at 797 (commenting on plaintiffs' data from eight elections). [2] Cumulative voting was used in 1995 for the City of Roscoe elections. [3] In 1993, there were one Anglo candidate and one Mexican-American candidate running against Villafranea.