Smith v. Clinton

ARNOLD, Circuit Judge. The issue in this case is whether at-large voting for two seats in the Arkansas House of Representatives from dual-member District 48-49 in Crittenden County denies black voters an equal opportunity to participate in the political process, in violation of section 2 et seq. of the Voting Rights Act of 1965, as amended, 42 U.S.C. § 1973 et seq. 1 We have made a careful, practical assessment of the impact of the challenged multi-member structure, “in the light of past and present reality, political and otherwise.” See White v. Regester, 412 U.S. 755, 769-70, 93 S.Ct. 2332, 2341, 37 L.Ed.2d 314 (1973). We now hold that the plaintiffs have demonstrated a violation of their rights under federal law. The crux of our decision is this: at-large election of representatives in this multimember structure so dilutes the voting strength of black residents of the district as virtually to guarantee that no black person will ever be elected State Representative in Crittenden County. The Court also finds that it is feasible to divide District 48-49 into two contiguous, rationally sized legislative districts, with black voters in the majority in one of the districts. Accordingly, we hold that the March 8, 1988 primary election for State Representative from the district must be set aside, and a plan to divide District 48-49 into two districts must be implemented in time for a special primary election to be held before the November 1988 general election. The order of this Court is intended to affect no portion of the Arkansas legislative apportionment structure except that in District 48-49. I. Members of the Arkansas House of Representatives are elected every two years for two-year terms. Ark. Const. Art. V, § 2. Candidates for the Legislature must obtain a majority of the vote in order to win party nomination. Ark.Code Ann. §§ 7-7-102(a), 7-7-304(a)(2) (1987). If no candidate receives a majority of votes in the primary, a run-off election is held. Ark.Code Ann. § 7-7-304(a)(3) (1987). The parties have stipulated that seventy-four of the one hundred members of the State *1312 House of Representatives are elected from single-member districts. House District 48-49 is a single geographic entity from which two State Representatives are elected through at-large voting. Candidates designate District 48 or District 49 as the seat for which they choose to run. District 48-49 is located entirely within Crittenden County, and encompasses most of the county’s area and ninety-four percent of its population. There is no subdistrict-residency requirement for State Representative candidates. According to estimates based on 1980 census data, approximately forty-two percent of the population of District 48-49 is black, and about thirty-seven percent of the district residents who are of voting age are black. No black candidate has ever been elected State Representative from District 48-49; nor has any black person been elected to the Arkansas General Assembly from Crittenden County in the Twentieth Century. Plaintiffs Elbert Smith and others, black registered voters in District 48-49, filed this action on January 15, 1988, challenging on statutory and constitutional grounds the multimember structure of the district. The defendants include the Governor, Secretary of State, and Attorney General of Arkansas (members of the State Board of Apportionment), and the principal officers of the Democratic and Republican parties in the State. Plaintiffs alleged that the multi-member legislative district adversely affects the opportunity for black residents to participate in the political process, in violation of the Voting Rights Act of 1965, the Fourteenth and Fifteenth Amendments to the United States Constitution, and 42 U.S. C. § 1983. The plaintiffs also moved to enjoin the March 8, 1988 primary election for State Representatives from the district. On February 16, 1988, this Court held a hearing on the plaintiffs’ motion for a preliminary injunction. After hearing evidence for a day, we denied the motion, stating our reasons from the bench. The plaintiffs, we said, had shown a strong likelihood of success on the merits, but we could not say at that early stage of the case that defendants’ chances were negligible. Candidates had already filed and done a good deal of campaigning. It seemed best to let the election be held, subject always to our ability to set it aside, if plaintiffs should win at the trial on the merits, in time to re-run the election in single-member districts in 1988. In making this decision we were heavily influenced by a reluctance to interfere with state election processes on less than a full record. So the primary election 2 was held on March 8. State Representative Lloyd McCuiston, who is white, ran for one seat without opposition, and he was renominated. State Representative James Stockley, who is white, defeated Ben McGee, who is black, for the nomination for the other seat. The trial on the merits of this case followed on April 25 and 26. II. Initially, we address the defendants’ contention that we need not reach the merits of this case. The defendants suggest that the plaintiffs are barred because of the passage of time since the State Board of Apportionment formulated the present state-wide plan in 1981. The necessary premise of this argument is that unless voters challenge an apportionment plan promptly upon its being put into effect, they are barred from contesting a district until a subsequent reapportionment. 3 We conclude that neither the doctrine of laches nor the expiration of an applicable limitations period presents an obstacle to the plaintiffs’ case under the Voting Rights *1313 Act. First, the injury alleged by the plaintiffs is continuing, suffered anew each time a State Representative election is held under the multimember structure. Second, there have been significant developments since the 1981 Arkansas reapportionment. The Voting Rights Act was amended in 1982, and the Supreme Court’s interpretation of the statute in Thornburg v. Cringles, 478 U.S. 30, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986), lays down a rather uncompromising structure for the application of the law in vote-dilution cases. Third, we note that the plaintiffs are required to prove that, as a result of the challenged structure, a white majority bloc is usually able to defeat the preferred candidates of the minority. Of course, evidence of this circumstance would be unavailable unless the structure had been in place for some time. It is suggested that the complaint should be dismissed because indispensable parties have not been joined, those parties being the members of the State Legislature affected by the case. There is no question that they are affected, but we do not believe that they are indispensable in the legal sense. It was open to members of the Legislature to intervene and become parties, if they wished to do so. Both of them have been in the courtroom throughout this case, and one of them testified. In any event, their legal interests have been considered just as fully as if they were parties. The defendants next contend that the plaintiffs may not challenge one district out of the whole state reapportionment plan. Plaintiffs, they say, must either challenge the whole plan or state that they have no objection to any part of it other than District 48-49. This argument, too, is without merit: assuming that there is a State policy that would require a suit filed in State court to challenge the whole plan or not at all, see Bizzell v. White, 274 Ark. 511, 625 S.W.2d 528 (1981) (per curiam) (4-3 decision), such a policy would not govern the application of federal law by a federal court. If the plaintiffs are injured by a particular aspect of the plan and are not affected at all by the rest of it, there is no reason why they should not be allowed to challenge the aspect of the plan that affects them. 4 Indeed, they would presumably lack standing under Article III to challenge aspects of the plan (for example, multimember districts in other parts of the State) that do not affect them at all. 5 Accordingly, we proceed to discuss the law applicable to the merits of this case. III. Section 2 of the Voting Rights Act of 1965, as amended and codified, reads as follows: (a) No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color, or in contravention of the guarantees set forth in section 1973b(f)(2) of this title, as provided in subsection (b) of this section. *1314 (b) A violation of subsection (a) of this section is established if, based on the totality of the circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of a class of citizens protected by subsection (a) of this section in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office in the State or political subdivision is one circumstance which may be considered: Provided, That nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population. 42 U.S.C. § 1973. Congress amended Section 2 in 1982 to make clear that a violation of the Act may be established by a showing of discriminatory effect stemming from an electoral mechanism; it is not necessary that minority voters prove that the system was adopted or has been maintained for a discriminatory purpose. Thornburg, 106 S.Ct. at 2759; S.Rep. No. 417, 97th Cong., 2d Sess. 28 (hereafter S.Rep.), reprinted in 1982 U.S.Code Cong. & Admin.News 177, 205. In determining whether the Act has been violated, we may consider the following typical factors suggested by the Senate Judiciary Committee Report: 1. the extent of any history of official discrimination in the state or political subdivision that touched on the right of the members of the minority group to register, to vote, or otherwise to participate in the democratic process; 2. the extent to which voting in the elections of the state or political subdivision is racially polarized; 3. the extent to which the state or political subdivision has used unusually large election districts, majority vote requirements, anti-single shot provisions, or other voting practices or procedures that may enhance the opportunity for discrimination against the minority group; 4. if there is a candidate slating process, whether the members of the minority group have been denied access to that process; 5. the extent to which members of the minority group in the state or political subdivision bear the effects of discrimination in such areas as education, employment and health, which hinder their ability to participate effectively in the political process; 6. whether political campaigns have been characterized by overt or subtle racial appeals; 7. the extent to which members of the minority group have been elected to public office in the jurisdiction. S.Rep. at 28-29, reprinted in 1982 U.S. Code Cong. & Admin.News at 206-07. Other factors we may consider include whether there is a significant lack of responsiveness to the particularized needs of the minority group and whether the policy underlying the election practice or procedure is tenuous. Id. This Court is required to consider all of the circumstances and applicable factors to answer the ultimate question: “whether[,] ‘as a result of the challenged practice or structure^] plaintiffs do not have an equal opportunity to participate in the political processes and to elect candidates of their choice.’ ” Thornburg, 106 S.Ct. at 2763 (quoting S.Rep. at 28, reprinted in 1982 U.S.Code Cong. & Admin.News at 206). Plaintiffs are not required to prove the existence of any certain number of the factors listed above, nor are they required to show that a majority of the factors “ ‘point one way or the other.’ ” Id. at 2764 (quoting S.Rep. at 29, reprinted in 1982 U.S.Code Cong. & Admin.News at 207). We must conduct a “searching practical evaluation of the ‘past and present reality.’ ” S.Rep. at 30, reprinted in 1982 U.S. Code Cong. & Admin.News at 208. In the context of vote-dilution cases, this evaluation is shaped by consideration of three circumstances. First, the black voters *1315 must show that their numbers are sufficiently large and geographically compact to constitute a majority in a single-member district. Thornburg, 106 S.Ct. at 2766. Second, the plaintiffs must show that the group to which they belong is politically cohesive. Id. at 2767. Third, the black voters must show that “the white majority votes sufficiently as a bloc to enable it — in the absence of special circumstances, such as the minority candidate running unopposed ... usually to defeat the minority’s preferred candidate.” Ibid. The latter two of these factors may be demonstrated by a showing that voting in the jurisdiction is highly racially polarized. Id. at 2769. With this legal framework in mind, we turn now to the facts developed during the trial of this case. IV. A. The plaintiffs have submitted a proposed plan for division of District 48-49 into two separate districts, in one of which blacks would constitute about sixty-one percent of the voting-age population and sixty-five percent of the overall population. The defendants do not contest the fact that such a district could be created. Accordingly, we find that the black population in District 48-49 is sufficiently large and compact so that a separate district with black voters in the majority could be created. At this time we express no view as to the merit, relative to any other plan that might be developed, of the plan submitted by the plaintiffs. B. The plaintiffs have offered the report and testimony of an expert witness, Dr. Allan Lichtman, as to the extent of racially polarized voting in the district and county. The report summarizes a study of nine of the elections involving black candidates in Crittenden County since 1978. Using regression analysis 6 Dr. Lichtman examined contests between black and white candidates for State Representative from District 48-49 in 1978, 1986, and 1988. He also studied the results from Crittenden County in the 1988 Democratic presidential primary, and mayoral elections in the cities of West Memphis (1982 and 1986) and Earle (1985 and 1986). Among black voters, the mean level of support for the black candidates in these nine races was eighty-nine percent. The mean level of support among whites for these black candidates was nine percent. Excluding the mayoral contests, the mean level of support for black candidates was ninety-seven percent among black voters; the mean level of support for black candidates among white voters was nine percent. The levels of support for black candidates by black and white voters in three State Representative races from District 48-49 are of particular interest to the Court. In the 1988 primary election for State Representative, an estimated one hundred percent of black voters supported the black candidate, McGee; ninety-three percent of white voters supported the white candidate, Stockley. In 1986 ninety-five percent of black voters supported the *1316 black candidate, and eighty-five percent of white voters supported the white candidate. In 1978 the black candidate collected the votes of ninety-nine percent of black voters and the white candidate got the votes of ninety-two percent of white voters. The evidence also indicates that the Rev. Jesse Jackson, the black candidate in the Democratic presidential primary, in 1988 received ninety-four percent of the votes of blacks in the county and five percent of the votes of whites. In races for mayor of West Memphis, the largest city in the district, the black candidate received ninety-six percent of votes cast by blacks in 1986 and an estimated one hundred percent of the votes cast by blacks in 1982. In both of these races, the estimated percentage of whites voting for the black candidate was zero. Supplementing his regression analysis with extreme-case, or homogeneous-precinct analysis, Dr. Lichtman testified that in precincts with ninety percent or more black population, the black State Representative candidates received ninety-five percent of the vote in 1988, eighty-nine percent of the vote in 1986, and eighty-seven percent in 1978. In precincts that were ninety percent or more white, black candidates received fourteen percent of the vote in 1988, twenty-one percent in 1986, and nine percent in 1978. Dr. Harold Stanley, the expert witness for the defendants, testified that the figures derived from his own regression analysis of voting returns in Crittenden County were “substantially close” to those collected by Dr. Lichtman for the same elections, those between black and white candidates. Dr. Stanley further testified, however, that focusing only on races involving black candidates against whites does not present an accurate picture of voting in the jurisdiction. Accordingly, the defendants offered evidence indicating that (1) considering most county-wide races and races for mayor in West Memphis and Earle in the last twelve years, blacks and whites most often voted together to elect candidates of their collective choice; and (2) the preferred candidate of black voters most often finished first within the jurisdiction. Dr. Stanley analyzed sixty-five races between 1976 and 1988 and determined that in fifty-four of these elections the candidate receiving a majority of the votes cast by blacks placed first in Crittenden County. He also calculated that in all of these elections combined, first-place finishers received sixty-five percent of the vote of blacks; by comparison, fifty-six percent of the votes cast by whites went to the ultimate firstplace finisher within the county. On the basis of this study, Dr. Stanley testified that the overall pattern in Critten-den County is that blacks and whites usually vote together, and the candidates preferred by blacks often place first. The defendants also presented evidence that, in State Representative campaigns between white candidates, a majority of black voters supported the winning candidate on four of five occasions. The plaintiffs have objected to the admissibility of much of the defendants’ evidence, on the grounds of relevance. In particular, they say that only evidence involving races between black and white candidates can be considered. We assume without deciding that all of the evidence offered by the defendants is admissible and properly to be considered. We make this assumption because the result in this case would be the same either way. First, we believe it is proper to give considerable weight to the evidence of polarization in elections between black and white candidates. In Thornburg, the Supreme Court relied heavily on such evidence. 106 S.Ct. at 2771-72. Further, in a functional assessment of the political process, one of the most important factors is the extent to which members of the minority group have been elected to office in the jurisdiction. Id. at 2766 n. 15. Whether our assessment of the political realities in Crittenden County is expressed in terms of “black candidates” or “candidates preferred by black voters,” our conclusion is still that “minority group members prefer certain candidates whom they could elect *1317 were it not for the interaction of the [multi-member] structure with a white majority that votes as a significant bloc for different candidates,” id. at 2776 (opinion of Brennan, J.). Second, the most probative indication of vote dilution stemming from the multimember structure of the district is the results of the State Representative elections. This case is about a particular electoral structure and its effect on minority participation. Certainly, there is evidence that white candidates preferred by black voters sometimes win in elections involving only whites. The evidence of polarized voting in State Representative elections involving blacks against whites is so strong, however, that it cannot be overcome even when all reasonable inferences are accorded to the evidence of elections involving only white candidates. The defendants have offered evidence of other reasons, such as incumbency, for the choices made by voters in the district. The materiality of this evidence is questionable. See Thornburg, 106 S.Ct. at 2773 (focus of case under Voting Rights Act is difference in choices made by voters, not reasons for this difference) (opinion of Brennan, J.). In any event, we assume for present purposes that this evidence may have some bearing on whether the particular election results on which we focus stem from the multi-member structure of the district. We find this evidence insufficient to compel a different result in this case, given the sharp polarization in races involving black candidates for State Representative. Nor are we persuaded by evidence of some success by black candidates in school-board elections or other local races, because the electoral structure at issue here has no effect on these candidates. We find, therefore, that there is racially polarized voting in the district, despite the fact that blacks and whites often prefer the same candidate in races involving only whites. The purpose of our inquiry into polarized voting is twofold: “to ascertain whether minority group members constitute a politically cohesive unit and to determine whether whites vote sufficiently as a bloc usually to defeat the minority’s preferred candidates.” Id. at 2769. Upon consideration of all the evidence offered by both parties, we find that there is racially polarized voting in Crittenden County and District 48-49; black voters usually vote cohesively, as a unit; and white voters have the strength under the present multi-member structure to enable them to frustrate the choices made by black voters. C. We make the following additional findings: (1) The Court takes judicial notice that there is a history of racial discrimination in the electoral process in Arkansas. See Perkins v. City of West Helena, 675 F.2d 201, 211 (8th Cir.), aff'd mem., 459 U.S. 801, 103 S.Ct. 33, 74 L.Ed.2d 47 (1982). We do not believe that this history of discrimination, which affects the exercise of the right to vote in all elections under state law, must be proved anew in each case under the Voting Rights Act. (2) We further find that the history of discrimination has adversely affected opportunities for black citizens in health, education, and employment. 7 The hangover from this history necessarily inhibits full participation in the political process. (3) State law has no provision forbidding “bullet” or “single-shot” voting, but the *1318 effectiveness of any such practice would be negated by the requirement that candidates run for designated seats. Arkansas law requires candidates for state representative to secure a majority of votes in a primary election in order to become the party nominee. This district does not seem to us unusually large, as multimember districts go. There is no subdistrict residency requirement. No evidence as to any slating process has been presented. (4) The plaintiffs have offered no evidence of discriminatory intent in the formation of District 48-49, and we necessarily find on this record that there was no such intent. Plaintiffs have not proved a claim under the Constitution. (5) Neither side has presented convincing evidence about the voter-registration rates of the respective races. (6) The plaintiffs have not attempted to demonstrate lack of responsiveness to the needs of black citizens on the part of officials in the district. The defendants have presented evidence which indicates that county officials have been responsive to some of the concerns particular to blacks and have made attempts to make the political process accessible to blacks. Accordingly, we find no present lack of responsiveness to the needs of blacks has been shown. State policy with respect to multi-member districts (26 members of the House chosen from such districts, and no Senators) is not tenuous, but neither is it especially strong. (7) The evidence does not indicate that overt or subtle racial appeals have been made. V. The Supreme Court has held that the most important factors in a Voting Rights Act challenge to a multimember district are racial polarization in voting and whether minority-group members have won elections in the jurisdiction in question — here, District 48-49. Thornburg, 106 S.Ct. at 2766 n. 15. Other factors listed in the Senate Report, if present, “are supportive of, but not essential to, a minority voter’s claim.” Ibid, (emphasis in original). Plaintiffs have presented strong, perhaps overwhelming, proof of the essential elements of their claim. Some of the other factors favor plaintiffs, some do not, and some are just not in the case. On balance, we find that the multimember configuration of District 48-49 deprives black citizens of an equal opportunity to participate in the political process and to elect candidates of their choice. So long as the present situation is allowed to continue, there will probably never be a black State Representative from Crittenden County, even though, as detailed above, black citizens are numerous enough to have a clear majority in a single-member district. Candidates favored by blacks can win, but only if the candidates are white. Accordingly, the result of the March 8, 1988, primary election for State Representative in District 48-49 is hereby set aside. Defendants are restrained and enjoined from certifying the winners of that primary as party nominees. The defendant members of the State Board of Apportionment and the plaintiffs are ordered to submit proposed plans for separate Districts 48 and 49 on or before May 27, 1988. 8 This Court will convene on June 6, 1988, to consider these plans. It is so ordered. Footnotes [1] On April 26, 1988, at the end of a two-day trial, we announced findings and conclusions from the bench, stating that a formal opinion would be filed in due course. This is that opinion. [2] We refer to the Democratic primary. No one filed for the Republican nomination for either seat. [3] The defendants have not indicated what they believe is the applicable limitations period. Arkansas law provides that the State shall be reapportioned in the year following each federal census, and that this apportionment is effective unless contested in the State Supreme Court within 30 days. Ark. Const. Art. VIII, §§ 4, 5. We do not believe this time frame for a special remedy under state law is of consequence under the Voting Rights Act, and the defendants do not so contend. [4] The Arkansas Supreme Court’s opinion in Biz-zeil, in any case, rests in part on one factor not present here. The challenge made there was to a single Senate District, but if the challenge had been upheld, the boundary lines of that district would have had to be re-drawn, necessarily producing a "ripple effect” on some other districts. No such argument is possible here. Plaintiffs seek relief only within District 48-49. The outer boundaries of that district, which will now become two districts, will not be affected. [5] Defendants moved for a continuance one week before trial. We denied the motion. Defendants had ample notice of the trial setting, and time was important in getting this case decided during 1988. The main ground of the motion was the need for more time to develop expert testimony. At trial, defendants in fact presented creditable and reputable expert testimony. Their lack of success is due to the legal standard laid down in Thornburg. We do not believe that additional time would have enabled defendants to improve their case materially. Their side was well and fully presented, within the rather strict limits of the law. We also deny plaintiffs’ motion for class certification. It was not filed within ninety days of the complaint, as required by E.D. Ark. R. 24. This action makes little if any practical difference. [6] Dr. Lichtman described regression analysis as the "standard method for inferring the behavior of population groups from data collected for political units.” Defendants do not dispute this. The role of this statistical methodology in Voting Rights Act cases has been explained in this manner: Correlation and regression analysis is used to determine the degree of relationship between two variables, such as the racial composition of the precincts in a city or district and the electoral support provided a particular candidate or group of candidates within those precincts. The independent variable in the analysis is a summary measure of the race of the voters in each precinct (usually the percentage of blacks among the total population, voting age population, or registered voters). The dependent variable is the percentage of voters in each precinct casting ballots for the specific candidate or group of candidates. ... Correlation and regression analysis provides statistical coefficients that summarize the consistency and strength of the relationship between the two variables. Engstrom & McDonald, Quantitative Evidence in Vote Dilution Litigation: Political Participation and Polarized Voting, 17 Urb. Lawyer 369, 374-75 (1985). [7] The plaintiffs’ evidence shows, for example, that the per capita income of blacks in the county was $2,293 according to the 1980 census, about one-third that of white residents. In 1980, 58.6% of blacks in the county had income below the poverty level (and 10.7% of whites); 19.6% of blacks over 25 were high school graduates (and 61.4% of whites); and the median value of homes owned by blacks was $16,700 (and $39,200 for whites). ”[C]ourts have recognized that political participation by minorities tends to be depressed where minority group members suffer effects of prior discrimination such as inferior education, poor employment opportunities, and low incomes.” Thornburg, 106 S.Ct. at 2776 (citations omitted) (opinion of Brennan, J.). [8] An order to this effect was entered on May 2, 1988. At trial counsel for defendants stated that ten days would be enough time for them to prepare a plan.