Full opinion text
FINDINGS OF FACT AND CONCLUSIONS OF LAW GLEESON, District Judge: This is an action brought under Section 2 of the Voting Rights Act of 1965, 42 U.S.C. § 1973, and the First, Thirteenth, and Fourteenth Amendments to the United States Constitution. Plaintiffs, African-American citizens of the Town of Hempstead (the “Town”), challenge the at-large voting practice used in the Town to elect the six Council-members who sit on the Town Board of the Town of Hempstead (the “Town Board”). Plaintiffs contend that this system dilutes the voting strength of the minority population in the Town in violation of the Voting Rights Act, and seek an order directing the implementation of a single-member district election system. The complaint was filed on August 8,1988. On November 16, 1994, the ease was reassigned to this Court. Discovery was completed in February 1996. Defendants moved for summary judgment, which motion was argued and denied on April 5, 1996. A three-week bench trial was held in July 1996, and closing arguments by counsel were heard on August 14,1996. This decision is arranged in three parts. Part I sets forth a brief overview of the legal principles governing claims under Section 2 of the Voting Rights Act. Part II sets forth my findings of fact. Part III sets forth my conclusions of law. I conclude that plaintiffs have established a violation of Section 2 and are entitled to the relief they seek. I. OVERVIEW OF THE LEGAL PRINCIPLES GOVERNING CLAIMS UNDER SECTION 2 OF THE VOTING RIGHTS ACT Section 2 of the Voting Rights Act of 1965, as amended, reads in pertinent part 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____ (b) A violation of subsection (a) of this section is established if, based on the totality of 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 (1994). There is no dispute that the at-large system for electing the members of the Town Board is an electoral practice that is subject to challenge under Section 2. In addition, it is clear that the 1982 amendments to Section 2 rejected the plurality opinion in Mobile v. Bolden, 446 U.S. 55, 100 S.Ct. 1490, 64 L.Ed.2d 47 (1980), which sought to impose on Section 2 plaintiffs an obligation to prove that the challenged practice was adopted or maintained with the intent to discriminate against minority voters. Thornburg v. Gingles, 478 U.S. 30, 43-44, 106 S.Ct. 2752, 2762-63, 92 L.Ed.2d 25 (1986). In Gingles, the Supreme Court identified three “preconditions” to a successful challenge to multimember districts under Section 2: (1) the minority group must be sufficiently large and geographically compact to constitute a majority in a single-member district; (2) it must be politically cohesive; and (3) the white majority must vote sufficiently as a bloc to enable it, in the absence of special circumstances, to defeat the minority’s preferred candidate. 478 U.S. at 50-51, 106 S.Ct. at 2766-67. “Regarding the third prong, Gingles instructed that ‘in general, a white bloc vote that normally will defeat the combined strength of minority support plus white “crossover” votes rises to the level of legally significant white bloc voting.’ ” NAACP v. City of Niagara Falls, 65 F.3d 1002, 1007 (2d Cir.1995) (quoting Gingles, 478 U.S. at 56, 106 S.Ct. at 2769). The satisfaction of these three preconditions, which have come to be known as the “Gingles factors,” is necessary but not sufficient to prove a Section 2 violation. More is required; if the preconditions are met, a district court must then consider whether, under the totality of the circumstances, the challenged practice impairs the ability of the minority voters to participate equally in the political process. City of Niagara Falls, 65 F.3d at 1007 (citing Gingles, 478 U.S. at 80, 106 S.Ct. at 2781-82). Thus, proof of the Gingles preconditions does not always portend liability under Section 2, Johnson v. De Grandy, 512 U.S. 997, 1012 n. 10, 114 S.Ct. 2647, 2657 n. 10, 129 L.Ed.2d 775 (1994), but “it will only be the very unusual case in which the plaintiffs can establish the existence of the three Gingles factors but still have failed to establish a violation of § 2 under the totality of circumstances.” City of Niagara Falls, 65 F.3d at 1019 n. 21 (citing Jenkins v. Red Clay Consolidated School Dist. Bd. of Educ., 4 F.3d 1103, 1135 (3d Cir.1993), cert. denied, 512 U.S. 1252, 114 S.Ct. 2779, 129 L.Ed.2d 891 (1994)); Teague v. Attala County, 92 F.3d 283, 293 (5th Cir.1996); Clark v. Calhoun County, 21 F.3d 92, 97 (5th Cir.1994); see also Johnson v. De Grandy, 512 U.S. at 1012 n. 10, 114 S.Ct. at 2657 n. 10 (noting that § 2 challenges to multimember districts are likely to be easier plaintiffs’ cases than challenges to electoral practices in single-member districts). In assessing the totality of the circumstances in which a challenged electoral practice is employed, a district judge must conduct a “searching practical evaluation of the ‘past and present reality.’ ” Gingles, 478 U.S. at 45, 106 S.Ct. at 2764 (quoting S.Rep. No. 417, 97th Cong., 2d Sess. (1982) (“Senate Report”) at 30, reprinted in 1982 U.S.Code Cong. & Admin.News (“U.S.C.C.A.N.”) 177, at 208). This requires a fact-intensive, functional view of the political process, in an effort to determine whether the Town’s political processes are equally open to participation by its black citizens. Id. at 45-46, 106 S.Ct. at 2763-64. Gingles identified as relevant to this inquiry the factors set forth in the Senate Report (which accompanied the 1982 amendments to Section 2), which are as follows: 1.the extent of any history of official discrimination in the state or political subdivision that touched 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; 8. whether there is significant lack of responsiveness on the part of elected officials to the particularized needs of the members of the minority group; and 9. whether the policy underlying the state or political subdivision’s use of the challenged voting qualification, prerequisite to voting, or standard, practice or procedure is tenuous. See Senate Report at 28-29, 1982 U.S.C.C.A.N. 206-07, cited in Gingles, 478 U.S. at 36-37, 44-45, 106 S.Ct. at 2758-59, 2762-64; see also City of Niagara Falls, 65 F.3d at 1007-08. The factors set forth in the Senate Report are not exclusive, and are not to be applied mechanically. A failure by plaintiffs to establish a particular factor “is not rebuttal evidence of non-dilution.” Senate Report at 29 n. 118, 1982 U.S.C.C.AN. at 207. Nor should the outcomes of these factors be tallied, as though a majority of them, or particular ones among them, must weigh in plaintiffs’ favor before a Section 2 violation is established. Id. Rather, these factors are simply guideposts in a broad-based inquiry in which district judges are expected to roll up them sleeves and examine all aspects of the past and present political environment in which the challenged electoral practice is used, “because ‘ultimate conclusions about equality or inequality of opportunity were intended by Congress to be judgments resting on comprehensive, not limited, canvassing of relevant facts.’” City of Niagara Falls, 65 F.3d at 1008 (quoting De Grandy, 512 U.S. at 1011, 114 S.Ct. at 2657). II. FINDINGS OF FACT A. The Town Of Hempstead And Its Characteristics The Town of Hempstead is the most populous town in the United States. Measured by population, it is larger than all but 13 cities in the United States, and is larger than six states. It is 25% larger than New York’s third and fourth congressional districts, which both include substantial portions of the Town as well as surrounding areas. The Town is apparently among the largest undivided election districts in the country. According to the United States Census, the total population in the Town is 725,639 (as of 1990), of which 12.1% are black and 83% are white. The voting age population is 11.2% black and 84.8% white. Of this voting age population, 48.6% are registered Republicans, 29.6% are registered Democrats, and 21.8% are either unregistered or are registered as something other than Republican or Democrat. Since 1960, the black population has grown steadily from 3.4% of the total population to 5.8% (1970) to 9.3% (1980) to the 1990 level of 12.1%. As of 1990, the black population in the Town was 87,644, and there is ample reason to conclude that it has continued to grow as a percentage of the total population during the past seven years. The Town of Hempstead is located in Nassau County. The Town includes twenty-two villages and thirty-four unincorporated areas. The county, the Town, and the villages each have their own local governments. The responsibilities of these governments are in some respects distinct. Who fixes a damaged road, for example, depends on whether it is a state road, a county road, a town road or a village road. In other respects, the responsibilities of these levels of government overlap. The Town is governed by a Town Board, a legislative body that is required under New York Town Law to have a Supervisor and at least four Town Councilmembers. The number of Council seats may be increased to six by a referendum vote of the Town residents. Prior to 1967, the Town Board consisted of a Supervisor, a Presiding Supervisor and four Town Councilmembers. In 1967, the number of Councilmembers was increased by referendum to six. Since that time, the six Coun-cilmembers have been elected in staggered four-year terms. Every two years, in the odd-numbered years, three Councilmembers are elected at-large. Each- eligible voter in the Town is able to cast three votes, one for each vacant seat on the Board. “Single-shot” voting (voting for only one candidate) is permitted, and cumulative voting (casting multiple votes for one candidate) is prohibited. The three candidates who obtain the most votes win the three seats on the Board; there is no requirement that a candidate receive a majority vote to win a seat. This system is used in every town on Long Island and in all but approximately ten of the 900 towns in New York State. A referendum to change the electoral system in the Town of Hempstead from an at-large to a councilman-ie district system was rejected in 1967 by the Town residents. From 1967 until January 1996, the Town Board also included a Supervisor and a Presiding Supervisor, both of whom sat on the six-member Nassau County Board of Supervisors. After a successful challenge to the constitutionality of that body, see Jackson v. Board of Supervisors, 818 F.Supp. 509 (E.D.N.Y.1993), and as a result of a countywide vote and state legislation, the Board of Supervisors was replaced with a nineteen-member county legislature in 1994. The Presiding Supervisor position was eliminated in 1996. Thus, the Town Board now consists of six Councilmembers, who are part-time Town employees earning $45,000 per year, and the Town Supervisor, a full-time employee who is the chief executive and administrative officer of the Town and a voting member of the Town Board. Throughout the trial of this case, defendants have asserted that the Town Board has only a limited governmental function. They correctly point out that Town residents are affected by a complex web of governments— state, county, town and village. Nonetheless, the evidence establishes that the Town Board exercises substantial authority, both formally and informally, in a wide variety of areas of importance to its residents. The formal powers and responsibilities of the Town Board include controlling Town finances, acquiring and selling real property, maintaining Town property (including its various parks), filling vacancies in Town offices, removing fire and health hazards, awarding and executing Town contracts, cleaning and repairing Town streets and roads, maintaining adequate lighting on Town roads, collecting garbage in certain areas of the Town, and promulgating zoning laws and hearing applications for rezoning areas of the Town. The Town has numerous departments, including the Public Safety Department, the Department of Sanitation, the Department of Water, the Department of Parks and Recreation, the Department of Senior Enrichment, the Department of Planning and Economic Development, the Department of General Services, the Department of Highways, the Department of Industry and Commerce, the Department of Conservation and Waterways, the Department of Occupational Resources, and the Department of Buildings. The Board appoints the members of the Board of Zoning Appeals, and hears appeals from its denials of zoning variances. The current Town budget is $290 million, and there are approximately 2500 Town employees. The formal authority of the Town Board is supplemented by the political influence its members wield. For example, there is no Town Police Department (the Public Safety Department’s 60 “safety officers” are not peace officers), but a Councilmember need only pick up the telephone in order to influence the actions of the Nassau County Police Department. Similarly, the Town Board has the power to pass resolutions that can influence other governing bodies. In these and other ways, Town Councilmembers exercise considerable influence over the operations of the other governmental entities that affect the Town’s residents. The main political parties in the Town are the Republican Party and the Democratic Party. Other political parties that have fielded candidates include the Conservative Party, the Liberal Party and the Right to Life Party. The Republican Party has had a lock on Town-wide elections since the inception of Town government. Indeed, since 1907, the victor of every Town-wide election (Presiding Supervisor, Supervisor, Councilmember, and Town Clerk) — 71 elected officials in all — has been a Republican. With the exception of Curtis Fisher, who was elected in 1993, all have been white. B. The First Gingles Precondition: A Sufficiently Large And Geographically Compact Minority Group Dr. Andrew Beveridge, an expert in the field of districting, created a map establishing six single-member districts in the Town. He followed four principles in creating the six districts depicted in Plaintiffs’ Exhibit (“PX”) 4D, which is appended to this decision at A-l: (1) each district should have as close to an equal number of residents as possible, with a maximum variance of 5%; (2) the boundaries of each district should conform, to the extent possible, to the existing political geography; (3) the districts should be reasonably compact; and (4) to the extent possible, where the first three criteria have been met, the black population should be included in one district. Dr. Beveridge thereby proposed a hypothetical black-majority district without subordinating traditional districting principles to race. For Census purposes, areas within the Town are designated either villages, Census Designated Places (“CDPs”) or unincorporated areas. A CDP is an unincorporated area that is considered a community. There are approximately 10,000 Census blocks and 747 block groups in the Town, from which Dr. Beveridge was able to measure racial composition and draw the six proposed districts, each of which has close to 120,000 people. The largest district is only 1.67% over 120,-000, and the smallest, District 3, was only 2.53% less than 120,000. District 3 is a majority-minority district. Its boundaries follow village and CDP boundaries except in two places: it cuts through the village of Freeport and through the West Hempstead CDP. Where it cuts through Freeport, it crosses south of Sunrise Highway and takes in additional areas with minority residents. The proposed District 3 is very similar to Assembly District 18 — a majority-minority district for election of a member of the New York State Assembly. A comparison of District 3 and Assembly District 18 is depicted in Plaintiffs’ Exhibit 4F, a copy of which is appended at A-2. District 3 includes all four of the predominantly black communities in the Town: Lakeview, Roosevelt, Freeport and Hempstead. The total population of District 3 is 55.45% black; the voting age population is 52.57% black. The proportions of blacks and whites in District 3 are within tenths of a percentage point of their proportions in Assembly District 18. There are some communities that Dr. Bev-eridge split in his proposed map that might naturally be expected to be included in the same district. For example, Hewlitt, one of the five areas that make up a community known as the “Five Towns” (the other four are Inwood, Lawrence, Cedarhurst, and Woodmere) is not in the same proposed district as the other four. Also, Lido Beach is closest to District 4 but is included in District 6. Dr. Beveridge also split what is known as the “tri-community” of Málveme, Lakeview, and Lynbrook, including only Lakeview in District 3. Nonetheless, the proposed districts are compact. Dr. Beveridge measured compactness by dispersion in the district, which is calculated by the ratio of the area of the actual district to the area of the ideal district, where the ideal district is a circle the radius of which is the line from the center of the district to the farthest point in the actual district. The most compact district would thus have a dispersion measurement of one. The average compactness for the districts Dr. Beveridge created was .44, and the compactness for proposed District 3 was .38. District 3 was thus more dispersed than the average dispersion of all six districts. However, in 1990 the average compactness for United States Congressional districts was .36, and the average compactness for New York Congressional districts was .30. Moreover, the compactness for New York State Assembly District 18 (the majority-minority district) is .29. Dr. Beveridge also measured the compactness of his proposed districts by dividing the perimeter of the ideal district by the perimeter of the actual district. Again, the perfectly compact district has a perimeter ratio of one. Proposed District 3 has a perimeter ratio of .27, which compares favorably with the average perimeter ratios of the majority-minority Assembly District 18 (.26) and the U.S. Congressional districts as measured in 1990 (.24), and it exceeds the 1990 average perimeter ratio of New York Congressional districts (.20). The compactness of District 3 is further demonstrated the old-fashioned way — by looking at a map. The black population in the Town is clearly confined, in large measure, to a geographic block in its center. Plaintiffs Exhibit 4D, appended at A-l, depicts this visually. In fact, the black population is even more compact than the map suggests, as the northern end of District 3, though shaded in its entirety, includes Eisenhower Park, in which very few people reside. C. The Second Gingles Precondition: Political Cohesion Of The Minority Group There is no dispute that the black voters in the Town are politically cohesive. Although defendants have challenged the ability of plaintiffs’ proposed District 3 to perform by electing a minority-preferred candidate, they agree that the black voters consistently vote cohesively for the Democratic candidates, who always lose the Town Board elections. Plaintiffs’ expert, Dr. Michael D. McDonald, conducted a “performance analysis” to determine whether, given turnout rates and the degree of racially polarized voting, one could reasonably expect that a candidate preferred by black voters could win in a single-member district system. Dr. McDonald concluded that based on the level of black support for the minority-preferred candidate, estimated using a double regression analysis of the election districts in the 18th Assembly District (“A.D.”), blacks would have been able to elect their preferred candidate had they been voting in a single-member district with a voting age population that was 47.8% black in the 1991 Town Council election, 44.6% black in the 1993 Town Council regular election, and 34.2% black in the 1993 Town Council special election. It can be expected, and I find, that blacks in proposed District 3 would be able to elect their preferred candidate. D. The Third Gingles Precondition: White Bloc Voting Both plaintiffs and defendants presented expert testimony at trial regarding the voting patterns among blacks and whites in the Town. Plaintiffs expert, Dr. McDonald, analyzed Town Board elections from 1983 to 1993. The first stage of his analysis was a comparison of the percentages of the vote each candidate received in the Town-wide election with the percentages received from voters residing in the 18th A.D., the majority-black district for election to the New York State Assembly. From this analysis, Dr. McDonald drew the preliminary conclusion that the black-preferred candidates would have fared better than they did in the Town-wide elections if the Town Council elections had been held in a majority-black district. Indeed, in the regular elections in 1991 and 1993 and in the special election in 1993, all but one of the seven winning candidates would have been different had the election been held within the 18th A.D. However, this analysis is not alone sufficient to determine whether there is a pattern of racially polarized voting in the Town of Hempstead, because it does not measure the degree of difference in support for certain candidates among whites and blacks. Dr. McDonald thus conducted a further analysis to estimate these levels of support for each candidate. He used methods of statistical analysis called “ecological correlation and regression analysis” (or “ecological regression”) and “extreme case analysis.” Both methodologies are designed to measure the strength of the relationship between changes in two variables. Ecological regression and extreme ease analysis are standard techniques used to estimate voting behavior among distinct population groups. Gingles, 478 U.S. at 52-53, 106 S.Ct. at 2767-68. Extreme case analysis predicts the percentage of voter support for certain candidates among the racial groups by using the support given to the candidates in geographic areas where the voting population is racially homogeneous, which was defined in this case as more than 90% white or more than 90% black. Ecological regression analysis correlates the percentage of voters who voted for a particular candidate with the percentage of blacks in the voting age population in a particular geographic area. Because both extreme ease analysis and ecological regression analysis provide only estimates and are subject to error, Dr. McDonald performed both among voters of various areas to determine whether they produced consistent estimates of the racial groups’ support for each candidate. Dr. McDonald performed four analyses to measure white voting: (1) an extreme case analysis of the seven A.D.s (or parts thereof) within the Town that are over 90% white; (2) an extreme case analysis of the eight election districts within the 18th A.D. that are over 90% white; (3) a simple regression analysis using results from all election districts within the 18th A.D.; and (4) a double regression analysis using results from all election districts within the 18th A.D. To measure black voting, Dr. McDonald did three analyses: (1) an extreme case analysis of the seven election districts within the 18th A.D. that are over 90% black; (2) a simple regression analysis using the results from all election districts within the 18th A.D.; and (3) a double regression analysis using results from all election districts within the 18th A.D. In his ecological regression analysis, Dr. McDonald measured the relationship between the racial composition of the Town’s A.D.s and election districts and the level of support given to particular candidates within those units in elections from 1983 to 1993. He placed the independent variable (the percentage of black or white registered voters in an election district within the 18th A.D.) on the horizontal axis. Dr. McDonald then placed the dependent variable (the percentage of voters in a particular A.D: who voted for a particular candidate) on the vertical axis. He then plotted data points on the graph that represented the percentage of black or white voters in each A.D. who supported a particular candidate. Finally, Dr. McDonald drew a line through the data points which best fit the points on the graph, or the line from which there would be the least variance from each data point. The slope of this line represents the increase (or decrease) in support for a particular candidate when the racial composition of a geographic area changes. By extending the line through the vertical axes on the left and right of the graph, Dr. McDonald estimated the level of white and black voter support for a particular candidate. Dr. McDonald’s ecological regression analysis consisted of both a simple regression and a double regression. The simple regression measured the percentage of white and black voters of the total voting age population who supported a particular candidate. The double regression measured the percentage of support among voters who actually voted. Both analyses of white and black voter support for certain candidates were based on the support in each election district within the 18th A.D. However, Dr. McDonald chose to use the double regression results since they were a more accurate measurement of voter support in each race. Dr. McDonald computed these figures by considering the percentage of voters in each election district within the 18th A.D. who voted for a candidate and the percentage of voters in each election district who did not vote for that candidate. He then added these figures to determine the percentage of voters in each election district who actually voted in that election. With these figures, he was able to estimate the support for particular candidates in each election district from the voters who actually voted in that election, rather than the support for each candidate as a percentage of the total voting age population in the election district. Thus, this method took account of any differences in the turnout rate between blacks and whites. Dr. McDonald was then able to compare the white voter support with the black voter support. He made the following findings: 1988 Town Council (8 candidates elected) Whites Blacks Candidate Mondello (R,W) 60.6% 25.1% Peterson (R,W) 60.0% 26.1% Cairo (R,W) 59.6% 22.8% Bernstein (D,W) 28.2% 52.8% Lawrence (D,B) 26.5% 53.2% Reddan (D,W) 25.7% 49.5% 1985 Town Council (8 candidates elected) Whites Blacks Candidate Gaurdino (R,W) 60.0% 29.1% Bernstein (R,W) 58.7% 28.9% Weisbein (R,W) 57.9% 26.0% Lawrence (D,B) 26.4% 44.9% Magaliff (D,W) 25.8% 43.3% Guise (D,W) 25.6% 43.3% 1987 Town Council (8 candidates elected) Whites Blacks Candidate Cairo (R,W) 61.7% 27.3% Zagarino (R,W) 61.0% 23.4% Cullin (R,W) 59.6% 26.1% Davidson (D,W) 25.3% 45.4% Brewington (D,B) 24.0% 57.8% DelMastro (D,W) 24.7% 43.0% 1989 Town Council (8 candidates elected) Whites Blacks Candidate Levy (R,W) 54.4% 25.3% Guardino (R,W) 53.5% 25.2% Kearney (R,W) 53.0% 23.3% DelMastro (D,W) 33.1% 49.5% Labelson (D,W) 32.2% 50.4% Mosely (D,B) 32.0% 53.9% 1991 Town Council candidates elected) Whites Blacks Candidate Cairo (R,W) 50.6% 15.3% Cullin (R,W) 49.2% 17.3% Zagarino (R,W) 49.3% 15.0% Kalastein (D,W) 35.4% 49.5% Grabinger (D,W) 34.7% 51.5% Ford (D,B) 33.3% 58.4% 1993 Town Council candidates elected) Whites Blacks Candidate Fisher (R,B) 50.1% 28.6% Kearney (R,W) 50.1% 15.4% Santino (R,W) 50.0% 14.6% Davidson (D,W) 32.5% 51.3% DeJong (D,W) 32.1% 52.4% Dorsky (D,W) 32.0% 51.2% 1993 Town Council candidate elected) Special Election (1 Whites Blacks Candidate Blakeman (R,W) 61.3% 20.8% DeFreitas (D,B) 37.6% 78.4% Dr. McDonald concluded that each of these Town Board elections exhibited racially polarized voting. Indeed, all forms of analysis of the voting patterns consistently reveal a clear pattern of racially polarized voting. As a result, Dr. McDonald concluded, black voters have never been able to elect their preferred candidate to the office of Town Board Couneilmember. Dr. McDonald further observed that: One cannot help but notice, of course, that the racial polarization apparent in these elections is associated with a form of partisan polarization. Black voters behave in a way indicating their preferred candidates are Democrats. In the face of the white majority’s support for Republicans, black voters do not have much, if any, opportunity to elect candidates of their choice. Defendants’ expert, Dr. Harold W. Stanley, also analyzed Town Board elections from 1983 to 1993. In addition, Dr. Stanley analyzed the Town Supervisor and Presiding Supervisor elections from the same years. Like Dr. McDonald, Dr. Stanley used extreme case analysis and ecological regression analysis to determine the levels of white and black support for each candidate. However, Dr. Stanley used all of the election districts within the Town, rather than simply those within the 18th A.D., to conduct his analyses. As a result, he was able to use over 500 data points in his regression analysis, rather than just the 74 upon which Dr. McDonald relied. For his extreme case analysis, Dr. Stanley was able to use 9 extreme case black election districts and 342 extreme case white election districts. Dr. Stanley found racial polarization similar to that found by Dr. McDonald in the Town Council races. While the racial polarization Dr. Stanley found was slightly less in some instances than that found by Dr. McDonald, the differences are not meaningful. Both sides agree that blacks and whites vote differently. Blacks support Democratic candidates. Whites support Republican candidates. Democratic candidates, whether white or black, have received similar levels of support from black voters. Defendánts have not seriously contested the existence of racial polarization in the voting patterns in the Town. Indeed, they argue that most elections are racially polarized, including every Presidential election since 1948 (with the exception of 1964). However, they contend that the racial polarization lacks legal significance because partisan polities, not racial bias, best explains the divergent voting patterns. E. The Senate Report Factors 1. History Of Official Discrimination There has been no official policy of discrimination against blacks in the state or in the Town. Nor is there any basis to conclude that the Town’s initial adoption of its at-large system, or its residents’ 1967 decision to retain it, were motivated by racial animus. Plaintiffs contend, however, that the past use of a literacy test and a purging provision of the New York Election Law had a discriminatory impact on the ability of blacks to register to vote, remain registered to vote, and otherwise participate in the political process. From 1922 to 1969, a literacy test was used in Nassau County. In order to register to voté, a citizen was required to present a certificate of literacy issued by the New York State Education Department. Plaintiffs contend that the retention of this test for four years after the enactment of the Voting Rights Act, coupled with the evidence that blacks in the Town are more likely than whites to be undereducated, constitutes historical discrimination cognizable under Section 2. In fact, recent data suggest that blacks in the Town are significantly overrepresented at the low end of the education spectrum, and that a literacy test would almost certainly constitute a greater impediment to black voter registration than to white voter registration. Although defendants challenge the assertion that the same would be true in 1969, and correctly point out the absence in this case of socio-economic data from that period, I find the inference that the literacy test had a discriminatory impact on blacks to be a fair one. On the other hand, this adverse impact is blunted by the fact that blacks constituted a considerably smaller percentage of the Town population in 1969— less than 6%. The purging of voters from the voting rolls of Nassau County was more recent and is more significant. Section 5-406 of the New York Election Law, enacted in 1949, required the purging of voters who had failed to vote for four successive years. Plaintiffs have demonstrated, through expert testimony based on both ecological regression and extreme case analysis, that the statute produced very substantial differences in the rates at which black and white voters were subject to this “non-voting purging.” The purge rate was 56% greater in predominantly minority election districts than in predominantly white election districts. A black registrant in the 18th A.D. had a 70% greater chance of being purged for non-voting than did a randomly selected registrant from one of the remaining Assembly Districts in the Town. The number of purged voters was substantial; according to Allan J. Lichtman, whose affidavit was received in evidence by agreement, 25,786 voters in the Town were purged. Defendants have not contested these facts, but instead have emphasized that the purge was conducted pursuant to state, rather than Town, legislation. They further contend that in 1990, when a judge in this district temporarily restrained the implementation of the purge provision because of its disparate impact on blacks, the Town Board voted to consent to further injunctive relief, and the purging ceased. Although I agree that these facts preclude a finding that the purging evidences discriminatory intent on the part of the Town Board, the undisputed discriminatory impact of the purging on black voters in the Town is an appropriate consideration in evaluating the totality of the circumstances. Finally, plaintiffs offered testimony that blacks encountered racial hostility at the polls in the Town. Hazel N. Dukes, President of the New York State Conference of the NAACP and a member of its national board of directors, testified to her personal observations at two polling places in the Town. Although I credit the testimony of Ms. Dukes, it did not in my view describe unfair or discriminatory treatment of black voters. In essence, Ms. Dukes found that the Republican inspectors at two polling places (each polling place had an inspector from each major party) were insufficiently helpful to apparently unregistered voters (e.g., they failed to advise them of challenge ballots or potential court challenges) and more “aggressive” than Ms. Dukes felt was appropriate. This was not discrimination, invidious or otherwise. 2. The Degree Of Racially Polarized Voting From 1983 to 1993, the successful candidates in every Town Board election consistently garnered between 15% and 29% of the black vote and between 49% and 62% of the white vote. The difference between black and white voter support for the victorious candidate was consistently over 20 percentage points and in most cases over 30 percentage points. In addition, in every election for Town Councilmember, the majority of white voters withheld their support for the minority-preferred candidate. While other cases have involved a greater disparity between the white vote and the black vote, see, e.g., City of Niagara Falls, 65 F.3d 1002 (2d Cir.1995) (where frequently less than 10% of whites voted for a candidate supported by substantial majorities of black voters), there is nonetheless a persistent and significant degree of racial polarization in Town Board elections. 3. Electoral Mechanisms That Enhance Vote Dilution While there is no “anti-single shot” rule or majority vote requirement in the Town Board elections, the district is unusually large and there is a prohibition on cumulative voting. These circumstances, and particularly the extraordinary size of the district, significantly reduce the chances of a minority-preferred candidate being elected. As noted, the Town’s three-quarters of a million residents apparently make it one of the largest undivided districts in the country. Campaigning is far more difficult than it would be under a single-member districting system. There are upwards of 50 communities in the Town, and the issues of importance to people in those communities vary widely in different parts of the Town. A successful campaign requires the resources to be able to visit, distribute literature to, and address the issues in all areas of the Town. This requires a level of financing that a candidate for Town Board is not likely to achieve. Black Republicans and Democrats alike testified to the difficulties in running in the at-large system due to the sheer size of the jurisdiction. The difficulties faced by the black candidates in these Town-wide elections contrast sharply with their enhanced ability to campaign effectively in smaller districts, such as the 18th A.D. and the new Nassau County legislative districts. 4. Access To Candidate Slating Process The Republican Party has had unblemished success in electing its candidates to the Town Board. For nearly a century, every Town Councilmember has been a Republican. The Republican Party endorsement is essential to winning an election in the present at-large system. Indeed, there is a well-founded and widely-held belief that, as defendant Councilmember Curtis Fisher put it, “the Republicans seem to control everything in Nassau County.” In testimony I credit, witnesses testified that the Republican Party is so powerful that registration in it is a virtual prerequisite to obtaining public employment at any level — from a laborer with the Town, to a school principal, to a Nassau County police officer. There are no impediments to blacks joining the Republican or Democratic Parties in the Town, and blacks are not prevented from registering to vote or exercising their right to vote in the Town. Moreover, both the Democratic and Republican clubs in the Town are multi-racial organizations and welcome membership and participation by interested blacks. There are Republican clubs in all of the communities in the Town. These clubs are generally the social arm of the party, and are the gathering points for active party members. The lowest-level governing body of the party is the committee. Each community has one, and it consists of two committee members from each election district. For example, the East Meadow committee has 52 seats, 2 for each of the 26 election districts in East Meadow. Each of those committees elects an executive leader. There are 69 executive leaders in Nassau County, three of whom are black. These 69 executive leaders, together with approximately 20 vice-chairs of the Nassau County Republican Party, meet from time to time with Joseph Mondello, who has been the county chairman for 13 years. Those meetings are called executive committee meetings. There is a formal procedure for selecting Republican candidates for the Town Board and other elective offices in the Town and the County of Nassau. The county chair will recommend a candidate to the executive committee, which votes on whether to recommend that person to the party’s convention, which consists of all of the approximately 2,000 committee members in the county. The convention then votes on the recommendation. The voting system is weighted in accordance with each area’s performance in the previous election. In practice, the slating process is more simple. The executive committee never rejects or even debates the merits of the county chairman’s recommendations. It always and unanimously recommends Mondello’s selections to the convention, which always and unanimously accepts them. The weighted voting system has never been used, and there has never been a Republican primary for a Town-wide office. In short, while the executive leaders, and presumably others as well, may lobby Mondello in support of proposed candidates, as a practical matter, Mon-dello selects the party’s nominees. In addition to the many Republican clubs in the villages and unincorporated areas, there is a prominent black Republican organization, the Nassau County African-American Republican Coalition (“A.A.R.C.”), which includes various black political leaders (some of whom are clergy) in the Town. The purpose of the coalition is to enhance the role of black Republicans in county, town and village governments in Nassau County. The A.A.R.C. has lobbied Mondello to nominate black Republicans for certain elective offices. These have generally been recommendations for candidates to run on the Republican ticket in majority-minority districts, such as the recommendation of Andrew Hardwick to run in the state’s 18th Assembly District. Other successful recommendations included Judge Carnell Foskey, who was appointed a District Court Judge. However, the Town Board and the Republican Party have not been receptive to the idea that a black should occupy what is perceived in the Town as a far more significant office: Councilmember on the Town Board. To the contrary, the Republican Party has failed to produce a black candidate for Town Board, with the sole exception of Curtis Fisher in 1993. That exception does not alter my finding that blacks have been denied access to the Republican Party slating process for Town Board elections. In 1987, two vacancies arose on the Town Board. Patrick Zagarino and Angie Cullen, who are both white Republicans, were appointed to the vacancies. In 1989, there were two more vacancies on the Town Board. When it became known that two more white Republicans would be appointed to fill them, there was a march on Town Hall by blacks seeking the appointment of a black Republican. The protesters included the leaders of the A.A.R.C. This public display of frustration by the most prominent black Republicans in Nassau County did not derail the appointment to the Town Board of two white Republicans, David Levy and Joseph Kear-ney. In an apparent effort to placate the A.A.R.C., Mondello appointed Curtis Fisher, a black Republican, to the position of Executive Assistant to the Town Board. Fisher lives in the white community of Baldwin Harbor, and had been for many years Mondello’s close personal friend and tennis partner. In 1993, a vacancy arose on the Town Board. When that occurred, the A.A.R.C. recommended Lance Clark, a black attorney who is the Deputy Mayor of the Village of Hempstead. Clark has been a pillar of the black communities of Hempstead and Lake-view. He was active in the NAACP branches in both communities as well as the Hundred Black Men of Nassau, Suffolk, Inc., and Operation Get Ahead, a youth organization. He has been a Republican committeeman in the Village of Hempstead for 14 years. Clark, like many black Republicans, felt that the at-large system deprived the African-American community of representation on the Town Board by a person who lived in that community and had to deal with the day-to-day issues there. The A.A.R.C.’s recommendation to Mon-dello that Lance Clark be appointed to the Town Board was rejected. Indeed, Clark was not even interviewed. Instead, without any input from the A.A.R.C., Mondello appointed his close friend Curtis Fisher instead. Fisher had been found less qualified to represent the African-American community than Clark by the A.A.R.C., which had interviewed both before recommending Clark to Mondello. The parties have vigorously disputed the relative qualifications of Clark and Fisher to be appointed to the Town Board. The plaintiffs correctly observe that Clark had far more involvement in both the community and the Republican Party than Fisher, and was the more qualified and likely candidate for the Town Board, especially considering the emphasis within the party on “paying dues” through loyal and lengthy public service before ascending to higher office. Defendants counter by citing Fisher’s background as a star athlete and experience as an educator, and further point out that there is nothing wrong or unusual about rising in politics by playing tennis with a power broker. Although the plaintiffs have the better of this argument, the issue before me is not whether the Nassau County Republican Party appoints or nominates the best available Republicans, or the most deserving ones, but whether its slating process for Town Board seats is truly open to blacks. Despite the appointment of Curtis Fisher — indeed, in part because of it — the events of 1989 and 1993 cause me to conclude that it is not. I am further persuaded of this by the testimony of the Town’s prominent black Republicans regarding their efforts to penetrate the appointment and slating process for Town Board seats. There is no dispute that, as noted above, black Republicans fervently wanted representation on the Town Board, and their frustration at the party’s failure to provide it in 1989 had bubbled over to produce the march on Town Hall. It is also clear that the A.A.R.C. had enjoyed success in recommending to Mondello blacks for appointment to or nomination for lesser positions and for offices in majority-minority districts. However, the A.A.R.C. had not, in either 1987 or 1989, recommended anyone to Mondello for the Town Board. The testimony of prominent black Republicans revealed inconsistent explanations for that. Robert Francis, the Town’s Commissioner of Planning and Economic Development, attributed it to the lack of black Republicans interested in the position. Briding Newell, another prominent black Republican who currently works as an analyst with the county legislature, testified that the A.A.R.C. decided in 1993 that its longtime lobbying for a Town Board seat would be stronger if it finally “attached a name to the process.” It is possible to make too much of this conflict, which I attribute to the palpable discomfort these otherwise impressive witnesses exhibited in testifying about their own party’s treatment of black Republicans’ efforts to obtain representation on the Town Board. They were not alone in this regard. Lance Clark himself — the A.A.R.C.’s choice for Town Board, who was flatly rejected by Mondello — testified unconvincingly that the Nassau County Republican Party is receptive to the recommendations of the .AA.R.C., which Clark described as Mondello’s “African-American community advisors.” In any event, I conclude that Town Board seats were sufficiently beyond the realm of possibility for black Republicans that they never even attempted to obtain one through the normal party mechanisms until 1993. And then they were simply shut out. Gregory Peterson, a Republican who is the Town Supervisor, testified that the party sought in 1993 to show that it was open to everyone, to “reaeh[] out to the African-American community.” I do not credit this testimony, because the facts show that rather than appoint the person recommended by the party’s own black members, a “safe” black Republican was appointed instead. 5. Discrimination In Other Areas Which Hinders Blacks’ Ability To Participate In Political Process Plaintiffs presented the expert testimony of Dr. William O’Hare regarding the socioeconomic status of blacks in the Town. Dr. O’Hare obtained data from the U.S. Census Bureau and a variety of academic and professional publications to examine the relationship between the socioeconomic status of blacks in the Town and their participation in the political process. There is an income differential among blacks and whites in the Town. The median household income of blacks is $45,418, as compared to $53,574 for whites. There is a higher per capita income among whites and a higher unemployment and poverty rate among blacks. With respect to the type of jobs held by blacks and whites, 60% of blacks and 73% of whites were in white-collar jobs. There is also a disparity in the educational levels attained by blacks and whites in the Town. For example, 17% of blacks have at least a four-year college degree, as compared to 29% of whites; and 76% of blacks have a high school degree, as compared to 85% of whites. Finally, blacks are less likely than whites to own a home, a car, and a telephone, and the homes that are owned by blacks have a lower median value than the homes owned by whites. These differences notwithstanding, on the whole, blacks in the Town of Hempstead have fared well. The median household income of black families in the Town far exceeds that of (a) blacks in the United States; (b) blacks in New York State; (c) whites in the United States; and (d) whites in New York State. Forty-four percent of the blacks in the Town earn more than $50,000 per year. Home ownership by blacks (62%) in the Town exceeds the national average, as does the median value of owner-occupied homes ($162,800). Indeed, in an unrelated study of the best places for blacks to live, Dr. O’Hare rated the Nassau-Suffolk area, which embraces the Town, first among 48 metropolitan areas in the country. In sum, as in most metropolitan areas, blacks in the Town have a lower socioeconomic status than whites. As a result, there are mild differences in the abilities of blacks and whites to participate in the political process. For example, it is more difficult for people employed in blue-collar jobs where they are paid by the hour to take time off from work to go to the polls. Also, those with lower incomes obviously have less money to contribute to political campaigns. However, in the Town of Hempstead, these differences in the socioeconomic status of blacks do not significantly impair their relative ability to participate in the political process. 6. Racial Appeals In Political Campaigns The western part of the Town borders on the New York City borough of Queens. Several of the Queens neighborhoods along that border are predominantly black. Until fairly recently, the Elmont and North Valley Stream CDPs, like the other parts of the Town bordering Queens, were predominantly white. In the past fifteen years, black flight from Queens to the Nassau County suburbs has caused some racial tension in the Town. This tension resulted in some troubling appeals to racism in campaign literature. In a late 1970s campaign for a State Senate seat from an Assembly District within the Town, the incumbent Republican appealed to the fears of Town residents that black students from Queens would be bused to schools in the Town. The campaign literature used pictures of black children in school buses to convey the message that voting for the Democratic opponent would result in such busing. In 1987, the racial composition of the communities of Elmont and North Valley Stream was in transition. Although certain neighborhoods in these communities are now predominantly black (see A-l), they were then still predominantly white, but were showing the effects of the influx of blacks from Queens. The campaign literature of incumbent Republican Town Councilman Joseph Cairo sought to appeal to the anti-black sentiments aroused by this migration. Specifically, Cairo’s brochure, under the heading “Preventing Urban Encroachment,” stated that he was “well aware of his community’s proximity to the City of New York,” and the attendant risk of “urbanization.” Cairo claimed to be “a major opponent of those who seek to ‘Queensify’ North Valley Stream and its environs.” Under the heading “Crime,” Cairo’s brochure warned of the possibility of “urban crime spilling over the county border.” He claimed to be the “unofficial liaison with the Nassau County Police Department,” in which capacity he “sensitized local patrolmen to the special concerns of the community.” These statements were contained in a brochure that included seven photographs of Cairo with approximately three dozen members of the Elmont and North Valley Stream communities; none of the photographs included a black person. There was evidence, which I credit, that the “sensitizing” of the police department resulted in an unofficial border patrol policy: if black youths from Queens ventured across the border, the police were to stop them, find out their business and ensure that they “go back where they belong.” Specifically, Melvin Boone, a registered Republican and a detective with the Nassau County Police Department, was a uniformed officer in 1986 in the Fourth Precinct. That precinct included Inwood, which, like Elmont and North Valley Stream, borders Queens and included neighborhoods that were becoming increasingly black. Detective Boone testified that, although there was no explicit directive, the “overwhelming feeling” within the department was that its job was to keep Queens blacks out of the border neighborhoods, and border patrols were conducted accordingly. It is true, as defendants contend, that the language in Cairo’s campaign brochure is subject to different interpretations. One reading of it is that Cairo would simply work to keep the suburban areas from becoming too commercialized and crime-ridden. The plaintiffs, supported by the testimony of witnesses, argue that a different meaning was intended. For example, if the brochure was truly addressing a crime problem, plaintiffs contend, why does it mention “urban” crime?; if commercialization was the concern, why use “Queensification?” They argue that, in the western part of the Town in the late 1980s, these were thinly-veiled code words for anti-black sentiment. Although some light could presumably be shed on these issues by Cairo himself, who is a sitting Town Councilman and a defendant in this case, he elected not to testify. I find that in the particular circumstances at that time and in those communities, Cairo’s campaign literature was perceived, reasonably, as a promise to- stem the tide of blacks across the Queens border into Elmont and North Valley Stream. As noted, other evidence supports plaintiffs’ claims that the Town Board — perhaps through Cairo himself— used its unofficial influence over the police department to further this goal. There was testimony, which I credit, that this particular piece of campaign literature in Cairo’s 1987 campaign was part of a larger pattern of Republican campaign materials that were characterized by subtle racial appeals, which sought to depict the integration of white areas as a threat, and to prey on the fear of such a threat. 7. Minorities Elected To Public Office Under the functional view of the political process mandated by Section 2, the most important factors bearing on a challenge to a multimember system are the extent of racial polarization and the extent to which minority group members have been elected to public office in the jurisdiction. Gingles, 478 U.S. at 48 n. 15, 106 S.Ct. at 2765 n. 15 (citing Senate Report at 28-29). Between 1907 and 1993, every member of the Town Board was a white Republican. In 1993, Curtis Fisher, a black Republican, was appointed to a vacancy on the Board, and he successfully ran for election later that year. To this day, he is the only nonwhite member in the history of the Town Board. To be sure, black Republicans have been appointed or elected to other positions in the area. Alfred Robbins, Carnell Foskey and Leonard Clark were elected to Nassau County District Court, and Gerald Carter was appointed to that position. James Garner, Lance Clark, Jacqueline Jackson and Ralph Smith have been elected to governmental positions in the villages of Hempstead and Freeport. In addition, black Republicans have been appointed to policymaking positions in the Town: Robert Francis (Commissioner of Planning and Economic Development); Clinton Boone (Commissioner of Occupational Resources); and Briding Newell (Commissioner of the Department of Drug and Alcohol). The defendants further rely on the appointment of some black Republicans to other Town and party positions, but the most critical fact in this regard is that until the election of Curtis Fisher in 1993, no African-American was ever elected to the legislative body at issue in this case. As for the election of Fisher, plaintiffs contend that it should be disregarded entirely. The Senate Report acknowledged the risk that “the majority citizens might evade the section[,] e.g., by manipulating the election of a ‘safe’ minority candidate.” Senate Report at 29, n. 115, 1982 U.S.C.C.A.N. at 207. Plaintiffs contend that the appointment of Fisher, followed by his inevitable election to the position, was such a manipulation. For the reasons set forth in the next section, I find this argument persuasive. 8. Responsiveness By Elected Officials To Minority Needs The plaintiffs contend that the Town Board has been unresponsive to various needs of the black communities in the Town. Although their claims are not borne out by the evidence in all respects, there is nevertheless substantial evidence that the Town Board is unresponsive to the particularized needs of the black communities in the Town. In this regard, evidence of tangible efforts of elected officials to respond to the needs of the minority group is relevant. Senate Report at 29 n. 116, 1982 U.S.C.C.A.N. at 207; City of Niagara Falls, 65 F.3d at 1023 n. 24. In City of Niagara Falls, for example, the court found relevant that the defendant city had established a Human Rights Commission, which began an Affirmative Action Task Force to increase minority enlistment in the police and fire departments. The city farther established programs to promote the integration of its schools, community policing in black neighborhoods and loans to minority business owners. 65 F.3d at 1023. This case stands in stark contrast. When Gregory Peterson, the Town Supervisor, was questioned about the fact that no affirmative action resolution had ever been proposed by Curtis Fisher, he explained that none was necessary because “We have an affirmative action policy” and “an affirmative action officer.” Fisher himself testified that he was appointed Affirmative Action Officer in 1989, succeeding Carnell Foskey. However, there was no affirmative action policy at all. The only policy was a prohibition of overt racial animus in Town government, and both Peterson and Fi