Full opinion text
OPINION AND ORDER SEYBERT, District Judge: Plaintiffs, African-American citizens of the Town of Babylon in Suffolk County, Long Island, New York, challenge the town’s at-large election system for its Town Board, charging that the system violates Section 2 of the Voting Rights Act of 1965, as amended (the “Act”), 42 U.S.C. Section 1973, and the First, Thirteenth, Fourteenth and Fifteenth Amendments of the United States Constitution. Plaintiffs allege that the at-large method of electing the Town Board dilutes the voting power of the Town of Babylon’s African-American citizens. Plaintiffs seek to have the present method of election declared invalid and to have it replaced by single-member districts. Having considered the testimony and exhibits presented during the course of a six- and one-half day non-jury trial, discovery materials and the parties’ written submissions following trial, the Court concludes that plaintiffs have failed to demonstrate either a violation of the Act or of their constitutional rights. FINDINGS OF FACT A. The Parties 1. Plaintiffs United States District Judge Korman certified the plaintiff class on April 27, 1989 to represent African-American and Hispanic citizens of the Town of Babylon. At no time during the trial, however, was any evidence offered that plaintiffs, all of whom are African-American, represented the interests of the Town’s Hispanic population. Consolidating minority groups is permissible where the statistical evidence is that the minority groups vote cohesively for the same candidates. See League of United Latin American Citizens v. Clements, 999 F.2d 831, 863-64 (5th Cir.1993) (en banc), cert. denied, — U.S. -, 114 S.Ct. 878, 127 L.Ed.2d 74 (1994). In the case at bar, plaintiffs did not offer even evidence that Hispanic voting strength was being diluted by the challenged method of election, much less attempt to show cohesive voting across ethnic groups. In fact, in examining the voting preferences of the Town of Babylon’s African-American citizens, plaintiffs combined the votes of Hispanic voters with those of other non-African-American citizens. Plaintiffs’ counsel, Margaret Ford, essentially conceded that the lawsuit no longer sought to protect the voting rights of the Town’s Hispanic community, noting beginning of the trial that “we have not talked about the rights of Latinos in this particular lawsuit ...” (Tr. at 144.) A district court may decertify a class pursuant to Fed.R.Civ.P. 23(e)(1). This rule provides that “as soon as practicable after the commencement of an action brought as a class action, the court shall determine by order whether it is to be so maintained. Any decision under this subdivision may be conditional, and may be altered or amended before the decision on the merits.” Fed.R.Civ.P. 23(c)(1); see In re Prudential Securities Inc. Limited Partnerships Litigation, 158 F.R.D. 301, 304 (S.D.N.Y.1994); Richardson v. Byrd, 709 F.2d 1016, 1019 (5th Cir.) (“The district judge must define, redefine, subclass, and decertify as appropriate in response to the progression of the case from assertion to facts.”), cert. denied, 464 U.S. 1009, 104 S.Ct. 527, 78 L.Ed.2d 710 (1983); In re Harcourt Brace Jovanovich, Inc. Securities Litigation, 838 F.Supp. 109, 115 (S.D.N.Y.1993) (“The court may modify the class, establish subclasses, or decertify as appropriate in response to factual development.” (citations omitted)). The issue of whether a class should be decertified may be raised by the Court acting sua sponte. In re Prudential Securities, 158 F.R.D. at 304; Gerstle v. Continental Airlines, Inc., 466 F.2d 1374, 1377 (10th Cir.1972). Given the complete lack of evidence of Hispanic vote dilution, the Court on its own motion decertifies the class in so far as it purports to be made up of Hispanic citizens. Plaintiffs’ claims with respect to Hispanic citizens are dismissed in their entirety. The class certification remains effective for African-American citizens of the Town of Babylon. 2. Defendants Defendants are the Town of Babylon and the current members of the Town Board, sued in their official capacities as Board members. B. The Current Method of Election The Babylon Town Board consists of four board members and the Town Supervisor, who is a full voting member of the Town Board. N.Y. Town Law § 60 (McKinney 1987). The four board members are elected at large to four-year, staggered terms (two in each election cycle). N.Y. Town Law § 24 (McKinney 1987 and Supp.1995). The Supervisor is elected at large to a two-year term. N.Y. Town Law § 24 (McKinney 1987 and Supp.1995). In addition to being a voting member of the Council, the Supervisor is also a full-time employee of the Town and its chief executive and administrative officer. N.Y. Town Law §§ 52, 60 (McKinney 1987). Under New York law, neither the number of Board seats nor the method of election may be changed without submitting the issue to the voters in a referendum. N.Y.Town Law § 81 (McKinney 1987). Moreover, New York law specifically provides only for the number of board seats to be changed from four to either to six or two. N.Y. Town Law § 81 (McKinney 1987). Town Board elections are partisan. A candidate may appear on the ballot under more than one party label. For example, the Republican nominee is often also the nominee of the Right to Life Party. There is no majority vote requirement. The top two finishers among the Town Board candidates are elected. Single shot, or bullet, voting, referring to a voting practice in which a voter is allowed to cast fewer than all of his or her votes, thus enhancing the significance of each vote she or he easts, is permitted. As of the date of trial, three members of the Town Board, including the Supervisor, were Democrats and two were Republicans. None is African-American. In the past twenty years, only one African-American, Myrna Taylor in 1993, has run for the Board. In November 1995, both the Democratic Supervisor, Richard Schaffer, and the two Democratic members of the Town Board were reelected. C. Population and Geographical Information The 1990 U.S. Census of Population reports the following population information for the Town of Babylon: Population Voting Age Population Total 202,889 154,988 mite 172,605 (85.07%) 134,306 (86.66%) African- 30,284 (14.93%) American 20,681 (13.34%) The 1990 U.S. Census of Population reports a small percentage of Hispanics for Babylon and even smaller percentages of other groups. The Town of Babylon is contained within Suffolk County. Within the Town of Babylon itself are incorporated and unincorporated villages and hamlets, including, based on a 1987 map of the Town that defendants admitted into evidence and plaintiffs did not contest, Wyandanch, Amityville, North Ami-tyville, Deer Park, Lindenhurst, North Lin-denhurst, East Farmingdale, Wheatley Heights, Babylon, North Babylon and Co-piague. These villages and hamlets appear to command allegiance from their residents, who tend to identify their homes as being in one of the villages, for instance, Wyandanch, rather than in the Town of Babylon. The Town of Babylon contains three areas of compact African-American population. The largest concentration is in the Wyandanch and Wheatley Heights area and is composed of Voter Tabulation Districts 115, 135, 145, 74, 14, 141, 56, 147 and 124. This concentration is on the northern border of the town, roughly in the center. Its population includes 11,864 African-Americans, using 1990 Census figures. The second concentration is in the area of North Amityville, which is on the western border of the Town, slightly north of the center point of the border. This concentration is composed of Voter Tabulation Districts 129, 72, 42, 65 and 140. Its population includes 9,215 African-Americans, using 1990 Census figures. The third concentration is on the eastern border in the hamlet of Deer Park. This concentration is identified as Voter Tabulation District (also election district) 60. Approximately 1,809 African-Americans live in this third concentration, again using 1990 Census figures. The total African-American population in these three concentrations is 22,888. Each of these population groupings is separated from the others by significant white populations. Moreover, these concentrations are geographically divided. Between Wyan-danch and North Amityville are an industrial park, the area of East Farmingdale, Republic Airport and numerous large cemeteries. Between Wyandaneh and Voter Tabulation District 60 is the remainder of the hamlet of Deer Park which is predominantly white and residential. Voter participation in the predominantly white region of Deer Park is high. D. Use of Statistical Methods to Estimate Voter Behavior As an understanding of the voting behavior of the African-American and white populations within the Town of Babylon is critical to determining whether the African-American vote in the Town is being diluted in an at-large election system, both plaintiffs and defendants presented expert testimony at trial regarding voting patterns, dating back to 1983. Lay testimony from both sides on election returns and party affiliation supplemented this expert testimony. Defendants’ expert on voting behavior was Dr. Harold Stanley, Associate Professor in Political Science at the University of Rochester, New York. Plaintiffs’ key expert in this area was Franklin Lewis, an independent consultant in political redistricting, economic modeling and statistical analysis. Both experts relied on the statistical methodologies of ecological regression analysis and extreme case analysis, approved and utilized in the landmark vote dilution case of Thornburg v. Gingles, 478 U.S. 30, 61, 106 S.Ct. 2752, 2771-72, 92 L.Ed.2d 25 (1986), to estimate voter behavior in the Town of Babylon. Using these methodologies, Lewis estimated how the Town’s African-American and white voters voted in the 1988 Democratic presidential primary, the 1989, 1990, 1991 District Court elections, the 1989 Supreme Court election and the 1993 Town Board election. Stanley applied these methods of analysis to all Town Board and Town Supervisor elections in the period from 1983 to 1993. Moreover, Stanley took the additional step of using these methodologies to estimate more broadly the number of African-American and white registered voters in the Town of Babylon and the political parties to which each group belonged. 1. Ecological Regression Analysis Ecological regression analysis is the standard technique used to infer voting behavior among distinct population groups. Gingles, 478 U.S. at 52-53 and n. 20, 106 S.Ct. at 2767-68 and n. 20. It would be extremely difficult to interview every voter, classify each by race or ethnicity and determine how each voted and the party to which they belong. Regression analysis allows the parties to surmount this proof-gathering burden by making reasonably accurate estimates of majority and minority voting behavior from demographic data and, depending on whether voting in a specific election or party affiliation is being estimated, election returns and party registration data, respectively. Both plaintiffs’ and defendants’ experts employed the double-equation ecological regression method which generates an estimate of the voting behavior of the electorate by race as well as an estimate of voter turnout by race. See Bernard Grofman et al., Minority Representation and the Quest for Voting Equality 86 (Cambridge Univ.Press 1992) (Pl.Exh. 9) [hereinafter “Grofman ”]. In utilizing double-equation ecological regression analysis to determine the political party affiliation of minority and majority groups, the dependent variable, plotted on the vertical axis, would be the percentage of African-American or white registered voters in a given election district who are Democrats/Republicans. If voting patterns in particular elections were being studied, the dependent variable would be the percentage of African-Americans or white registered voters who voted for a specific candidate. The percentage of the election district’s voting age population that is African-American or white would be the independent variable plotted on the horizontal axis. Two regressions would be performed for each candidate running for election or for each political party, depending on whether voting or party affiliation was being studied. One regression would compare the African-American voting age population in each election district with the support that a candidate or particular political party received in that district. The second regression would compare the white voting age population in each election district with the support that a candidate or particular political party received in that district. These data are then used as coordinates to plot points on a graph. A line is drawn which best fits the points on the graph. The place at which the line crosses the left vertical axis of the graph may be interpreted as representing the voting pattern or political affiliation, as the case may be, in an average election district with no African-Americans while the point at which the line crosses the right vertical axis may be interpreted as representing the voting pattern or political affiliation, as the ease may be, in an all-African-American election district. The regression line will have a slope which measures the degree of change in the dependent variable produced by one unit change in the independent variable. 2. Extreme Case Analysis Extreme case or homogeneous precinct analysis involves examining election results and party affiliation in election districts where the voting age population is primarily either African-American or white. This analytical method has the advantage of simplicity and directness over ecological regression analysis in that it assumes that the voting patterns or party affiliations of a particular racially homogeneous election district are likely to be a fair indicator of the ethnic group’s voting pattern or party affiliation more generally. For instance, if white candidates get most or all, and African-American candidates get few or no votes in homogeneous white election districts, that is evidence that whites may be voting along racial lines. One of the obvious problems with homogeneous analysis is that the voting patterns of voters in a racially homogeneous area may not, for a variety of economic, social or political reasons, be typical of voters of the same racial group located in a racially-mixed district. See id. at 85. It is therefore recommended that results be compared to those from ecological regression analysis. Id. Only election districts where the race or ethnicity of the voting age population is at least 90% homogenous should typically be subjected to extreme case analysis. In the case at bar, however, plaintiffs expert, Lewis, indicated that only one election district was at least 90% African-American in voting age population. He therefore expanded his analysis to include five additional election districts whose voting age population was at least 85% African-American. Stanley, on the other hand, found at least eight election districts in the period prior to 1987 to be 90% or greater in African-American population and four Voter Tabulation Districts from 1987 onward that were at least 90% African-American in voting age population. Lewis attempted to account for this discrepancy by claiming that Stanley may have counted as African-American the portion of the Hispanic community reported by the U.S. Census Bureau as black Hispanic. The inconsistency in the number of districts having a population at least 90% African-American may also be attributable, in the case of pre-1987 elections, to Stanley’s use of total population figures from the 1980 Census of Population rather than 1990 Census voting age data, to Lewis’s erroneous voting age population figures or to simple error by one of the parties. The discrepancies were never adequately explained to the Court. Both experts did concur that, assuming close to 100% of the African-American community’s support, a candidate would need the votes of at least 48% to 45% of the white voting age population to be elected. 3. Other Methods for Evaluating Voting Behavior John Flateau, Executive Director of the DuBois Bunehe Center for Public Policy at Medgar Evers College, City University of New York, also testified on plaintiffs’ behalf as a political expert. Flateau examined the election results in predominantly African-American and white election districts when an African-American candidate was seeking office. Flateau did not use the statistical techniques of ecological regression analysis and extreme case analysis. He simply visually studied election results and concluded that the minority candidate usually won in election districts that were predominantly African-American but lost in election districts that were predominantly white. Simply because the minority candidate usually won in predominantly African-American election districts and lost in white districts does not, a fortiori, mean that he or she had weak or insufficient support from the white community. In fact, Flateau’s very definition of a predominantly African-American election district, an election district having an African-American population greater than 50%, makes it difficult to estimate even whether the minority candidate received his or her support from the minority or majority community. A number of the election districts in which the minority candidate succeeded had in fact a significant white constituency from which the candidate could have received votes. The Court, accordingly, finds that Flateau’s conclusions are too attenuated to be given weight of any significance. Even plaintiffs’ other expert, Lewis, stated that a mere visual examination of election data would not allow for accurate interpretation. E. Registration and General Election Data New York maintains registration data by party, but provides no racial breakdown. Stanley testified without contradiction that as of October 1993, African-Americans .comprised approximately 12.8% of all registered voters in the Town of Babylon, a percentage slightly less than their percentage of the voting age population. Stanley also provided registration by race and party as of October 1993: Party Total population registered as: African-American registered voters registered as: White registered voters registered as: Republicans 40,892 (41.0%) 5.0% 46.3% Democrats 31,955 (32.1%) 83.3% 24.5% Others 26,817 (26.9%) 11.7% 29.2% Stanley arrived at these estimates of white and African-American registration by political party by performing ecological regression and extreme case analysis using New York State registration data from October 1993 and 1990 Census data. Plaintiffs’ expert Blake testified to slightly different figures regarding party affiliation in the Town of Babylon but did not disclose the source of his data. According to Blake, 42.3% of registered voters were Republican, 31.4% were Democrats and 26.3% were independent. Analysis of election returns and testimony from two lay witnesses with long involvement in local politics, Richard Schaffer, the Town Supervisor since 1992 and a member of the Democratic Party, and Sondra Bachety, currently co-chairperson of the Babylon Democratic Party, indicate that a majority of voters who register as independents or as members of parties other than the Democratic or Republican parties tend to vote Republican in local elections. Bachety testified that voters classifying themselves as independents or members of minor parties generally split their votes on a 60%-40% basis between Republican and Democratic candidates, respectively. Despite their minority status among registered voters, Democrats have controlled the Town Board since 1987. Two Democrats were elected to Board seats in 1987 and reelected in 1991. Democrats also captured the Supervisor’s seat in 1987 and have held it since that time. Historically, however, Republicans have dominated Town Board elections. Democrats have won only 7 of the 22 council seats elected since 1975. For all but three years from 1967 to 1987, Republicans held the Supervisor’s seat. Republican domination of local politics extends to the Suffolk County legislature which is selected on the basis of single-member districts. Elections involving judicial officers are quasi-partisan under state law and have been Republican-dominated. F. Voter Turnout The Court’s own analysis of the parties’ data and Stanley’s expert testimony reveal that African-American voter turnout is lower than their rate of voter registration and then-proportion of voting age population. The 1993 Town Board election represented the highwater mark in African-American voter turnout: African-Americans constituted 10.7% of voters at the polls. The Court’s calculations show the following turnouts for other elections: Election African-American Voters as % of Voters at Polls 1987 Town Board Contest 73% 1989 Town Board Contest 8.36% 1991 Town Board Contest 7.21% 1989 District Court Contest 7.14% 1989 Supreme Court Contest 6.6% 1991 District Court Contest 8.7% Voter turnout among all races and ethnic groups as compared to total voting age population (154,988, using 1990 Census figures) was generally low in the contests specified above, being consistently at about 30%. In the 1993 Town Board election, this percentage was as low as 26.7%. G. Individual Election Contests 1. The 1988 Democratic Presidential Primary The earliest election that Lewis analyzed was the 1988 Democratic presidential primary in which the Reverend Jesse Jackson was a candidate. Lewis concluded that a majority of African-American voters supported Jackson and that a majority of white voters supported the several white candidates seeking the nomination. Lewis’s ecological regression estimates of the percentage of the white vote that the candidates other than Jackson received, however, differed markedly, by as much as 19.8%, from the total actual votes these candidates received. 2. Judicial Contests All except one of the local contests that Lewis examined were judicial elections. Judicial elections tend to be low-profile contests in which voters are likely to vote along party lines. Town Board elections are somewhat similar in that, according to Bachety, unless a candidate is a well-known person or has a “very high profile campaign” (Tr. at 657), which she defined as a “well-financed, well-run campaign” (Tr. at 669), the candidate would have a difficult time attracting the attention of voters to the same extent as candidates for the Town Supervisor’s or Suffolk County Executive’s race. In Baehety’s view, if voters in the Town were unfamiliar with the candidates, they would be more likely to vote along party lines. Judicial elections do differ from Town Board contests, however, in that judicial candidates are more limited in their ability to campaign. Candidates, therefore, do not have the same opportunity to make themselves stand out above the others. There was also uneontroverted testimony regarding the critical importance of cross-endorsements in judicial elections, a factor absent in Town Board races. Lewis’s review of judicial contests had other problems. His regression analysis revealed that in the 1990 District Court election, African-American support for Francel Maria Trotter-Bellinger, the African-American candidate, was zero, a figure obviously in error since Lewis’s extreme case analysis for the same election indicated that 90% of the voters in predominantly African-American election districts supported Trotter-Belling-er. Moreover, as in the case of the 1988 Democratic primary, Lewis’s regression analysis of this election, the 1989 and 1991 District Court contests and the 1989 Supreme Court race produced estimates of African-American and white support for the various candidates that did not closely correlate with the actual votes received by these candidates. Following are charts reflecting these discrepancies: Actual Votes Received (Top four candidates) Candidate Lewis (Party-Race) Estimate 1989 District Court Contest Trotter-Bellinger D-B 16,503 15,694 (off by 809) (5.1%) Klein R-W 23,701 20,703 (off by 2,998) (14.5%) Barton R-W 25,317 23,285 (off by 2,032) (8.7%) Braslow D-W 16,482 15,653 (off by 829) (5.3%) 1990 District Court Contest Trotter-Bellinger D-B 13,601 15,751 (2,150) (13.6%) Loughlin R-W 23,701 22,069 (1,632) (7.4%) 1991 District Court Contest Trotter-Bellinger D-B 20,467 19,296 (off by 1,171) (6.1%) Costello R-W 24,375 22,571 (off by 1,804) (8%) 1989 Supreme Court Contest (Top six candidates) Floyd K=B 26,013 24,216 (off by 1,797) (7.4%) Freidenberg D-W 15,113 14,506 (off by 607) (4.2%) O’Brien R-W 24,240 22,531 (off by 1,709) (7.6%) Segal R-W 22,220 20,743 (off by 1,477) (7.1%) Seidell D-W 17,984 17,005 (off by 979) (5.8%) Winick D-W 14,650 13,902 (off by 748) (5.4%) Lewis and Stanley disagreed as to the degree of discrepancy between estimated and actual election results that would render an expert’s ecological regression analysis unreliable. Stanley testified that the percentage difference between estimated votes calculated through regression analysis and actual election results should be less than %%, as a greater discrepancy would denote a mistake in data transposition or in presentation of the expert’s results. Lewis indicated that a discrepancy of less than 10% was not problematic, citing as authority Grofman, supra, at 92 (“if at the jurisdictionwide level, estimates of known values show a discrepancy of more than, say, 10 percent, there is reason to scrutinize the results with special care”). Only Lewis’s regression analyses of the 1988 Democratic primary and the 1989 and 1990 District Court campaigns yielded possible error in excess of 10%. Although not as severe, the mismatch between actual election data and Lewis’s estimates in the remaining judicial contests were all in excess of 4%. Lewis made little effort to justify the discrepancies. He stated only that, “[t]he differences — I am not sure exactly what the causes of the differences are between my estimated vote totals and the actual total votes. For just about every case, they are within the accepted ten percent range.” (Tr. at 1046.) This response is troubling, as even the source that Lewis cites does not state that discrepancies under 10% are acceptable but rather that those over 10% should be viewed with “special care.” That does not mean that possible error under 10% requires no explanation. See Cousin v. McWherter, 904 F.Supp. 686, 693 (E.D.Tenn.1995) (where expert noted that accuracy of regression analysis evidenced where data are within a “few percentage points or two” of actual votes that candidates received). In light of Lewis’s inability to explain the discrepancies in his results and Stanley’s testimony that regression estimates should closely correlate with known data, the Court has difficulty relying on Lewis’s regression estimates. Nevertheless, the Court will proceed in examining Lewis’s estimates for the 1989 and 1991 District Court elections and the 1989 Supreme Court election. The Court declines to consider more closely the returns from the 1990 election since, given Lewis’ clearly erroneous attribution of zero African-American votes to Trotter-Bellinger, it is impossible to assess whether the votes that she did receive came from white or African-American voters. In the 1991 District Court contest which involved only a single judicial seat, Lewis’s regression analysis reveals that 40.5% of white voters supported the African-American Democrat, Trotter-Bellinger, and 59.5% of white voters supported the white Republican candidate, Ralph Costello. 100% of African-Americans voting were estimated to have supported Trotter-Bellinger. Lewis’s extreme ease analysis shows that 59% of voters in predominantly white districts supported Costello and 94% of voters in predominantly African-American districts supported Trotter-Bellinger. Lewis did not produce an extreme case analysis of white support for the African-American candidate and African-American support for the white Republican candidate. If one assumes that votes not east for the candidate reported were cast for other candidate, then predominantly white districts provided 41% support to Trotter-Bellinger and predominantly African-American districts provided 6% support to Costello. The remaining judicial election contests that Lewis considered were ones in which more than one seat was up for election. In the 1989 Supreme Court judicial election, which involved three seats, Lewis’s regression analysis indicates that Judge Floyd’s share of the white vote was virtually indistinguishable from that of the two white Republicans, Edward O’Brien and Marvin Segal, who were running in the same contest. Lewis estimated that each Republican received approximately 20% of the white votes cast for a candidate. Judge Floyd was elected along with O’Brien and Segal. Lewis’s regression analysis estimated that Judge Floyd was in third place among African-Americans. Judge Floyd received an estimated 400 votes less than two of the three white Democrats but about 350 votes more than the third Democrat. Lewis’s extreme ease analysis were generally consistent with these findings, although that analysis ranked Judge Floyd as the first choice among African-American voters. In the remaining multi-seat judicial election that Lewis studied, the 1989 District Court contest involving two seats, Lewis estimated, using extreme case analysis, that African-American election districts gave their greatest support to the African-American candidate Trotter-Bellinger (50.96% of votes cast for a candidate), substantial, but significantly less, support to the white Democrat, Steven Braslow (38.28% of votes cast for a candidate), and very little support to the Republicans and other party candidates. White election districts gave their greatest support to the two white Republican candidates, Patrick Barton and Joseph Klein, Jr., (31.99% and 29.91% of votes cast for a candidate, respectively), and lesser, but roughly equal, support to Trotter-Bellinger and Braslow (16.77% and 18.00% of votes cast for a candidate, respectively). Regression analysis of the 1989 District Court election attributed 58.3% of the African-American vote to Trotter-Bellinger, with the remaining 41.66% of the African-American vote going to Braslow. Lewis’s regression analysis of white voting patterns in this election more closely corresponded with his extreme case analysis data. Lewis again found that white voters split their vote, with the Republican candidates receiving between 30.3% and 32.4% of white votes and Trotter-Bellinger and Braslow, the two Democrats, receiving 16.2% and 17.6%, respectively, of white votes. The Republicans, Barton and Klein, were the victors in this race. 3. 1993 Town Board Election Both Lewis and Stanley analyzed the 1993 Town Board election, the first such election in which an African-American was on the ballot. As evidenced by the chart below, the same discrepancies between the estimated and actual votes received by a candidate that existed in Lewis’s regression analysis of judicial elections are present in Lewis’s analyses of this election. Candidate Actual Votes Lewis "RAPATtfAfl Estimate Taylor D-B 17,615 (off by 1,250) (7.1%) 18^65 Horsley D-W 18,038 (off by 1,344) (7.4%) 19,382 Brown R-W 20,786 (off by 1,569) (7.5%) 22,355 Tafuri R-W 19,041 (off by 1,428) (7.5%) 20,469 Again, the Court will proceed, despite this problem, to consider both experts’ conclusions. The Court has the additional comfort in this ease that Stanley’s and Lewis’s estimates were comparable. Stanley’s ecological regression and extreme case analyses found that 97% and 81.8%, respectively, of African-American voters at the polls supported the election of Myma Taylor, the African-American candidate. Lewis’s regression and extreme case analyses attributed 57% of African-American votes cast for a candidate to Taylor and 49.4% of such votes to her white Democratic running mate, Wayne Horsley, respectively. Taylor testified during trial that she had not performed as well as Horsley in Babylon’s white communities. This claim is not borne out by either expert’s findings. According to Stanley, Taylor and Horsley garnered the support of an estimated 36% and 40% of white voters, respectively. Lewis estimated that Taylor and Horsley commanded 20% and 22% of white votes cast for a candidate, respectively. Thus, both experts’ analysis showed that white voters favored Taylor at nearly the same rate as Horsley. The Republican candidates in the 1993 Town Board contest fared considerably better in the white regions of the Town than the Democrats. Stanley estimated that Anthony Tafuri and Francine Brown garnered support from 52% and 56.5% of white voters, respectively. Lewis found Tafuri and Brown to have received between 28% and 31% of white votes cast for a candidate, respectively. This white support led Tafuri and Brown to victory- 4. Town Board and Town Supervisor Elections: 1983-1993 During the period from 1983 through 1993, there were twelve Democratic candidacies for Town Board seats. Republicans won eight out of the twelve seats. Taylor received more support on the Democratic line from white voters than six of the eleven white Democratic candidates running in the period from 1983 to 1993. In 1987, Democrats gained a majority of the Town Board for the first time since at least 1967. Schaffer and Bachety testified to the causes leading to the defeat of the incumbent Republican Town Supervisor and his two incumbent Republican running mates competing for the Town Board seats. A private citizen unhappy with a Board decision affecting his business launched a well-financed campaign to defeat the incumbents. The personal popularity of the Democratic candidate for County Executive, who won a landslide victory in that election, further aided the Democratic Town Board candidates. In the next election in 1989, the now incumbent Democratic Supervisor, Arthur G. Pitts, was able to gain reelection but his personal popularity was insufficient to carry his running mates for the Board seats, both of whom lost to the Republican incumbents. In the 1991 election where the Supervisor, Pitts, and both Democratic Board members were running as incumbents, the Supervisor was reelected, finishing first among, but without majority support from, white voters. The two incumbent Democratic Board members lost among white voters, but were reelected by virtue of overwhelming support from African-American voters. In the 1993 election, Schaffer, as the Democratic candidate for Town Supervisor, gained for the first time ever for a Democratic candidate running for Town office, majority support from the white community, earning between 59.5% and 59.7% of the white vote. Schaffer’s popularity, however, was unable to carry the other Democratic candidates for the Town Board to victory. 5. African-American Support for Democratic Candidates In all but one of the elections that Lewis and Stanley examined, the African-American community in Babylon provided majority support to the Democratic candidates, many of whom were white. The one exception to this pattern is that the African-American community supported Judge Floyd’s 1989 campaign for New York State Supreme Court over at least one and possibly all (depending on whether Lewis’s regression or extreme case analysis is relied on) of his white Democratic opponents. Lay testimony regarding other elections not studied by Lewis and Stanley indicates that this African-American support of an African-American Republican over a Democratic candidate may be atypical. Schaffer and one of the named plaintiffs, Dr. Eugene T. Reed, testified that in a Suffolk County legislative contest, African-American voters, including Reed himself, supported the highly popular white Democrat Maxine Postel over the politically inactive African-American Republican candidate, Ron Travis, who was from North Amityville. H. History of African-Americans in the Town of Babylon No African-American witnesses at trial testified to having experienced difficulty in registering to vote. Nor was any witness aware that his or her parents had experienced difficulty. In fact, in response to interrogatories during the discovery phase of this litigation, plaintiffs stated that no member of their class had ever been prohibited from voting, from running for office, from registering in the political party of their choice, from participating in the internal affairs of the political parties, from attending meetings of political parties or from attending conventions at which candidates for local election were selected. Plaintiffs’ interrogatory responses also failed to identify any differences in the Town’s registration and voting procedures for African-American and other citizens. Plaintiffs further represented in their interrogatory responses that African-Americans in the Town of Babylon have sought and been elected to offices in both the Democratic and Republican parties. At trial, Reed reported that in 1979, he assisted in an African-American voting registration drive that added 2,000 new voters to the Suffolk County registration rolls. He testified that Suffolk County police officers subsequently appeared at the homes of the new registrants to verify registration information. Plaintiffs did not provide the names of any registrants who were contacted. Plaintiffs did offer into evidence a letter, dated May 22, 1981, from the United States Attorney’s Office in the Eastern District of New York to the Board of Election of Suffolk County which referred to “numerous Civil Rights complaints surrounding the verification procedures utilized by the Suffolk County Police Department.” This letter was never authenticated in accordance with Federal Rule of Evidence 901. Given that the authority and signature of the letter’s signatory was never verified that the letter is dated May 22, 1981, approximately two years after the incident about which Reed testified, that it is somewhat unclear whether the letter is referring to the same incident and that the letter assumes knowledge of facts not in the Court’s possession, the Court is unable to give significant weight to the letter. Plaintiffs testified to a number of election campaigns that they believed were characterized by racial appeals or resulted in discriminatory conduct. Reed stated that during his 1979 campaign for a Suffolk County legislative seat, fliers with his picture were circulated in the white community. The caption on the fliers read something to the effect of “The truth is out: If he gets elected, he will only care about North Amityville and Wyan-danch.” Plaintiffs never supplied the Court with a copy of the flier or specified who was responsible for its distribution. After the 1979 campaign, in which Reed was unsuccessful, the legislative districts were redrawn so that North Amityville and Wyandaneh were no longer part of the same district. Taylor, the African-American candidate in the 1993 Town Board race, testified that during her campaign, the Democratic party’s general campaign manager, Leonard Friese, instructed her to concentrate her efforts in the African-American community and not to campaign in the white community. Given the seriousness of this allegation, the Court would have expected plaintiffs to have subpoenaed Friese to testify or given some indication as to his unavailability rather than rely on Taylor’s hearsay testimony. While the Court does not doubt that Taylor perceived Friese as having discouraged her from campaigning in the white districts, the Court cannot assess, without Friese’s own testimony, whether Friese actually intended to discourage Taylor or whether Taylor may have misconstrued his advice. That Taylor could have misconstrued Friese’s advice is buttressed by Schaffer’s testimony that Taylor was included in invitations to white neighborhood meetings. Moreover, Taylor herself readily conceded that the Democratic party leadership had actively recruited her to run and included her in campaign activities. Taylor further testified that racial appeals were used in a 1984 School Board contest in which the NAACP sponsored three candidates, two of whom were African-American. Separate polling places were apparently established for this contest so that African-Americans firing in northern Amityville voted at a junior high school in North Amityville while whites firing in southern Amityville voted at the high school. Taylor claimed that campaign literature was also circulated stating that if an African-Ameriean were elected to the School Board, property values in the region would decline. Plaintiffs did not provide the Court with a copy of this campaign literature or identify who was responsible for its dissemination. Reed described the general climate in the Babylon community as discriminatory. He began by recalling plans in the 1950’s to build two same grade schools on opposite ends of the town. In Reed’s view, this plan, had it been carried out, would have resulted in de facto segregation of the African-American and white school populations. In the end, however, this plan was scrapped. Reed also implied that discrimination against the African-American community was evident in the tracking that occurred in Babylon schools, whereby students, presumably with more advanced skills, are allowed to take classes commensurate with their abilities. Reed, however, never produced more than a few anecdotes of how tracking in Babylon schools has a discriminatory effect. African-Americans in the community of Babylon had difficulty through the 1960’s entering unions, obtaining employment in the private sector and with the Town government and becoming teachers in Suffolk County schools. The extent to which these barriers persisted after the early 1960’s was never clearly addressed. Plaintiffs did indicate that African-Americans presently have adequate representation as teachers in the Suffolk County schools. Reed testified, however, that African-American police officers in the Suffolk County Police Department, which is responsible for patrolling the Town of Babylon, have “had very big problems,” resulting in an unspecified number of discrimination suits. (Tr. at 55.) Plaintiff witnesses testified at length as to the location of the Town’s undesirable facilities in the African-American residential districts. Among the facilities plaintiffs cited as being located near African-American residences were the Town’s landfill, a crematorium, an animal shelter and high voltage electrical wires. The landfill, located in a large industrial zone in the West Babylon/Wyan-danch region, was constructed in the 1940’s. No residential population fives immediately next to the landfill. The area near the landfill became residentially populated after the landfill was constructed. Moreover, both African-American and white residents appear to be affected by the landfill’s plume. With respect to the crematorium, plaintiffs did not specify what it was about this facility that they perceived as undesirable. The crematorium itself appears to be located in the industrial zone and thus away from residential neighborhoods. Neither party indicated clearly on a map where the Town’s high voltage electrical wires were located. Schaf-fer did testify, however, that white neighborhoods, including his own, were in proximity to high voltage electrical wires. As to the animal shelter, it was unclear from the testimony whether the African-American community had ever voiced concern to the Town Board about its location. Schaffer further noted that an undesirable not mentioned by plaintiffs, the Town’s sewer, was located in a white residential neighborhood. The Town of Babylon is improving services and expand community development in the African-American communities of the Town. In 1994, 74% of federally funded community development dollars were spent on projects in Wyandaneh and North Amityville, predominantly African-American regions of the Town. Over the six-year period ending in 1994, the Town spent 43.7% of federally funded community development dollars on such projects in these communities. In the last year, Schaffer also has discussed with a major supermarket chain the possibility of locating a store in Wyandaneh. Plaintiffs claim that the Town has not adequately policed the African-American communities, resulting in growing crime rates. No statistical support, in the form of higher crime rates, delays in answering 911 calls or number of officers staffing these communities, was offered to buttress this assertion. Reed and Burnett seemed to attribute the problem not to the absence of police officers but to the style of policing. Reed testified that while police officers do patrol in the predominantly African-American regions, there are fewer African-American police officers from those communities assigned to work there than in the past, resulting in Reed’s view, in higher crime rates. The Town is making efforts to improve policing in the predominantly African-American regions. The Town is building a new police substation for the communities of North Amityville and Wyandaneh. Schaffer also testified to the Town’s attempts to get the County to improve policing in the minority areas of Town. Plaintiffs also sought to prove that the Town’s resources are unequally distributed through testimony that the African-American communities were the last to obtain access to the Town’s water supply. Schaffer himself, however, indicated that he had just been hooked into the Town’s water system. The Town, moreover, is making efforts to hook up the areas currently without public water. None of these areas are at risk of landfill contamination. African-American citizens of Babylon appear to have ready access to the current Town Supervisor and serve in a variety of policy-making positions in the Town. Reed is the Deputy Town Supervisor, an appointed political patronage position. His office is next to that of the Town Supervisor. Eugene Burnett, a citizen of the Town, is co-chairperson of the Town’s Democratic Party and is consulted by the Supervisor on a variety of matters affecting the Babylon community. Moreover, Burnett, who is also chairperson of the African-American Caucus, frequently makes recommendations to the Town Board of African-Americans to fill vacancies in the Town government and on municipal boards and committees. The Town Board has followed many of these recommendations. Until just prior to trial, Burnett held a paid position on the Industrial Development Council. His replacement on the council is also African-American Taylor, the African-American candidate for the Town Board in 1993, is the Town’s Commissioner of Human Resources. Named plaintiffs Lillian Scott, Bernice Burnett and Gwendolyn Brown also hold board positions. African-Americans occupy a share of Town administrative positions and board appointments roughly proportional to their population in the Town. CONCLUSIONS OF LAW I. Section 2 Claim A. Applicable Standards Congress enacted Section 2 of the Act to eliminate discrimination in voting present since the end of the Reconstruction period in the 1870’s. The Act was intended to enforce the Fifteenth Amendment’s guarantee that no citizen’s right to vote would “be denied or abridged ... on account of race, color, or previous condition or servitude.” U.S. Const.Amdt. 15; see Voinovich v. Quilter, 507 U.S. 146, 152, 113 S.Ct. 1149, 1155, 122 L.Ed.2d 500 (1993). In 1982, Congress amended the Act to make clear that a finding of a positive intent to discriminate was not required to make out a Section 2 violation. S.Rep. No. 417, 97th Cong., 2d Sess. 2, 27 (1982) [hereinafter “S.Rep.” or “Senate Report”], reprinted in 1982 U.S.C.C.A.N. 177, 179, 205. These amendments were a repudiation of the Supreme Court’s decision in City of Mobile v. Bolden, 446 U.S. 55, 100 S.Ct. 1490, 64 L.Ed.2d 47 (1980), which had held that plaintiffs could successfully challenge a voting-scheme under the Act only if they could establish discriminatory intent. Gingles, 478 U.S. at 34, 106 S.Ct. at 2758. The 1982 amendments firmly applied the “results” test employed previously in White v. Regester, 412 U.S. 755, 93 S.Ct. 2332, 37 L.Ed.2d 314 (1973) and Whitcomb v. Chavis, 403 U.S. 124, 91 S.Ct. 1858, 29 L.Ed.2d 363 (1971) in place of the Bolden standard. See S.Rep. at 2, 20-23, 32-33, 1982 U.S.C.C.A.N. 179, 197-201, 209-11. “Congress substantially revised § 2 to make clear that a violation could be proved by showing discriminatory effect alone and to establish as the relevant legal standard, the ‘results test,’ applied by this Court in White v. Regester, 412 U.S. 755, 93 S.Ct. 2332, 37 L.Ed.2d 314, ... and by other federal courts before Bolden, . . . .” Gingles, 478 U.S. at 35, 106 S.Ct. at 2758; see also id. at 97, 106 S.Ct. at 2790-91 (O’Connor, J., concurring). Section 2 of the Voting Rights Act provides: (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 ... to vote on account of race or color ... (b) A violation of subsection (a) 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 (emphasis in original). The Supreme Court’s first interpretation of Section 2, as amended, came in Gingles, 478 U.S. at 34-5, 106 S.Ct. at 2758, a case challenging certain multimember districts, plus one single-member district, in North Carolina’s post-1980 legislative apportionment plan. Gingles held that “[t]he essence of a § 2 claim is that a certain electoral law, practice, or structure interacts with social and historical conditions to cause an inequality in the opportunities enjoyed by African-American and white voters to elect their preferred representatives.” Id. at 47, 106 S.Ct. at 2764; see also Nipper v. Smith, 39 F.3d 1494, 1511 (11th Cir.1994) (en banc), cert. denied, - U.S. -, 115 S.Ct. 1795, 131 L.Ed.2d 723 (1995). The Court provided three threshold conditions that must be established in order to prove a Section 2 violation: First, the minority group must be able to demonstrate that it is sufficiently large and geographically compact to constitute a majority in a single-member district. ... Second, the minority group must be able to show that it is politically cohesive.... Third, the minority must be able to demonstrate that the white majority votes sufficiently as a bloc to enable it — in the absence of special circumstances, ... — usually to defeat the minority’s preferred candidate. Gingles, 478 U.S. at 50-51, 106 S.Ct. at 2766 (citations omitted); see also Johnson v. DeGrandy, — U.S. —, — - —, 114 S.Ct. 2647, 2657-58, 129 L.Ed.2d 775 (1994) (emphasizing continued appropriateness of Gin-gles preconditions); Growe v. Emison, 507 U.S. 25, 39-40, 113 S.Ct. 1075, 1084, 122 L.Ed.2d 388 (1993) (stating that three Gin- gles preconditions are applicable both to challenges to multimember and single-member districts). Plaintiffs cannot prevail unless they prove each of these preconditions by a preponderance of the evidence. Gingles, 478 U.S. at 50, 106 S.Ct. at 2766. The first two preconditions, requiring geographic eompactness/numerosity and minority political cohesion, are necessary “to establish that the minority has the potential to elect a representative of its own choice in some single-member district.” Growe, 507 U.S. at 40, 113 S.Ct. at 1084. These preconditions essentially ask “whether the court can fashion a remedy for a demonstrated abridgement” of Section 2 of the Act. Nipper, 89 F.3d at 1510-11. No remedy would exist under Section 2 for a group that lacks the population or political cohesiveness, or is too geographically dispersed, to benefit from single-member districts. See McNeil v. Springfield Park Dist., 851 F.2d 937, 942 (7th Cir.1988), cert. denied., 490 U.S. 1031, 109 S.Ct. 1769, 104 L.Ed.2d 204 (1989). The second Gingles precondition is also needed, together with the third, to determine whether “the challenged districting thwarts a distinctive minority vote by submerging it in a larger white voting population.” Growe, 507 U.S. at 40, 113 S.Ct. at 1084; see Monroe v. City of Woodville, Mississippi, 897 F.2d 763, 764 (5th Cir.) (per curiam), cert. denied, 498 U.S. 822, 111 S.Ct. 71, 112 L.Ed.2d 45 (1990); Collins v. City of Norfolk, Virginia, 883 F.2d 1232, 1237 (4th Cir.1989), cert. denied, 498 U.S. 938, 111 S.Ct. 340, 112 L.Ed.2d 305 (1990). The second and third preconditions thus relate to liability, that is, whether the “electoral scheme is abridging the right of the plaintiff minority group to vote on account of race or color” in violation of Section 2. Nipper, 39 F.3d at 1510-11 (internal quotations omitted). Critical terms in determining whether the preconditions have been met are minority voter “political cohesiveness” and majority “bloc voting.” These terms refer to the relationship between the race or ethnicity of the voter and the way in which the voter easts his or her ballot. See Gingles, 478 U.S. at 53 n. 21, 106 S.Ct. at 2768 n. 21. For instance, “that a significant number of minority group members usually vote for the same candidates is one way of proving the political cohesiveness necessary to a vote dilution claim.” Gingles, 478 U.S. at 56, 106 S.Ct. at 2769; see also Monroe, 897 F.2d at 764. Although lay testimony may evidence political cohesiveness or racial bloc voting, statistical evidence in the form of double-equation ecological regression analysis and extreme case analysis, presented through expert testimony is typically also offered. Gingles, 478 U.S. at 53 n. 20, 106 S.Ct. at 2767; Monroe, 897 F.2d at 763 (“Statistical proof of political cohesion is likely to be the most persuasive form of evidence, although other evidence may also establish this phenomenon”). The effects of racial bloc voting can be most severe in at-large elections, where the voting strength of even a substantial minority group may be minimized or canceled by the concerted action of the majority group. See Gingles, 478 U.S. at 47,106 S.Ct. at 2764 (quoting Burns v. Richardson, 384 U.S. 73, 88, 86 S.Ct. 1286, 1294, 16 L.Ed.2d 376 (1966)) and n. 13; Chapman v. Meier, 420 U.S. 1, 16, 95 S.Ct. 751, 760-61, 42 L.Ed.2d 766 (1975) (quoting Whitcomb, 403 U.S. at 158, 91 S.Ct. at 1877). Nevertheless, an at-large voting system does not constitute a per se violation of Section 2 and is still subjected to the three Gingles preconditions. See Gingles, 478 U.S. at 48-9, 106 S.Ct. at 2765-66. Satisfaction of the Gingles preconditions is essential to a Section 2 vote dilution claim, DeGrandy, — U.S. at -, 114 S.Ct. at 2657, but is not conclusive of the existence of dilution, id. at -, 114 S.Ct. at 2657; see NAACP v. Niagara Falls, 65 F.3d 1002, 1019 (2d Cir.1995); Clements, 999 F.2d at 849; Baird v. Consolidated City of Indianapolis, 976 F.2d 357, 359 (7th Cir.1992), cert. denied, 508 U.S. 907, 113 S.Ct. 2334, 124 L.Ed.2d 246 (1993). But if Gingles so clearly identified the three as generally necessary to prove a § 2 claim, it just as clearly declined to hold them sufficient in combination, either in the sense that a court’s examination of relevant circumstances was complete once the three factors were found to exist, or in the sense that the three in combination necessarily and in all circumstances demonstrated dilution.... [T]he ultimate conclusions about equality or inequality of opportunity were intended by Congress to be judgments resting on comprehensive, not limited, canvassing of relevant facts. Lack of electoral success is evidence of vote dilution, but courts must also examine other evidence in the totality of circumstances, including the extent of the opportunities minority voters enjoy to participate in the political process. DeGrandy, — U.S. at -, 114 S.Ct. at 2657. DeGrandy thus affirmed the Supreme Court’s indication in Chisom v. Roemer, 501 U.S. 380, 396-97, 111 S.Ct. 2354, 2364-65, 115 L.Ed.2d 348 (1991), that lack of opportunity to participate is as critical to a plaintiff class’s Section 2 claim as the opportunity to elect. In Chisom, the Supreme Court had explained that “the inability to elect representatives of their choice is not sufficient to establish a violation unless, under the totality of the circumstances, it can also be said that the members of the protected class have less opportunity to participate in the political process.” 501 U.S. at 397, 111 S.Ct. at 2365; see Niagara Falls, 65 F.3d at 1019-20; Nipper, 39 F.3d at 1512-13; Clements, 999 F.2d at 843; Baird, 976 F.2d at 359. The Court must therefore also engage in a careful examination of the “totality of the circumstances” surrounding a Section 2 vote dilution claim to determine whether plaintiffs have proven that they have “less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice” to the Town Board. 42 U.S.C. § 1973; see DeGrandy, — U.S. at — - —, 114 S.Ct. at 2657-58; Gingles at 80, 106 S.Ct. at 2782; White, 412 U.S. at 766, 93 S.Ct. at 2339 (citing Whitcomb, 403 U.S. at 149-50, 91 S.Ct. at 1872-73); Niagara Falls, 65 F.3d at 1008; Nipper, 39 F.3d at 1513. Relevant to evaluating the totality of the circumstances are nine factors set out by the Senate Judiciary Committee in the Senate Report accompanying the 1982 amendments to the Voting Rights Act (the “Senate Report factors”). DeGrandy, - U.S. at -, n. 9, 114 S.Ct. at 2656, n. 9; Gingles, 478 U.S. at 44-5, 106 S.Ct. at 2762-64. These factors are: 1. The extent of any history of official discrimination in the state or political subdivision that touched the right of members of the minority group to register, to vote, or otherwise to participate in the political 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 the 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 28-9, 1982 U.S.C.C.A.N. at 206-7 (footnotes omitted); DeGrandy, — U.S. at -, n. 9, 114 S.Ct. at 2656, n. 9; Gingles, 478 U.S. at 44-5, 106 S.Ct. at 2762-64. Two additi