MEMORANDUM OF DECISION

PONSOR, District Judge.

I. INTRODUCTION

Plaintiffs, two Hispanic community organizations and eight Hispanic citizens of the City of Holyoke, challenge the city’s election system for School Committee and City Council, charging that its at-large components violate the federal Voting Rights Act of 1982. 1 This statute prohibits electoral mechanisms that provide minorities “less opportunity *913 than other members of the electorate to participate in the political process and to elect representatives of their choice.” 42 U.S.C. § 1973(b).

Having considered the testimony and exhibits presented over eleven days of non-jury trial, the extensive written submissions following trial and the final argument, this court concludes that the plaintiffs have failed to demonstrate a violation of the Act as regards the School Committee. The School Committee comprises ten members, of whom only two (not counting the city’s mayor, who sits automatically as chairperson) are elected at-large. No Hispanic person has ever run for election to either of the School Committee’s two at-large seats. At the same time, however, Hispanic citizens have had the opportunity to elect, with fair consistency during the short relevant period, representatives of their choice to the School Committee’s district seats. As of . today, they have the power to elect two representatives, or twenty percent of the School Committee, very close to their proportion of voting age population. While this proportionality is not dispositive in itself, in combination with other factors discussed below it is sufficient to tip the scales in defendants’ favor.

With regard to the City Council, however, the court does find a violation of the Voting Rights Act. A flat majority, eight out of fifteen members of the City Council, is elected at-large. While qualified Hispanic candidates have run for the City Council at-large, and Hispanic voters have supported them, their election has been foreclosed by consistent non-Hispanic white bloc voting. Viewed in the totality of the circumstances in Ho-lyoke, the at-large system for electing a majority of the City Council can only be seen as depriving the city’s Hispanic minority of a fair opportunity to participate in the political process. The court will therefore prohibit further elections for the City Council under the existing system, and will set the matter down for further proceedings on the proper remedy.

The main body of the opinion below will be presented in three parts. The first will discuss the standards to be applied in a ease of this sort, as contained both in the Voting Rights Act itself and in the decisions of the Supreme Court. The second will recount the court’s findings of fact. In the third section, the court will apply the pertinent law to the facts and state its conclusions.

II. LEGAL STANDARDS

Nearly three decades ago, Congress enacted § 2 of the Voting Rights Act of 1965 (the “Act”), as amended, 42 U.S.C. § 1973, to eliminate discrimination in voting that had plagued the nation since the end of Reconstruction in the 1870’s. Congress, together with the President, crafted the Act to enforce the Fifteenth Amendment’s guarantee that no citizen’s right to vote shall “be denied or abridged ... on account of race, color, or previous condition of servitude.” U.S. Const, amend. XV, § 1; see also Voinovich v. Quilter, — U.S.-,-, 113 S.Ct. 1149, 1155, 122 L.Ed.2d 500, 510 (1993). In order to effectuate its purpose, the statute is to be given “the broadest possible scope.” Chisom v. Roemer, 501 U.S. 380, 403, 111 S.Ct. 2354, 2368, 115 L.Ed.2d 348 (1991), quoting Allen v. State Board of Elections, 393 U.S. 544, 567, 89 S.Ct. 817, 832, 22 L.Ed.2d 1 (1969).

Heavily underlining the statute’s original goal, Congress in 1982 amended the Act to make it clear that a finding of a positive intent to discriminate was not required to make out a Section 2 violation. Latino Political Action Committee v. City of Boston, 784 F.2d 409, 412 (1st Cir.1986). These amendments were designed to repudiate 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 placed on plaintiffs the virtually impossible burden to show discriminatory intent before a voting scheme could be successfully challenged under the Act. The 1982 amendments firmly installed a “results” test in place of the Bolden standard.

The full text of the amended Section 2 reads as follows:

(a) No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of *914 the right of any citizen of the United States to vote on account of race or color, or in contravention of the guarantees set forth in section 1973b(f)(2) of this title, as provided in subsection (b) of this section, (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 (as amended Pub.L. 97-205, § 3, 96 Stat. 134.)

Section 2 now prohibits any practice or procedure that “interacting with social or historical conditions” has the effect of impairing the opportunity for a protected class to elect its candidate of choice on an equal basis with other voters. Voinovich v. Quitter, — U.S. at-, 113 S.Ct. at 1156, 122 L.Ed. at 512, quoting Thornburg v. Gingles, 478 U.S. 30, 47, 106 S.Ct. 2752, 2764, 92 L.Ed.2d 25 (1986).

As has been frequently noted, the language of Section 2 quoted above embodies a Congressional compromise, highlighted in the final sentence of the provision, that has made construction and application of the statue difficult at times.

“There is an inherent tension between what Congress wished to do and what it wished to avoid” — between Congress’ “in-ten[t] to allow vote dilution claims to be brought under § 2” and its intent to avoid “creating] a right to proportional representation for minority voters.”

Holder v. Hall, — U.S.-,-, 114 S.Ct. 2581, 2613, 129 L.Ed.2d 687 (1994), (Ginz-berg, J., dissenting and quoting Justice O’Connor’s separate concurrence in Thornburg v. Gingles, 478 U.S. 30, 84, 106 S.Ct. 2752, 2783, 92 L.Ed.2d 25 (1986)).

As Justice Ginzberg has noted, however, this difficult task “is one that the courts must undertake because it is their mission to effectuate Congress’ multiple purposes as best they can.” Id. — U.S. at-, 114 S.Ct. at 2625.

The Gingles decision remains the starting place for this effort. In the Gingles case, plaintiffs challenged certain multi-member districts and one single-member district in North Carolina’s post-1980 legislative apportionment plan. Justice Brennan’s decision for the Court, largely affirming the district court’s holding in favor of the plaintiffs, provides the analytical structure still to be used in a claim under § 2. Johnson v. De Grandy, — U.S.-,-, 114 S.Ct. 2647, 2656, 129 L.Ed.2d 775 (1994).

Gingles sets forth three now-familiar threshold conditions that must be established to prove a § 2 violation: (1) that the minority group is “sufficiently large and geographically compact to constitute a majority in a single-member district”; (2) that the minority group is “politically cohesive”; and (3) that “the white majority votes sufficiently as a bloc to enable it ... usually to defeat the minority’s preferred candidate.” Thornburg v. Gingles, 478 U.S. at 46, 50-51, 106 S.Ct. at 2764, 2766-2767; see also Johnson v. De Grandy, — U.S. at-, 114 S.Ct. at 2656 (reiterating the Gingles threshold conditions).

The first and second preconditions — geographic compactness/numerousness and minority political cohesion — are needed “to establish that the minority has the potential to elect a representative of its own choice in some single member district.” Growe v. Emison, — U.S.-,-, 113 S.Ct. 1075, 1084, 122 L.Ed.2d 388 (1993). Minority political cohesion and majority bloc voting, the second and third elements, taken together, are necessary to establish that “the challenged districting thwarts a distinctive minority vote by submerging it in a larger white voting population.” Id. There can be *915 no “wrong nor can [there] be a remedy” unless these three preconditions are met. Id.

Key terms in this analysis, minority voter “cohesion” and majority “bloc” voting, refer to a consistent relationship between the race or ethnicity of the voter and the way in which that voter casts his or her ballot. Gingles, 478 U.S. at 53 n. 21, 106 S.Ct. at 2768 n. 21. Such bloc voting behavior may be proved by lay testimony, but in most cases (including this one, as will be seen below), statistical evidence presented through expert testimony is offered to demonstrate, or disprove, racial bloc voting in a given jurisdiction. Id.

The effects of bloc voting obviously can be most severe in ab-large elections, where even a substantial minority may be entirely excluded from any representation on a political body by the concerted action of the majority group. Thus, an at-large voting system, while it does not constitute a per se violation of § 2, is suspect because it may “operate to minimize or cancel out the voting strength of racial minorities in the voting population.” Thornburg v. 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).

Even with an at-large system, however, plaintiffs cannot prevail in a § 2 claim simply by establishing the existence of the three Gingles threshold preconditions. More must be shown. The trial court’s scrutiny must penetrate beyond the three Gingles factors, because other considerations may show that the minority, despite meeting the three prerequisites, still has an undiminished right to participate in the political process. Johnson v. De Grandy, — U.S. at-n. 10, 114 S.Ct. at 2657 n. 10, quoting Baird v. Consolidated City of Indianapolis, 976 F.2d 357, 359 (7th Cir.1992).

[L]ack of equal electoral opportunity may be readily imagined and unsurprising when demonstrated under circumstances that include the three essential Gingles factors, [but] that conclusion must still be addressed explicitly, and without isolating any other arguably relevant facts from the act of judgment.

Id. — U.S. at-, 114 S.Ct. at 2657.

For this reason, a district court must always engage in an intensive examination of the “totality of circumstances” surrounding a § 2 claim to determine if minority group members have been deniéd an equal opportunity to “participate in the political process and to elect representatives of their choice.” 42 U.S.C. 1973(b). Some of the factors to be considered in this analysis are contained in the Senate Report accompanying the 1982 amendments to the Voting Rights Act. This report sets out nine factors that the Senate Judiciary Committee considered relevant to the “totality of circumstances.” Rep. No. 417, 97th Cong., 2d Sess. 27-30 (1982) U.S. Code Cong. & Admin.News 1982 pp. 177, 204-208. See also Gingles, 478 U.S. at 44-55, 106 S.Ct. at 2762-2769; Johnson v. De Grandy, — U.S. at-n. 11, 114 S.Ct. at 2658 n. 11. 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 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 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;

*916 7. The extent to which members of the minority group have been elected to public office in the jurisdiction.

Two additional factors [that may be probative] are:

8. Whether there is a significant lack of responsiveness on the part of elected officials to the particularized needs of the member of the minority group; and

9. Whether the policy underlying the state or political subdivision’s use of such voting qualification, prerequisite to voting, standard, practice or procedure is tenuous.

S.Rep. No. 417 at 28-29; Gingles, 478 U.S. at 44 — 45, 106 S.Ct. at 2762-2764.

Even this listing of nine factors is not exhaustive; other considerations may be relevant in a particular case. Ultimately, the court’s evaluation of a voting rights claim requires “an intensely local appraisal of the design and impact of the ... multi-member district in the light of past and present reality, political and otherwise.” Gingles, 478 U.S. at 78, 106 S.Ct. at 2780, quoting White v. Regester, 412 U.S. 755, 769-70, 93 S.Ct. 2332, 2341, 37 L.Ed.2d 314 (1973).

One consideration, always relevant but recently highlighted by the Supreme Court, is proportionality. A defendant may cast doubt on a § 2 claim by showing that the “percentage of single-member districts in which minority voters form an effective majority mirrors the minority voters’ percentage of the relevant population.” Johnson v. De Grandy, — U.S. at -, 114 S.Ct. at 2660. While the existence of such proportionality is not automatically fatal to a Voting Rights claim — any more than the absence of proportionality automatically entitles a minority group to a remedy — it is a highly relevant factor, as Justice Souter’s opinion in De Grandy makes clear.

Treating political opportunity as the focus of the enquiry, we do not see how these district lines, apparently providing political effectiveness in proportion to voting-age numbers, deny political opportunity.

Id., at-, 114 S.Ct. at 2658.

Although De Grandy involves a purely district-based electoral system, its holding with regard to proportionality has broader implications and was presaged by the Seventh Circuit’s decision in Baird v. Consolidated City of Indianapolis, 976 F.2d 357 (7th Cir. 1992), cert. denied, — U.S.-, 113 S.Ct. 2334, 124 L.Ed.2d 246 (1993). Baird upheld summary judgment for defendants in a case where the minority voters controlled a proportional share of the City-County Council under a hybrid district/at-large system similar in certain respects to the one at issue here. Similarly, in Latino Political Action Committee v. City of Boston, 784 F.2d 409 (1st Cir.1986), examining another hybrid system, the Court of Appeals found rough proportionality between minority voting age population and representation on the School Committee and City Council. The court concluded that these results “while not necessarily in and of themselves proof of nondiscrimination, at least indicate that the district plan works to achieve Black representation, roughly, as its sponsors said they intended.” Id. at 413. (Breyer, J.)

Finally, in weighing the totality of the circumstances the Supreme Court has instructed trial courts to consider “a reasonable alternative practice as a benchmark against which to measure the existing voting practice.” Holder v. Hall, — U.S. ——, --, 114 S.Ct. 2581, 2582, 129 L.Ed.2d 687 (1994). Justice O’Connor has suggested that “[i]n a challenge to a multimember at-large system, for example, a court may compare it to a system of multiple single-member districts.” Id., at -, 114 S.Ct. at 2589 (O’Connor, J., concurring).

In summary, while neither the Supreme Court nor Congress have mandated a single universally applicable standard, a fairly clear outline of the appropriate methodology in these cases has emerged. The trial court must begin with the three threshold factors set forth in Gingles. If the plaintiffs clear this hurdle, the judge then moves to an assessment of the totality of circumstances, taking as helpful examples the factors set forth in the Senate Report, but maintaining a broad perspective to include any local facts that may illuminate the fundamental question: do minority voters “have less opportunity to participate in the political process and to elect representatives of their choice?”

*917 III. FINDINGS OF FACT

A. Historical Background

The City of Holyoke is located just north of Springfield, between the mountains of western Massachusetts and the Connecticut River. Until the mid-nineteenth century, Holyoke was a farming community. However, its location on the Hadley Falls, where the river drops almost sixty feet, made it a prime location for the textile and paper mills that transformed this rural backwater into a hub of New England’s industrial economy.

Beginning in the 1850’s, mills began to line the waterways of Holyoke, providing jobs for the skilled and non-skilled alike. Irish, Freneh-Canadian and German immigrants entered the textile and paper mills, swelling the population of Holyoke. Later in the century, Polish and Italian immigrants joined the still-expanding workforce and further diversified the city’s ethnic makeup. Immigrants who made Holyoke their home did much to fuel the industrial engine that powered America’s rise as a manufacturing giant through the nineteenth and into the twentieth centuries. The city prospered, and many of the city’s inhabitants along with it. Ho-lyoke’s graceful houses and comfortable neighborhoods grew up, and form a substantial part of the city to this day.

Of course, industrial life in nineteenth century urban America had many grim features as well. Holyoke, a small city, had its share. Life for immigrants living in the tenements that dominated large areas of what is now Wards One and Two, known as the “Flats” and South Holyoke, included twelve-hour plus workdays, an acute housing shortage, inadequate sanitary conditions and high death rates. Overcrowded tenements, poor health, poverty and other social ills persisted well into the twentieth century, despite the efforts of organized labor, charitable organizations and social reformers.

Holyoke experienced another wave of immigration beginning in the 1970’s. Local agricultural interests encouraged rural Puer-to Ricans experienced in tobacco farming to migrate to western Massachusetts to pick tobacco and harvest other crops. Soon, Puerto Rican families, disillusioned with the low pay and harsh conditions of farm labor, began looking for other jobs and settled in Holyoke. As the community took root, Hispanics from New York City and Hartford, Connecticut joined them. The low-cost, aged housing in the Flats and South Holyoke, historic neighborhoods to poverty stricken immigrants and. their first generation offspring, became the barrios for most Hispanics who made Holyoke their home.

Unfortunately, by this time, the paper and textile industries in New England had already experienced a steep decline. Largely gone were the skilled and unskilled jobs that provided an economic mainstay for previous generations of immigrants who labored in Holyoke. In some ways history seems to have repeated itself for the majority of Hispanic residents now in Holyoke, who find themselves confronting a present-day version of the complex of economic, housing, educational and health problems endured by generations of immigrants who have come to Holyoke over the past century and a half.

B. Holyoke’s Hispanic Community: Economic and Social Conditions.

Today the economic disparities between Holyoke’s Hispanic and non-Hispanic white populations are unquestionably dramatic. Approximately 59% of Hispanic persons live below the poverty level, compared with 13.7% of the city’s non-Hispanic white residents. Two-thirds of Hispanic families have incomes of less than $15,000 per year, compared with about 30% of non-Hispanic white households. The 1990 census data reveal an unemployment rate for Hispanics more than two and a half times that of non-Hispanic whites. Ninety-five percent of Holyoke’s Hispanic households are renters; approximately one-half of the non-Hispanic white families own their own homes.

In the area of health care, the Hispanic population also suffers disproportionately. As late as the mid-1980’s, Holyoke’s Hispanics experienced one of the highest rates of infant mortality in the Commonwealth, and, currently, less than 40% of Hispanic children under five years of age are properly immunized, as compared to 65% of Hispanic children throughout the state. The AIDS rate *918 for Hispanics in the city is the highest for any discrete group in Massachusetts. Seventy percent of all AIDS cases in Holyoke are among Hispanics.

The testimony revealed that Holyoke’s municipal government often responded weakly to the health needs of the Hispanic community. For example, the task force that eventually succeeded in lowering the mortality rate for Hispanic infants in Holyoke received no more than token support from Holyoke’s Board of Health. Despite the problems with immunizations of Hispanic children, no efforts were made to retain Spanish-speaking school nurses until 1993, and even today none has been hired. The Massachusetts Department of Health provided Nueva Esperanza, a local agency based in Ward 2, with funds to set up a free community clinic that offered childhood immunizations. The City contributed no money to this effort. However, it did provide Nueva Esperanza with bilingual notices to publicize this service. In addition, the Board’s director has worked to combat the spread of vaccine-resistant tuberculosis related to high rates of HIV infection. In the health area, the evidence has shown a significant degree of neglect and inconsistency on the part of the city in addressing the needs of the Hispanic community.

The schools have exhibited perhaps the most dramatic change as a result of the growth of the Hispanic community in Ho-lyoke. In 1980, approximately 2000 of the city’s 7000 students were Hispanic. As the Hispanic community grew in the next decade, a large percentage of the non-Hispanic white students left the system for parochial and other private schools. During this time, a desegregation lawsuit claiming improper concentrations of minority students led to a Consent Decree that attempted to even the distribution of minority students in the system. For the 1993-94 school year over three-quarters of the students in Holyoke’s schools, 78%, were Hispanic.

In the late 80’s and into 1990, Holyoke attempted to address the changed composition of the student body by, for example, hiring minority faculty and implementing language programs. These efforts came to a virtual standstill, however, as a result of a fiscal crisis in 1991, when state budget cuts resulted in a reduction of the school budget by more than one-third. As defendants pointed out, this crisis was to some extent outside the control of local government, affected all areas of municipal spending and injured Hispanic and non-Hispanic students alike. Even so, plaintiffs offered convincing testimony that the impact of the cuts fell disproportionately upon Hispanic students. Bi-lingual programs suffered and Hispanic drop-out rates rose. Moreover, large majorities of Holyoke’s non-Hispanic white voters twice rejected modest budget overrides under Proposition 2^ that would have softened the blow to Holyoke’s schools substantially. At the same time, they approved a comparable override for trash collection. A report at this time by a committee of the Massachusetts Board of Education criticized Holyoke and Lawrence as examples of “community neglect of education as a responsibility of local government.” Plaintiff’s Exhibit 52, at 3.

In Holyoke, a pronounced disparity exists in the level of educational attainment of Hispanics and non-Hispanic whites. As of 1990, 33.5% of Hispanics attained less than a ninth grade education compared to only 11% of the white population. Approximately half as many Hispanics as whites have a high school diploma — 30.6% compared to 58.4%.

In the area of housing, Hispanics again suffer disproportionate disadvantages. Eighty percent of the Hispanic population of Holyoke lives in one contiguous geographic area of the city, the vast majority within four census tracts commonly known as the Flats and South Holyoke. Between 1968 and 1983 more than 4000 housing units were demolished in the city, most of them in the Flats and South Holyoke, where more than half the taxable housing stock was built prior to the Depression. As a result, overcrowding has increased in these poorer areas. In 1980, the vacancy rate in the Flats and South Holyoke was 17%, and overcrowding occurred in 7% of units. Ten years later, vacancies dropped to 9.6% and overcrowding increased to 13%, while the rates in other parts of the city remained relatively constant. During the same period rents increased dra- *919 matieally in the Flats and South Holyoke, both in absolute terms and as a percentage of monthly income.

As with its response to other social problems besetting the Hispanic community, Ho-lyoke’s response to housing needs has been mixed. In 1982, the city was harshly criticized by the Massachusetts Commission Against Discrimination for its failure to comply with federal civil rights laws as a condition to receiving Community Development Block Grants (CDBGs) from the federal Department of Housing and Urban Development (HUD). Through the years, the city’s support of any program advocating fair housing or attempting to enforce laws prohibiting discrimination in housing has been tepid or non-existent. In 1985, the mayor of Holyoke refused to authorize a request for twenty-five federally funded Section 8 housing certificates on the ground that the City was already saturated with public housing. In the same vein, the City resisted, this time unsuccessfully, the rehabilitation of row houses in the downtown area that benefitted Hispanic residents. However, during the same time the evidence demonstrated that Holyoke has devoted the bulk of its CDBG funds — 87% between 1985 and 1993 — to Hispanic areas of the city.

The recent experience of Nueva Esperanza, an established Hispanic community development agency, typifies the tug of war between the City and the Hispanic community agencies. In 1994, the organization sought $210,000 in CDBG funds to rehabilitate the aged and dilapidated Skinner Community Center in South Holyoke. The proposal was rejected by the mayor and the City Council Redevelopment Committee. Mayor William Hamilton stated that he felt that Hispanics were already “well taken care of’ in Holyoke. At the same time, however, the mayor did recommend funding $90,000 worth of housing, arts and substance abuse programs housed in the Skinner Center.

In summary, Hispanic households in Ho-lyoke indisputably experience very substantially lower levels of income, employment, health, education, and housing as compared to non-Hispanic white families. While the City, especially in recent years, cannot be said to have turned its back on its Hispanic citizens, its response to the needs of the Hispanic community has often been slow and halting, and at times egregiously neglectful.

C. Election History Since 1983.

The election system for Holyoke’s School Committee assumed its current form over 120 years ago, in 1873, with ten members: two elected at-large, one from each of seven districts, and the mayor sitting as chairperson. The Holyoke City Council, formerly called the Board of Aldermen, took on its present composition in 1963,. with fifteen members — eight elected at-large and one from each of the city’s seven districts.

According to the 1980 census, Holyoke had 44,678 total inhabitants, of which 6,165, or 13.8%, described themselves as Hispanic. Ten years later, by the time of the 1990 census, the City’s population had dropped somewhat to 43,704, while the Hispanic community had more than doubled, to 13,573 or 31.06% of Holyoke’s total population. 2 The percentages of voting age population are slightly different, reflecting the younger age of the Hispanic community, with Hispanics making up approximately 21.89%, and non-Hispanic whites 74.90%, of the voting age total in Holyoke.

In May 1992, the. Holyoke City Council adopted a redistricting plan, redrawing then-existing lines for the city’s seven wards. 3 Under the 1992 plan now in effect, Hispanics comprise 70.9% of Ward 1 and 76.51% of Ward 2, with 59.43% and 66.65% of the voting age population respectively. The 1990 census figures reveal that more than 85% of Holyoke’s Hispanic population is concentrated in these two wards and Ward 4, which is 38.48% Hispanic. The City’s remaining four *920 wards contain small or negligible percentages of Hispanics.

All municipal elections in Holyoke are nonpartisan, and the requirements to run for office are minimal. None of the suspect devices often used historically to exclude minority candidates, such as majority vote requirements or anti-single shot provisions, has ever existed in Holyoke.

Since 1981, Hispanic candidates have been elected to seven positions on the Holyoke City Council or School Committee. Four of these elections were uncontested, and all seven seats were won in Ward 2. As of the last elections, held in November, 1993, fourteen of the fifteen seats on the City Council, and ten of the eleven seats on the School Committee, were held by non-Hispanic whites. Currently, Hispanics occupy the City Council seat and School Committee seat from Ward 2. Both ran for office unopposed.

As noted above, eight of the fifteen City Council seats and two of ten School Committee seats (not counting the mayor as chairperson and tenth member) are elected at-large. All of the at-large seats in both bodies become vacant at the same time. That is, in each at-large election, voters in all wards go to the polls at the same time and may vote for up to eight candidates for the City Council and two for the School Committee. As noted in the introduction, all Hispanics running at-large for the City Council have lost, and none has run at-large for the School Committee.

In order to weigh the sufficiency of the plaintiffs evidence with regard to the second and third Gingles criteria, minority voter cohesion and majority bloc voting, the court must analyze carefully the actual election results in Holyoke. See Thornburg v. Gingles, 478 U.S. at 63, 106 S.Ct. at 2772. Two complementary methods — bivariate ecological regression analysis and extreme case analysis — have been recognized as proper to determine whether majority and minority voters differ in their voting behavior. Gingles 478 U.S. at 52-53, 106 S.Ct. at 2767-2768.

Ecological regression analysis is the standard technique for inferring the voting behavior of distinct population groups. Since it is obviously impossible to interview every voter, classify each by race or ethnicity, and determine how every ballot was cast, regression analysis attempts to make reasonably accurate estimates of majority and minority voting patterns by inferring voting behavior from demographic data and election returns. Regression methodology generates prediction equations that show how voting results vary depending on the proportion of minority and majority racial or ethnic groups in a given precinct. From this precinct information it is possible to generalize — that is, to estimate the overall average voting of the minority and majority groups in the election district being analyzed.

Extreme case analysis compares voting precincts or electoral units that are composed overwhelmingly of one ethnic or racial group. If voting is polarized along ethnic lines, extreme case analysis shows the differences in percentages of votes going to the candidates favored by each ethnic group in the precincts that are predominantly composed of a single ethnic group. This mode of analysis makes intuitive sense. If a particular precinct is composed 99% of a single ethnic group, the election results in that precinct probably are a fair indicator of the cohesiveness of that particular ethnic group’s voting pattern.

Plaintiffs expert, Dr. Alan Liehtman, evaluated all City Council and School Committee at-large elections between 1983 and 1993 where Hispanic candidates ran for office. Liehtman applied ecological regression analysis to precinct election return data to test for a pattern of bloc voting by Hispanic and non-Hispanic white voters. Liehtman defended his decision to focus on those elections where Hispanics ran for office, on the ground that (in his view) the best test of the adequacy of the minority’s opportunity to elect their chosen candidates was whether they could elect persons of the same ethnic background. While the facts of this case indicate that in one election Holyoke’s Hispanics supported a sympathetic non-Hispanic candidate more enthusiastically than an Hispanic, this does not undermine the overall validity of his anal-yses of the elections he did look at. See *921 Thornburg v. Gingles, 478 U.S. at 68, 106 S.Ct. at 2775 (“it is the status of the candidate as chosen representative of a particular racial group, not race of candidate, that is important.”) For this reason, this court has mainly adopted Liehtman’s testimony.

The court’s confidence in Liehtman’s findings was buttressed by the fact that he crosschecked his results under ecological regression analysis, by performing an extreme case analysis in wards or precincts that were overwhelmingly composed of either white non-Hispanic or Hispanic persons. Extreme case analysis was used to evaluate election results in these precincts from 1983 through 1993. This methodology was especially useful to corroborate white bloc voting as the majority of non-Hispanics live in precincts that are 90% or more composed of whites. Lichtman also performed extreme case analysis on precincts that were 79% or more Hispanic, using the available data for the years 1987 through 1993. Finally, Lichtman also used ecological regression analysis to evaluate the voting behavior of Hispanics and non-Hispanic whites on referenda questions in two elections in 1991.

An examination of Holyoke’s elections since 1983 reveals a discernible pattern of minority voter cohesion and majority bloc voting, manifesting with varying degrees of intensity from one election to another.

1. 1983 Elections

Thirteen non-Hispanic whites, two Hispanics and one African-American candidate ran for Alderman in the 1983 at-large general election. The Hispanic candidates, Cruz and Gomez, ranked, respectively, first and third among Hispanic voters, and Little, the African-American candidate, came in second. Among Hispanics, there was a high degree of bullet voting, especially for Cruz — that is, Hispanic voters cast less than their eight allowable ballots, focusing their vote on their preferred candidate or candidates.

By contrast, non-Hispanic white voters ranked Cruz fifteenth and Gomez sixteenth in the 1983 general election. Little ranked fourteenth among non-Hispanics, placing him along with Cruz and Gomez at the bottom of the list. The extreme case analysis performed on election returns from precincts that were 90% or more non-Hispanic confirmed these results, placing Cruz, Gomez and Little at the bottom of the list for non-Hispanic white voters.

In this election, Hispanic voters exhibited a clear pattern of cohesion behind minority candidates; whites voted en masse against them. Turnout, however, was low, with 301 votes for Cruz and 230 for Gomez out of a total city-wide Hispanic voting age population of 2835.

2. 1985 Elections

No Hispanics ran at-large for either the School Committee or Board of Aldermen seats in 1985. However, Betty Medina Lichtenstein, an Hispanic, ran against Elaine Pluta, a non-Hispanic, to win the School Committee seat in Ward 2. In this historic election, Lichtenstein became the first Hispanic elected to political office anywhere in Massachusetts.

The precinct returns in this election showed a high level of minority cohesion behind Lichtenstein. In precinct 2A, with a 10% Hispanic voting age population, Lichtenstein ran second behind Pluta, receiving 30% of the votes cast. In the predominantly Hispanic precincts, 2B (80% Hispanic) and 2C (77% Hispanic), Lichtenstein garnered 80% and 66% of the vote respectively. The total number of votes Lichtenstein received was 563 compared to 459 for Pluta.

Expert and lay witnesses from both parties applauded Lichtenstein’s campaign and, in essence, agreed that Hispanic political cohesion resulted in her election. At the same time, it is evident that there was significant white bloc voting. Given the composition of the ward, the white bloc vote was unable to defeat Lichtenstein.

Hispanic voter preference for an Hispanic candidate was not automatic, however. In the same election, also in Ward 2, another Hispanic candidate, Cruz, was soundly defeated by a non-Hispanic in the contest for alderman. There was no minority cohesion with respect to Cruz’s candidacy.

*922 Hispanic voter turnout in 1985 was 22.8% of voting age population, the highest of any year studied, which undoubtedly helped carry Lichtenstein into office. These results came on the heels of a successful community-based voter registration drive in Wards 1 and 2 carried out by deputized volunteer registrars affiliated with the Holyoke Rainbow Coalition, a multi-racial community organization inspired by Jesse Jackson’s presidential bid.

3. 1987 Elections

In 1987, an Hispanic, Orlando Isaza, ran for Alderman at-large, coming in first among Hispanic voters. If every single Hispanic voter had cast a ballot for him as one his or her eight votes, Isaza could have expected to garner one out of eight of the Hispanic votes cast. His actual total of 48.5% of all Hispanic votes cast in that election demonstrated, again, a very high level of bullet voting by Hispanics for an Hispanic candidate. Nevertheless he received only 4.7% of non-Hispanic white votes, ranking eleventh in the field of fourteen candidates and fading to gain a seat on the Board. Citywide Hispanic turnout for Isaza was relatively high, at 19.3%.

Extreme case analysis confirmed a definite pattern of white bloc voting in this election. The 90%-plus non-Hispanic precincts ranked Isaza eleventh, giving him only 15% of the vote. By comparison, in these precincts, the eight candidates who won their elections received between 25% and 30.4% of the votes cast.

By all accounts Isaza’s was the most successful city-wide campaign ever run by an Hispanic in Holyoke. He received the endorsement of two major newspapers and public support from a large number of influential non-Hispanic community leaders. Despite excellent credentials and serious outreach efforts, Isaza’s campaign was unable to translate this support into a sufficient number of votes from non-Hispanic white voters in Holyoke to win the election.

Juan Cruz lost the 1987 race for the Ward 2 City Council election. The parties’ expert witnesses basically agreed that the statistical analysis of the election results was not strongly suggestive of Hispanic cohesion behind Cruz. Testimony indicated that organized campaigning by Hispanics that year was centered on Isaza’s bid for an at-large seat on the City Council. Lack of Hispanic voter cohesion, not white bloc voting, appears to have caused Cruz to lose his bid for City Council.

The heated mayoral campaign between Marty Dunn and Ernest Proulx was prominent throughout Holyoke in 1987. Subtle racial appeals permeated the campaign literature of both candidates, reflecting what the plaintiffs aptly characterize as a “we and they” approach to politics in the City. The racial subtext of the mayoral campaign may have affected other electoral contests that year.

Dunn’s campaign literature featured the slogan “It takes guts,” coupled with a teach the “Spanish” English theme as an answer to increasing crime and vandalism. One ad featured a large picture of an Hispanic young man, cigarette dangling from his lips and the caption “The people who really should read this, can’t.” Ex. 54.

A second ad presented a picture of a building with graffiti spray-painted on its wall. In large letters it stated “Mrs. Richardson shouldn’t have been the one who had to move.” In smaller letters the text explains that the Richardson business “was forced to leave the neighborhood.” Dunn proclaimed that he would not “stand still while vandalism and crime drive our people and our businesses out.” Other Dunn ads contained similar sentiment, expressing concern that “our” residents were being mistreated. The “us” was fairly clearly the longtime white residential community, the “them” the more recent Hispanic minority.

Proulx, for his part, attacked Dunn for not calling for a moratorium on all subsidized housing programs in Holyoke. Proulx explained that he supported such a moratorium with one important exception — subsidized elderly housing. The vast majority of government subsidized elderly housing in Holyoke was occupied by white non-Hispanic senior citizens.

*923 4. 1989 Elections

In 1989 no Hispanic candidate ran at-large for either the City Council or the School Committee. Hispanic candidates did, however, run for the Board of Alderman in Ward 2 and School Committee in Wards 1, 2 and 7. In these, the evidence of voter cohesion was mixed.

In the Ward 2 race for Alderman, the Hispanic candidate, Lopez, was defeated by Pluta, a non-Hispanic who had lost to Lichtenstein in the 1985 Ward 2 School Committee race. Lopez lost in precinct 2A, but received 38% of the vote — extremely close to the Hispanic voting age population of the precinct. He carried precinct 2B, but with a margin of victory that was too low to elect, receiving only 52% of the vote in this heavily Hispanic area. Clearly, Pluta was supported by a significant number of Hispanic voters. The total vote for Pluta was 510 votes compared to 471 votes for Lopez. A clear pattern of racial bloc voting cannot be inferred from these results.

The Ward 2 School Committee seat was again taken by Lichtenstein, an Hispanic, who ran unopposed. However, the number of blank ballots cast for this seat exceeded the number of votes Lichtenstein received. Apparently, non-Hispanic whites in this ward expressed their dissatisfaction with Lichtenstein by casting blank ballots. The analysis of plaintiffs expert confirmed that Hispanic voters backed her overwhelmingly.

In this same year, two other Hispanic candidates ran for ward seats on the School Committee. Lillian Santiago-Gareia, a newcomer to Holyoke, was defeated by an established non-Hispanic incumbent for the School Committee seat in Ward 1. In Ward 7, which contained a small Hispanic population, the Hispanic candidate, Carlos Vega, was knocked out of the School Committee race in the primaries by two white candidates. No significant degree of minority voter cohesion in the School Committee races in Wards 1 and 7 can be discerned.

5. 1991 Elections

In district elections this year Lopez defeated a non-Hispanic candidate for the Ward 2 aldermanic seat. The vote totals in this race closely correlated with the ethnic composition of the precinct. Lopez received 53.2% of the vote in Precinct 2A (40.9% Hispanic) and 75.6% of the vote in 2B (74.9% Hispanic). Lichtenstein again ran unopposed for School Committee gamering 537 votes, only slightly above the 516 blank ballots east. Hispanic voters obviously cohered solidly behind Lopez and Lichtenstein.

Also this year, Hispanic candidate Rodriguez-Ortiz ran at-large for City Council and was the second choice among Hispanic voters. A non-Hispanic candidate, Elaine Pluta, ranked first among Hispanics. Pluta received 53.2% and Rodriguez-Ortiz 33.9% of the votes cast by Hispanics, again indicating a high degree of bullet voting. Only one other candidate received a statistically recordable number of Hispanic votes. Pluta was ranked fifth among non-Hispanic voters and was elected to office. Rodriguez-Ortiz lost.

This at-large election reflects the somewhat complex quality of the evidence in this ease. Pluta, a non-Hispanic, was most favored among Hispanics, and won with solid support from non-Hispanics as well. At the same time, as plaintiffs point out, Pluta would have won even if Hispanics had completely withdrawn their support, simply based on her non-Hispanic votes. Pluta’s election suggests that, in at-large City Council elections at least, Hispanic voters can sometimes elect the candidate of choice, provided he or she is not Hispanic. Rodriguez-Ortiz, though he ranked high among the Hispanics who did vote, did not inspire a substantial turn-out and, apparently because of his conservative political stance, was not favored by many Hispanics, including plaintiffs’ witnesses Santiago-Gareia and Isaza, who both said they did not support him.

6.1991 Ballot Referenda

In August and November of 1991, referen-da questions were on the Holyoke ballot to determine the level of property taxes to be allocated for the schools, police and fire departments. The results of these referenda are further indicia of the distinct political *924 preferences of Hispanic and non-Hispanic voters.

Hispanic voters were solidly in favor of additional funding for schools, police and fire departments. By contrast, non-Hispanic white voters rejected supplemental funds for the schools, despite the fiscal crisis, but favored more taxes for the police and fire departments. In these same referenda, non-Hispanic white voters also approved additional funds for trash collection, the Council on Aging and the local war memorial.

7. 199S Elections

In 1993, Rodriguez-Ortiz again ran for an at-large City Council seat. He was the only Hispanic in a field of twelve. Although he ranked first among the Hispanic voters who did go to the polls, the extremely low turnout, only 2% of the Hispanic voting age population, led experts for both plaintiffs and defendants to disregard these election results entirely.

In Ward 2 that year Hispanic candidates ran unopposed for City Council and School Board. Again, in each race a large number of blank ballots were cast, roughly proportional to the percentage of white voters in each precinct. In Ward 1, an Hispanic candidate lost the primary race for School Committee finishing third behind two non-Hispanic white candidates.

8. Turnout and Minority Cohesion

Defendants’ expert, Dr. Ronald Weber, examined many of the same election data summarized above and compiled additional statistics for overall Hispanic and non-Hispanic white turnout for all elections for 1983-93, regardless of whether an Hispanic candidate ran. He opined that the low Hispanic voter turnout in most elections, and the small number of precincts with a majority Hispanic voting age population, rendered any conclusion as to Hispanic voter cohesion unreliable. The court has rejected this opinion; the far more credible evidence is that there is a significant degree of minority voter cohesion in Holyoke.

Nevertheless, the question of turnout is serious and should be addressed. The issue has troubled judges reviewing decisions in voting rights eases. In Jones v. City of Lubbock, 730 F.2d 233 (5th Cir.1984), the court discounted a plaintiffs expert opinion based upon 261 votes out of a total minority population of 4,735, roughly 5.5%, calling the expert’s conclusion based on such data a “great leap” that “undermines the credibility of his study.” Id., at 236 (P. Higginbotham, J., concurring). In Gomez v. City of Watsonville, 863 F.2d 1407 (9th Cir.1988), cert. denied, 489 U.S. 1080, 109 S.Ct. 1534, 103 L.Ed.2d 839 (1989), however, the Court of Appeals reversed a district court’s conclusion that low minority voter turnout precluded a finding of cohesion, where it was clear that the Hispanic voters who did go to the polls consistently lined up behind Hispanic candidates.

The district court erred by focusing on low minority voter registration and turnout as evidence that the minority community was not politically cohesive. The court should have looked only to actual voting patterns rather than speculating as to the reasons why many Hispanics were apathetic.

Id., at 1416 (emphasis in original).

The Voting Rights Act presumes lower voter turnout and diminished political participation when depressed socio-economic conditions and lower levels of education handicap the minority population. Thornburg v. Gingles, 478 U.S. at 69, 106 S.Ct. at 2776. Nationwide, Hispanic voter registration and turnout rates lag behind both the general population and the African American population. Rodolfo O. de la Garza and Louis DeSipio, Symposium: Regulating the Electoral Process: Save the Baby, Change the Bathwater, and Scrub the Tub: Latino Electoral Participation After Seventeen Years of Voting Rights Act Coverage, 71 Tex.L.Rev. 1479, 1499 (1993). De la Garza and DeSipio cite social research on 1988 Census data which indicates that Hispanic voter registration rates are almost 50% below that of the general population. Id. (summarizing U.S. Dept. of Commerce 1988 Census Election Report, Series p-20, No. 440 (1989)). Hispanic voting rates are even lower, standing at 29% among Hispanics as compared to 57.4% for the general population. Id. at 1499. *925 Across the country, the relative youth, poverty and lower educational levels that characterize the Hispanic population interact to depress electoral participation. All of these factors are at work in Holyoke.

Aside from these well recognized economic and social factors, specific obstacles in Ho-lyoke impeded the Hispanic vote. After a burst of voter registration in Hispanic areas in 1985, the use of special deputy registrars was cut back by city officials. From 1988-93 the problem of recruiting sufficient bilingual poll workers was addressed with studied ineptness by the city, leading to many polling stations in Hispanic areas having no Spanish-speaking workers. In the most recent election the difficulty was finally eliminated by the simple expedient of advertising in the newspaper for bi-lingual workers. Moreover, before April of 1993 the city had no Spanish-speaking clerk to handle voter registration at the City Clerk’s office. During the same time, Hispanies who failed to receive, fill out and return the census forms, which are entirely in English, were removed from the voting lists. Prior to 1984, the forms were not even accompanied by Spanish instructions.

The July 1991 referendum presented a good example of the hurdles confronting Hispanic voters. With nine referenda items on the ballot, city officials viewed the text as too long to print on the ballot in English and Spanish. The ballots and polling place informational postings were therefore printed only in English, with a separate printed handout translating the referenda questions into Spanish. However, only 200 copies of the handout were printed, and the city prepared no advance publicity on the use of the handout. In at least two Hispanic precincts, some of the bi-lingual poll workers failed to appear. As the crowning indignity, the public postings of the ballot questions, in English only, carried a warning against defacement— in Spanish but not English. See, Ex. 57.

There may be voting rights cases where extremely low voter turnout renders conclusions about minority voter cohesiveness impossible, but this is not such a case. See e.g., NAACP v. City of Columbia, S.C., 850 F.Supp. 404 (D.S.C.1993). Hispanic turnout rates in Holyoke have varied from 22% to as low as 2% over a ten-year period, obviously differing considerably from election to election and from precinct to precinct. Even ignoring the social and economic conditions faced by Hispanies nationally, and disregarding the specific obstacles in Holyoke, this sample of voting patterns is more than sufficient to anchor conclusions with regard to cohesion -with a reasonable degree of certainty. Of course, viewed in context, the evidence becomes even stronger.

IV. CONCLUSIONS OF LAW

A. The Cingles criteria.

The evidence presented by plaintiffs is sufficient to satisfy all three of the Gingles threshold criteria: (1) numbers and geographical compactness, (2) minority voter cohesion and (3) majority bloc voting.

The first is not vigorously disputed. Plaintiffs need only show that the minority voting age population is sufficient to comprise a majority in at least one single-member district in the electoral unit in dispute. Here, it is conceded that Hispanies constitute a majority in both Wards 1 and 2. It may be questionable how significant this criterion is in a case involving a hybrid system where only the at-large component is challenged, but the issue in any event is academic.

As already noted, the defendants hotly dispute the existence of minority voter cohesion in Holyoke’s elections, basing this position primarily on low Hispanic voter turnout and partly on some inconsistencies in the voting patterns, such as the preference for Pluta over Rodriguez-Ortiz. Given the evidence in this case, however, a factfinder would have to willfully ignore the perfectly plain reality, to reach any other conclusion but that, overwhelmingly, Hispanic voters in Holyoke have substantially cohered behind a small number of preferred candidates. With one exception, all these candidates have themselves • been Hispanic. Larger voter turnout or some other circumstance might have made the picture even clearer, and cases may be imagined where the evidence of cohesion is even stronger, but here it is certainly enough.

*926 In view of the number of elections analyzed, it would be ridiculous to suggest that perhaps the Hispanies who cast their ballots were not typical of the Hispanic voting age population as a whole — that is, to speculate that at some hypothetical election where one hundred percent of the voters, minority and majority, went to the polls, it would emerge, mirabile dictu, that Holyoke’s Hispanies east their ballots in much the same pattern as the non-Hispanic white voters. This outcome is theoretically possible, but it defies common sense and the available evidence.

Similarly, it is incorrect to suggest that once the number of voters supporting a particular candidate falls below a certain level, no cohesion can be found, even if all the voters turning out do support that candidate. For one thing, in the Isaza and Lichtenstein campaigns, Hispanic voters did go to the polls in significant numbers. For another, as Gomez v. City of Watsonville, 863 F.2d 1407 (9th Cir.1988), cert. denied, 489 U.S. 1080, 109 S.Ct. 1534, 103 L.Ed.2d 839 (1989), points out, it is usually not proper to decline to find cohesion merely on the basis of low numbers. Id. at 1416. See also, Whitfield v. Democratic Party of Arkansas, 890 F.2d 1423, 1431 (8th Cir.1989), cert. denied, 498 U.S. 1126, 111 S.Ct. 1089, 112 L.Ed.2d 1193 (1991) (district court improperly assumed that lack of motivation caused low voter turnout — focus should be only on actual voting patterns); United States v. Dallas County Com., 739 F.2d 1529 (11th Cir.1984) (existence of apathy not a matter of judicial notice). To repeat, while perhaps the evidence is not spectacular, it is more than sufficient to show legally significant minority voter cohesion.

Defendants argue that Hispanic lack of success in at-large elections can be explained by factors other than the at-large system itself, including voter apathy, unattractive candidates, poor campaign organizations and the like — all evidenced by low voter turnout. Even if this argument were true, it would probably not undermine a finding of minority voter cohesion. More importantly, it is not true.

Polities is the most human of pursuits and, like all humans, the Hispanic voters and candidates in the elections described in this memorandum were not perfect. As always, more could have been done, and it could have been done better. But these imperfections do not explain the consistent lack of success of Hispanic candidates in the City Council at-large elections in Holyoke. While the School Committee might be different, at least for the City Council it is difficult to imagine any Hispanic candidate, no matter how attractive and no matter how effective at bringing out the Hispanic vote, having a fair opportunity to win in any at-large election in Holyoke during this period. Having listened to the witnesses, it is the court’s opinion that few citizens of Holyoke, Hispanic or non-Hispanic, would dispute this proposition — certainly none did at trial.

This lack of opportunity with respect to at-large City Council seats is explained by the third Gingles criterion, majority bloc voting, which is not seriously disputed by the defendants. Beyond a handful, non-Hispanic white voters in Holyoke have not voted for at-large Hispanic candidates, period. In any City Council at-large election, such as Isa-za’s, this segment of the electorate can and does simply wall out any successful Hispanic campaign effort. Given this reality, the dominant at-large component of the City Council structure works to dilute the vote of Hispanics and prevent them from participating equally in the political process. The existence of a majority of precincts in Holyoke with over 90% white non-Hispanic voters has permitted particularly accurate testing on this point. Through the so-called extreme case analysis, majority bloc voting has been confirmed in all the at-large City Council elections in which an Hispanic candidate has run. The statistical evidence also confirms white bloc voting in Wards 1 and 2. At trial, no alternative explanation of these statistics was seriously attempted.

B. Totality of the Circumstances.

The totality of circumstances test is flexible and fact intensive. There is no requirement that any particular number of factors lean in plaintiff’s favor in order to conclude that Hispanies do not have an equal opportunity to participate in the political process. In fact, some of the factors suggested *927 by the Senate Judiciary Committee’s report on the Voting Rights Act favor defendants in this case. But, when the factors are considered altogether it is manifest that local historical, social and economic conditions have interacted with the electoral system to substantially undermine Hispanic opportunity to participate in the local political process.

On the one hand, there is no evidence that Holyoke has ever had any official discrimination in voting. Its election system does not include the traditional features that bar or discourage minority participation. Indeed, in one of the city’s election districts, Ward 2, Hispanic candidates have consistently won in the past several elections both for School Committee and City Council. In another, Ward 1, Hispanics have a large voting age majority, more than adequate to elect a representative, of whatever ethnic background, sensitive to the needs of Hispanic citizens.

However, a host of factors cut the other way. As noted, voting in Holyoke has often been racially polarized in the sense that “there is a consistent relationship between the race of the voter and the way in which the voter votes_” Gingles, 478 U.S. at 53, n. 21, 106 S.Ct. at 2768, n. 21. In addition, the 1987 mayoral campaign at least was characterized by subtle, and not-so-subtle racial appeals. The Hispanic community does bear the effects of discrimination in such areas as housing, employment and health, and there has been a significant degree of unresponsiveness on the part of elected officials to the needs of the Hispanic community. As noted, no Hispanic has ever run for an at-large seat on the School Committee, but all those who have run at-large for the City Council have lost. While there has been no institutionalized discrimination in voting, a fair amount of bureaucratic interference and passivity has definitely hampered the Hispanic electorate.

The court has also considered two other circumstances specific to this case, stressed by the defendants. First, the relatively recent arrival to Holyoke of a substantial portion of the Hispanic community, including some of the candidates, is significant. In considering this, the court does not mean to suggest that new arrivals have fewer rights than long term residents. But building a political base does take time, and it is understandable that a well-established incumbent may find it easier to attract votes than a candidate, such as Santiago-Garcia, who only lived in the community a short time before seeking office. Second, the court has considered that Holyoke during the past decade has been experiencing a period of economic distress. In time this phase will pass, as similar periods have passed in the City’s history. But economic uncertainty can generate class or ethnic tensions completely independent of electoral systems.

The recent arrival of Hispanics does contribute to explaining why the Hispanic political community in Holyoke has, at least for part of the past decade, been at an embryonic stage. However, it must also be recognized that economic uncertainty in Holyoke has often affected Hispanic residents disproportionately and contributed to depressed minority participation in politics. In any ease, under the Act’s functional, result-oriented test for minority vote dilution, the factors emphasized by the defendants do not offset the strong evidence offered by plaintiffs, at least with regard to the City Council elections.

Finally, the court has, as De Grandy requires, considered the question of proportionality. Hispanics in Holyoke consistently elect Hispanic district representatives from Ward 2; they have the power to elect a candidate of their choice from Ward 1. Comprising slightly more than 20% of the voting age population of Holyoke, Hispanics have the power to elect that proportion of the School Committee. The calculus is different for the City Council, for two reasons. First, the Hispanic opportunity to elect two out of fifteen members is not proportional; second, the majority of the council is permanently locked away from Hispanics through the at-large system. Given the plaintiffs satisfaction of the Gingles criteria and the weight of evidence favoring plaintiffs with regard to the totality of the circumstances, the conclusion that the at-large system for City Council elections violates the Voting Rights Act is inevitable.

*928 It is equally clear that the election system for the School Committee falls on the other side of the line for several reasons.

First, the at-large component to the School Committee election system affects only a small minority of the Committee. It is the tail and not the dog. As noted already, the existence of this component does not interfere with the ability of Holyoke’s Hispanics to elect School Committee members of their choosing in proportion to their percentage of the voting age population. While courts must carefully scrutinize at-large systems, these systems in themselves violate neither the Constitution nor the Voting Rights Act. Something can be said for maintaining a modest number of seats for persons who must be responsible to the community as a whole. A conclusion that this relatively minor aspect of the system is illegal would amount almost to a rejection of at-large systems per se, something the Supreme Court has declined to do.

Second, unlike the City Council, no Hispanic has ever tried to seek election at-large to the School Committee. While the plaintiffs’ evidence is sufficient to justify a finding in their favor in City Council elections, where real races give the court concrete data on the issues and the candidates’ reception, the evidence with respect to the School Committee fades as a blind inference. In fact, there is a good' argument that an Hispanic candidacy for School Committee, given the majority Hispanic composition of the student body, might well stand on somewhat stronger legs than a candidacy for City Council.

Third, even successful lawsuits win by different margins. Here, while the plaintiffs have prevailed with regard to the City Council, their evidence cannot be said to be overwhelming. The social and political landscape in Holyoke contains more shades of grey than either black or white. In Gingles itself, Justice Brennan distinguished between different districts, finding a violation in some but not all. In voting rights cases, more than most cases, courts should forbear acting except upon clear evidence. Here the evidence balances with such delicacy that the scales tip one way for the City Council, the other for the School Committee.

Finally, in Holder v. Hall, — U.S.-, 114 S.Ct. 2581, 129 L.Ed.2d 687 (1994), the Supreme Court advised courts handling voting rights cases to consider a benchmark, a standard against which an allegedly unfair system may be compared. Id. at-, 114 S.Ct. at 2585 (citing Justice O’Connor’s concurrence in Gingles, 478 U.S. at 88, 106 S.Ct. at 2785). The School Committee election system provides one possible benchmark against which the City Council’s system can be compared. This 120-year-old election mechanism permits a healthy degree of at-large electioneering, while insuring a fair opportunity for an existing minority at a particular time to participate in the political process. In the totality of the circumstances, the court cannot say that the School Committee election system violates the Voting Rights Act.

V. CONCLUSION

For the foregoing reasons, the court hereby finds in favor of the plaintiffs with regard to the City Council election system, and in favor of the defendants with regard to the School Committee election system. Before ordering the clerk to enter a final judgment, however, the court will receive submissions and hear argument from counsel regarding an appropriate remedy with regard to the City Council elections. A separate order will issue.

Notes

. The lawsuit originally also challenged the configuration of the districts used to elect a portion of the School Committee and City Council, but this claim has been resolved.

. Non-Hispanic Blacks constituted 2.62% of the population and Asians .72% of Holyoke's citizenry-

. Plaintiffs claim that this lawsuit triggered the new plan, which was adopted to bring the City's wards into compliance with the one-person one-vote requirements of the Fourteenth Amendment. Defendants vigorously deny that this lawsuit had any influence in the City's decision.