UNITED STATES DISTRICT COURT EASTERN DISTRICT OF NEW YORK
August 21, 1984
BEVERLY MORRIS, JOY CLARKE HOLMES, JOANNE OPLUSTIL, Plaintiffs, against THE BOARD OF ESTIMATE, THE CITY OF NEW YORK, EDWARD I. KOCH, individually and as Mayor of New York, CAROL BELLAMY, individually and as City Council President, HARRISON J. GOLDIN, individually and as Comptroller for the City of New York, HOWARD GOLDEN, ANDREW STEIN, STANLEY SIMON, DONALD MANES, ANTHONY GAETA, each individually and as Borough Presidents of the boroughs of the City of New York, Defendants, and FRANK V. PONTERIO, Intervenor-Defendant, ROBERT A. STRANIERE, individually and as a member of the New York State Assembly, Intervenor-Defendant.
The opinion of the court was delivered by: NEAHER
MEMORANDUM AND ORDER
NEARHER, District Judge.
In December, 1981, plaintiffs filed suit charging that the New York City Board of Estimate ("Board") deprived them of equal protection under the fourteenth amendment to the United States Constitution. More precisely, as Brooklyn residents, plaintiffs maintained that sections 61 and 62 of the New York City Charter, outlining the Board's legislative plan, devalued their ballots by violating the one person, one vote rule. See 1 New York City Charter & Ad. Code Ann. §§ 61-62 (Williams Press 1976 & Supp. 1982-83).
In December, 1982, however, the court held that the Board was not subject to that constitutional principle. The court, as a consequence, granted summary judgment to defendants -- the Board and its members ("city defendants") -- as well as to intervenor-defendant Frank V. Ponterio, a Staten Island resident. Morris v. Board of Estimate, 551 F. Supp. 652, 657 (E.D.N.Y. 1982).
In May 1983, the Second Circuit reversed.
"[W]e conclude tha the Board . . . is selected by popular election and performs general governmental functions. The principle of one person, one vote is therefore applicable."
Morris v. Board of Estimate, 707 F.2d 686, 690 (2d Cir. 1983). The Second Circuit, furthermore, remanded with instructions:
"[T]he district court must determine the degree of malapportionment present (after deciding on the appropriate methodology for doing so) and rule on the policies and interests which the Supreme Court has held may justify deviations from the literal one person, one vote formula."
Id. (footnote omitted).
On return, plaintiffs moved for summary judgment; but, after conferring with the court, the parties stipulated that the proceedings should be bifurcated. In the first stage, the most suitable mathematical measure would be identified and superimposed over the Board's electoral scheme to ascertain the malapportionment. If more than a minor deviation were disclosed, the Supreme Court approved policies and interests served by the Board would be discerned in the second stage to settle the variance's acceptability.
In accordance with that agreed course, the court has reviewed the submissions, the record and the law. Based on that examination, the court holds, as to the first stage issues, that the Supreme Court's methodology in Abate v. Mundt, 403 U.S. 182, 184 & n.1, 29 L. Ed. 2d 399, 91 S. Ct. 1904 (1971), is the appropriate standard and, when applied, reveals a 132.9% maximum deviation. The reasons impelling these holdings will be set forth after a brief relation of the limited facts pertinent to this suit's present posture.
The court and the Second Circuit have detailed the uncontested facts in the prior opinions. To forego needless repetition, reference is made to those previous treatments. Morris v. Board of Estimate, 551 F. Supp. at 653-55, rev'd, 707 F.2d at 687-88. Accordingly, only those facts particularly germane to the instant issues will be recounted.
The Board is one of New York's governing bodies and consists of eight members, who -- importantly -- area mixture of at-large (city-wide) and district (borough) representatives. As delineated in sections 61 and 62, the Mayor, Comptroller and City Council President possessing two Board votes each, are selected by the metropolitan electorate as a whole. The Borough Presidents of Brooklyn, Queens, Manhattan, Bronx and State Island (Richmond), having one Board vote each, are chosen by their respective borough constituencies.
As manifest from the 1980 federal census, the boroughs range appreciably in population.
Manhattan 1,427,53 3
Staten Island 352,121
Yet, notwithstanding that sweep were Brooklyn as the largest has more than six times the population of Staten Island as the smallest, each borough has but one assigned Board vote. That difference in population but equivalence in franchise is the primary basis for plaintiffs' charge that, as Brooklyn residents, the Board's legislative design violates their right to equal protection.
A. The Abate Test
Broadly grouped, reapportionment disputes fall into two categories: quantitative and qualitative.
"In [a strictly quantitative] case, there are a number of coordinate districts . . . and voters in larger districts allege that their votes are devalued in comparison to those of voters in smaller ones. The issue in a typical reapportionment case, therefore, is whether population deviations from the average district are impermissbly large. . . . The comparison is one based purely on population figures; no showing of discrimination along racial, ethnic, or political lines need be shown."
Nevett v. Sides, 571 F.2d 209, 215 (5th Cir. 1978), cert. denied, 446 U.S. 951, 64 L. Ed. 2d 807, 100 S. Ct. 2916 (1980) (emphasis added).
A distinctly qualitative suit, in contrast, transcends straightforward computations.
"[T]he theory of the qualitative reapportionment action is more subtle and abstruse. This action alleges that district lines were drawn or erased so as to diminish the political input of a cognizable element of the voting population."
Corder v. Kirksey, 585 F.2d 708, 710 n.2 (5th Cir. 1978), cert. denied, 460 U.S. 1013, 103 S. Ct. 1253, 75 L. Ed. 2d 482 (1983). See also The Supreme Court 1982 Term - Elections, 97 Harv. L. Rev. 135, 139 (1983) ("Elections") ("[F]air apportionment has two dimensions, one "quantitative" and the other "qualitative". The quantitative dimension -- captured by the one-person-one-vote standard -- requires that each individual's vote be of equal weight, and is compromised when districts are of unequal population. Mere numerical equality among voting districts, however, "does not guarantee "substantive equality in the sharing of power." The qualitative dimension, therefore, requires that individuals be able to act in effective voices in representation.") (emphasis in original) (footnotes to citations omitted).
Here, though, plaintiffs have limited their attack to the quantitative aspect by intentionally omitting qualitative contentions. See, e.g., Pl. Br. at 28 n.65
("[Plaintiffs"] argument is not, at this time, . . . a r acial discrimination claim under the Equal Protection Clause."). Necessarily then, as a quantitative action, this suit's outcome is governed mainly by the Board's variance from the one person, one vote principle. See, e.g., Andrews v. Koch, 528 F. Supp. 246, 252 (E.D.N.Y. 1981), aff'd, 688 F.2d 815 (2d Cir. 1982), aff'd sub nom. Giacobbe v. Andrews, 459 U.S. 801, 103 S. Ct. 32, 74 L. Ed. 2d 46 (1982) ("The issue . . . does not implicate questions of political, racial or ethnic discrimination. Simply stated, the question is whether the allocation of an equal number of at-large Council seats to each of the five boroughs, without regard to substantial population disparities, comports with the constitutional rule of substantial numerical equality. . . .").
While perhaps dispositive in a quantitative suit such as this, deviation from population equality is "fundamental in reapportionment litigation" in general. League of Women Voters of Nassau County v. Nassau County Board of Supervisors, 737 F.2d 155, 169 (2d Cir. 1984). As the Supreme Court has pronounced:
"Population is . . . the starting point for consideration and the controlling criterion for judgment in legislative apportionment constroversies."
Reynolds v. Sims, 377 U.S. 533, 567, 12 L. Ed. 2d 506, 84 S. Ct. 1362 (1964) (quoted in League of Women Voters of Nassau County v. Nassau County Board of Supervisors, 737 F.2d 155, 169 (2d Cir. 1984) (emphasis added and footnote omitted by the Second Circuit). See Abate v. Mundt, 403 U.s. at 185 ("[E]lectoral apportionment must be based on the general principle of population equality and . . . this . . . applies to state and local elections."). See also Note, The Reapportionment Dilemma: Lessons from the Virginia Experience, 68 Va. L. Rev. 541, 569 (1982) ("Reapportionment Dilemma") ("Population equality remains the single most important factor in judicial evaluation of apportionment schemes.").
After that elemental investigation of the deviation, any subsequent steps depend on the variation exposed. In Brown v. Thomson, 462 U.S. 835, , 103. S. Ct. 2690, 2696, 77 L. Ed. 2d 214 (1983), the Supreme Court described that disparity's contingent effect on a more searching probe.
"[W]e have held that "minor deviations from mathematical equality among state legislative districts are insufficient to make out a prima facie case of invidious discrimination . . . so as to require justification by the State." [Citation omitted.] Our decisions have established . . . that an apportionment plan with a maximum population deviation under 10% falls within this category of minor deviations. . . . A plan with larger disparities . . . creates a prima facie cse of discrimination and therefore must be justified by the State."
Discovering where the Board's legislative plan sets within the Brown variational framework begins with selecting a methodology. Fortunately, the Supreme Court has eased that choice by its demonstrated impatience with mathematical metaphysics.
"These are the figures found by the District Court. Appellee . . . argues [for] another method of computation. . . . The State and the city . . . disputed that the deviation . . . was as much as claimed. . . . We decline to enter this imbroglio ofmathematical manipulation and confine our consideration to the figures actually found by the court. . . ."
Mahan v. Howell, 410 U.S. 315, 319 n.6, 35 L. Ed. 2d 320, 93 S. Ct. 979 (1973) (emphasis added). See Boyer v. Gardner, 540 F. Supp. 624, 628 (D.N.H. 1982) ("[Mahan ] indicate[s] a reluctance to involve the courts in statistical complexities, at least absent reasons to indicate that a more complicated approach is required. . . .").
Consistent with that perceived disinclination to entertain esoteric, result-oriented equations, the Supreme Court has approved a simple methodology -- the Abate test -- names for the case in which it was sanctioned.
"An accepted measure . . . is that of maximum percentage deviation from the population mean.Such a formula was approved by the Supreme Court in Abate v. Mundt, 403 U.S. 182, 29 L. Ed. 2d 399, 91 S. Ct. 1904 (1971). The maximum percentage deviation is determined by adding the percentage deviation above the population mean of the district with the greatest number of voters to the percentage deviation below the population mean of the district with the fewest number of voters."
Flateau v. Anderson, 537 F. Supp. 257, 261 n.7 (S.D.N.Y. 1982). See Brown v. Thompson, U.S. at , 103 S. Ct. at 2701 ("We have come to establish a rough threshold of 10% maximum deviation . . . (adding together the deviations from average district size of the most underrepresenteed and most over-represented districts). . . .") (Brennan, J., dissenting). See also Andrews v. Koch, 528 F. Supp. at 250 ("This is the formula devised by the Supreme Court to measure the extent to which population disparities among legislative districts either conform or fail to conform to the constitutional requirement of substantial population equality."). See generally Guido, Deviations and Justifications: Standards and Remedies in Challenges to Reapportionment Plans, 14 Urb. Law. 57, 58 (1982) ("Guido") ("Variances are most commonly expressed as a percentage range, which is the sum of the percentage by which the most populous district exceeds the mathematical ideal and the percentage by which the least populous district falls short of that ideal.").
Besides being simple, the Abate test is also adaptable to disparate representational designs. To illustrate, in Abate itself, the multi-member voting plan of New York State's Rockland County was studied. Abate v. Mundt, 403 U.S. at 184. In White v. Weiser, 412 U.S. at 785, congressional districting in Texas was gauged with that standard.
Assuredly, the Abate test has calibrated apportionment forms more intricate than the Board's. In Andrews v Koch, 528 F. Supp. at 250, this court modified that measure to isolate the disparity of New York City Council, which presented a blend of single-member districts, floterial-member districts, at-large member boroughs and an at-large member city.
In that same respect, the New York Court of Appeals in Franklin v. Krause, 344 N.Y.S.2d at 887 used the Abate test to review the Nassau County Board of Supervisors' weighted voting plan. See also League of Women Voters of Nassau County v. Nassau County Board of Supervisors, 737 F.2d 155, slip op. at 3891 (2d Cir. 1984) (surveying with the Abate test that same legislative model after a 1980 census-induced reapportioning).
In brief, the Board's mixed structure of five borough members with one vote each and three at-large members with two votes each certainly does not preclude employing the Abate test, given its proven versatility.
B. Defendants' Methodologies
Despite the settled Abate test's simplicity and adaptability, defendants contend that it is unserviceable here due to the Board's three at-large members. The Second Circuit, also, presumed in dicta that their presence posed unique questions.
"[T]he fact that three members . . . are elected at large . . . raises novel issues as to how an individual voter's opportunity to elect these members should be assessed in conjunction with his opportunity to elect his borough representative."
Morris v. Board of Estimate, 707 F.2d at 690 n.3.
Actually, the Board's essential electoral design is unexceptional and does not present first-impression measurement problems. Indeed, the Board's at-large and borough amalgam compares to other governing bodies exercising legislative functions, whose apportionments have been assessed utilizing the Abate test. See, e.g., Latino Political Action Committee, Inc. v. City of Boston, 568 F. Supp. 1012, 1015 (D. Mass. 1983), stay denied, 716 F.2d 68 (1st Cir. 1983), stay denied, 463 U.S. 1319, 104 S. Ct. 5, 77 L. Ed. 2d 1421 (Brennan, Circuit Justice, 1983) (City Council and School Committee plans of Boston, Massachusetts, with each having four at-large and nine district members); Cohen v. Maloney, 410 F. Supp. 1147, 1149 & n.2, 1150 (D. Del. 1976) (City Council of Wilmington, Delaware, with four at-large and five district members). See generally Markoe v. Legislature of the Virgin Islands. 592 F.2d 188, 190-91 & n.2 (3rd Cir. 1979) (Senate of Virgin Islands with one at-large -- a St. John's Island resident -- and fourteen members even distributed between two districts); Perry v. City of Opelousas, 515 F.2d 639, 641 & n.2 (5th Cir. 1975) (Board of Aldermen of Opelousas, Louisiana, with one at-large and five district members); Martin v. Venables, 401 F. Supp. 611, 613, 614 & n.2 (D. Conn. 1975) (Town Council of Stratford, Connecticut, with one at-large and ten district members).
Like those polities, the Board's basis structure is a composite pattern.And as those polities have been, the Board also can be gauged by the Abate test -- irrespective of the at-large representation.
In short, sufficient decisional precedent exists to avert -- as unwarranted -- an unchartered departure from the Supreme Court's approved standard. See id. ("[T]he Supreme Court has indicated that total maximum deviatioin is the measure to be used and this Court is obliged to use it."). See also Dodge & McCauley, Reapportionment: A Survey of the Practicality of Voting Equality, 43 U. Pitt. L. Rev. 527, 536 (1982) ("[T]he . . . Supreme Court, the lower federal courts, and the various state courts have developed a relatively precise and simple mathematical model for the measurement of population eqaulity. The basic standard . . . is determined by the relationship between the population of the largest and smallest election district. . . .").
Concomitantly, the court rejects defendants' flawed tests. To particularize, city defendants would forego all mathematical analysis.
"[T]he Court's methodology . . . should be functional. . . . First, the Court should identify the constitutional objectives of the one person, one vote principle. Second, the Court should undertake a qualitative inquiry into the Board's scheme of representation."
City Def. Br. at 5.
That procedure is not the law. Abate aside, as discussed, population equality is the "starting point" and "controlling criterion." Reynolds v. Sims, 377 U.s. at 567. See Cohen v. Maloney, 410 F. Supp. at 1150 ("[S]tate and local governments have a duty . . . to make population the prime criterion in any apportionment. . . ."); Grisbaum v. Jefferson Parish Council, 336 F. Supp. 267, 271 (E.D. La. 1971) ("The . . . Supreme Court has consistently held that the controlling criterion . . . is that the Constitution permits no substantial variation from equal population. . . ."). Cf. Connor v. Finch, 431 U.S. 407, 409-10, 52 L. Ed. 2d 465, 97 S. Ct. 1828 (1977) ("[B]oth the Senate and the House reapportionments . . . fail to meet The most elemental requirement. . . -- that legislative districts be "as nearly of equal population as is practicable.") (quoting Reynolds v. Sims, 377 U.S. at 577) (emphasis added).
While city defendants might prefer that it not do so, the law does require that population equality be the reapportionment referrant and, consequently, deviations therefrom be marked off. Plainly stated, that inequality audit cannot be foregone.
Intervenor defendant Ponterio, on the other hand, does not neglect the variance's ascertainment but advances a specially tailored methodology.
"I have developed an analysis which both recognizes and the Board's overlay of city-wide and borough votes and examines every voting rule under which the Board functions."
September 30, 1983, Aff. of I. Def. Ponterio at 8.
That analysis is premised on equating the Board to a floterial district.
"The Board may be looked upon as a single floterial district, the super-district being the City and the sub-districts the five boroughs. Each Borough President is a sub-district representative and the city-wide members are the super-district representatives."
Id. at 9.
Therein lies the foundational fault.
By analogy or otherwise, the Board is not akin to a "single floterial district." In Davis v. Mann, 377 U.S. 678, 686 n.2, 12 L. Ed. 2d 609, 84 S. Ct. 1441 (1964), the Supreme Court defined that model.
"[F]loterial district" . . . refer[s] to a legislative district which includes . . . several separate distraicts . . . which independently would not be entitled to additionalrepresentation but whose conglomerate population entitles the entire area to another seat in the particular legislative body being apportioned.
See, e.g., Boyer v. Gardner, 540 F. Supp. at 626 ("When any town . . . [etc.] . . . has fewer than the number of inhabitants necessary to entitle it to one representative, the legislature is to create representatives [floterial] districts with enough inhabitants to entitle the district to one or more representatives.").
No definitional plasticity can liken the Board to an immense floterial district. The Board is an entire legislative model while a floterial district is but a component. See, e.g., Mahan v. Howell, 410 U.s. at 318-19 ("The statute apportioning the [Virginia House of Delegates] providced for a combination of 52 single-member, multimember, and floater delegate districts from which 100 delegates would be elected."); Boyer v. Gardner, 540 F. Supp. at 626 (A plan which reapportioned the New Hampshire House of Representatiaves "establish[ed] 162 electoral districts, of which 17 were so-called floterial districts.") (footnote omitted).
Additionally, Ponterio's matching the dissimilar leads him to questionable "accommodations".September 30, 1983, Aff. of I. Def. Ponterio at 10. Weighting each delegate's vote is one such problematical adjustment.
"Each city-wide member's vote is weighted by the population he represents . . . viz., the total city population of 7,071,030. Each Borough President's vote is weighted by his borough's population.
Id. at 11.
That accommodation distorts the city-wide members' voting power. Despite in reality having only a 6-5 ballot advantage, the at-large have a 3-1 Board dominance (by individual) under Ponterio's methodology.
Described another way, Ponterio augments the Board's city-wide total by the metropolitan population three times, whereas he enhances the Board's borough total by the metropolitan population just once (since the five boroughs' aggregate population equals one times the metropolitan population).
Increasing the city-wide members' strength to that extent demonstrably decreases the found deviation. To elaborate, since Ponterio computes the combined effect of the city-wide and borough representation, the bolstered at-large presence so overwhelms the borough delegation that the population inequalities among the boroughs are not even roughly translated into the ascertained disparity.
Comparing results illustrates the deviation's marked diminution and its unrelatedness to the unequal borough populations with Ponterio's approach. As noted, Brooklyn has more than six times Staten Island's population. When the disproportion between the two (as the most underrepresented and overrepresented boroughs) is calculated by either the Supreme Court's Abate test or a (suggested) modification thereof, the yield is 132.9% or 78%, respectively. See infra p. 31 and Pl. Br. at 32-33. Those figures vary substantially from one another but both are credibly high, reflecting the considerably unequal borough populations, which are at the base of each computation and at the heart of this lawsuit.
In contrast, when calculated by Ponterio's methodology, the deviation plummets to 16.4%. See September 30, 1983, Aff. of I. Def. Ponterio at 18 (Table IX).
That percentage has no plausible relationship to the indisputable inter-borough population differences. To the contrary, as partially a product of the weighting accommodation, which itself flows from the unsound floterial premise, that figure mutes the discrepancies among the boroughs.
But even apart from such flaws, the court would reject Ponterio's methodology as it would city defendants". Neither is needed. As seen, the Supreme Court sanctioned Abate test has assessed apportionments of comparable and of greater complexity. That being so, promulgation of a special formula is uncalled for in this case.
C. The Unmodified Abate Test
The remaining methodological question, then, is whether the Abate test must be modified to account for the at-large representation. Plaintiffs, as mentioned, do proffer one such adjusted measure similar to that used in Andrews v. Koch, 528 F. Supp. at 250. See Pl. Br. at 32-33. The court, nonetheless, determines conversion unnecessary.
At the time of Andrews, the City Council consisted of forty-six seats: one (the President) was city-wide; thirty-five were from districts and ten were from borooughs (i.e., two from each). Andrews v. Koch, 528 F. Supp. at 248.
In fact, three of those thirty-five districts were floterial.Id. at 250 n.6. Yet, despite the confluence of several representational types, the equal protection challenge was aimed solely at the disparity caused by the five borough delegate pairs.
Notwithstanding that focus, the court rejected the depiction of the five boroughs as multimember districts distinct from those councilmanic districts within the boroughs or shared by two boroughs. The court, instead, decided that the total electoral dynamic within each borough had to be surveyed in arriving at the real deviation. Id. at 250 n.7.
To accomplish that, the court modified the Abate test to encompass the full complement of each borough's councilpersons, viz., the two borough-wide delegates, the district delegates from within each borough and the floterial district delegates by proportional share. Id. at 250.
In this case, the variety of borough delegate layers is missing. Thus, while the one person, one vote challenge is also aimed at the borough representation, only one such delegate tier, in fact, sits on the Board -- the Borough Presidents. So, differing from Andrews which required a multidimensional gauge for measuring the combined representational voices from each borough, this case necessitates only a linear methodology. Put differently, each borough voter has one designated borough agent on the Board -- the Borough President. That straight line, one-to-one relationship easily lends itself to an unmodified Abate analysis.
Nor does the city-wide membership subvert the Abate test. Courts faced with mixed designs have simply excluded the at-large positions from the malapportionment mathematics. See, e.g., Latino Political Action Commmittee, Inc. v. City of Boston, 568 F. Supp. at 1015 (In figuring the 23.6% deviation from the Boston, Massachusetts' City Council and School Commmittee plans with thirteen members each, the court measured only the respective nine single-member districts, disregarding the respective four at-large seats.); Cohen v. Maloney, 410 F. Supp. at 1150 (In calculating the 18.3% disparity for Wilmington, Delaware's City Council, the court only assessed thte eight single-member districts, without considering the five at-large slots.); Perry v. City of Opelousas, 515 F.2d at 641 & n.2 (In determining the 6.2% disproportion for Opelousas, Louisiana's Board of Alderman, the court focused on the five single-member districts, without factoring in the one at-large member.); Martin v. Venables, 401 F. Supp. at 613, 614 & n.2 (In arriving at the 50.7% variance for Stratford, Connecticut's Town Council, the court audited the ten single-member districts, omitting the one at-large position.).
Significantly, when charges of one person, one vote violations have been brought in other suits involving these same boroughs, the purported impacts of the city-wide delegates have not been considered. In Oliver v. Board of Education, 306 F. Supp. 1286 (S.D.N.Y. 1969), a challenge rooted in the boroughs' unequal population dispersal was made to New York City's Board of Education, consisting of five electees from the boroughs and two at-large appointees selected by the Mayor.
In holding that the legislative scheme was constitutionally infirm, the court concentrated only on the inter-borough population discrepancies without taking into account any influence the mayoral at-large appointees might have had on the malapportionment.
"[F]ive members . . . are to be elected, one from each county. The other two are to be appointed by the mayor. It is clear beyond question that plaintiff's vote for the Kings County [Brooklyn] member will carry less weight than the votes of residents of other counties for the representatives of those counties."
Id. at 1289.
Similarly, as already noted, this court in Andrews v. Koch, 528 F. Supp. at 248, 250, did not include in the New York City Council's measurement the one actual at-large representative -- the President. Instead, the equal protection inquest was focused completely at the boroughs -- the locus of the inequality.
While the official's tie-breaker status constituted limited voting power, see 1 New York City Charter & Ad. Code Ann. § 23(d) (Williams Press 1976 & Supp. 1982-83), making any assumed effect on the boroughs' disparity probably de minimus and certainly difficult to accurately ascertain, the pertinence for this case is that no such measurement was attempted. The City Council President's presence relative to the malapportionment was disregarded.
See generally Avery v. Midland County, 390 U.S. 474, 476, 485, 20 L. Ed. 2d 45, 88 S. Ct. 1114 (1968) (In reaching its decision that the one person, one vote rule applied to local governments and, hence, to the Commissioners Court of Midland County, Texas, the Supreme Court marked the "vast imbalance" of population among the four single-member districts without discussing the one at-large representative's impact other than noting that the person "vote[d] only to break a tie.").
To be sure, in Oliver and Andrews as well as in the cases cited from other jurisdictions, whether to exclude the at-large representation was not an explicit issue. Nor should it have been. In an ideal plan, a voter regardless of location has as equal an opportunity to elect an at-large delegate (and thereby be heard) as any other voter.
See City of Mobile v. Bolden, 446 U.S. 55, 78, 64 L. Ed. 2d 47, 100 S. Ct. 1490 (1980) ("There can be . . . no claim that the "one person, one vote" principle has been violated . . . because . . . Mobile is a unitary electoral district and the Commission elections are conducted at large."). See generally Dusch v. Davis, 387 U.S. at 115-16 (In approving the electoral design for Virginia Beach, Virginia's City Council consisting of seven at-large members, who were residents of particular boroughs, and four other at-large members, who were chosen without regard to residency, the Supreme Court stated that, "The Seven-Four Plan makes no distinction on the basis of race, creed, or economic status or location. Each of the 11 councilmen is elected by . . . all the electors in the city. . . . If a borough's resident on the council represented in fact only the borough . . ., different conclusions might follow.") (emphasis added).
By parity of reasoning, at-large members in an untainted representational scheme are beholden to no individual district (or borough) but serve the electorate as a whole.
Cf. Fortson v. Dorsey, 379 U.S. at 438 (In finding no constitutional defect in Georgia's state senatorial reapportionment plan requiring county-wide elected senators to be residents of districts within certain counties, the Supreme Court said that, "The statute uses districts . . . as the basis of residence . . . not for . . . representation. Each district's senator must be a resident of that district, but since his tenur depends upon the county-wide electorate he must be vigilant to serve . . . all the people in the county. . .; thus . . . he is the county's and not merely the district's senator.") (emphasis added); Markoe v. Legislature of the Virgin Islands, 592 F.2d at 190 (In reviewing the apportionment of the Virgin Islands' fifteen-person Senate where one at-large electee had to be a St. John's Island resident, the court applying Fortson concluded that, "[T]he senator at large must be counted as a representative of the entire territory. . . .").
Hence, at least in principle, a quantitative vote devaluation such as occurs among substantially unequal single-member districts is absent in an at-large design.
See City of Mobile v. Bolden, 446 U.S. at 66 ("The constitutional objection to multimember districts . . . cannot be that . . . they depart from apportionment on a population basis. . . ."). See also Packer, Tracking the Court Through a Political Thicket: At-large Election Systems and Minority Vote Dilution, 23 Urb. L. Ann. 227, 258 (1982) ("By definition, an at-large district treats all voters equally. . . . At-large districting does not involve the individual deprivation characteristics of a fundamental rights violation.") (footnote omitted).
So, failing to assess the at-large membership's impact on the deviations in the cited cases conformed to the theoretical relationships that those representatives had with their electorates. Quite correctly then, the measurements converged on where the inequities were -- among the districts (the Board's borough equivalents).
Besides being in harmony with those theoretical relationships, deleting the at-large tacitly accorded with a common sense mathematical limitation. To the point, evaluating the effect -- if any -- of at-large representation on inter-district inequality means speculatively assigning those at-large votes among the districts. If that step is taken and the limitations thereby ignored, a deviational figure of doubtful validity is received.
With respect to this case, Professor Steven J. Brams, who specializes in applying mathematical models to legislative patterns, stresses the infecting discretion inbred in such an assignment and advances a preventive approach.
"[A]ny quantitative method for analyzing the Board that attempts to allocate the citywide members' motes among the boroughs is inevitably arbitrary and ad hoc.. . .In my opinion, the most appropriate methodology for analyzing the representation of individual citizens on the Board . . . is one which treats the two groups of representatives separately. "
App. to Pl. Br., Aff. of Professor Brams at 5 (emphasis in original).
The court accepts that common sense view and, extending therefrom, perceives that the absence of the at-large exclusion issue in the cited decisions was not merely due to advocate oversight. On the contrary, as shown by the cited authority, the caselaw appears too consistent to be the product of such persistent inadvertence.
Indeed, as opposed to oversight, certain cases demonstrate a conspicuous purposefulness in omitting the at-large from the measurement, as here, of hybrid legislative models. See Markoe v. Legislature of the Virgin Islands, 592 F.2d at 190-91 & nn. 2-3 (The court reviewed two apportionment plans for the Virgin Islands' fifteen-person Senate. Under the proposed "8-7" plan, the larger of the two districts would have eight senators and the smaller seven. Under the "7-7-1" scheme, which was in effect, each district had seven; and, in addition, one senator was elected at-large, although required to be a resident of the smaller district. In isolating the respective 9.9% and 3.5% disparities, the court included all fifteen members in the "8-7" plan's measurement but only fourteen members in the "7-7-1" scheme's assessment, specifically excluding the one at-large senator.); Grisbaum v. McKeithen, 336 F. Supp. at 272 (The Jefferson Parish Council of New Orleans, Louisiana, had four representatives from single-member districts, two serving two districts each and one elected at large. In separately calculating a 60% variance among the single-member districts and a 40% deviation between the sets of multiple districts, the court did not gauge the impact of the at-large on either.).
Thus, independent of not being explictly raised, the at-large exclusion issue's nonappearance in the noted decisions seems traceable, in part, to an unstated understanding that infusing the malapportionment formula with the electorate-wide representation would have been conjectural adulteration. But, whether or not that caselaw distillation is correct, this case explicitly raises that issue; and this court recognizes the speculative contamination inherent in injecting the city-wide membership into the inequality calculus. At the least, the prior decisions are consistent with that recognition.
Finally, deleting the at-large in those previous cases accords with the Supreme Court's example. For, as mentioned, the Supreme Court has exhibited a
"reluctance to involve the courts in statistical complexities, . . . absent reasons to indicate that a more complicated approach is required. . . ."
Boyer v. Gardner, 540 F. Supp. at 628. Cf. League of Women Voters of Nassau County v. Nassau County Board of Supervisors, 737 F.2d 155, slip op. at 3917 (2d Cir. 1984) ("[T]he Supreme Court's review of State and local government legislative plans has been noteworthy for the nonpromulgation of strict mathematical tests.") (emphasis in original). The at-large's surmised influence on district disproportions would not have been a reason to engage more sophisticated methodologies (of suspect reliability).
tion, unlike the New York City Council's more intricate electoral design confronted in Andrews, the Board's simpler at-large and borough legislative plan is accessible to the unmodified Abate test. Nor does the city-wide presence dictate adjusting that methodology. As prior decisions properly have done, the at-large representation should be eliminated from the deviation computation. That approach comports with the Board's existing theoretical relationships, with the common sense limitation on valid mathematical assessment and, ultimately, with the Supreme Court's guidance.
D. The Unmodified Abate Test Applied
Therefore, the Abate test is not only applicable, but applicable without modification. As thus applied, the individual inequalities are:
Borough Population Deviation
Brooklyn 2,230,936 -57.7%
Queens 1,891,325 -33.7%
Man hattan 1,427,533 - 0.9%
Bronx 1,169,115 .4%
Staten Island 352,121 .2%
Computing the -57.7% variance of Brooklyn as the most underrepresented borough with the .2% variance of Staten Island as the most overrepresented borough, the maximum deviation is 132.9%.
The present voting allocation would appear to be unconstitutional. See Abate v. Mundt, 403 U.S. at 187 (The Supreme Court approved an 11.9% justified deviation but added that "[N]othing we say today should be taken to imply that even these factors could justify sustantially greater deviations. . . ."); Mahan v. Howell, 410 U.S. at 329 (The Supreme Court accepted a 16.4% variance, which was also excused, but remarked that "While this percentage may well approach tolerable limits, we do not believe that it exceeds them.").
Nevertheless, this court has been instructed to "rule on the policies and interests which the Supreme Court has held may justify deviations. . . ." Morris v. Board of Estimate, 707 F.2d at 690. Moreover, addresssing such possible justifications conforms with the parties' stipulation bifurcating these proceedings. Accordingly, the court will now outline the areas of inquiry and initial approach for this second stage.
Since the Second Circuit's opinion, the Supreme Court has further elaborated on specific, legitimate justifications. In Karcher v. Daggett, 462 U.S. 725, , 103 S. Ct. 2653, 2663-64, 77 L. Ed. 2d 133 (1983), within the context of the more stringently scrutinized congressional districting,
the Supreme Court advised:
"Any number of consistently applied legislative policies might justify some variance, including, for instance, making districts compact, respecting municipal boundaries, preserving the cores of prior districts, and avoiding contest between incumbent Representatives. As long as the criteria are nondiscriminatory, . . ., these are all legitimate objectives that . . . could justify all legitimate objectives that . . . could justify minor population deviations."
(Emphasis added.) See Elections, 97 Harv. L. Rev. at 141 ("[Karcher] does substantially broaden -- at least in dicta -- the range of interests that the [Supreme] Court will consider.") (emphasis in original).
In Brown v. Thomson, U.S. at , 103 S. Ct. at 2696, rendered the same day, the Supreme Court restated certain considerations noted in Karcher and previously reckoned as acceptable for state and local governments.
"We have recognized that some deviations from population equality may be necessary to permit the States to pursue other legitimate objectives such as "maintain[ing] the integrity of various political subdivisions' and "provid[ing] for compact districts of contiguous territory."
(Citation omittted; brackets by Brown Court.) See Elections, 97 Harv. L. Rev. at 137 n.14 ("Brown illustrates the [Supreme] Court'sgreater willingness to tolerate population deviations in state legislative apportionments when such deviations are justified by an appropriate state policy. The [Supreme] Court upheld . . . a Wyoming statute providing that each county was to be given at least one representative in the state house. . . . Wyoming's longstanding . . . policy of using counties as representatitve districts was supported by . . . legitimate . . . concerns and . . . the incremental dilution of the challengers' voting power resulting from . . . granting . . . a representative to the least populous county was de minimus.").
During the justification inquiry then, the parties are to look to those areas mentioned in Karcher and Brown -- if appropriate -- plus any other pertinent areas reviewed by the Supreme court in earlier cases. See, e.g., Abate v. Mundt, 403 U.S. at 186 ("The need for intergovermental coordination is often greatest at the local level . . . But because almost all governmental entities are interrelated . . ., we would be hesitant to accept this justification by itself. To us, therefore, it is significant that Rockland County has long recognized the advantage of having the same individuals occupy the governing positions of both the county and its towns.").
Besides express Supreme Court statements, guidance in establishing or refuting proposed interests also may be sought from other judicial precedent in general accord with the Supreme Court's pronouncements. See, e.g., Andrews v. Koch, 528 F. Supp. at 251 ("[T]he interests defendants urge in support of the status quo appear to be essentially political."); Cohen v. Maloney, 410 F. Supp. at 1154 ("[N]atural boundaries may sometimes be regarded as a legitimate consideration justifying a particular apportionment plan. . . . [S]uch boundaries may create communication and association barriers of substantial significance in the administration of local government. . . . The facts of this case, however, do not suggest that the Brandywine [River] is a natural boundary which has traditionally been of significance in the administration of local government.").
The parties, in addition, are to examine two particular concerns already raised in the record. First, the Board's conceded effectiveness is to be addressed.
See generally Franklin v. Krause, 32 N.Y.2d 234, 344 N.Y.S.2d 885, 888-89, 298 N.E.2d 68 ("It has been argued to us, without material opposition, that the small board . . . is the most efficient form of government, and has proved to be such over the years.").
Second, the parties are to explore the alleged meaningful participation by the lesser populated boroughs in Board affairs under its present structure.
See Reynolds v. Sims, 377 U.S. at 565-66 ("[T]he achieving of fair and effective representation for all citizens is . . . the basic aim of legislative apportionment. . . .") (emphasis added); Travis v. King, 552 F. Supp. at 560 ("[T]he . . . rationale advanced by the state [of Hawaii] . . . is its desire to provide each . . . island . . . with meaningful representation. . . . Based on the unique geographic and economic insularity of the . . . island[s] . . ., along with the state's simplified and centralized . . . government, we find that the . . . policy is . . . rational. . . .") (footnote omitted). Cf. Morris v. Board of Estimate, 707 F.2d at 691 ("Whether the present allocation of seats . . . is a constitutionally permissible way of blending majority control with minority representation . . . is one issue to be examined on remand.").
Ascertaining at this time whether significant input by the smaller boroughs is a valid concern fostered by the Board may be especially important if the Board is eventually reapportioned, since certain electoral models "assure that sparsely populated areas . . . have a voice in the councils of government." Greenwald v. Board of Supervisors of the County of Sullivan, 567 F. Supp. 200, 203 (S.D.N.Y. 1983) (upholding the constitutionality of an adjusted weighted voting plan) (footnote omitted). See also Baker v. Regional High School District No. 5 476 F. Supp. 319, 321, 323 (D. Conn. 1979) (In tentatively approving an "ingenious" condition for a three-town board of education preventing majority votes from being carried by the largest town alone or by the two smallest towns together, the court described the total plan as an apparently "conscientious effort to comply with constitutional requirements while providing sufficient checks on majority action. . . ."). See generally Sailors v. Board of Education, 387 U.S. 105, 110-11, 18 L. Ed. 2d 650, 87 S. Ct. 1549 (1967) ("Viable local governments may need many innovations, numerous combinations of old and new devices, great flexibility in municipal arrangements to meet changing urban conditions. We see nothing in the Constitution to prevent experimentation.").
But future forms aside, investigating justifications here is to be in accordance with the inquiry outlined in Mahan as lately summarized in Brown:
"whether the legislator's plan "may reasonably be said to advance [a] rational state policy" and, if so "whether the population disparities among the districts that have resulted from the pursuit of this plan exceed constitutional limits."
Brown v. Thomson, U.S. at , 103 U.S. at 2696 (quoting Mahan v. Howell, 410 U.S. at 328) (brackets by Brown Court).
Finally, the court emphasizes that nothing in the aforesaid is to be construed as a predetermination that the Board does or does not advance specific, valid policies and interests. In that vein, the mentioned areas of inquiry are not by limitation. The number of such putative considerations to be probed is up to the parties as guided by the designated caselaw.
Consistent with the foregoing, the court directs the parties to begin conferring by September 14, 1984. The product of that good faith exchange will be a joint stipulation for submission no later than October 5, 1984, listing with conciseness and specificity:
(1) those agreed valid policies and interests presently served by the Board; and
(2) those disputed policies and interests which at least one defendant maintains are valid and are presently served by the Board. For each such disputed consideration, plaintiffs are to explain the basis of their objection.
After reviewing that joint stipulaiton, the court will schedule a conference with the parties to discuss the subsequent procedure for bringing this litigation to an expeditious and equitable end.
Copies of this Memorandum and Order are being forwarded to counsel for the parties, to the intervenor-defendants, and to the amicus curiae.