Opinion
March 13, 1969
Weil, Gotshal Manges for Rita H. Satz and others, plaintiffs.
Fiore, Acton Smith for Town of Greenburgh and another, plaintiffs.
Gordon Miller, County Attorney, for Board of Supervisors of Westchester County, defendant.
Arthur H. Ellis, Corporation Counsel, for City of Mount Vernon, defendant.
Francis S. Claps, Corporation Counsel, for City of New Rochelle, defendant.
Leone Pecoraro, Corporation Counsel, for City of Peekskill, defendant.
Anthony T. Antinozzi, Corporation Counsel, for City of Rye, defendant. William McKinley, Corporation Counsel, for City of White Plains, defendant.
Francis B. Cline, Corporation Counsel, for City of Yonkers, defendant.
Roger Gofton, Town Attorney, for Town of Yorktown, plaintiff-intervenor.
Frederick A. Judy, Town Attorney, for Town of Scarsdale, plaintiff-intervenor.
Louis J. Lefkowitz, Attorney-General, for Secretary of State of State of New York, defendant.
In connection with the present application for further action by the court in reapportioning the county legislative body, certain occurrences since the court's interim decision of January 24, 1969 (N.Y.L.J., Feb. 5, 1969, p. 20, col. 6) deserve mention. The two proposed local laws described in the former opinion, both based on a division of the county into 17 districts, came before the Board of Supervisors on February 17, and both were rejected by a majority vote of the board. One of these plans, calling for the election of one legislator from each of the 17 districts, has nevertheless been adopted by the League of Women Voters (with one modification) in preference to the plan previously proposed by the League. Since the January 24 decision, the court has also heard additional oral argument, received briefs and taken further testimony. The matter is now ready for determination.
As the court indicated in its prior decision in the present proceeding, discussion here is confined to plans under which the county is divided into districts. Weighted voting plans have been twice overwhelmingly rejected by popular vote. No party to the action now advocates weighted voting. The Board of Supervisors no longer proposes it. In the numerous communications received by the court from cities, towns, villages, political parties and their leaders, public officials and private citizens, it receives scarcely any support. Weighted voting has served as a useful stopgap, but it is now apparent that it cannot win popular acceptance in this county. It is worth noting that even the interim plan of weighted voting effectuated by the court's decision of July 6, 1966 ( 51 Misc.2d 168), under which the board is still acting, was later held to be invalid insofar as it related to the Cities of Yonkers and New Rochelle ( 53 Misc.2d 88). The alternative to weighted voting can only be a plan under which legislators are elected from districts by constituencies of substantially equal population. This does not mean that existing municipalities are to be ignored; but it does mean that they are to be subordinated to equality of population, which is the mainspring of the whole line of reapportionment decisions beginning with Baker v. Carr ( 369 U.S. 186) and continuing through Gray v. Sanders ( 372 U.S. 368), Reynolds v. Sims ( 377 U.S. 533) and the numerous decisions, both Federal and State, which followed them.
It is a singular fact that a plan to replace the Board of Supervisors of Westchester County with a smaller board selected independently of the existing towns and cities was devised many years ago by those who framed the county's charter. The State of New York provides by general law that the supervisors of the several cities and towns of each county shall constitute the Board of Supervisors (County Law, § 150), and a similar provision appears in the Westchester County Administrative Code (art. 3, § 12; L. 1948, ch. 852, as amd.). Nevertheless the Westchester County Charter, adopted in 1937, contains in article II a detailed plan for the replacement of the Board of Supervisors by a "County Board" (upon approval by popular referendum), consisting of 10 to 12 members elected from districts having a population of 50,000 or major fraction thereof. Such a plan was actually submitted to the voters, and defeated, in 1941. It was never again submitted, but it deserves mention as evidence of the fact that long before any reapportionment of the county legislative body was mandated on constitutional grounds the feasibility of replacing the city and town supervisors with a smaller body chosen by districts was specifically contemplated.
Various plans of proposed apportionment have been submitted to the court for consideration — including plans of proportional representation, plans based on the number of actual recent voters rather than population, plans based on preserving the identity of villages and assuring them a voice in county government, and various others. Some of these plans display great ingenuity, and no doubt all are well intentioned. All such proposals have received the attention of the court, but in the end a choice must be made between the three plans mentioned in the January 24 decision, namely: (1) An apportionment based upon the 7 assembly districts into which the county was divided by the Marden Commission in its report to the Court of Appeals in Matter of Orans ( 17A N.Y.2d 7, 19); (2) an apportionment based upon a division of the county into 11 districts but calling for a county board of 21 members, under what has come to be known as the "Greenburgh plan"; and (3) an apportionment based upon a division of the county into 17 districts and calling for a county board of either 17 or 34 members, the plan having originated in the legislative committee of the Board of Supervisors but failed of acceptance by the board as a whole. These plans require separate discussion.
(1) The Assembly District Plan. The great advantage of a plan of apportionment based on the county's 7 assembly districts is that these districts were laid out by the Marden Commission in such a way as to contain equality of population with almost mathematical precision ( Matter of Orans, supra). We can at least be reasonably confident that any plan providing for the election of one or more county legislators from each of these 7 districts satisfies constitutional standards. As assembly districts, the apportionment has been approved by the Court of Appeals in the Orans case; and as county legislative districts, the same apportionment was approved, as to its constitutionality, by the Appellate Division in this very case ( Town of Greenburgh v. Board of Supervisors, 30 A.D.2d 708; see, also, 57 Misc.2d 1008). Whatever defects the plan may have as the basis for a county board, it can scarcely be doubted that it complies with the mandate of "one person, one vote," and this is true even though it is based on the 1960 census.
Nevertheless the court has determined, after mature reflection, to reject the assembly district plan. To begin with, it is not advocated by any party to the present action. In the second place, it fragments existing municipalities to a greater extent than either of the other plans. In the third place, it seems like dubious wisdom to embark upon a new apportionment scheme based on population figures which are already known (through the special census of 1965) to be seriously out of joint. Most important of all, in the court's view, is the fact that a plan of reapportionment based on the same assembly districts was roundly defeated at the polls at the last election. Perhaps we could speculate that the reason for the defeat was that the plan contemplated the election of 5 representatives at large from each of the 7 assembly districts, to serve for a term of four years; and we could theorize that the voters looked upon the arrangement as too many for too long. We could suppose further that the public might feel differently about a board of 14 or 21, with a two-year term of office. But all this would be mere guesswork. The only certain fact is that the proposition was defeated by better than a two-to-one margin, and the court has no information enabling it to look with confidence behind the election returns. As in the case of weighted voting, the court is reluctant to order into effect a plan of apportionment which bears the handicap of popular disapproval.
(2) The Greenburgh Plan. This plan proposes a division of the county into 11 districts, each containing populations of about 40,000 or multiples thereof. It contemplates the election of single representatives in districts of about 40,000, two representatives in districts of about 80,000, and three representatives in districts of about 120,000, and would result in a board of 21 members. More specifically, Districts I and XI would elect 1 representative each, District IX would elect 3, and the remaining eight districts would elect 2 apiece. The plan has certain important advantages, the most notable of which is its relatively close approach to population equality. The total population of the county under the 1965 census was 853,198, which means that ideally each of 21 representatives would represent 40,628 people. Under this plan the greatest departures from the norm would be in District VIII, where the 2 legislators would represent 83,249 people or 41,624 each, and in District XI, where the single legislator would have a constituency of 39,708 people. In every case the variation from the average of 40,628 would be less than 1,000, and the ratio of the largest constituency to the smallest would be 1.05 to 1. Another advantage of the Greenburgh plan is that with few exceptions it preserves the territorial integrity of the various towns and cities, which is a matter of primary concern to local governmental units.
Of course, the plan has its disadvantages too. The inevitable result of dividing the county into districts bounded by town and city lines, each containing 40,000 people or a multiple thereof, is to produce some rather odd-looking districts. For example: District VI consists of the City of New Rochelle (and the adjoining Village of North Pelham) and contains 80,528 of population, making a two-member district. This district is shaped somewhat like a handsaw, with the handle to the south. The result of keeping New Rochelle intact is that it thrusts itself up through the middle of District VII, which resembles an inverted "V", with the Town of Eastchester running along the toothed side of the saw to the west, the Town of Mamaroneck on the straight edge to the east, and the small end of the saw pushing its way into the Town of Scarsdale to the north. This means that most of District VII is bisected and separated by District VI. District VIII, made up of the Towns of Rye, Harrison and North Castle and the City of Rye, wanders up the easterly side of the county in a fashion which could scarcely be called compact. But the worst example of an unnatural grouping is District IX. This district consists of the City of White Plains and the Towns of Mount Pleasant, New Castle and Bedford, with a combined population of 119,478. It is the only district which would elect three members to the board. The most noticeable objection to it is that the City of White Plains does not touch any of the three towns to which it is proposed to be joined. It is obvious that a district consisting of disconnected parts could not be justified, and Greenburgh seeks to overcome the difficulty by including in District IX a narrow and uninhabited strip of its own land, nearly a mile long and some 200 feet wide, lying between White Plains and Mount Pleasant. With the aid of this slender corridor the district, which starts at the southern end of White Plains, is carried continuously further north through the towns of Mount Pleasant, New Castle and Bedford, projecting itself into the heart of District XI near the northern end of the county. Electoral districts should be reasonably compact and contiguous ( Reynolds v. Sims, 377 U.S. 533, 578, supra; Swann v. Adams, 385 U.S. 440, 444). The device of joining up a disconnected district by utilizing an uninhabited corridor was condemned in Wells v. Rockefeller ( 273 F. Supp. 984, 987).
The Greenburgh plan as a whole is a combination of single-member and multi-member districts. Such a plan may be fully in accord with the Constitution ( Reynolds v. Sims, supra, p. 579; Fortson v. Dorsey, 379 U.S. 433; Kilgarlin v. Hill, 386 U.S. 120), but that holding is accompanied by a caveat. An apportionment based on multi-member districts may be unacceptable if it is calculated to produce invidious discrimination either racially or politically ( Burns v. Richardson, 384 U.S. 73, 88). Of course, there is no evidence that the present plan would have any such effect. There could be no such evidence since the plan is still nothing more than a proposal. But the court mentioned the possibilities in the Burns case ( supra, p. 88): "It may be that this invidious effect can more easily be shown if, in contrast to the facts in Fortson [ 379 U.S. 433], districts are large in relation to the total number of legislators, if districts are not appropriately subdistricted to assure distribution of legislators that are resident over the entire district".
Here the court is not passing on the validity of an existing plan, but is attempting to create a method of apportionment which will remain in effect until it is amended according to law. It seems to the court that the Greenburgh plan may possibly contain the seeds of unconstitutionality in the future by authorizing a concentration of representation in selected areas and possibly creating the conditions quoted above. No such danger can be apprehended from a plan of equally populated districts electing single members. The court believes also that districts of the latter sort would lend themselves more readily to the adjustments which will have to be made after the 1970 census and periodically thereafter. There is finally the practical consideration that candidates for office in the multi-member districts would be obliged to solicit the support of two or three times as many constituents as candidates in single-member districts, and would be responsible to an equally greater number for their performance in office.
The court is of the opinion that the single-member district plan is preferable, provided its own defects are not so great as to raise serious doubts about its constitutionality or its practical advantages.
(3) The Seventeen-District Plan. This plan, originating in the legislative committee of the Board of Supervisors, has two possible forms. Under one, a single representative would be elected from each of 17 districts, and under the other each district would elect two members. As between these two possibilities the court concludes that the single-member district plan is preferable. It is difficult to believe that a board of 34 members will be needed to perform the duties of a county legislature, none of whose members will have any other public duties to perform, as is now the case with the Town Supervisors. It is to be remembered that the 18 Town Supervisors will continue to function as town officers, and if to those 18 officials we were now to add 34 county legislators, the result of reapportionment would be a net increase of town and county officials from 45 to 52. One lesson to be learned from the results of the two referenda is that the public does not favor a numerical increase. The revision contemplated by article II of the County Charter would have reduced the board to 10 or 12 members; but since that number was based on one member for each 50,000 in population, it is evident that the population increase in the last 30 years would now suggest a board of 17 members. That is the theory underlying the plan now under discussion, which divides the county into 17 districts.
In its single-member form, this plan has been indorsed by the League of Women Voters, with a single modification which in itself seems to be noncontroversial. Since the Board of Supervisors has repudiated the plan, it has come to be referred to as "the League Plan."
The plan has several obvious defects, and the court's real task has been to reach a conclusion as to whether this plan is, on balance, more or less desirable than that advocated by Greenburgh. It is certain that from the important standpoint of population equality the League plan is the more imperfect of the two. As heretofore noted, the total population of the county under the most recent census is 853,198, so that 50,188 should be the average population of each of the 17 districts. The furthest departures from the average are in District 7, with 53,523 in the Town of Mamaroneck, the City of Rye, and a portion of the Town of Rye; and in District 11, with 47,079 in the first and fourth wards of New Rochelle and the villages of Pelham and North Pelham. The two extremes represent a deviation of 6.6% in the one case, and 6.2% in the other. The ratio between the largest constituency and the smallest is 1.14 to 1.
Whether these disparities are too great to constitute a valid apportionment is a question which cannot be answered without considering the other circumstances. It has been plainly said that "A state may legitimately desire to maintain the integrity of various political subdivisions," and that "Indiscriminate districting, without any regard for political subdivision or natural or historical boundary lines, may be little more than an open invitation to partisan gerrymandering" ( Reynolds v. Sims, 377 U.S. 533, 578-579, supra). Of course, the Supreme Court has emphasized the fact that substantial equality of population is the overriding objective. But we are told too, in the same leading case (p. 577) that "Mathematical exactness or precision is hardly a workable constitutional requirement." So the rule would appear to be that some departure from mathematical exactness in district populations is permissible in the legitimate preservation of existing municipal boundaries.
It is clear that the authors of both the League plan and the Greenburgh plan have attempted to follow the present municipal boundary lines, as far as possible, and that such departures from mathematical perfection as we find in both plans are caused by their adherence to such lines. Greenburgh would undoubtedly say that it has been more successful in both aspects, that is, in keeping population disparities to a minimum, and in avoiding the splitting of towns and cities. But it has done so by the device, which the court is unwilling to adopt, of creating a number of multi-member districts which lack compactness and (in one case) contiguity. The League plan is not beyond criticism in the latter respect. Its District 3 is no model of symmetry, and shows the same meandering tendency as Greenburgh's District VIII. The splitting of New Rochelle into three districts under the League plan is regrettable, but, except through the Greenburgh plan, the court knows of no way to avoid it. Both the Greenburgh plan and the League plan fragment the Town of Pelham; and there is no way to avoid that either, except through the assembly district plan which fragments everything else in southern Westchester.
Of course population disparities transcend all other factors in importance, and the question is whether under all the circumstances the maximum deviation of 6.6% from the norm under the League plan, and the ratio of 1.14 to 1 between the largest and smallest districts, invalidate the whole scheme. The court thinks not. There is simply no way to get a better result except (1) through the introduction of multi-member districts, as advocated by Greenburgh, or (2) by ignoring virtually all city and town lines for the sake of population equality. The court cannot believe that either of these expedients must be resorted to in order to overcome the disparities found here. Our Court of Appeals has approved an apportionment of the New York Legislature drawn along county lines ( Matter of Orans, 17 N.Y.2d 107). The variations in the Senate were 9.4% plus and 6.2% minus. The variations in the Assembly were 8.8% plus and 10.3% minus (17A N.Y.2d 10). The ratio of the largest senate district to the smallest was 1.17 to 1, and the ratio of the largest assembly district to the smallest was 1.21 to 1 (17A N.Y.2d 10, 16, 23). These, of course, are greater discrepancies than appear here, and they have thus far been sustained as justifiable under the circumstances.
In Swann v. Adams ( 385 U.S. 440, supra) the court disapproved a legislative reapportionment in the State of Florida in which the population variations were as high as 30% in the senate districts and 40% in the house. It was said that such variations would not be approved "without a satisfactory explanation grounded on acceptable state policy" (p. 444). among the factors relevant to State policy on this subject it was said that the integrity of political subdivisions was one consideration. So here the State of New York has demonstrated its policy of respecting county lines in the establishment of legislative districts, and for that purpose has approved of population variations substantially less than in Swann v. Adams. By similar reasoning, it would seem that a county which, by accommodating its apportionment as far as possible to municipal boundaries, has achieved still smaller percentages of variation, has acted both constitutionally and sensibly.
The court approves of and adopts the League plan, but calls attention to the following specific factors: The Village of Mount Kisco should be included wholly within District 2, rather than partly in District 4 as shown by County Attorney's Exhibit 1. This change will not only accord with the expressed wish of the Village Manager, but will more nearly equalize the populations of the two districts. The boundary line between Districts 8 and 9 in the Town of Greenburgh and the line between Districts 12 and 13 in the City of Mount Vernon are not drawn along the lines of presently existing election districts. The judgment should direct that, to the extent that election districts do not follow these lines, Greenburgh and Mount Vernon shall change the boundaries of the appropriate election districts.
The court will again retain jurisdiction for all purposes, but, in order to avoid the necessity for immediate application to the court, the following explanations are made now with respect to the new board and its members: The board shall be known as the "County Board," and its members as "County Legislators." Such legislators shall be nominated and elected in 1969 and take office on January 1, 1970, on which date the County Board shall replace the Board of Supervisors as the legislative body of the county, (County Law, § 150-a), and shall continue thereafter until validly superseded. The qualifications of candidates shall be as provided in section 3 Pub. Off. of the Public Officers Law. The County Legislators shall serve for a term of two years. Their salaries shall be as fixed by section 15 of article III of the Westchester County Administrative Code, unless altered by appropriate legislation. Such County Legislators shall be county officers. No provision of a city charter with respect to city supervisors shall be applicable to such County Legislators. Vacancies in the office of County Legislator shall be filled by appointment by the County Board until the next general election, at which time they shall be filled for the unexpired term. All the provisions of the Westchester County Charter and Westchester County Administrative Code now applicable to the Board of Supervisors shall be applicable to the County Board and its members, except as to any provision thereof which may be inconsistent with this decision.
The court expresses its appreciation to the Deputy County Attorney who has appeared throughout this proceeding, and to counsel for the Town of Greenburgh and for the League of Women Voters, for their great helpfulness in the performance of this difficult task.
The judgment, to be submitted on notice, should contain a description of the districts, sufficient for identification, in accordance with County Attorney's Exhibit 1 as modified with respect to the Village of Mount Kisco.