Opinion
May 10, 1967
Anthony J. De Staffan for plaintiff.
Louis J. Lefkowitz, Attorney-General ( Robert W. Imrie of counsel), for State of New York, defendant.
J. Willis Barrett for Board of Supervisors of County of Wayne, and others, defendants.
Harold J. Stiles for Town Board of Arcadia, defendant.
Charles J. Dittmar for Town Board of Williamson, defendant.
The plaintiff, a qualified citizen of the Town of Arcadia, brings this action seeking a declaratory judgment that the apportionment of the Board of Supervisors of the County of Wayne is unconstitutional (N.Y. Const., art. I, § 11) and in violation of the Fourteenth Amendment to the Constitution of the United States.
Wayne County, New York, has 15 towns. Each town is represented by one Supervisor. The towns vary in population, according to a special census enumeration taken in 1966, from the Town of Huron with 1,570 persons to the Town of Arcadia with a population of 12,544 persons, exclusive of 2,620 duly committed, mentally retarded patients at the Newark State School under the jurisdiction of the New York State Department of Mental Hygiene. For reasons subsequently expressed, these patients are not considered in the population of the Town of Arcadia for the purposes of apportionment under the "one person, one vote" principle.
The eight smallest towns in Wayne County represent approximately 30% of the population of the county and constitute a majority of the Board of Supervisors, as it is now organized. It is accordingly held that, under the principles enunciated in Reynolds v. Sims ( 377 U.S. 533) and Seaman v. Fedourich ( 16 N.Y.2d 94), the Wayne County Board of Supervisors is at present not organized so as to give equal representation on the board to the citizens of the county.
On February 3, 1967, after this action was commenced, the Supervisors tacitly recognized the malapportionment of the board, as alleged in the complaint, by adopting Resolution No. 18, which provided for "a weighted voting plan".
The board's weighted voting plan is directly proportional to the population of the several towns based on the 1966 special census, except in the case of the Town of Arcadia from which the Newark State School patient population of 2,620, previously mentioned, has been deleted (see Table I). All but 115 patients in the institution are nonresidents of Wayne County, according to a certification by the director of the school in answer to an inquiry by the court, on stipulation of all counsel. I am persuaded that the board was correct in its decision not to include these patients in the population. In contrast to what was apparently held to be the situation in the Binghamton State Hospital (cf. Seaman v. Fedourich, 16 N.Y.2d 94, supra), all of the patients at Newark State School have been duly committed as mental defectives and are ineligible to vote (Election Law, § 152, subd. 6). Furthermore, more than 95% of these patients are nonresidents of the county, and were they to be accorded representation on the "one person, one vote" principle, their number alone would in effect disenfranchise the residents of any one of four towns in the county, each of which has a citizen population less than the patient population of the Newark State School.
The following Table I shows in Column "A", the 1966 special census of the towns of the county; Column "B", the weighted vote fixed by the board pursuant to Resolution No. 18; Column "C", the proportion of the total weighted vote proposed to be cast by each Supervisor:
Table I Town A B C Arcadia ..................................... 12,544 125 17.532 Butler ...................................... 1,587 16 2.244 Galen ....................................... 4,500 45 6.311 Huron ....................................... 1,570 16 2.244 Lyons ....................................... 5,868 59 8.275 Macedon ..................................... 4,770 48 6.732 Marion ...................................... 3,229 32 4.488 Ontario ..................................... 5,366 54 7.574 Palmyra ..................................... 6,852 69 9.677 Rose ........................................ 2,230 22 3.086 Savannah .................................... 1,744 17 2.384 Sodus ....................................... 7,851 79 11.080 Walworth .................................... 3,534 35 4.909 Williamson .................................. 6,082 61 8.555 Wolcott ..................................... 3,490 35 4.909 ______ ___ _______ Totals .................................... 71,217 713 100.000 Newark State School ......................... 2,620 ______ 1966 census total ........................... 73,837 The premise of the board appears to be that by allocating votes to each Supervisor in direct proportion to the population of the town he represents, each will then have proportionate and effective voting power, thus complying with the "one person, one vote" principle required by Reynolds v. Sims ( 377 U.S. 533, supra).It seems appropriate, in a discussion of the basis for any type of apportionment plan, to consider what voting power is in terms of legislative functioning.
Banzhaf in his article, "Weighted Voting Doesn't Work: A Mathematical Analysis" (19 Rutgers L. Rev. 317,331 [1965]), states: "Since the determination of legislative outcomes is the only legitimate use and purpose of legislative power and since usually an individual legislator can only determine outcomes when the others are closely divided, the appropriate measure of a legislator's power is simply the number of different situations in which he is able to determine the outcome."
This definition of legislative or voting power accurately sets forth a logical standard against which various methods can be measured.
The requirements which any definition of voting power must satisfy are said to be as follows (19 Rutgers L. Rev., supra, p. 330): "In the first place, the definition must provide an objective measure — that is, that any number of qualified observers using it will arrive at the same conclusions concerning the same situation and that it is independent of such things as the personality of the representatives, party affiliations, and previous voting records. Secondly, the definition should be believable and reasonable to the average reader and not make use of unreasonable assumptions about the political process * * * Finally, it would seem that the definition should be based only on the realistic and effective use of the voting power — the legislator's ability to affect the outcome of legislative determinations — and that it yield a system of analysis which can be used to make meaningful calculations in actual legislative situations." (Italics supplied.)
This analysis of the basic purpose of a weighted vote is fundamental to the problem of equalizing the effective voting power of each Supervisor, whether the ultimate method used be reapportionment through regrouping towns into units having substantially equal population; a weighted voting plan, as proposed here; fractional voting plans; multimember districts; or multimember districts combined with weighted voting (modified weighted voting).
In Town of Greenburgh v. Board of Supervisors of Westchester County ( 53 Misc.2d 88), decided February 3, 1967, Mr. Justice NOLAN had a digital computer analysis available of the voting power of the 54 legislators in Westchester County, under a multimember modified weighted voting plan then adopted by the Board of Supervisors of that county (see Table II of that opinion). Since the weighted vote of Yonkers and New Rochelle under the plan was based on the number of registered voters in each ward, instead of the population therein, Judge NOLAN disapproved the plan because the citizens of those two cities, which constituted some 33% of the population of Westchester County, did not thereby have "one person, one vote" representation.
The computer analysis in that case fortified the Banzhaf premise that effective voting power for each legislator proportional to the population represented is the goal toward which legislative bodies should aspire in order to attain the principle of "one person, one vote" now mandated by Reynolds v. Sims ( 377 U.S. 533, supra).
Accordingly, I undertook to have such an analysis made of the Wayne County plan embodied in Resolution No. 18 of the Board of Supervisors. The analysis was made by the New York Scientific Center, a division of International Business Machines Corporation, on an IBM 360 Computer, Model 40. This analysis (Column "C" thereof) showed the number of times each town, through the use of its weighted vote, could effectively apply its voting power in all combinations of voting situations, when the issue was close, to either pass or defeat legislation, depending on whether the Supervisor of such town voted "yes" or "no" on a given vote.
Table II below is the analysis of the plan prescribed by Resolution No. 18. Column "A" shows the towns; Column "B", the weighted votes for each town, as fixed by the board; Column "C", the number of effective votes each Supervisor can cast; Column "D", the percentage of the total weighted vote cast by each Supervisor; Column "E", the percentage of the total effective votes cast by each Supervisor; Column "F", the ratio of the percentage of effective votes (Column "E") to the percentage of the weighted vote cast by each Supervisor (Column "D"):
Percentage Ratio of Column "B" to sum of Column "B".
Percentage Ratio of Column "C" to sum of Column "C".
This analysis shows that Arcadia (125 votes), with 17.53% of the population of Wayne County, can effectively control (i.e., pass or defeat) legislation 19.61% of the time; whereas Savannah (17 votes), with 2.384% of the population, can effectively control legislation only 2.250% of the time. In other words, Arcadia with 125 votes has effective voting power proportionate to the population of the county 111.9% of the time (or Column "E" ÷ Column "D"), whereas Savannah with its 17 votes is only 94.4% effective. To put it another way, Arcadia with 125 weighted votes has effective voting power ranging from 12.3% more than Sodus with 79 weighted votes to 17.5% more than Savannah. Although each town under the plan has a weighted vote directly proportional to its respective population, the Supervisor from Arcadia has an average among the towns of 14% more effective voting power to determine the passage or defeat of legislation by virtue of his weighted vote. This is a substantial disparity and overweight in effective voting power in favor of the largest town to the detriment of the other towns.
Plans involving weighted voting, in order to meet the test of constitutionality, should effectively carry out the "one person, one vote" principle. That is to say the plan should allocate effective voting power to legislators in nearly equal proportion to the population that each represents. It is clear that the present plan gives the Supervisor from Arcadia from 12.3% to 17.5% more effective voting power over each other Supervisor on the board. Excluding Arcadia, the difference between the other town Supervisors is not substantial. The weighted vote of Sodus, for instance, is effective 99.6% of the time, whereas the weighted vote of Savannah is effective 94.4% of the time. This represents but a 5.2% disparity between these two towns. The disparity between all other towns is less.
Therefore, I hold that the present plan is unconstitutional because it does not provide that the "vote of any citizen is approximately equal in weight to that of any other citizen" ( Reynolds v. Sims, 377 U.S. 533, 579, supra).
In Town of Greenburgh v. Board of Supervisors of Westchester County ( 53 Misc.2d 88, 98-99), it is well said "that the voting power of a legislator, and consequently the effective legislative representation of his constituents should be measured not by the number of votes which he casts, but by the number of times, in all possible voting combinations, that the legislator can cast an effective vote, that is, the number of times when the votes of all other members of the body may be so evenly divided that he can change the result of the vote, on a pending measure by changing his vote from aye to no, or from no to aye. Weighted voting plans should not be condemned, however, because weighted voting may produce inequitable results under some circumstances. Weighted voting will, or will not produce effective legislative influence, proportional to population according to the circumstances which may exist, in any given case, and the plan provided by the local law, if otherwise acceptable would, in the instant case, produce that result. Mathematical computations, as shown by Table II, which is reproduced at the end of this opinion, demonstrate that, under the plan each legislator would have not only a weighted vote, but also an effective vote which would be proportional to the number of his constituents."
In the Town of Greenburgh case, it was the improper allocation of weighted votes in the Cities of Yonkers and New Rochelle, where the population was submerged, which prevented approval of that plan. In weighting the votes of Yonkers and New Rochelle, the distribution was on the basis of the number of registered voters and not on the population of the wards. Thus the disparity between ward populations based on registered voters in the wards was entirely inconsistent with any rational plan for population based representation because the population was submerged.
A weighted voting plan should allocate effective voting power to legislators in nearly equal proportion to the population that each represents, and if the plan does not do this, it is not constitutionally acceptable. In the plan here under consideration, the block of 125 votes assigned to the Town of Arcadia overweights its vote inequitably over that of the other towns, and therefore, this would be invidious discrimination, which is unconstitutional ( Williamson v. Lee Optical Co., 348 U.S. 483, 489; Town of Greenburgh v. Board of Supervisors, 53 Misc.2d 88, 99, supra).
For the information of the parties and counsel, I had a further digital computer analysis made of effective voting power to determine whether it is practicable and possible to adjust the weighted votes of the respective towns so that each could have an effective vote reasonably proportional to its population, and thus substantially comply with the "one person, one vote" principle.
The result of this particular study is set forth in Table III below. The mathematical computations show that if Arcadia, representing 17.5% of the county population, were given 112 weighted votes in a total of 700 allocated to all the towns, then the effective voting power of Arcadia would be 17.48% and substantially proportionate to its population.
In adjusting the weighted vote of each town, in order to provide for an effective vote reasonably proportional to the actual percentage of population which each Supervisor represents, three changes were made in the weighted vote assigned to Arcadia, Sodus and Rose. Arcadia's weighted vote was reduced to 112, the weighted vote of Sodus was reduced one to 78, and the weighted vote of Rose was increased one to 23. Thus the total weighted votes cast by all towns was 700, instead of 713. Column "E" in Table III shows the effective weighted vote of the respective Supervisors, as adjusted. Column "F" shows the ratio of the effective vote, after adjustment, to the weighted vote cast by each Supervisor.
It is noted that by adjusting the weighted vote of Arcadia, Sodus and Rose, the disparity of effective voting power and the actual percentage of population between towns would be a maximum of 3.34%.
Percentage ratio of Column "C" to sum of Column "C".
Further study and analysis of this problem could be made, but this is a matter for the Board of Supervisors to promptly undertake in order to propose a plan which will be constitutionally acceptable.
I am not unmindful of the Third Department's decision in Iannucci v. Board of Supervisors of County of Washington ( 27 A.D.2d 346, 347). The weighted voting plan there was deemed by the majority of the court to be unconstitutional for the reason that: "Representatives of the larger districts, because of the weighting of their votes, necessarily have greater influence over the passage of legislation. The result is discrimination against the smaller districts."
I agree with this conclusion.
The court then says in effect that further consideration of weighted voting plans, including adjustments thereof, evades or sidesteps the issue because this "will cause the continued submission of various plans which will only delay the solution of the problem."
I suggest that, although computer analysis of this problem may well be complicated, so also is a constitutional solution where a county does not lend itself to division into geographical districts or units of approximately equal population.
For instance, consideration was also given to redistricting Wayne County into units of towns geographically contiguous so that each unit would have approximately equal population. This might be accomplished by dividing the county into either five or six districts. However, such regrouping, no matter how the towns are aligned or combined, would require that at least five or six towns in the eastern end of the county be joined into one district. These towns represent about one third of the area of the county and approximately 20% of the population.
I note that in Iannucci v. Board of Supervisors of County of Washington ( 27 A.D.2d 346, 347, supra), the court says: "Districting results in a relatively simple apportionment, which should be readily adjustable * * * We would further note that with the districting system there should be little difference in the actual number of people in each district. * * * In all reapportionment proceedings on the county level the problem will be simplified and adjustable by resorting to a proper district plan, providing for electoral districts of equal population."
Wayne County, however, does not seemingly lend itself to such redistricting.
I realize that I have explored certain areas of this problem beyond the specific plan submitted to me in the county's answer, but I do so because of a sense of urgency which pervades the situation.
A judgment may be submitted on notice:
(1) declaring that the apportionment of the Wayne County Board of Supervisors is unconstitutional;
(2) that the proposed plan contained in Resolution No. 18 of said board is disapproved in its present form and that the same is unconstitutional;
(3) that the Board of Supervisors continue to function as such until a temporary or permanent plan is approved by this court; and
(4) that the court will retain jurisdiction of the action, with leave to any party to apply for further relief on or after a date to be fixed therein or upon settlement of the judgment.