Rockefeller v. Powers, 909 F. Supp. 863 (E.D.N.Y. 1995)

*864 Richard D. Emery, New York City, Burt Neuborne, Democracy Project, Brennan Center for Justice, New York University School of Law, New York City, for plaintiffs.

Lawrence A. Mandelker, Kantor, Davidoff, Wolfe, Rabbino, Mandelker & Kass, P.C., New York City, Joel Graber, Office of Attorney General, State of New York, New York City, Peter S. Kosinski, New York State Board of Elections, Albany, New York, Kevin C. Reilly, Attorney General State of N.Y., New York City, Robert J. Cimino, Suffolk County Attorney by Robert J. Cabble, Ass't. County Attorney, Hauppauge, New York, Nancy Ledy-Gurren, Ledy-Gurren & Blumenstock, L.L.P., New York City, for defendants.

KORMAN, District Judge.

The purpose of this memorandum is to set out in somewhat more detail the reasons I stated on the record on November 27, 1995, for granting a preliminary injunction in this civil rights action brought pursuant to 42 U.S.C. § 1983. Plaintiffs are registered Republican voters residing in the 1st, 3rd, 4th, 10th, 11th, 12th, 14th, 28th, and 29th congressional districts, all of whom intend to vote in the Republican Primary on March 7, 1996, at which ninety-three delegates to the Republican National Convention will be selected. Although their complaint has a broader scope, by order to show cause plaintiffs sought injunctive relief with respect to the "petitioning phase" of the Republican presidential primary in New York. The process involves the circulation of petitions containing the names of delegates committed to a particular presidential candidate whose name appears on the petition and on the ballot. Plaintiffs contend that the process mandated by New York law and Republican Party rules violates the Equal Protection Clause of the Fourteenth Amendment and the First Amendment of the United States Constitution.

The general framework for the New York Republican primary was decided on at the 1992 National Convention. At that time, the Convention determined that for the 1996 Convention the delegates would be allocated nationwide as follows: three delegates per congressional district, six additional delegates per state, and some number of "bonus" delegates for each state based on the state's success in supporting and/or electing Republicans.

On August 8, 1995, the Governor of New York signed into law a statute providing two alternative means by which a delegate candidate could have his or her name placed on the primary ballot. See 1995 N.Y.Laws Ch. 586. Under one alternative, delegates are required to obtain signatures from 5% or 1250 of the enrolled party members in their districts, whichever is less. See id. Ch. 586(2). Under the other alternative, the requirements are substantially less burdensome: 0.5% or 1000, whichever is less. See id. Ch. 586(3). While the legislation offers each political party the option of which alternative to choose, it has been the practice in recent presidential election years for the Legislature to pass an authorizing statute that merely codifies each party's preferred method of organizing its primary. What appear to be the options from which the parties may choose are in fact the choices they have already made. The option requiring 0.5% or 1000 is the one requested by the Democratic Party and the option requiring 5% or 1250 is the one requested by the Republican Party.

On October 31, 1995, the last day on which to exercise its option, the Republican State Committee formally notified the New York Board of Elections that it had chosen the alternative that would impose the 5% or 1250 requirements on candidates for delegate. Because of the uneven distribution of Republicans in New York State, the 5% or 1250 requirement imposes very different burdens on different districts. Of the State's thirty-one congressional districts, only six contain fewer than 25,000 Republicans and are therefore forced to comply with the full 5% requirement. The other twenty-five districts *865 receive the benefit of the 1250 cap, which corresponds to a percentage substantially below 5%. Indeed, the uncontested data submitted by plaintiffs reveal that in twenty of the thirty-one districts, the operative percentage is 1.54% or less and in sixteen it is 1.41% or less. In eight districts it is less than 1% and in the district with the most Republicans it is 0.79%. The following chart shows the number of enrolled Republicans in each congressional district ("CD") and the percentage actually required. The asterisks indicate districts that are too small to benefit at all from the 1250 cap.

CD   #/Republicans   #/Signatures    %
 3      158,097          1250        .79
22      153,824          1250        .81
27      148,319          1250        .84
 4      148,013          1250        .84
23      143,555          1250        .87
31      137,952          1250        .91
24      136,788          1250        .91
 1      129,111          1250        .97
25      124,260          1250       1.01
26      116,360          1250       1.07
19      114,008          1250       1.10
29      113,561          1250       1.10
28      112,797          1250       1.11
 2      112,442          1250       1.11
20      102,232          1250       1.22
13       88,350          1250       1.41
21       86,470          1250       1.45
 5       85,213          1250       1.47
18       85,062          1250       1.47
30       80,970          1250       1.54
14       52,555          1250       2.38
 9       52,511          1250       2.38
 7       46,303          1250       2.70
 8       33,247          1250       3.76
 6       26,457          1250       4.72
17       23,090          1155       5.00*
12       20,215          1011       5.00*
10       17,114           856       5.00*
15       15,561           778       5.00*
16       14,852           743       5.00*
11       11,814           591       5.00*

When viewed against these figures, the option chosen by the Republican Party can be more accurately described as requiring 1.54% or less in two-thirds of the congressional districts and a far greater percentage in most of the remaining congressional districts, including six where the percentage increases to 5% as the number of enrolled Republicans decreases. The highest percentage, which means the greatest burden, applies to districts with the fewest enrolled Republican voters. The burden of obtaining the requisite number of signatures is made even more difficult by the rule precluding an enrolled Republican from signing more than one petition and by a host of arcane technical requirements. Cf. Jenness v. Fortson, 403 U.S. 431, 442, 91 S. Ct. 1970, 1976, 29 L. Ed. 2d 554 (1971) ("The 5% figure is ... apparently somewhat higher than the percentage of support required to be shown in many States as a condition for ballot position, but this is balanced by the fact that Georgia has imposed no arbitrary restrictions whatever upon the eligibility of any registered voter to sign as many nominating petitions as he wishes.").

Plaintiffs' attack on the constitutionality of imposing such different burdens implicates two distinct lines of analysis. First, the burden imposed on plaintiffs in a particular district can be examined under the Due Process Clause to determine whether it places too great a restriction on ballot access, regardless of the burdens imposed in other districts. Second, the difference between the burdens placed on plaintiffs and voters in other districts must be analyzed under the Equal Protection Clause.^[1]

Regulations that impose burdens that are generally applicable to all voters are usually treated deferentially under a balancing test. As the Supreme Court held when reviewing a Hawaii law that prohibited voting for write-in candidates at primary and general elections:

the rigorousness of our inquiry into the propriety of a state election law depends upon the extent to which a challenged regulation burdens First and Fourteenth Amendment rights. Thus, as we have recognized when those rights are subjected to "severe" restrictions, the regulations must be "narrowly drawn to advance a state interest of compelling importance." But *866 when a state election law provision imposes only "reasonable, nondiscriminatory restrictions" upon the First and Fourteenth Amendment rights of voters, "the State's important regulatory interests are generally sufficient to justify" the restrictions.

Burdick v. Takushi, 504 U.S. 428, 434, 112 S. Ct. 2059, 2063, 119 L. Ed. 2d 245 (1992) (internal citations omitted); see also Unity Party v. Wallace, 707 F.2d 59, 61-62 (2d Cir. 1983).

In contrast, the cases dealing with requirements that treat similarly situated voters differently have imposed a more probing, less deferential level of scrutiny under the Equal Protection Clause. The analysis in these cases may assume that, if the restrictions in question were evenly applied, they would meet the balancing test articulated in Burdick. The critical inquiry in these cases is whether the difference in the burdens imposed can be justified. Where the difference is de minimis, only a rational basis is required. Where it "impairs the voters' ability to express their political preferences," heightened scrutiny is required. Illinois State Bd. of Elec. v. Socialist Workers Party, 440 U.S. 173, 184-85, 99 S. Ct. 983, 990-91, 59 L. Ed. 2d 230 (1979); see also Unity Party, 707 F.2d at 63.

A simple hypothetical illustrates the distinction between the two lines of analysis. If New York, which now opens the polls for sixteen hours in a general election, reduced the number to one hour, that would clearly constitute an undue burden on the right to vote, even if applied uniformly across New York State. If the number of hours was ten, that figure could arguably survive a Due Process/First Amendment challenge under the balancing test articulated in Burdick and other cases. The ten-hour voting day would, however, be subject to further scrutiny if it applied in New York City while the sixteen-hour day applied in the rest of New York State. The scrutiny would be heightened if it could be shown that the distinction was deliberately drawn to advantage one party or another. Simply stated, a restriction that would withstand a Due Process/First Amendment challenge, could nonetheless fail to survive a challenge made under the Equal Protection Clause, if not applied uniformly.^[2]

The issue presented here is whether the scheme the Legislature has put in place can survive the scrutiny mandated by the Equal Protection Clause. The Legislature has essentially said that any independent state interest in regulating ballot access could be satisfied by the 0.5% requirement contained in the alternative option it gave the Democratic and Republican parties and which the Democratic Party has adopted. I assume for present purposes that New York State may defer to the judgment of the Republican Party that more than 0.5% should be required as minimal support for a delegate candidate. Nonetheless, because the state "collaborates in the conduct of the primary, and puts its power behind the rules of the party," Gray v. Sanders, 372 U.S. 368, 374, 83 S. Ct. 801, 805, 9 L. Ed. 2d 821 (1963), the Republican Party is a state actor when it selects ballot access rules and it cannot promulgate rules that offend the Equal Protection Clause.

The leading case analyzing ballot access signature requirements under the Equal Protection Clause is Illinois State Bd. of Elec. v. Socialist Workers Party, 440 U.S. 173, 99 S. Ct. 983, 59 L. Ed. 2d 230 (1979), in which Justice Marshall examined a state ballot access law requiring candidates for statewide office to obtain 25,000 signatures but requiring candidates for citywide office to obtain signatures from 5% of the number of voters who voted at the previous election. As applied, the law required candidates for citywide office in Chicago to obtain nearly 36,000 signatures. In approaching this incongruous result, Justice Marshall noted that ballot access restrictions implicate the right to vote because "`voters can assert their preferences only through candidates or parties or both.' By limiting the choices available to voters, the State impairs the voters' ability to express their political preferences." Id. at 184, 99 S. Ct. at 990 (quoting Lubin v. Panish, 415 U.S. 709, 716, 94 S. Ct. 1315, 1320, 39 L.Ed.2d *867 702 (1974)). "When such vital individual rights are at stake," Justice Marshall continued, "a State must establish that its classification is necessary to serve a compelling interest." Id.

Justice Marshall acknowledged that candidates seeking a place on the ballot could be required to make "a preliminary showing of a `significant modicum of support'" in order to obtain a place on the ballot. Id. 440 U.S. at 185, 99 S. Ct. at 990 (quoting Jenness, 403 U.S. at 442, 91 S.Ct. at 1976). These requirements serve, at the very least, the state's interest "in avoiding confusion, deception, and even frustration of the democratic process in the general election." Jenness, 403 U.S. at 442, 91 S. Ct. at 1976. States may, therefore, seek "to assure that the winner of an election is `the choice of a majority, or at least a strong plurality, of those voting, without the expense and burden of runoff elections.'" Socialist Workers Party, 440 U.S. at 185, 99 S. Ct. at 990 (quoting Bullock v. Carter, 405 U.S. 134, 145, 92 S. Ct. 849, 857, 31 L. Ed. 2d 92 (1972)).

Nonetheless, in setting the modicum of support necessary to obtain a place on the ballot, "a State may not choose means that unnecessarily restrict constitutionally protected liberty." Id. 440 U.S. at 185, 99 S. Ct. at 990 (quoting Kusper v. Pontikes, 414 U.S. 51, 58-59, 94 S. Ct. 303, 308-09, 38 L. Ed. 2d 260 (1973)). States must instead choose "the least drastic means to achieve their ends." Id. Justice Marshall reasoned that requiring 36,000 signatures from candidates seeking office in Chicago was "plainly not the least restrictive means of protecting the State's objectives" because "[t]he Illinois Legislature has determined that its interest in avoiding overloaded ballots in statewide elections is served by the 25,000 signature requirement." Id., 440 U.S. at 186, 99 S. Ct. at 991. Given the lower requirement placed on candidates in a statewide election, where a far greater number of voters were eligible to sign petitions, Justice Marshall found "no reason, much less a compelling one, why the State needs a more stringent requirement for Chicago." Id. at 186, 99 S. Ct. at 991; see also Norman v. Reed, 502 U.S. 279, 293-94, 112 S. Ct. 698, 707-08, 116 L. Ed. 2d 711 (1992) (striking down Illinois law for same "constitutional flaw" at issue in Socialist Workers); cf. Bullock v. Carter, 405 U.S. 134, 148, 92 S. Ct. 849, 858, 31 L. Ed. 2d 92 (1972) (finding no justification for filing fees in party primary where "candidates for offices requiring statewide primaries are generally assessed at a lower rate than candidates for local offices").

The same equal protection analysis led the Seventh Circuit to strike down unequal ballot access requirements in Gjersten v. Board of Elec. Comm'rs, 791 F.2d 472, 474 (7th Cir. 1986). At issue there was a state law regulating the internal structure and election procedures of political parties. The law required candidates for the Cook County central committee from the city's "wards" to collect signatures from 10% of the party's primary electors in the ward, but required candidates from the suburban "townships" to obtain signatures from only 5% of their electors. See id. at 474. In striking down the added burden on the wards, the Seventh Circuit recognized that the ballot access law manifested

the explicit legislative judgment that a five-percent signature requirement was sufficient to protect the state's interest in an election for the same office in suburban Cook county. Although provided ample opportunity, the defendants presented no justification for the ten-percent requirement. In the absence of such a justification, the equal protection analysis developed in Socialist Workers Party does not require the court to further balance interests of the state against the rights of the candidates and voters.

Id. at 477. The holding in Gjersten makes it plain that whether the number of signatures is expressed as a percentage or as a raw number makes no difference in the analysis. The relevant comparisons are between the actual burdens imposed in different districts.

The Second Circuit has not yet had the occasion to apply the equal protection analysis developed by the Supreme Court in Socialist Workers Party and applied by the Seventh Circuit in Gjersten. It has, however, recognized that the analysis of ballot access restrictions under the Equal Protection *868 Clause must carefully examine the impact of those restrictions as applied:

An analysis of statutory language alone will not reveal the true dimension of a claimed denial of [voting] rights. We must also examine the nature, extent and likely effect of the law on the interest of those claiming to be fenced out by it. Only then can it be determined whether there exists a significant burden on fundamental rights or on a protected class such that heightened scrutiny is mandated. When the full scope of the obstacle to the exercise of these rights and the burden on the class is revealed, then the inquiry turns to the State's interests that purport to justify it.

Unity Party, 707 F.2d at 61-62 (internal citations omitted).

Under this approach, the ballot access requirements adopted by New York State and the Republican State Committee cannot be sustained because, as applied, they unreasonably discriminate based on the number of Republicans enrolled in a congressional district. In those districts in which it is easier to find a needle in haystack than an enrolled Republican, they require signatures from a percentage of enrolled Republicans three times greater than is required in the large majority of congressional districts, where the number of enrolled Republicans approaches or substantially exceeds 100,000 and the difficulty of finding them is greatly reduced.

The different burdens the requirements impose raise exactly the same question for the defendants here as in Socialist Workers Party and Gjersten: If the Republican Party, as a state actor, is willing to accept a relatively low burdenrepresented by a low percentage in most districts, then what is the justification for imposing a much higher burden represented by a 5% requirementin other districts where it is already more difficult to find potential petition signers? The only justification appears to be a desire to increase the advantage already enjoyed by the presidential candidate favored by the Party organization. The favored candidate has the benefit of an in-place organization that can gather the required signatures for delegates with relative ease. Differences in the burdens imposed in different districts are unlikely to have any significant practical effect on the ability of delegates pledged to the favored candidate to obtain the requisite number of signatures. In contrast, such differences may have a substantial impact on candidates who are independent of the Party organization. Because of shortages of time and resources, these candidates, engaging in national campaigns, may be forced to forego petitioning in many small districts and focus only on districts, or other states, where ballot access is relatively easy. In the congressional districts in New York with the fewest Republicans, the Party's candidate will be locked in and voters who would have voted for other candidates will be locked out.

The harm caused by this scheme is not limited to the particular voters whose right to vote in a particular election is rendered meaningless. Overly restrictive ballot access laws have the more generalized effect of stifling political expression because they tend to deprive candidates of a significant vehicle for disseminating ideas as well as attaining elective office. As Justice Marshall observed in Socialist Workers Party:

The State's interest in screening out frivolous candidates must be considered in light of the significant role that third parties have played in the political development of the Nation. Abolitionists, Progressives, and Populists have undeniably had influence, if not always electoral success. As the records of such parties demonstrate, an election campaign is a means of disseminating ideas as well as attaining political office. Overbroad restrictions on ballot access jeopardize this form of political expression.

Socialist Workers Party, 440 U.S. at 185-86, 99 S. Ct. at 990-91 (internal citations omitted). In the context of a major party's primary, candidates who are independent of the party organization are, in many ways, analogous to third parties at a general election. Although at the primary all candidates are members of the same party, they represent different political ideas and have different qualifications for national and party leadership. Such candidates, like third parties, have "undeniably had influence, if not always electoral success." Id. at 185-86, 99 S.Ct. at *869 990-91.^[3] Primaries in which they can meaningfully participate often provide the only means for changing the direction and assuming the leadership of the major political parties. As Theodore White observed in his history of the 1968 presidential election:

In hindsight one can see that [Nelson] Rockefeller was challenging the entire structure of his Partyas were [Eugene] McCarthy and [Robert] Kennedy on the other side; his resources were greater than those of the two Democrats, certainly in statehouse support if not in troops. But structures do not yield except under pressure. In politics, one challenges establishments in primaries, not elections. John F. Kennedy had done so in 1960, McCarthy was doing so in 1968, Rockefeller himself had made a near thing of it in 1964.

Theodore H. White, The Making of the President 1968, at 230 (1969) (emphasis added).

If discriminatory requirements prevent "independent" candidates from obtaining the requisite number of signatures to place on the ballot delegates pledged to them, then the primary becomes little more than a statesponsored endorsement of the candidate of the party leadership. While the New York State Republican Party may choose an alternative method of delegate selection that would effectively permit its leadership to select delegates to the Republican National Convention, that objective cannot justify discriminatory ballot access requirements when the Party has chosen to have its delegates selected in a primary election conducted under the auspices of the State. As Professor Tribe has summarized the applicable law:

[W]hile the Supreme Court has never required a state to hold elections for any particular office, "once the franchise is granted to the electorate, lines may not be drawn that are inconsistent with the Equal Protection Clause of the Fourteenth Amendment." Such "inconsistent lines" may be found in state laws effecting unequal absolute deprivations on the right to vote, or in the state laws responsible for only marginal, relative inequalities in the distribution of the right to vote.

Laurence H. Tribe, American Constitutional Law § 16-11, at 1461-61 (2d ed. 1988) (foot-notes omitted) (quoting Harper v. Virginia Bd. of Elec., 383 U.S. 663, 665, 86 S. Ct. 1079, 1080, 16 L. Ed. 2d 169 (1966)).

Under any teststrict scrutiny, balancing of interests, or rational basisthe rules at issue here, which have the practical effect of limiting voters' choices and curtailing political expression, cannot be sustained. The only legitimate objective that they further requiring a minimum modicum of support could be achieved in a far more rational and less restrictive way by a scheme that did not unjustifiably impose such different burdens in different districts.

While uniform burdens could result in higher access requirements rather than lower ones, the more uniform the burden the less likely it is to be overly burdensome. Those in the majority are rarely willing to achieve equality by submitting themselves to the burdens they readily place on small minorities. Accordingly, it is unlikely that the Republican State Committee would impose a burden as high as 5% if it affected more than a small minority of districts in which Republicans constitute a small percentage of the voting population. This reality reflects the full dimension of the equal protection guarantee.

Because challenges to ballot access restrictions often occur at the eleventh hour, the Court of Appeals for the Second Circuit has not addressed the possible justifications that a state may have for adopting signature requirements that, when applied, do not fall uniformly on all voters and candidates. There are, however, four thoughtful opinions, relied upon by defendants here, that address this issue in related contexts. See Hewes v. Abrams, 718 F. Supp. 163 (S.D.N.Y.) (Haight, J.), summarily aff'd, 884 F.2d 74 (2d Cir. 1989); Mahon v. Abrams, No. 88 Civ. 1745 (S.D.N.Y. Mar. 21, 1988) (Sand, J.), aff'd without op., 847 F.2d 835 (2d Cir. 1988); *870 McGee v. Board of Elec. of New York, 669 F. Supp. 73 (S.D.N.Y.1987) (Weinfeld, J.); Meyerson v. Board of Elec. of New York, No. 87 Civ. 6168 (S.D.N.Y. Sept. 3, 1987) (Leisure, J.). While each of these cases is distinguishable from the present case, they warrant some discussion here.

Both McGee and Meyerson rejected challenges to the portion of New York's ballot access scheme that required candidates for county offices in Bronx County to obtain signatures from 5000 enrolled voters but required candidates for county offices in Nassau and Suffolk Counties to obtain only 2000 signatures.^[4] Even though the population of all three counties was roughly the same, the disparity was considered to be rational because the higher population density in the Bronx made it easier to obtain 5000 signatures in the Bronx than in Nassau and Suffolk Counties. The requirement of 2000 signatures in the two less densely populated counties simply equalized the ballot access burdens. As Judge Leisure observed in Meyerson: "In light of this difference in population density, it is wholly rational for the New York Legislature to conclude that two thousand signatures in Nassau County represents approximately the same degree of support for a candidate as five thousand signatures in Bronx County due to the greater ease with which signatures may be gathered in the Bronx because of greater population density." Meyerson, No. 87 Civ. 6168, at 14.

Both Meyerson and McGee recognize that, in comparing different burdens, the most important consideration is not the absolute number or the percentage of signatures required, but the extent of the actual, practical burden placed on candidates and voters. Under certain circumstances, raising or lowering the total number of signatures and/or the percentage of signatures required in certain districts equalizes rather than amplifies differences among districts. This rationale would not, however, explain the system reflected by the record here because it deliberately places a heavier burden on districts where the fewest number of enrolled Republicans reside.

In Mahon, Judge Sand rejected a challenge to the 1988 equivalent of the New York Session Law challenged by plaintiffs here. He found that plaintiffs had raised a fair basis for litigation but denied the motion for a preliminary injunction. To the extent the denial rested primarily on the untimeliness of the plaintiffs' lawsuit, which was instituted after the petitioning phase and sought access to the ballot for candidates in districts where they had gathered no signatures, the case is clearly distinguishable. To the extent Judge Sand preliminarily addressed the case on its merits, his opinion reflects my first impression that a numerical cap of 1250 that eases the requirements in the districts that would be most heavily burdened by a 5% signature requirement is not unconstitutional on its face. I also agree with Judge Weinfeld that a state can decide that, although a certain number of signatures "is the general requirement for showing the modicum of support to be placed on the ballot, a lesser showing will be allowed in specific contests." McGee, 669 F. Supp. at 80. In practice, however, the 1250 cap applies not only to the largest districts, but to nearly all of the districts. The generally applicable modicum of support required is, therefore, not 5%, but some percentage based on the 1250 requirement. In two-thirds of the districts the percentage is 1.54 or less. Under the circumstances, the cap has created an exception that swallows almost all of the rule and renders irrational what remains of it.

Judge Haight's opinion in Hewes is more troubling. In that case, the Right to Life Party's candidate for mayor of New York City objected to a ballot access law requiring candidates to obtain signatures from 5% of the enrolled members of their respective parties or 10,000, whichever was less. In New York City, the law in practice required Democrats to obtain signatures from 0.5% of their party, Republicans to obtain signatures from 2.4% of their party, and Right to Lifers to obtain the full 5% because their party had *871 only 6,000 members. See Hewes, 718 F. Supp. at 164. The plaintiff sought an injunction requiring that the signature requirement for all parties be reduced to 0.5%, the lowest percentage required of any party. Judge Haight denied the injunction and dismissed the complaint. He held that the statute needed only to survive rational basis review and that it did so in part because the cap was a legitimate means of avoiding requiring large parties to gather an unreasonably high number of signatures. See id. at 167.

I agree that once a state has selected a percentage that represents the appropriate modicum of support it may institute a cap to liberalize ballot access. What I do not agree with is the inference, accepted by Judge Haight in Hewes, that 5% was the modicum of support actually required by the ballot access scheme at issue. The scheme in Hewes required only primary candidates from the established minor parties to obtain the signatures of 5% of the enrolled party members, while it required less than half that percentage from the Republican Party and one tenth of that from the Democratic Party.^[5] Moreover, if no Right to Life candidate met the disproportionately high burden and the Right to Life Party was left without a nominee, the effect would be to eliminate an independent party from the ballot.

In any event, Hewes is distinguishable because it addressed signature requirements that differed as between political parties. These parties were of vastly different sizes. A major party's candidate for mayor in New York City, despite the low percentage requirement, would still be required to gather more signatures than there were voters in the Right to Life Party. Given the advantage that such a candidate would enjoy over candidates from smaller parties at the general election, a state might rationally require small party candidates to show stronger intra-party support than candidates from larger parties in order to be placed on the primary ballot. I need not now consider whether this reasoning permissibly protects the ballot from unnecessary clutter or impermissibly limits voters' choices. I need only note that the same rationale would not explain requirements that are the same for everyone within a district but vary so significantly from district to district for the same party position.

The foregoing discussion reflects my conclusion that plaintiffs have established a likelihood of success on the merits. Nonetheless, I decline to foist on the Republican Party the relief requested by plaintiffs, namely, an order requiring no more than 0.79% of the enrolled Republicans to place a delegate candidate on the ballot. This is the percentage required under the current scheme in the congressional district with the largest number of enrolled Republicans. The underlying purpose of the Legislature was to grant the Republican Party latitude in fixing the minimum modicum of support it deemed appropriate. According the same deference to the desire of the Republican Party to fix whatever number it deems appropriate, any relief should permit the Party to define a "modicum of support" at the highest percentage consistent with the Equal Protection Clause. Consistent with the analysis in Socialist Workers Party, I determined that level based on the minimum percentage of signatures required in a majority of congressional districts. The data furnished to me by the plaintiffs, which no one has disputed, demonstrate that if the requirement was 1.41% or 1250, 16 of 31 districts would have to meet the percentage requirement and 15 of 31 would benefit from the cap. The 1.41% requirement could then rationally be characterized as a generally applicable rule and the 1250 cap as an exception that liberalizes ballot access.

The scope of my order is necessarily limited. Because this case was not brought as a class action I have only the specific plaintiffs before me. The plaintiffs include voters in four districts that would benefit from a 1.41% requirement: the 10th, 11th, 12th, and 14th districts. Voters from the other districts *872 that would benefit are not represented here and plaintiffs here have no standing to object to their potential lack of choices on primary day. I, therefore, ordered the defendants not to require otherwise valid signatures of more than 1.41% of enrolled Republicans from delegates seeking access to the ballot in the Republican primary in the 10th, 11th, 12th, or 14th districts.

I have granted relief because plaintiffs have shown a likelihood of success on the merits and of irreparable harm on the day of the primary, but I have also considered the public interest, which is always "a factor to be considered in the granting of a preliminary injunction." Carey v. Klutznick, 637 F.2d 834, 839 (2d Cir.1980). The relief I have granted threatens no harm to the public interest because, if I am correct that the ballot access restrictions chosen by the Republican Party have the effect of reducing the choices available to voters in many districts, the only possible effect of my order may be to increase the choices available to voters in the affected districts. These choices will still be limited to candidates who have shown the level of support that the party itself considers sufficient to justify ballot access in most other districts and who are able to overcome other burdens that are not at issue here. Indeed, for this reason, even if this were a class action, the preliminary injunction would have either little or no effect in two-thirds of the congressional districts. Because there is not even a rational basis for requiring a far greater level of support in the minority of congressional districts with the fewest enrolled Republicans, the public interest is not disserved by granting injunctive relief.

[1] By distinguishing these two analytical inquiries, I do not mean to obscure the fact that the categories can overlap and are often intertwined or left undistinguished in the relevant case law. See Norman v. Reed, 502 U.S. 279, 288 n. 8, 112 S. Ct. 698, 705 n. 8, 116 L. Ed. 2d 711 (1992).

[2] The reverse is also true. A uniformly applied ballot access restriction could survive an equal protection challenge and nonetheless fail to withstand a due process challenge.

[3] Senator Eugene McCarthy's near victory in the 1968 New Hampshire primary and his anticipated victory in the Wisconsin primary played a major role in bringing down a sitting President of his own party over an issue of national policy. This story is chronicled in Theodore H. White, The Making of the President 1968, at Chs. 3-4 (1969).

[4] The overall scheme imposed a signature requirement of 5% of enrolled voters for all county offices. The law then capped the number of signatures required at 5000 for counties in New York City but at only 2000 for all other counties with more than 250,000 inhabitants.

[5] Even if the different requirements at issue in Hewes violated the Equal Protection Clause, I would not have granted the remedy he sought. Under the reasoning I discuss below, I would have set the percentage as high, but no higher, than would be needed to subject a majority of parties to the same general rule.