Reset A A Font size: Print

United States Court of Appeals, First Circuit.

LIBERTARIAN PARTY OF NEW HAMPSHIRE, Plaintiff, Appellant, v. WILLIAM M. GARDNER, New Hampshire Secretary of State, in his official capacity, Defendant, Appellee.

No. 15-2068

Decided: December 02, 2016

Before Lynch, Thompson, and Kayatta, Circuit Judges. William E. Christie, with whom Shaheen & Gordon, P.A., Gilles R. Bissonnette, and American Civil Liberties Union of New Hampshire were on brief, for appellant. Laura E. B. Lombardi, Senior Assistant Attorney General, New Hampshire Department of Justice, with whom Stephen G. LaBonte, and Joseph A. Foster, New Hampshire Attorney General, were on brief, for appellee. Mark W. Freel and Locke Lord LLP on brief for Libertarian Association of Massachusetts, Green-Rainbow Party, United Independent Party, Maine Green Independent Party, and Moderate Party of Rhode Island, amici curiae in support of appellant. Bruce I. Afran, Mark R. Brown, and Oliver B. Hall on brief for the Center for Competitive Democracy, amicus curiae in support of appellant. Gordon J. MacDonald, Holly J. Barcroft, and Nixon Peabody LLP on brief for the Republican National Committee, amicus curiae in support of appellee.

Like most states, the State of New Hampshire has long required that political parties seeking to have their nominees listed on statewide election ballots first demonstrate a sufficient modicum of support among registered voters. New Hampshire law deems that a party has made such a demonstration if, in the most recent prior statewide election, one of its candidates received at least four percent of the statewide vote for Governor or United States Senator. N.H. Rev. Stat. Ann. § 652:11. Otherwise a party need submit nomination papers signed by a number of registered voters at least equal to three percent of the total votes cast in the most recent state general election. Id. § 655:42(III).

In this lawsuit, the Libertarian Party of New Hampshire (“LPNH”) focuses on the time period during which New Hampshire law allows parties to gather nomination signatures and submit nomination papers. Prior to 2014, this period ran roughly twenty-one months from the prior November election to early August of the pertinent election year.1 In 2014, New Hampshire reduced this time period to a bit more than seven months (hereinafter “seven months”) by delaying the start date to January 1 of the pertinent election year. See 2014 N.H. Laws § 29:1 (“HB 1542”) (codified at N.H. Rev. Stat. Ann. § 655:40-a) (effective July 22, 2014). In the wake of this shortening of the time period within which it could gather nomination signatures, LPNH promptly filed this lawsuit claiming that the new restriction violated its rights under the First and Fourteenth Amendments to the United States Constitution. LPNH now appeals the district court's summary judgment decision to the contrary. For the following reasons, we affirm.

I. Background

LPNH qualified for the ballot in New Hampshire as a “political party” under state law in 1992, 1994, and 1996, based on the electoral performance of its gubernatorial candidates in prior elections. In 2000, LPNH managed to gather enough qualifying signatures to secure a place on the statewide ballot for all of its nominees. But no LPNH statewide candidate secured as much as four percent of the vote, and LPNH offers no evidence that any of its local candidates fared better. In 2002, 2004, 2006, 2008, and 2010, LPNH did not gain party access to the statewide ballot. LPNH offers no evidence suggesting that New Hampshire law posed any unreasonable impediment to qualifying during those years. Apparently, some of LPNH's candidates, including its presidential candidate in 2004, also sought access through the individual nomination process, whereby an individual who gathers just 3,000 signatures is listed on the statewide ballot.

In 2012, LPNH itself again reached the New Hampshire general ballot via the nomination papers route. The record of that successful effort reveals that the gathering of signatures on nomination petitions is largely a paid, professional undertaking. LPNH tells us (and defendants do not dispute) that “LPNH, like other minor political parties seeking ballot access or advocacy groups seeking to certify a ballot question, rel[ies] on outside professional petitioners to collect signatures.” LPNH retained one local paid “petitioner” to gather signatures at a fee of $1 per signature. LPNH also apparently paid a national outfit $2 per signature to gather roughly 13,100 signatures during August and September of 2011 plus roughly 1,700 signatures on a single day in July 2012.2 It appears that unpaid volunteers also gathered roughly 3,000 to 4,000 additional signatures. LPNH ultimately spent approximately $40,000 to gather 19,000 signatures in 2012, overshooting the mark (of roughly 14,000 qualifying signatures) because not all signatures submitted were likely to be certified.

Getting nominating signatures in 2012 turned out to be easier than getting votes. LPNH's gubernatorial nominee received 2.8% of the vote, its presidential nominee received 1.2%, and, in a state with hundreds of state legislative races, LPNH recruited only ten other candidates, just one of whom reached ten percent of the vote. LPNH apparently made no effort to get on the statewide ballot in 2014.

Confronted with the shortened signature-gathering window in 2016, LPNH decided to “put all party-petitioning efforts-including fundraising for those efforts-- on hold until this litigation ends, as the outcome of this litigation would dictate whether [LPNH] would even go through the party-petitioning process during the 2016 general election.“ LPNH estimates that compliance with the new law would increase the cost of gathering sufficient signatures, because paid petitioners generally charge more during an election year. The current chairperson of LPNH testified that an election year paid-petition drive “would probably be a $50,000 effort,” that is to say, about $10,000 more than LPNH spent in 2012.

LPNH alleges that the shortened window for gathering signatures “facially” violates the Fourteenth Amendment's guarantee of equal protection and the freedom of association secured by the First Amendment. After discovery, the parties agreed that the case could be decided on cross motions for summary judgment. After holding a hearing to gather more evidence on LPNH's prior efforts to secure ballot access, the district court agreed that there were no genuine disputes of material fact. Neither party challenges this conclusion on appeal. Rather, they limit their arguments to contesting whether the district court made an error of law in concluding that the undisputed facts did not establish a violation of LPNH's asserted constitutional rights because HB 1542 “imposes only a reasonable burden on ballot access that is outweighed by the State's interest in avoiding ballot clutter.” Libertarian Party of N.H. v. Gardner (“Gardner“), 126 F. Supp. 3d 194, 210 (D.N.H. 2015). The parties agree that our review of this conclusion should be de novo.

II. Analysis

A. Mootness?

Even if we were to decide that signatures gathered prior to January 1, 2016, must be counted by the state, LPNH offers no evidence that it gathered any such signatures or, for that matter, that it gathered any material number of signatures this election cycle at all. And the 2016 election itself is now history. We must therefore ask whether this case is moot. See Barr v. Galvin, 626 F.3d 99, 104 (1st Cir. 2010).

We conclude that it is not. Statewide elections will regularly occur after this year. LPNH has a demonstrated (albeit episodic) record of seeking statewide ballot access in New Hampshire and elsewhere. In view of that record, we give the “benefit of the doubt” to LPNH's continuing practical interest in the resolution of its legal claim. Id. at 106. Moreover, a new suit to allow signature gathering in the pre-election years of 2017 and 2019 would need to start pretty much now anyhow to avoid the same type of arguable mootness. We thus conclude that there exists a sufficient probability that LPNH's challenge to New Hampshire's existing ballot-access regime is likely to reoccur, and is not now unripe. Jurisdiction therefore lies. See FEC v. Wis. Right To Life, Inc., 551 U.S. 449, 462 (2007) (noting that federal courts may hear otherwise moot controversies that are uniquely capable of repetition yet will often evade review); Barr, 626 F.3d at 106 (“[A]lthough the [capable of repetition yet evading review] exception has been applied frequently in election-related cases ․ not every election case fits within its four corners.”).

B. Merits


As we have described, LPNH objects to the effect of HB 1542 “not in the context of an actual election, but in a facial challenge.” Wash. State Grange v. Wash. State Republican Party, 552 U.S. 442, 449 (2008). Facial challenges to state laws are difficult to mount. See id. at 449–51. The Supreme Court has articulated two formulations of the standard for assessing facial challenges to statutes. See United States v. Stevens, 559 U.S. 460, 472 (2010) (“Which standard applies in a typical case is a matter of dispute that we need not and do not address ․”); see also Hightower v. City of Boston, 693 F.3d 61, 77 n.13 (1st Cir. 2012) (discussing the dual extant standards without “resolv[ing] the issue”). In United States v. Salerno, 481 U.S. 739 (1987), the Court held that a facial challenge can only succeed where the plaintiff “establishes that no set of circumstances exists under which the Act would be valid.” Id. at 745. In Washington State Grange, a facial challenge to a state ballot-access provision, “the Court noted that 'some Members of the Court have criticized the Salerno formulation,' but that 'all agree that a facial challenge must fail where the statute has a “plainly legitimate sweep.” ' ” Hightower, 693 F.3d at 77 n.13 (quoting Wash. State Grange, 552 U.S. at 449). We rely on this latter formulation of the standard.

The difficulty in mounting a facial challenge to a state law arises most notably in challenges to laws that, by their terms, leave room for discretion in their application, meaning a state official could “accord the law a limiting construction to avoid constitutional questions.” Wash. State Grange, 552 U.S. at 450. The state law at issue here, however, has no such play in the joints. It fixes its various requirements objectively and specifically, with the largely immaterial exception of the precise day in early August when papers must be submitted. It is also one-size-fits-all, for all comers. In short, the components of the burden it imposes are defined by its facial terms, not by any anticipated exercise of discretion in its application.

The facial nature of LPNH's challenge nevertheless still calls for some skepticism because it comes without the benefit of an actual demonstration of a failed effort to get on the ballot under the requirement being challenged. We are mindful that “[c]laims of facial invalidity often rest on speculation,” id., and we are reluctant to engage in the “premature interpretation of statutes on the basis of factually barebones records,” id. (quoting Sabri v. United States, 541 U.S. 600, 609 (2004)).


The Supreme Court first considered a constitutional challenge to state-enacted ballot-access regulations in Williams v. Rhodes, 393 U.S. 23 (1968). Such challenges implicate “two different, although overlapping, kinds of rights--the right of individuals to associate for the advancement of political beliefs, and the right of qualified voters, regardless of their political persuasion, to cast their votes effectively,” thereby triggering scrutiny under both the First and Fourteenth Amendments. Id. at 30. The Ohio ballot-access regulations challenged in Williams required a new party seeking to place its candidates on the statewide ballot to file, by the February preceding the November election, nominating petitions signed by a number of voters equal to at least fifteen percent of the total vote cast in the most recent gubernatorial election. The regulations further required that the party establish a party organization, and conduct primaries or national conventions. All in all, the Court found that the regulations “tend[ed] to give [the Republicans and Democrats] a complete monopoly,” id. at 32, closing off the ballot to a party that actually gathered 450,000 nominating signatures. Id. at 26. Noting that, at that time, “[f]orty-two states require[d] third parties to obtain the signatures of only 1% or less of the electorate” with no apparent problems, id. at 33 n.9, the Supreme Court rejected Ohio's stated justifications.

Three years later, the Court considered a similar challenge to Georgia's ballot-access laws. See Jenness v. Fortson, 403 U.S. 431 (1971). In brief, Georgia granted a party's nominees automatic access to the ballot only if a candidate of the party “received 20% or more of the vote at the most recent gubernatorial or presidential election.” Id. at 433. Otherwise, a candidate for an office needed to gather within 180 days signatures of five percent of the total number of voters eligible to vote in the prior election for that office, all of whom were eligible to sign such nominating petitions. Id. Pointing to what it regarded as “surely an important state interest in requiring some preliminary showing of a significant modicum of support before printing the name of a political organization's candidate on the ballot,” id. at 442, the Court found in Georgia's regime “nothing that abridges the rights of free speech and association secured by the First and Fourteenth Amendments,” id. at 440. It also concluded that the plaintiffs' “claim under the Equal Protection Clause of the Fourteenth Amendment fares no better.” Id.

Another three years later, the Court considered California's claim that its five-percent requirement was similarly valid. Storer v. Brown, 415 U.S. 724 (1974). In California, though, the party seeking ballot access could gather signatures only from persons who did not vote in a primary conducted by the major parties. As the Court observed in remanding the case for further factfinding, one would need to gather more than five percent of this restricted subset of eligible voters in order to equal five percent of the entire set of voters in the previous election. Id. at 739; see also Am. Party of Tex. v. White, 415 U.S. 767, 784 (1974) (finding that requiring signatures totaling one percent of the vote cast in the previous gubernatorial election to be gathered from only those who did not vote in a party primary “falls within the outer boundaries of support the State may require”).

On the two occasions when the Supreme Court has actually struck down five-percent requirements where the pool of those who could sign was not substantially restricted, it has done so not because it determined a five-percent requirement by its nature imposed too significant a burden, but because the state itself recognized it could achieve its goals without so high a signature requirement. In Illinois State Board of Elections v. Socialist Workers Party, 440 U.S. 173 (1979), for example, the Court struck down an Illinois signature requirement for ballot access in political subdivision elections that exceeded the signature requirements for ballot access in statewide elections. See id. at 186–87. It did the same in Norman v. Reed, 502 U.S. 279 (1992), explaining that Illinois's requirement had “unconstitutional breadth” because “a prerequisite to establishing a new political party in ․ multidistrict subdivisions [was] some multiple of the number of signatures required of new statewide parties.” Id. at 293.

Neither the Supreme Court nor any circuit court has struck down a statewide ballot-access regime on the grounds that a signature requirement of five percent (or less) is too much, or that six months (or more) is too little time within which to gather the signatures from a pool of substantially all voters. See, e.g., Rainbow Coal. of Okla. v. Okla. State Election Bd., 844 F.2d 740, 741–42, 747 (10th Cir. 1988) (finding a law requiring 45,497 signatures, or five percent of the number of voters in the previous election, in one year a “relatively high signature requirement” but not impermissible); Libertarian Party of Fla. v. Florida, 710 F.2d 790, 794 (11th Cir. 1983) (finding a law requiring 144,492 signatures, or three percent of the state's registered voters, in 188 days “not impermissibly burdensome”).

The New Hampshire combination of percentage and timeframe, while likely more demanding than the laws in many states, is markedly less burdensome than the regime at issue in Jenness. Ballot access under the actual New Hampshire requirement of three percent in seven months required approximately 2,114 valid signatures per month. By contrast, if applied to New Hampshire, the Georgia requirements of five percent in 180 days approved in Jenness would have required LPNH to gather well more than 4,111 valid signatures per month to gain ballot access in 2016.3 Moreover, the nominating petition in Georgia secured a place on the ballot only for the nominated candidate, not for a party's whole slate. Jenness, 403 U.S. at 432.

LPNH argues that Jenness blessed only Georgia's five-percent requirement, not its 180-day window. It is true that the specific challenge in Jenness focused on the five-percent requirement. But it is also true that in distinguishing Williams, Jenness contrasted “the totality of the Ohio restrictive laws taken as a whole,” id. at 437 (quoting Williams, 393 U.S. at 34) with Georgia's entire “statutory scheme,” id. at 438. Similarly, and as we have already noted, the differing results in Jenness and Storer hinged precisely on consideration of the manner in which the five-percent requirement need be satisfied. Notably, Jenness expressly described Georgia as allowing a nominee to “seek, over a six months' period, the signatures of 5% of the eligible electorate ․” Id.; see also Developments in the Law -- Elections, 88 Harv. L. Rev. 1121, 1143 (1975) (Jenness “specifically endorsed a comprehensive approach to evaluating the constitutionality of a state's ballot access restrictions.”).

It therefore follows that LPNH's challenge to New Hampshire's three-percent-within-seven-months requirement must fail unless either Jenness's approval of a more broadly applicable five-percent-within-180-days requirement is no longer good law, or this case is distinguishable on other grounds. LPNH does not argue that Jenness is no longer good law. Rather, it argues that, for a variety of reasons, this case presents materially distinguishing facts. We consider these reasons in turn.

First, LPNH correctly notes that, in sustaining Georgia's five-percent-within-180-days requirement, the Court in Jenness observed that Georgia's rules did not otherwise impose any “suffocating restrictions whatever upon the free circulation of nominating petitions.” 403 U.S. at 438. The Court similarly noted that “Georgia has imposed no arbitrary restrictions whatever upon the eligibility of any registered voter to sign as many nominating petitions as he wishes.” Id. at 442. New Hampshire voters, on the other hand, may sign only one petition per election cycle. This restriction, LPNH reasons, materially distinguishes the New Hampshire ballot-access regime from the Georgia regime sustained in Jenness. LPNH, though, offers no evidence of losing even a single signature opportunity due to that restriction, or that any other party even circulated petitions for party access or will be doing so. Moreover, even allowing for such a possibility, the incremental impact on LPNH's efforts would seem to be far less than the burden of gathering signatures from as large a percentage of the electorate as was sustained in Jenness.4

LPNH also argues that the New Hampshire regulations do not really allow seven (or even six) months as a practical matter because the time period includes the winter months when, LPNH says, it is much more difficult to find people in groups outside. Of course, hundreds of thousands of New Hampshire citizens streamed in and out of polling stations in the presidential primaries held last February. And even if we were to view seven New Hampshire months as the equivalent of five Georgia months, the cumulative burden in New Hampshire would still be less than that sustained in Jenness because of New Hampshire's low percentage requirement.

The record LPNH has compiled in this case also makes clear that the marginal impact of the difference between five, six, or seven months is quite small. With $50,000 in hand, a party can obtain 20,000 signatures within two or three months (or even seven days if we use LPNH's July 2012 experience as a gauge). Or, with the equivalent of twenty fulltime volunteers, a party can apparently gather enough signatures in a single month.5 Even reducing these experience-based estimates by half, one is still left to conclude that, once one has more than several months--and certainly five to seven months--the driving variables in determining success in gathering 20,000 signatures are money and volunteers, not time within which to gather signatures. The difference between seven months and more than seven months is therefore largely immaterial as long as the time for raising money and recruiting volunteers is not reduced.

Reframing the issue, LPNH next attempts to pitch at least a portion of its challenge as a challenge only to the January 1, year-of-election start date mandated by HB 1542. The logic here seems to be that even if the combined burden of signature amount and gathering time is not impermissible in the abstract, a limitation on when the gathering time must occur may be invalid. We accept the premise that the precise point on the calendar during which the signature-gathering window may remain open can pose a problem independent of the size of the window. For example, early dates may precede the events that give rise to third-party support, as in the presidential elections of 1892, 1912, 1924, and 1968, see generally Alexander Bickel, Reform and Continuity, The Electoral College, The Convention, and the Party System 87-88 (1968), as well as 1980, see Anderson v. Celebrezze, 460 U.S. 780, 790 (1983). In Anderson, the Supreme Court set aside an eight-month qualifying date requirement that ended in March of the pertinent election year as too early for just this reason. See id. at 791–92.

LPNH makes no claim that New Hampshire requires it to submit its nominating papers too soon. To the contrary, it argues that New Hampshire requires third parties to wait too long before they can gather signatures. The case for challenging a start date requirement as too late cuts against some of the reasons for striking down early end-date requirements. Moving ballot-access efforts closer to an election encourages third parties to first work within one of the two more established parties, advancing what the Supreme Court has called “the State's interest in preserving party harmony.” Id. at 805 (citing Bickel, supra, at 87-88). Similarly, as New Hampshire notes, signatures gathered nearer the end date are more likely to represent the views of the signers as of the end date than those signed well in advance. That being said, New Hampshire's January 1 start date is a requirement not imposed by the Georgia regime at issue in Jenness, so we need consider whether it renders the New Hampshire ballot-access regime more burdensome than the Georgia regime the Supreme Court there approved.

LPNH contends that the January 1 start date burdens a party seeking ballot access by requiring that it spend time on petitioning during the election year, rather than before the election year, thereby reducing the time available for “electioneering” when it is most effective. LPNH, however, simultaneously stresses that the gathering of nomination signatures is principally performed by paid professionals who are not otherwise engaged in the actual campaigns. The start date, in turn, does not apply to fundraising activities, hence LPNH can raise the funds to pay professional signature gatherers any time it wants.

LPNH did use some volunteer time in 2012 to gather signatures. That time seems to equal only two and one-half months of volunteer time, more or less. We have no doubt that the petition process itself takes also some of the time of campaign officials even if they do not themselves gather signatures. The record, however, provides no quantification suggesting that any such effort materially detracts from “electioneering,” especially keeping in mind that there is no claim that the candidates' efforts are impeded. All in all, this is small beer. According to LPNH's own estimates, the entire “lost” election year effort could be regained for $5,000 paid to professional signature gatherers. And the effort is not lost, but simply time-shifted so that its value is arguably reduced but certainly not eliminated.

Somewhat inconsistently, LPNH argues that the January 1 start date sidelines its attempts to solidify support in years when no election occurs by forcing it to sit idly by instead of strengthening its infrastructure by collecting signatures during those years. This argument relies on the false premise that if a political organization is not actively soliciting signatures for nomination papers, there is nothing it can do to attract supporters and strengthen its organizational apparatus. To the contrary, as the district court observed, “[e]ven with the January 1 start date in place, LPNH remains free to plan its election-year petition drive[,] ․ recruit volunteers[,] ․ [and] raise funds for the drive during the off year that it can then spend on paid petitioning during the election year.” Gardner, 126 F. Supp. 3d at 204–05.

LPNH also claims that the January 1 start date deprives it of the chance to gather signatures at city off-year elections in November, fairs in the fall, and other such similar outside events before the cold New England winter begins. As we noted above, with New Hampshire voters turning out in droves for party primaries in February, we doubt New Hampshire citizens hibernate as much as LPNH implies. In any event, New Hampshire traditionally holds town meetings in the spring, followed promptly by farmers' markets, parades, and ball games in the spring and summer months. In the words of a former LPNH gubernatorial candidate, spring gun shows are fruitful sources of signatures, and July 4th provides each year “the political mecca of parades.” LPNH itself gathered 1,700 signatures on a single day in July of 2012. All in all, we reject as greatly overstated LPNH's contention that the weather and calendar render the seven-month, January-to-August timeframe so unsuitable for signature gathering that we should regard it as more constraining than the 180-day window sustained in Jenness.

Relatedly, LPNH contends that the start date forces it to “stand still” in the year prior to the year of the election. If this is simply another way of saying that it has to hold off on signature gathering until after January 1, we have just addressed that argument. If, instead, LPNH is saying that the new start date somehow restricts it from other campaign undertakings prior to January 1, it seems instead to do just the opposite. The new start date eliminates signature gathering prior to January 1, thereby increasing resources available for other activities during that time. Either way, this argument adds nothing.

LPNH next argues that candidates will not be confident that they will make the ballot if the late start date prevents LPNH from collecting all the signatures it needs by June 30 of the election year. There is no evidence, though, that LPNH has ever placed any premium on concluding before June 30. The major parties' nominees for governor and state senate are not even selected until September. And there is no reason that the professional signature gatherers cannot conduct all their work well before June 30 if LPNH wants. Again, the controlling factor would seem to be money, not time.

But, says LPNH, the paid gatherers themselves cost more during an election year. The evidence of this is skimpy, to say the least, consisting solely of a back of the envelope guess of a $10,000 delta by an LPNH official. This is too little in both foundation and magnitude upon which to rest a facial challenge to New Hampshire's law.

LPNH argues, finally, that the January 1 start date delays any effort to determine what it calls the “verification rate,” meaning the percentage of signatures that are verified as accurate. This information apparently helps estimate how many new signatures need be gathered. LPNH makes no attempt to spell out what this actually means, or whether the resulting cost (if any) is already included in its $10,000 estimate.

Collectively, these arguments that the change in start date by itself imposes a substantial burden fail to convince us that New Hampshire's ballot-access regime is as burdensome as-much less more burdensome than--the Georgia regime upheld in Jenness. To the extent they represent any burden for a political party that has a sufficient modicum of support to mount statewide campaigns that contribute to the voters' understanding and meaningful options, the burden is minimal. With social media and other modern technology, finding and connecting with supporters can happen with greater expediency than ever before. Contacting many supporters to contribute to fundraising efforts is easier today than it has ever been. And, perhaps more importantly, $50,000 just isn't what it once was, especially in politics. In a state in which twenty-four individual candidates spent an average of over $57,000 each to mount successful local campaigns for state senate in 2010, see Election Overview: New Hampshire 2010, Nat'l Inst. on Money in State Politics, election-overview?s=NH&y=2010 (last updated June 13, 2016), it would be strange to say that a viable statewide political party cannot be expected to shoulder a $50,000 burden for statewide ballot access for its nominees.


Unable to mount a persuasive case that the burden New Hampshire imposes on third parties seeking statewide ballot access for their nominees even equals, much less exceeds, the burdens imposed by the Georgia regime sustained in Jenness, LPNH turns its attention to the other side of the equation. LPNH correctly notes that the burdens in Jenness were justified by Georgia's effectively asserted interest in demanding that a party demonstrate a modicum of support before receiving the right to have its nominees listed on the ballot. In this case, by contrast, New Hampshire has not, LPNH says, effectively asserted such a justification. Rather, in this case LPNH says that any reliance on the state's asserted interest in ensuring a baseline level of support among ballot-eligible parties is only a “new post-hoc justification” concocted mid-way through this litigation and a smokescreen for the legislature's real concerns in enacting the amendment. LPNH points to two New Hampshire legislative subcommittee reports bearing on HB 1542, both of which identify the facilitation of nomination-paper certification as the goal advanced by the amendment, and neither of which mentions an interest in protecting the integrity of elections by combating ballot overcrowding. Indeed, New Hampshire defended the new January 1 start date solely on the basis of its certification justification at an earlier stage in this litigation, not elaborating on the interest in limiting ballot access until after discovery had commenced.

Based on this evidence, LPNH urges us to rule that New Hampshire's stated interest is an illegitimate post hoc justification that ought not be credited. LPNH argues that, applying the sliding-scale test developed in Anderson, “the precise interests put forward by the State as justifications for the burden imposed by its rule” pale in comparison to “the character and magnitude of the asserted injury to the rights protected by the First and Fourth Amendments that the plaintiff seeks to vindicate.” 460 U.S. at 789.

Precedent, including Anderson, provides no direct guidance on whether we can rely on belated statements of interest first voiced in the course of litigation challenging a statute. LPNH and its amici urge us to adopt by analogy the approach used in equal protection cases, where a proffered justification cannot withstand a heightened degree of scrutiny if that justification is “hypothesized or invented post hoc in response to litigation.” United States v. Virginia, 518 U.S. 515, 533 (1996). They observe that some courts applying Anderson have compared the middle range of its sliding scale to “intermediate scrutiny,” see Guare v. State, 117 A.3d 731, 740 (N.H. 2015), and therefore urge us to ignore the state's asserted interest in ensuring a minimal level of support for parties appearing on the ballot.

The problem for LPNH is that the burden here caused solely by the start date itself (as opposed to the three-percent-within-seven-months requirement), for the reasons already stated, is minimal, placing it at the easier-to-justify end of Anderson's sliding scale. Applying LPNH's chosen analogy, we would accordingly find ourselves at the “rational basis” end of any equal protection analysis. And, at that end, we can rely on statements of the state's interest first identified in litigation briefs. Barr, 626 F.3d at 110 (finding that if the law imposes only a “modest” or “reasonable” burden, “there need be only a rational basis undergirding the regulation in order for it to pass constitutional muster”); cf. Dudum v. Arntz, 640 F.3d 1098, 1116 n.28 (9th Cir. 2011) (expressing doubt that “the normal ability of litigants to advance arguments justifying their out-of-court behavior is suspended in election challenges where, as here, the burden imposed on voting is minimal at best”).6


The foregoing leaves LPNH with one escape gambit: identify as a burden the cumulative weight of the ballot-access regime as shortened by HB 1542, characterize that cumulative burden as falling further along the Anderson sliding scale, and then weigh it against only the state interest identified by the legislature in passing HB 1542. We reject this approach as illogical. It mismatches the cumulative burden of the entire ballot-access regime with only the justification initially cited by the New Hampshire legislature for altering the start date (and thus the duration of petition gathering). We see no reason to require a state legislature to restate its facially obvious reasons for having a ballot-access law each time it enacts an amendment tweaking that law. Similarly, if the current regime would have withstood challenge had it been enacted now as a whole, we can see no reason why it should fall if developed over time by amendment.

These ballot-access cases, unlike most voting-access cases, pose an identity between burden and purpose. The obvious purpose of the regime, in toto, is precisely to create the burden itself, which in turn has the effect, at the least, of limiting voters' selection to those who can make “some preliminary showing of a significant modicum of support before printing the name of a political organization's candidate on the ballot ․” Jenness, 403 U.S. at 442; see also, e.g., Libertarian Party of Me. v. Diamond, 992 F.2d 365, 371 (1st Cir. 1993) (endorsing state interest in “avoiding overloaded ballots and frivolous candidacies, which diminish victory margins, contribute to the cost of conducting elections, confuse and frustrate voters, increase the need for burdensome runoffs, and may ultimately discourage voter participation in the electoral process”). In the words of the Supreme Court, this is “an important state interest.” Jenness, 403 U.S. at 442.

Of course, we are not so naïve as to reject the possibility that a state legislature dominated by the two major parties may well wish to overshoot the mark, closing the door to all competitors. Our ultimate decision must therefore depend in large part on what we deem to be an appropriate “modicum of support.” Do we mean, for example, that the party's nominees have enough support to make winning plausible? Or is it enough that the party's presence in the campaign will generate enough interest and support to help frame issues and introduce new ideas, affect the positioning of other candidates, and signal growing dissatisfaction with the dominant parties? In Anderson, the Supreme Court suggested that the hard-to-quantify spillover benefits of a third-party candidacy to the “diversity and competition in the marketplace of ideas,” implicated “the primary values protected by the First Amendment.” 460 U.S. at 794.

In this respect, it is fair to read in Anderson a greater appreciation for the benefits of third-party participation than is apparent on the face of the opinion in Jenness. It would be quite a stretch, though, for a circuit court to leverage such a change in nuance into a license to stray from a clear ruling sustaining as constitutional a burden demonstrably greater than that imposed by New Hampshire. See Sarzen v. Gaughan, 489 F.2d 1076, 1082 (1st Cir. 1973) (“When a ․ federal court has spoken, stability and stare decisis require that litigants and other courts take its pronouncement at face value until formally altered.”); cf. Eulitt ex rel. Eulitt v. Me. Dep't of Educ., 386 F.3d 344, 349 (1st Cir. 2004) (citing Sarzen, 489 F.2d at 1082) (noting that until a higher court revokes its binding precedent, a lower court “is hard put to ignore that precedent unless it has unmistakably been cast into disrepute by supervening authority”). And even were we to consider such an ambitious frolic, the record in this case provides little fuel for much of a journey. Neither LPNH nor its amici explain how easy or hard it is to gather $50,000 or a few dozen volunteers in today's world of social media. Are there examples of parties or candidates that cannot raise $50,000 statewide, yet can still mount viable campaigns? Or do the costs of staff and advertising and the thresholds set by debate organizers themselves require levels of support greater than that required by New Hampshire? Have any parties barely managed to crawl over access thresholds, yet ended up playing a substantial role other than as a spoiler? Or does an analysis of election results in New Hampshire show that it is easier to get on the ballot than it is to garner substantial support in the form of actual votes? If a truly viable third party emerges, would it not harm that party to have too low a threshold, resulting in many extra parties with no chance, rather than a single third party less easily lost in the clutter? And what effect would lower thresholds have on laws providing public finance for campaigns? Anyone gauging the dividing line between a properly qualifying threshold and a barrier to meaningful competition would presumably want to consider these and other questions. The absence of information along these lines restricts the scope and quality of the judicial judgments required in cases such as this. As the litigant with the burden of proof, see, e.g., Harley-Davidson Credit Corp. v. Galvin, 807 F.3d 407, 411 (1st Cir. 2015), LPNH may suffer from that lack of detail in the record.

III. Conclusion

With a cumulative burden well less than that found acceptable in controlling precedent, and with no other attributes that themselves pose significant barriers to access, New Hampshire's regulations stand as an admittedly robust but nevertheless constitutional exercise of the state's “'broad power to prescribe the “Times, Places and Manner of holding Elections for Senators and Representatives,” which power is matched by state control over the election process for state offices.”' Wash. State Grange, 552 U.S. at 451 (quoting Art. I, § 4, cl. 1; Clingman v. Beaver, 544 U.S. 581, 586 (2005)). We affirm the judgment of the district court granting New Hampshire's motion for summary judgment in this facial challenge to part of the state's ballot-access framework.


1.   The end date, which varies each year, is five weeks before the New Hampshire primary, N.H. Rev. Stat. Ann. § 655:41, which falls on the second Tuesday in August, id. § 653:8.

2.   LPNH offers no clarity on exactly what it paid even though it presumably knows. We nevertheless will rely for purposes of this opinion on the testimony of LPNH chairperson Richard Tomasso.

3.   The number of signatures required in New Hampshire under the Georgia five-percent requirement would almost certainly be much larger than 4,111 per month because Georgia applied its five-percent requirement to the number of eligible voters, see Jenness, 403 U.S. at 433, rather than (as in New Hampshire) to the smaller number of persons who actually voted.

4.   Nor does New Hampshire's law impose any of the possibly “suffocating restrictions,” Jenness, 403 U.S. at 438, that led to the remand in Storer. See Storer, 415 U.S. at 740. Any registered voter may sign a petition. Notarization is not required.

5.   Based on LPNH's experience in 2000, it appears that one fulltime volunteer can gather 1,000 signatures in a month.

6.   We therefore need not decide whether we agree with the district court that post hoc justifications may be considered whenever Anderson balancing applies. See Gardner, 126 F. Supp. 3d at 209. We do note, though, that in Crawford v. Marion CountyElection Board, 553 U.S. 181 (2008), six Justices--three using Anderson's “balancing approach,” id. at 190 (plurality opinion), to find that the law in question imposed a “limited burden,” id. at 202 (quoting Burdick v. Takushi, 504 U.S. 428, 439 (1992)), and three relying on a “two-track approach,” id. at 205 (Scalia, J., concurring in the judgment), to find the law's burden “minimal,” id. at 209--declined to consider the distinct but related argument made by challengers to Indiana's voter identification law “that the statute was actually motivated by partisan concerns” and not the concerns advanced by the state in litigation, id. at 191 (plurality opinion).

KAYATTA, Circuit Judge.

Copied to clipboard