NATIONAL ORGANIZATION FOR MARRIAGE, Plaintiff, Appellant/Cross–Appellee, AMERICAN PRINCIPLES IN ACTION, Plaintiff, v. Walter F. McKEE, in his official capacity as member of the Commission on Governmental Ethics and Election Practices, et al., Defendants, Appellees/Cross–Appellants, Matthew Dunlap, in his official capacity as Secretary of the State of Maine, Defendant.
This appeal requires us to address the constitutionality of several Maine election laws governing, inter alia, the registration of political action committees (“PACs”) and the disclosure and reporting of information about expenditures made for election-related advocacy.1 Appellant National Organization for Marriage (“NOM”), a New Jersey-based nonprofit corporation organized for the purpose of providing “organized opposition to same-sex marriage in state legislatures,” contends that Maine's laws are unconstitutionally vague and overbroad. Claiming a chill of its First Amendment-protected advocacy efforts in Maine, NOM brought a facial and as-applied challenge seeking an injunction against the laws' enforcement and a declaration of their unconstitutionality. On summary judgment, the district court largely rejected NOM's claims, agreeing only that the phrase “for the purpose of influencing,” which the court severed from the provisions in which it appeared, was unconstitutionally vague.2
NOM renews here its arguments challenging Maine's laws on vagueness and overbreadth grounds. NOM asks as well that we reverse a ruling by the district court unsealing the trial record. In turn, the defendants (various Maine officials) contend that the district court erred in finding vague, and severing from Maine's statutes, the phrase “for the purpose of influencing.”
After careful consideration of the parties' arguments and key precedents, we conclude that Maine's laws pass constitutional muster. Central to our holding is the nature of the laws NOM challenges here. These provisions neither erect a barrier to political speech nor limit its quantity. Rather, they promote the dissemination of information about those who deliver and finance political speech, thereby encouraging efficient operation of the marketplace of ideas. As the Supreme Court recently observed, such compulsory “transparency enables the electorate to make informed decisions and give proper weight to different speakers and messages.” Citizens United v. FEC, 130 S.Ct. 876, 916 (2010). While we acknowledge that disclosure can, in some cases, unduly burden or chill political speech, there is no evidence that the Maine laws at issue here have had such a deleterious effect on NOM or its constituents.
We agree with the appellees that the use of “for the purpose of influencing” in the statutes at issue, given the appropriately limited reading offered by Maine's Commission on Governmental Ethics and Election Practices, is not unconstitutionally vague, and therefore we vacate the district court's holding as to that phrase and the consequent severance of portions of Maine's statutes. We otherwise affirm the district court's judgment in its entirety.
I. Statutory and Procedural Background
A. Maine's Election Laws
Maine has enacted a comprehensive set of election laws that embraces, among other things, contribution limits, a public financing system for state-office candidates, and various reporting and disclosure requirements for those engaged in election-related advocacy. We have previously described the contribution limit and public financing aspects of Maine's regulation of elections—which are not at issue here—in some detail. See Daggett v. Comm'n on Governmental Ethics & Election Practices, 205 F.3d 445, 450–52 (1st Cir.2000). The provisions challenged here, all relating solely to reporting and disclosure, fall into three categories: rules governing PACs, rules governing “independent expenditures,” and general attribution and disclaimer requirements.
1. PAC Provisions
Maine's PAC provisions are, as the appellees aptly characterize them, “pure disclosure laws.” Maine imposes no limitation on the amount of money PACs may raise, nor does it cap the sum a PAC may spend independently of a candidate or candidate committee. If they contribute money directly to a candidate, PACs are subject to the same per-candidate contribution limits—$750 per election for gubernatorial candidates and $350 per election for legislative candidates—as any other donor. See Me.Rev.Stat. tit. 21–A, § 1015(1), (2). The only PAC-specific requirements relate to registration, recordkeeping, and reporting.
An organization may qualify as a PAC under Maine law in one of several ways, of which two are relevant here. The first pertains to so-called “major-purpose” PACs. An organization that “has as its major purpose initiating, promoting, defeating or influencing a candidate election, campaign or ballot question” must register as a PAC in Maine if it receives contributions or makes expenditures aggregating more than $1,500 in a given calendar year for that purpose. Id. §§ 1052(5)(A)(4), 1053. The second relates to “non-major-purpose PACs,” which are subject to a significantly higher contribution/expenditure threshold for registration. Specifically, Maine law requires that an organization register as a PAC if it “does not have as its major purpose promoting, defeating or influencing candidate elections but ․ receives contributions or makes expenditures aggregating more than $5,000 in a calendar year for the purpose of promoting, defeating or influencing in any way the nomination or election of any candidate to political office.” Id. §§ 1052(5)(A)(5), 1053.
Within seven days of exceeding the relevant contribution or expenditure threshold, a PAC must register with the Maine Commission on Governmental Ethics and Election Practices (the “Commission”). Id. § 1053. Registration requires that the organization supply a name and address for the PAC; identify its form of organization and date of origin; name its treasurer, principal officers, and primary fundraisers and decisionmakers; and indicate which candidates, committees, referenda, or campaigns it supports or opposes. Id. An organization need not make any formal changes, such as forming a separate legal entity or creating a segregated fund, to operate as a PAC in Maine.
Once registered, a PAC is subject to two ongoing obligations under Maine law. First, the PAC treasurer must maintain records of certain election-related expenditures and contributions for four years following the election to which the records pertain. Id. § 1057. Second, the PAC must electronically file campaign finance reports on a quarterly basis, with additional reports due eleven days before any primary or general election and forty-two days after. Id. § 1059. The contents of the report vary by type of PAC. A major-purpose PAC must report any contribution to the PAC of more than $50 (including the name, address, occupation, and place of business of the contributor), while a non-major-purpose PAC reports only those contributions made “for the purpose of promoting, defeating or influencing a ballot question or the nomination or election of a candidate to political office.” Id. § 1060(6). The reporting of expenditures breaks down along similar lines: major-purpose PACs report all expenditures, including operational and administrative expenses, whereas non-major-purpose PACs report “only those expenditures made for the purpose of promoting, defeating or influencing a ballot question or the nomination or election of a candidate to political office.” Id. § 1060(4), (5), (7).
Maine law also explicitly requires PACs that are organized in another state to comply with all applicable registration and reporting requirements. See id. § 1053–B. There is, however, a narrow exemption: out-of-state PACs may contribute to candidates, party committees, and PACs in Maine without registering with the Commission, provided that (1) such contributions are the out-of-state PAC's only financial activity within Maine and (2) the out-of-state PAC “has not raised and accepted any contributions during the calendar year to influence an election or campaign” in Maine. Id.
2. “Independent Expenditure” Provision
In addition to its PAC-specific requirements, Maine's election laws also require across-the-board reporting of certain “independent expenditures.” At a general level, an “independent expenditure” is any payment or obligation made “for the purpose of influencing the nomination or election of any person to political office” other than a direct contribution to candidates and their campaign committees. Id. §§ 1012(3), 1019–B(1). Maine law provides that any individual or entity making independent expenditures aggregating more than $100 over the course of a particular candidacy must file a report with the Commission. Id. § 1019–B(3). That report must simply identify the expenditures by date, payee, and purpose, state whether the expenditures were made in support of or opposition to the relevant candidate, and state under oath or affirmation whether the expenditures were coordinated with a candidate or candidate committee. Id.
An expenditure may qualify as an “independent expenditure” in one of two ways. First, an expenditure will fall within the independent expenditure reporting requirement where it is made to finance a communication that “expressly advocates the election or defeat of a clearly identified candidate” and it is not a direct contribution to a candidate or candidate's committee.3 Id. § 1019–B(1)(A). Second, certain expenditures for communications made close to an election-twenty-one days before a primary and thirty-five days before the general election-are presumed to be “independent expenditures.” Id. The presumption applies only to an expenditure “made to design, produce or disseminate a communication that names or depicts a clearly identified candidate” in a race where at least one candidate has accepted public financing. Id. § 1019–B(1)(B). The person making the expenditure is afforded a chance to rebut the presumption by filing a written statement with the Commission within forty-eight hours of the expenditure “stating that the cost was not incurred with the intent to influence the nomination, election or defeat of a candidate.” Id. § 1019–B(2). Once a rebuttal statement is filed, the Commission will determine by a preponderance of the evidence, after gathering relevant material, whether the expenditure was incurred with such an intent. Id.
3. Attribution and Disclaimer Requirements
Finally, Maine law also requires that political advertisements and certain other political messages contain statements of attribution and disclaimer. The governing statute provides that any “communication expressly advocating the election or defeat of a clearly identified candidate ․ clearly and conspicuously state” whether it has been authorized by the candidate (the disclaimer) and state the name and address of the person who financed the communication (the statement of attribution). Id. § 1014(1)-(2). These disclaimer and attribution statements must also be included in any communication shortly before an election4 that “names or depicts a clearly identified candidate,” unless the communication “was not made for the purpose of influencing the candidate's nomination for election or election.” Id. § 1014(2–A).
The Commission may level a variety of sanctions-primarily in the form of fines—for violations of the provisions discussed above. An entity that falls within the definition of a PAC but fails to register may be subject to a civil fine of $250, id. § 1062–A(1), and a PAC's failure to file reports within thirty days of a reporting deadline can result in a fine of up to $10,000 or a criminal misdemeanor charge. Id. § 1062–A(8). Likewise, violations of the independent expenditure reporting requirement are punishable by a civil fine of up to $5,000. Id. § 1020–A(5–A)(A). Finally, violations of the attribution and disclosure requirements are subject to lesser fines (up to $200 if made within 20 days before an election, and no more than $100 at other times), but may be punished by a special fine of up to $5,000 if the violation was committed with the intent to misrepresent the source or candidate authorization of the advertisement. Id. § 1014(4).
B. Procedural History
NOM filed the initial complaint in this case in October 2009, shortly before a referendum election in Maine on a raft of issues that included same-sex marriage.5 The complaint challenged the constitutionality of a provision relating to ballot question committees, Me.Rev.Stat. tit. 21–A, § 1056–B,6 and was accompanied by motions for a temporary restraining order and a preliminary injunction. Following an expedited hearing, the district court denied NOM's motion for a temporary restraining order. NOM subsequently amended its complaint to add the claims at issue here: those targeting the constitutionality of Maine's PAC registration, independent expenditure, and attribution and disclaimer laws.
The district court held a hearing on NOM's motion for a preliminary injunction, consolidated with a bench trial on the merits, on August 12, 2010.7 The parties stipulated to a joint trial record, which was submitted under seal. At the hearing, the district court expressed doubt about the basis for sealing the record, and subsequently issued an order to show cause why the trial record should be maintained under seal.
In a decision issued on August 19, 2010, the district court largely denied NOM's claims and upheld the constitutionality of the challenged statutes. See Nat'l Org. for Marriage v. McKee, 723 F.Supp.2d 245 (D.Me.2010). Finding that NOM had adequately demonstrated an interest in engaging in expressive activity that was deterred by the prospect of regulation under Maine's laws, the court held that NOM had standing to challenge the statutes at issue.8 Id. at 256–58. On the merits, the court found for NOM on only two points.
First, the court held the phrase “for the purpose of influencing” to be unconstitutionally vague, citing the treatment of similar language in the Supreme Court's opinion in Buckley v. Valeo, 424 U.S. 1 (1976). Nat'l Org. for Marriage, 723 F.Supp.2d at 261. The phrase (and variants thereof) appears in several places throughout the challenged statutes, including in the definition of a non-major-purpose PAC,9 the rebuttal provision for presumed independent expenditures (i.e., those made shortly prior to an election),10 and in the provisions defining which communications are subject to disclaimer and attribution requirements.11 The court determined that the appropriate remedy was to sever the phrase from the statutes. Id. The most significant impact of this holding was on the independent expenditure statute. Because the rebuttal procedure for presumed independent expenditures was dependent on the term “influencing,”12 the court's ruling severed the entire rebuttal procedure. Thus, the independent expenditure presumption became conclusive for expenditures for communications clearly identifying a candidate made shortly before an election. The court held that the provision, as altered, nonetheless passed constitutional muster, as recent Supreme Court decisions “have made the rebuttal exercise pointless.” Id . at 265.13
Second, the court held one of the implementing regulations for the independent expenditure statute unconstitutional, finding that it impermissibly burdened First Amendment speech. Id. at 266. The regulation was one of a pair governing the timing of reporting independent expenditures. The first, which the court upheld, required that independent expenditures of over $100 made within two weeks of an election be reported to the Commission within twenty-four hours. See 94–270–001 Me.Code R. § 10(3)(A). The second required the reporting within twenty-four hours of any independent expenditures aggregating over $250, regardless of when made. Id. § 10(3)(B). Finding the short reporting time frame mandated by both regulations to be burdensome, the court held that the second regulation, unlike the first, could not be justified by a close relationship to “the state's interest in providing information to voters at precisely the time that such information can be of greatest use.” Nat'l Org. for Marriage, 723 F.Supp.2d at 266. The defendants do not challenge this holding on appeal.
In addition to its merits holdings, the district court also ruled that the trial evidence must be unsealed. Explaining that it was “not willing to make a First Amendment decision based upon a sealed record,” the court ordered the parties to refile the record in publicly available form.14 Id. at 249 n.4.
This timely appeal followed.
We begin, as we must, with the defendants' argument that NOM lacks standing to prosecute some of its constitutional claims. We review a district court's ruling on the question of standing de novo. Sullivan v. City of Augusta, 511 F.3d 16, 24 (1st Cir.2007).
The standing requirement—or, more accurately, requirements, as standing “comprises a mix of constitutional and prudential criteria,” Osediacz v. City of Cranston, 414 F.3d 136, 139 (1st Cir.2005)—flows from the limited nature of federal court jurisdiction, and specifically from the grounding of the federal judicial power in “Cases” and “Controversies.” U.S. Const. art. III, § 2; Ariz. Christian Sch. Tuition Org. v. Winn, 131 S.Ct. 1436, 1441–42 (2011). The constitutional aspect of standing embraces three core requirements:
“First, the plaintiff must have suffered an ‘injury in fact’ ––– an invasion of a legally protected interest which is (a) concrete and particularized, and (b) ‘actual or imminent, not “conjectural” or “hypothetical.” ‘ Second, there must be a causal connection between the injury and the conduct complained of ––– the injury has to be ‘fairly trace[able] to the challenged action of the defendant, and not ․ th[e] result [of] the independent action of some third party not before the court.’ Third, it must be ‘likely,’ as opposed to merely ‘speculative,’ that the injury will be ‘redressed by a favorable decision.’ “
Ariz. Christian Sch., 131 S.Ct. at 1442 (quoting Lujan v. Defenders of Wildlife, 504 U.S. 555, 560–61 (1992)). The Supreme Court has overlaid these constitutional dictates with several prudential limitations on standing, including “ ‘the general prohibition on a litigant's raising another person's legal rights, the rule barring adjudication of generalized grievances more appropriately addressed in the representative branches, and the requirement that a plaintiff's complaint fall within the zone of interests protected by the law invoked.’ “ Osediacz, 414 F.3d at 139 (quoting Allen v. Wright, 468 U.S. 737, 751 (1984)). In certain facial First Amendment challenges to a statute, we may relax these prudential limitations, Osediacz, 414 F.3d at 141, but the constitutional requirements apply with equal force in every case, Sutliffe v. Epping Sch. Dist., 584 F.3d 314, 326 n.6 (1st Cir.2009).
Defendants do not seek to dispose of the entire suit on standing grounds, but instead surgically target NOM's standing to challenge Maine's PAC-related election laws.15 They assert that NOM lacks standing to bring the PAC claims because it failed to prove that it came within the reach of Maine's PAC laws, i.e., that NOM's activities would have qualified it as a PAC. In light of the fact that prudential limitations on standing may be relaxed in the context of First Amendment challenges-and because defendants cite the constitutional standards for standing in making their argument-we construe this argument as a challenge to the adequacy of NOM's Article III injury-in-fact showing.16
Preenforcement First Amendment challenges like this one occupy a somewhat unique place in Article III standing jurisprudence. By definition, such cases present us with situations where the government has not yet applied the allegedly unconstitutional law to the plaintiff, and thus there is no tangible injury. However, in these circumstances the Supreme Court has recognized “self-censorship” as “a harm that can be realized even without an actual prosecution.” Virginia v. Am. Booksellers Ass'n, 484 U.S. 383, 393 (1988); see also N.H. Right to Life Political Action Comm. v. Gardner, 99 F.3d 8, 13 (1st Cir.1996) (“[I]t is not necessary that a person expose herself to arrest or prosecution under a statute in order to challenge that statute in a federal court.”). The chilling of protected speech may thus alone qualify as a cognizable, Article III injury.
The mere allegation of a “chill,” however, will not suffice to open the doors to federal court. See Laird v. Tatum, 408 U.S. 1, 13–14 (1972) (“Allegations of a subjective ‘chill’ are not an adequate substitute for a claim of specific present objective harm or a threat of specific future harm․”). Where, as here, the plaintiff claims injury based on such a chilling of speech, the plaintiff must establish with specificity that she is “within the class of persons potentially chilled.” Osediacz, 414 F.3d at 142. This burden will be satisfied by record evidence supporting “an objectively reasonable possibility that she would be subject to the allegedly unconstitutional [law].” Id. at 143; see also N.H. Right to Life, 99 F.3d at 14 (“A party's subjective fear that she may be prosecuted for engaging in expressive activity will not be held to constitute an injury for standing purposes unless that fear is objectively reasonable.”).
NOM challenges three separate provisions of Maine's PAC laws: the major-purpose PAC definition, the non-major-purpose PAC definition, and the provision governing out-of-state PACs. We examine in turn whether the record supports an “objectively reasonable possibility” that each provision would be applied to NOM, Osediacz, 414 F.3d at 143, and then turn briefly to a related inquiry specific to NOM's vagueness challenges to the PAC statutes.
A. Standing to Challenge Major–Purpose PAC Provision
It is plain that NOM has no objectively reasonable apprehension of being regulated as a major-purpose PAC. Among other things, a major-purpose PAC must have “as its major purpose initiating, promoting, defeating or influencing a candidate election, campaign or ballot question” in Maine. Me.Rev.Stat. tit. 21–A, § 1052(5)(A)(4). NOM identifies itself as a nonprofit advocacy organization with a national scope, dedicated to providing “organized opposition to same-sex marriage in state legislatures.” NOM's advocacy efforts and expenditures have spanned the country, with significant expenditures in California, Iowa, New York, and New Hampshire, among others. In 2009, the year of NOM's largest expenditures in Maine (made to support repeal of Maine's same-sex marriage law), NOM spent $1.8 million in Maine out of $8 million in total expenditures for the year. In light of this record, NOM does not have as its “major purpose” election advocacy in Maine, and it is accordingly not subject to regulation as a major-purpose PAC. NOM therefore lacks standing to challenge § 1052(5)(a)(4).
B. Standing to Challenge Non–Major–Purpose PAC Provision
It is a closer question whether the record reveals an objectively reasonable possibility that NOM would be regulated as a non-major-purpose PAC under Maine law. To so qualify, NOM would have to anticipate receiving contributions or making expenditures of more than $5,000 in a year “for the purpose of promoting, defeating or influencing in any way the nomination or election of any candidate to political office.” Me.Rev.Stat. tit. 21–A, § 1052(5)(A)(5). The evidence is inconclusive as to whether NOM actually crossed the $5,000 threshold during the 2010 election cycle .17 However, we need not determine whether NOM in fact became subject to the provision during the relevant period, because NOM claims injury based upon self-censorship in anticipation of the law's application to it, and not upon the actual burdens of the law. NOM's executive director, Brian Brown, testified—consistently with the allegations in NOM's complaint—that NOM's fear of enforcement of Maine's election laws was curtailing NOM's speech, and that “[u]ntil Maine's law is changed,” NOM was “not going to expend precious resources” becoming involved in campaigns in the state. The appropriate inquiry, then, is whether it was objectively reasonable for NOM to believe that the non-major-purpose PAC provision might apply to it and that it would have to curtail its activities in Maine to avoid such a result.
The record evidence confirms that NOM's fears were objectively reasonable and led NOM to engage in self-censorship.18 The complaint, which was verified by Brown, explained that NOM sought to engage in a variety of forms of election-related speech, including “radio ads, direct mail, and publicly accessible Internet postings of its radio ads and direct mail.” NOM alleged that some portion of these advertisements would relate to “clearly identified candidates for state or local offices.” To this end, NOM discussed potential advertisements with a marketing vendor, and went so far as to have the vendor create three template advertisements (specifically, copy for two broadcast advertisements and one mailer) to be used in not-yet-identified candidate races. One such template, titled “Consequences,” raised fears that legalizing same-sex marriage would lead to schools teaching children about same-sex relationships, and concluded:
Legislator Z and some politicians in Maine can't fix the real problems in these troubled times, but they've got time to push gay marriage on Maine families? Call Legislator Z and tell him/her: “Don't mess with marriage.”
While the record does not indicate how much the contemplated advertisements would cost, NOM alleged generally that each of its communications costs more than $250. The advertisements were never used, in line with NOM's claim to have curtailed its planned speech. The record also contained evidence that NOM had made political expenditures in Maine in the past, including contributions of $1.8 million in 2009 to a committee opposed to Maine's same-sex marriage law.
We agree with the district court that, although NOM's “showing certainly could have been stronger,” Nat'l Org. for Marriage, 723 F.Supp.2d at 258, NOM has met its standing burden with respect to its challenge to § 1052(5)(A)(5). The burden of proving that one's speech was chilled is a modest one. See Osediacz, 414 F.3d at 143. The record evidence adequately establishes both “an objectively reasonable possibility” that NOM would be subject to Maine's requirements for non-major-purpose PACs if it engaged in its intended speech, and that NOM forwent political speech to avoid the alleged burdens (and possible penalties for non-compliance) attending the non-major-purpose PAC provision. Id. Such self-censorship in the face of possible legal repercussions suffices to show Article III injury. See N.H. Right to Life, 99 F.3d at 13 (“[A]n actual injury can exist when the plaintiff is chilled from exercising her right to free expression or forgoes expression in order to avoid enforcement consequences.”).
C. Standing to Challenge Out–of–State PAC Provision
We next examine NOM's standing to challenge § 1053–B, which provides generally that a “political action committee organized outside of [Maine] shall register and file reports with the [C]ommission” in accordance with Maine's PAC laws.19 Me.Rev.Stat. tit. 21–A, § 1053–B. The question of whether NOM might be considered a “political action committee” retreads ground we have just covered. Maine law defines “political action committee” to include, among other things, a non-major-purpose PAC. Moreover, the record shows that NOM, which operates from New Jersey, is organized as a Virginia nonprofit. Thus, there is no question that NOM is “organized outside of [Maine],” and there is a reasonable possibility that it would be considered a “political action committee” within the meaning of the statute. NOM therefore has standing to challenge the out-of-state PAC provision.
D. Standing to Bring Vagueness Challenge
We last address a standing-related argument specific to NOM's vagueness claims. Defendants argue that NOM cannot bring a vagueness challenge to the non-major-purpose PAC definition, as well as to its corresponding definition of the term “expenditure,”20 because NOM's advocacy efforts were clearly covered by the provisions' terms. In so arguing, defendants rely on the well-established proposition that a “ ‘plaintiff who engages in some conduct that is clearly proscribed cannot complain of the vagueness of the law as applied to the conduct of others.’ “ Holder v. Humanitarian Law Project, 130 S.Ct. 2705, 2719 (2010) (quoting Hoffman Estates v. Flipside, Hoffman Estates, Inc., 455 U.S. 489, 495 (1982)).21 Defendants contend that certain postcards NOM sent in September 2010 targeting state legislative candidates were unambiguously covered by the non-major-purpose PAC statute and expenditure definition, i.e., that they were “for the purpose of promoting, defeating or influencing in any way” a candidate election, Me.Rev.Stat. tit. 21–A, § 1052(5)(A)(5), and “for the purpose of influencing the ․ election of any person to political office; or for the initiation, support or defeat of a campaign,” id. § 1052(4)(A)(1).
The defendants' argument is off-target for at least two reasons. First, the question of whether the non-major-purpose PAC provisions clearly applied to NOM's September mailings is irrelevant to NOM's standing to bring its vagueness claims. Because this is a preenforcement challenge based on conduct forgone due to an alleged chill, the appropriate focus for the defendants' arguments would be on whether “the statutory terms are clear in their application to [NOM's] proposed conduct.” Humanitarian Law Project, 130 S.Ct. at 2720 (emphasis added). Moreover, the judgment in this case was entered and NOM's appeal was filed in August 2010, and thus evidence of NOM's September 2010 advocacy efforts is not properly part of the record on appeal.22 At the time of the hearing below, NOM had not yet engaged in any advocacy efforts in Maine in 2010.
Second, NOM's claim is not simply a challenge to the vagueness of the provisions as they would be applied to its actual or intended advocacy efforts; NOM also brings a facial challenge to the provisions under the First Amendment overbreadth doctrine. The bar against vagueness challenges by those whose conduct the law clearly proscribes is “relaxed ․ in the First Amendment context, permitting plaintiffs to argue that a statute is overbroad because it is unclear whether it regulates a substantial amount of protected speech.” United States v. Williams, 553 U.S. 285, 304 (2008).23 We thus see no bar to reaching the merits of NOM's vagueness challenge to the non-major-purpose PAC provisions.
III. First Amendment Overbreadth Challenges
Turning to the merits of NOM's constitutional challenges, we first address its First Amendment arguments that Maine's election laws are unconstitutionally overbroad, reviewing those claims de novo. United States v. Morales-de Jesús, 372 F.3d 6, 8 (1st Cir.2004) (constitutional challenges are reviewed de novo). The First Amendment's guarantee of free speech applies with special vigor to discussion of public policy and the qualifications of political candidates.24 Buckley v. Valeo, 424 U.S. 1, 14 (1976). Indeed, “there is practically universal agreement that a major purpose of [the First] Amendment was to protect the free discussion of governmental affairs.” Mills v. Alabama, 384 U.S. 214, 218 (1966). Accordingly, “[t]he First Amendment affords the broadest protection to such political expression in order ‘to ensure [the] unfettered interchange of ideas for the bringing about of political and social changes desired by the people.’ “ Buckley, 424 U.S. at 14 (alteration in original) (quoting Roth v. United States, 354 U.S. 476, 484 (1957)).
NOM has framed its First Amendment challenges to Maine's election laws as overbreadth claims, arguing that each law is unconstitutional on its face. Under the overbreadth doctrine, “a statute is facially invalid if it prohibits a substantial amount of protected speech.” Williams, 553 U.S. at 292. The overbreadth doctrine is “ ‘strong medicine’ “ that should be “employed ․ with hesitation, and then ‘only as a last resort.’ “ New York v. Ferber, 458 U.S. 747, 769 (1982) (quoting Broadrick v. Oklahoma, 413 U.S. 601, 613 (1973)). For that reason, courts “vigorously enforce[ ] the requirement that a statute's overbreadth be substantial, not only in an absolute sense, but also relative to the statute's plainly legitimate sweep.” Williams, 553 U.S. at 292.
A. Distinction Between Issue Discussion and Express Advocacy
We first address NOM's arguments that the statutes challenged here are overbroad because they may reach discussion of issues as well as express advocacy of a candidate's election or defeat. The division between pure “issue discussion” and “express advocacy” of a candidate's election or defeat is a conceptual distinction that has played an important, and at times confounding, role in a certain set of modern Supreme Court election law precedents. Though the contours (and significance) of the distinction have never been firmly fixed, the core premise is that regulation of speech expressly advocating a candidate's election or defeat may more easily survive constitutional scrutiny than regulation of speech discussing political issues more generally.
Because a number of NOM's arguments here raise, both directly and indirectly, this distinction between issue discussion and express advocacy, we pause briefly to describe how the distinction arose and developed. We ultimately conclude, however, that the distinction is not important for the issues addressed in this appeal.
1. Issue/Express Advocacy Distinction Generally
The issue discussion/express advocacy distinction has its roots in the Supreme Court's decision in Buckley v. Valeo. Perhaps the Court's seminal decision in the area of campaign finance regulation, Buckley resolved a wide-ranging series of challenges to provisions of the Federal Election Campaign Act of 1971 (“FECA”). One of those challenged provisions, of relevance to our discussion here, imposed an absolute cap on independent expenditures, stating that “ ‘[n]o person may make any expenditure ․ relative to a clearly identified candidate during a calendar year which, when added to all other expenditures made by such person during the year advocating the election or defeat of such candidate, exceeds $1,000.’ “ Buckley, 424 U.S. at 39 (alterations in original) (quoting 18 U.S.C. § 608(e)).
Reviewing this language, the Court first noted that the “use of so indefinite a phrase as ‘relative to’ a candidate” raised serious vagueness concerns. Id. at 41. The Court construed the phrase (by reference to its surrounding terms) as limited to expenditures “advocating the election or defeat of” a candidate. However, this construction, in the Court's estimation, merely “refocus[ed] the vagueness question.” Id. at 42. The Court's evident concern was that the statute, even as limited, failed to draw a sharp enough line between advocacy of a candidate's election and discussion of issues, and that the resulting uncertainty over what the statute covered would “ ‘compel[ ] the speaker to hedge and trim,’ “ id. at 43 (quoting Thomas v. Collins, 323 U.S. 516, 535 (1945)):
[T]he distinction between discussion of issues and candidates and advocacy of election or defeat of candidates may often dissolve in practical application. Candidates, especially incumbents, are intimately tied to public issues involving legislative proposals and governmental actions. Not only do candidates campaign on the basis of their positions on various public issues, but campaigns themselves generate issues of public interest.
Id. at 42. To avoid this uncertainty, the Court limited the scope of the statute to “expenditures for communications that in express terms25 advocate the election or defeat of a clearly identified candidate for federal office.” Id. at 44.26
The constitutional basis for this concern with distinguishing between laws that regulate advocacy of a candidate's election and those that regulate pure issue discussion has never been entirely clear. Buckley explicitly framed its discussion in terms of unconstitutional vagueness under the Due Process Clause, and there is, to be sure, a vagueness dimension to the analysis. See, e.g., FEC v. Wis. Right to Life, Inc., 551 U.S. 449, 497 (2007) (Scalia, J., concurring in part) (referring to the express advocacy portion of Buckley as the decision's “vagueness holding”). However, this interpretation has its limits; the mere fact that a statute may cover issue discussion as well as candidate advocacy does not alone render it vague under due process standards, provided that the statute is reasonably clear in its coverage.
Perhaps for this reason, there are hints in Buckley that the constitutional basis for the Court's concern lay more in overbreadth—i.e., that statutes that reached issue discussion might be deemed to regulate impermissibly a substantial amount of speech protected by the First Amendment—than in vagueness. See, e.g., 424 U.S. at 80 (limiting a second, disclosure-related provision of FECA to communications that “expressly advocate” a candidate's election to “insure that the reach of [the provision] is not impermissibly broad”). This reading finds considerable support in subsequent authority. See Osborne v. Ohio, 495 U.S. 103, 120 n.14 (1990) (describing Buckley as a “case where a law was construed to avoid potential overbreadth problems”); FEC v. Mass. Citizens for Life, Inc., 479 U.S. 238, 248 (1986) (stating that Buckley's “express advocacy” limitation was imposed to “avoid problems of overbreadth”); cf. McConnell v. FEC, 540 U.S. 93, 192 (2003) (noting that Buckley “narrowly read[ ] the FECA provisions ․ to avoid problems of vagueness and overbreadth”), overruled on other grounds by Citizens United, 130 S.Ct. 876. Regardless of its origins, the dividing line between issue discussion and express advocacy, as it evolved, came to be associated more strongly with First Amendment overbreadth analysis than with due process vagueness concerns.27 See, e.g., Wis. Right to Life, 551 U.S. at 457 (noting that the “law in this area requires us ․ to draw such a line, because we have recognized that the interests held to justify the regulation of campaign speech [under the First Amendment] ․ ‘might not apply’ to the regulation of issue advocacy” (quoting McConnell, 540 U.S. at 206 n.88)).
2. Application to Maine's Election Statutes
Drawing on these cases, NOM argues that the statutes before us are unconstitutionally overbroad because they reach issue advocacy as well as express advocacy of a candidate's election or defeat. NOM's argument presumes that the distinction between issue discussion and express advocacy is relevant to the review of the statutes here. That is not the case for a couple of reasons.
First, the issue/express advocacy dichotomy has only arisen in a narrow set of circumstances not present here. From the beginning, the distinction's primary purview has been cases scrutinizing limits on independent expenditures.28 The statute that prompted the Buckley Court to introduce the “express advocacy” construction was a blanket $1,000 limit on independent expenditures. 424 U.S. at 41–44. The more recent Supreme Court precedents to make use of the express/issue advocacy distinction addressed a narrower federal law prohibiting corporations and labor unions from employing general treasury funds to pay for “electioneering” communications targeting candidates for election. See McConnell, 540 U.S. at 189–209; Wis. Right to Life, 551 U.S. at 464–82. This line of cases came to a definitive end with Citizens United, which held limitations on such expenditures by corporations and unions to be unconstitutional, and thus effectively prohibited any law limiting independent expenditures regardless of the identity of the regulated entity. 130 S.Ct. at 896–913. As the present case does not involve a limit on independent expenditures, the relevance of these cases is limited at best.
Second, and more fundamentally, the Supreme Court has explicitly rejected an attempt to “import [the] distinction” between issue and express advocacy into the consideration of disclosure requirements. Id. at 915; see also id. (“[W]e reject Citizens United's contention that the disclosure requirements must be limited to speech that is the functional equivalent of express advocacy.”). The provisions before us are all effectively disclosure laws, in that they require the divulgence of information to the public or the Commission, but do not directly limit speech.29 We find it reasonably clear, in light of Citizens United, that the distinction between issue discussion and express advocacy has no place in First Amendment review of these sorts of disclosure-oriented laws. Accord Human Life of Wash. Inc. v. Brumsickle, 624 F.3d 990, 1016 (9th Cir.2010) (“Given the Court's analysis in Citizens United, and its holding that the government may impose disclosure requirements on speech, the position that disclosure requirements cannot constitutionally reach issue advocacy is unsupportable.”).
Thus, to the extent that NOM's overbreadth arguments turn on the distinction between issue discussion and express advocacy, we reject them.
B. Standard of Scrutiny
Since Buckley, the Supreme Court has distinguished in its First Amendment jurisprudence between laws that restrict “the amount of money a person or group can spend on political communication” and laws that simply require disclosure of information by those engaging in political speech. 424 U.S. at 19, 64. The Court has recognized that disclosure laws, unlike contribution and expenditure limits, “impose no ceiling on campaign-related activities,” id. at 64, and thus are a “less restrictive alternative to more comprehensive regulations of speech.” Citizens United, 130 S.Ct. at 915; see also Buckley, 424 U.S. at 68 (“[D]isclosure requirements certainly in most applications appear to be the least restrictive means of curbing the evils of campaign ignorance and corruption that Congress found to exist.”). For that reason, disclosure requirements have not been subjected to strict scrutiny, but rather to “ ‘exacting scrutiny,’ which requires a ‘substantial relation’ between the disclosure requirement and a ‘sufficiently important’ governmental interest.” Citizens United, 130 S.Ct. at 914 (quoting Buckley, 424 U.S. at 64, 66); see also Doe v. Reed, 130 S.Ct. 2811, 2818 (2010).30
While NOM concedes that exacting scrutiny applies to review of Maine's independent expenditure and disclaimer and attribution laws, it contends that Maine's PAC definitions are subject to strict scrutiny. In fact, NOM suggests that any law defining an organization as a PAC is subject to strict scrutiny, because, “[a]s a matter of law, not fact,” PAC status is burdensome and subjects an entity to “extensive regulations.” NOM's argument here reflects two contradictory points. On the one hand, NOM seeks to justify the application of strict scrutiny by reference to some undefined set of “full-fledged political committee burdens.” On the other, NOM disclaims any challenge to the disclosure requirements attendant to PAC status under Maine law—i.e., the actual burdens of registration and reporting imposed by the state's PAC provisions—but purports to challenge only the “PAC definition, through which Maine unconstitutionally imposes full-fledged political committee burdens.” NOM's point appears to be that “by giving government the power to license speech” by defining an entity as a PAC, whatever obligations are imposed on PACs “in effect are prior restraints.”
NOM's attempt to ascribe a free-standing significance to the PAC label is unpersuasive. It is not the designation as a PAC but rather the obligations that attend PAC designation that matter for purposes of First Amendment review. Those obligations-as well as the basic definition of a “PAC”—vary across the jurisdictions that regulate PACs. Maine's requirements are substantially different from those at issue in the cases NOM cites in support of its contention that PAC status is inherently burdensome. For example, in Citizens United, where, as NOM points out, the Supreme Court characterized federal-law PACs as “expensive to administer and subject to extensive regulations,”31 130 S.Ct. at 897, the Court was considering a regime that required corporations to set up a separate legal entity and create a segregated fund prior to engaging in any direct political speech. In addition, these federal-law PACs were subject to numerous obligations and restrictions, among them a prohibition on an organization soliciting contributions for its segregated fund from anyone except its “members,” which excluded “those persons who have merely contributed to or indicated support for the organization in the past .” Mass. Citizens for Life, 479 U.S. at 253–54 (citing 2 U.S.C. § 441b(b)(4)(A), (C)). In contrast, Maine's non-major-purpose PAC provision does not condition political speech on the creation of a separate organization or fund, establishes no funding or independent expenditure restrictions,32 and imposes three simple obligations on an entity qualifying as a PAC: filing of a registration form disclosing basic information, quarterly reporting of election-related contributions and expenditures, and simple recordkeeping.
Because Maine's PAC laws do not prohibit, limit, or impose any onerous burdens on speech, but merely require the maintenance and disclosure of certain financial information, we reject NOM's argument that strict scrutiny should apply. Accordingly, we review each of the laws at issue under the “exacting scrutiny” standard applicable to disclosure requirements.
C. Application of Exacting Scrutiny to Maine's Laws
As we have stated, we will consider a law constitutional under exacting scrutiny standards where there is a “substantial relation” between the law and a “ ‘sufficiently important’ governmental interest.” Citizens United, 130 S.Ct. at 914 (quoting Buckley, 424 U.S. at 64, 66). In Buckley, the Court recognized the goal of “provid[ing] the electorate with information as to where political campaign money comes from and how it is spent” to be such a “sufficiently important” governmental interest capable of supporting a disclosure law. 424 U.S. at 66 (internal quotation marks omitted). The Court's more recent decisions have continued to recognize the importance of this informational interest. See, e.g., Citizens United, 130 S.Ct. at 914–15; McConnell, 540 U.S. at 196.
Buckley tied the government's interest in the dissemination of information to the functioning of the electoral process, noting that “[i]n a republic where the people are sovereign, the ability of the citizenry to make informed choices among candidates for office is essential.” 424 U.S. at 14–15. The Court observed that disclosure has several benefits in this regard:
It allows voters to place each candidate in the political spectrum more precisely than is often possible solely on the basis of party labels and campaign speeches. The sources of a candidate's financial support also alert the voter to the interests to which a candidate is most likely to be responsive and thus facilitate predictions of future performance in office.
Id. at 67.
However, the informational interest is not limited to informing the choice between candidates for political office. As Citizens United recognized, there is an equally compelling interest in identifying the speakers behind politically oriented messages. In an age characterized by the rapid multiplication of media outlets and the rise of internet reporting, the “marketplace of ideas” has become flooded with a profusion of information and political messages. Citizens rely ever more on a message's source as a proxy for reliability and a barometer of political spin. Disclosing the identity and constituency of a speaker engaged in political speech thus “enables the electorate to make informed decisions and give proper weight to different speakers and messages.”33 Citizens United, 130 S.Ct. at 916; see also Cal. Pro–Life Council, Inc. v. Getman, 328 F.3d 1088, 1105 (9th Cir.2003) (recognizing that, in the “cacophony of political communications through which ․ voters must pick out meaningful and accurate messages [,] ․ being able to evaluate who is doing the talking is of great importance”). Additionally, in the case of corporate or organizational speakers, disclosure allows shareholders and members to “hold [them] accountable for their positions.” Citizens United, 130 S.Ct. at 916. In short, “[t]he First Amendment protects political speech; and disclosure permits citizens and shareholders to react to [that] speech ․ in a proper way.” Id.
In line with these precedents, defendants offer Maine's interest in disseminating information about political funding to the electorate in support of the laws challenged here.34 As the district court found, the interest is plainly a motivating factor behind Maine's laws, and “Maine, through its Commission website and otherwise, makes [the financial disclosure] information easily available to the public.” Nat'l Org. for Marriage, 723 F.Supp.2d at 263. We thus proceed under the exacting scrutiny framework to examine whether there is a “substantial relation” between Maine's informational interest and each of the laws at issue.
1. Non–Major–Purpose PAC Provisions
As we have described, Maine considers an entity to be a non-major-purpose PAC when it receives contributions or makes expenditures of more than $5,000 annually “for the purpose of promoting, defeating or influencing in any way” a candidate's election. Me.Rev.Stat. tit. 21–A, § 1052(5)(A)(5). Upon crossing that threshold, the newly-deemed non-major-purpose PAC must register with the Commission, maintain records of certain expenditures as well as donor contributions aggregating more than $50, and file reports both on a quarterly basis and shortly before and after each election. Id. §§ 1053, 1057, 1059–60. The reporting requirements are well tailored to Maine's informational interest, requiring disclosure only of the candidates or campaigns the non-major-purpose PAC supports or opposes, its expenditures made to support or oppose the same, and identifying information for any contributors who have given more than $50 to the PAC to support or oppose a candidate or campaign. Id. § 1060.
NOM does not challenge the substantive obligations attendant to non-major-purpose PAC status, nor contest that the registration, recordkeeping, and reporting requirements bear a substantial relation to Maine's informational interest. Instead, NOM contends that Maine's definition of a non-major-purpose PAC, standing alone, is unconstitutionally overbroad. In rejecting NOM's argument for strict scrutiny, we have already addressed the claim that PAC status is somehow inherently burdensome apart from the specific requirements it entails. However, there is a second aspect to NOM's argument. NOM contends that Supreme Court precedent sharply limits regulation of PACs to those that are under the control of a candidate or have as their “major purpose” the election of a candidate. By its very definition, Maine's non-major-purpose PAC provision covers entities that fall outside of that allegedly limited zone of permissible regulation, and thus, NOM contends, the provision is fatally overbroad. We disagree.
NOM extracts support for its argument from a dictum in Buckley, albeit a dictum that has had some reach. In Buckley, the Court concluded that the definition of expenditure used in connection with FECA's disclosure provision-and particularly the phrase “for the purpose of influencing”-raised significant line-drawing problems because it had the “potential for encompassing both issue discussion and advocacy of a political result.” 424 U.S. at 79. In the course of its discussion, the Court noted that FECA's definition of “political committees,” which, like the disclosure provision, was defined in terms of contributions and expenditures, “could raise similar vagueness problems.” Id. The provision escaped these concerns, the Court explained, because it could be construed more narrowly:
To fulfill the purposes of [FECA, political committees] need only encompass organizations that are under the control of a candidate or the major purpose of which is the nomination or election of a candidate. Expenditures of candidates and of “political committees” so construed can be assumed to fall within the core area sought to be addressed by Congress. They are, by definition, campaign related.
Id. Buckley's narrow reading of FECA's political committee definition, though dictum, appears to have been accepted by later opinions. See McConnell, 540 U.S. at 170 n.64; Mass. Citizens for Life, 479 U.S. at 252 n.6; cf. FEC v. Akins, 524 U.S. 11, 26–27 (1998) (noting dispute over extent of narrowing construction). NOM draws from this the conclusion that the First Amendment permits an entity to be designated a “PAC” only where it (1) “is under the control of a candidate” or (2) has as its major purpose “the nomination or election of a candidate.”
We find no reason to believe that this so-called “major purpose” test, like the other narrowing constructions adopted in Buckley, is anything more than an artifact of the Court's construction of a federal statute. See McConnell, 540 U.S. at 191–92. The Court has never applied a “major purpose” test to a state's regulation of PACs, nor have we. And, as we have discussed, the line-drawing concerns that led the Court to read FECA's definition of “political committee” narrowly are not relevant to our First Amendment review of Maine's statutes. Moreover, as the district court aptly observed, application of NOM's “major-purpose” test would “yield perverse results” here:
Under NOM's interpretation, a small group with the major purpose of re-electing a Maine state representative that spends $1,500 for ads could be required to register as a PAC. But a mega-group that spends $1,500,000 to defeat the same candidate would not have to register because the defeat of that candidate could not be considered the corporation's major purpose.
Nat'l Org. for Marriage, 723 F.Supp.2d at 264. We, like the district court, see no basis to conclude “that the First Amendment's protections should apply so unequally.” Id.
We therefore reject NOM's argument that the non-major-purpose PAC definition is unconstitutionally overbroad. Because we find a substantial relation between Maine's disclosure-oriented regulation of non-major-purpose PACs and its interest in the dissemination of information regarding the financing of political speech, we conclude that the law does not, on its face, offend the First Amendment.
2. Independent Expenditure Provision
We similarly find that Maine's independent expenditure reporting provision poses no First Amendment concerns. The law primarily obligates anyone spending more than an aggregate of $100 for communications expressly advocating the election or defeat of a candidate to report the expenditure to the Commission. Me.Rev.Stat. tit. 21–A, § 1019–B(1)(A), (3). Reviewing a prior, substantially similar version of this provision in Daggett v. Commission on Governmental Ethics and Election Practices, 205 F.3d 445, 466 (1st Cir.2000), we held that “the modest amount of information requested is not unduly burdensome and ties directly and closely to the relevant government interests.” We see no reason to depart from that conclusion here.
The independent expenditure law also presumptively requires a report of any expenditure over $100 for communications naming or depicting a clearly identified candidate within a set period prior to any election. Me.Rev.Stat. tit. 21–A, § 1019–B(1)(B), (3). Though we did not review this aspect of the law in Daggett, the Supreme Court upheld in Citizens United a similar provision of federal election law that required disclosure in connection with expenditures for electioneering communications (communications made shortly before an election that refer to a clearly identified candidate for federal office). 130 S.Ct. at 913–16. In so doing, the Court noted that “the public has an interest in knowing who is speaking about a candidate shortly before an election.” 130 S.Ct. at 915–16. The law here is perhaps more tailored than that at issue in Citizens United, as it offers an opportunity to rebut the presumption that a communication made shortly before an election and identifying a candidate had the “intent to influence the nomination, election or defeat of a candidate.” Me.Rev.Stat. tit 21–A, § 1019–B(2). Regardless, the information that must be reported under this subsection is, as Daggett found, “modest,” 205 F.3d at 466, and it bears a substantial relation to the public's “interest in knowing who is speaking about a candidate shortly before an election.” Citizens United, 130 S.Ct. at 915–16.
NOM argues that Maine lacks a “sufficiently important” interest in the $100 threshold at which the reporting requirement adheres, and, alternatively, that the threshold lacks a “substantial relation” to a sufficiently important governmental interest. NOM's argument operates from a mistaken premise; we do not review reporting thresholds under the “exacting scrutiny” framework. In Buckley, facing a similar challenge to a $10 threshold for a recordkeeping provision and a $100 reporting threshold, the Supreme Court noted that the choice of where to set such monetary thresholds “is necessarily a judgmental decision, best left in the context of this complex legislation to congressional discretion.” 424 U.S. at 83. The Court concluded that, although there was no evidence in the record that Congress had “focused carefully on the appropriate level at which to require recording and disclosure,” and despite the fact that the low thresholds might “discourage participation by some citizens in the political process,” it could not say that “the limits designated are wholly without rationality.” Id.; see also id. n.111 (“[W]hen it is seen that a line or point there must be, and that there is no mathematical or logical way of fixing it precisely, the decision of the legislature must be accepted unless we can say that it is very wide of any reasonable mark.” (quoting Louisville Gas Co. v. Coleman, 277 U.S. 32, 41 (1928) (Holmes, J., dissenting))). The Court thus upheld FECA's recordkeeping and reporting thresholds.
Following Buckley, we have granted “judicial deference to plausible legislative judgments” as to the appropriate location of a reporting threshold, and have upheld such legislative determinations unless they are “ ‘wholly without rationality.’ “ Vote Choice, Inc. v. DiStefano, 4 F.3d 26, 32–33 (1st Cir.1993) (quoting Buckley, 424 U.S. at 83). In Daggett, for example, we applied these standards in rejecting a challenge to the $50 reporting threshold in the prior iteration of Maine's independent expenditure law. 205 F.3d at 466 (“We remain unconvinced ․ that, if $100 was an appropriate threshold for requiring the reporting of independent expenditures in federal elections in Buckley, $50 is an illegitimate threshold for Maine elections.”).
Despite the fact that the threshold has been doubled since Daggett, NOM argues that we should find the line unconstitutional because it is not indexed to inflation. In so arguing, it relies on an observation in Randall v. Sorrell, 548 U.S. 230, 261 (2006), that “[a] failure to index limits means that limits which are already suspiciously low ․ will almost inevitably become too low over time.” The limits at issue in Sorrell, however, were substantive contribution limits, the setting of which presents different considerations than the determination of the threshold for a reporting requirement,35 and which is subject to different standards of review. Neither we nor the Supreme Court has ever second-guessed a legislative decision not to index a reporting requirement to inflation. Indeed, in Buckley, the Court acknowledged that Congress, in setting FECA's $100 reporting threshold, appeared to have simply adopted the threshold used in similar disclosure laws since 1910—i.e., over the course of more than sixty years, without any adjustment for inflation. 424 U.S. at 83. We thus reject NOM's argument that the $100 threshold is unconstitutional simply because it is static. Moreover, we cannot conclude that Maine's choice of a $100 threshold, double the amount we upheld just a decade ago in Daggett, is wholly without rationality.
3. Disclaimer and Attribution Provisions
Finally, we agree with the district court that “Citizens United has effectively disposed of any attack on Maine's attribution and disclaimer requirements.” Nat'l Org. for Marriage, 723 F.Supp.2d at 267. NOM argues that Maine's “attribution and disclaimer requirements are so great that the government's interest does not reflect the burden on speech,” as the required disclosures will “distract readers and listeners from NOM's message.” We disagree. The requirements are minimal, calling only for a statement of whether the message was authorized by a candidate and disclosure of the name and address of the person who made or financed the communication. Me.Rev.Stat. tit. 21–A, § 1014(1)-(2). These are precisely the requirements approved in Citizens United,36 see 130 S.Ct. at 913–14 (citing 2 U.S.C. § 441d), and they bear a close relation to Maine's interest in dissemination of information regarding the financing of political messages. The disclaimer and attribution requirements are, on their face, unquestionably constitutional.37
IV. Due Process Vagueness Challenges
Having found that each of the challenged statutes pass muster under the First Amendment, we turn next to NOM's contention that portions of the statutes are unconstitutionally vague.
The vagueness doctrine, a derivative of due process, protects against the ills of laws whose “prohibitions are not clearly defined.” Grayned v. City of Rockford, 408 U.S. 104, 108 (1972); see also Williams, 553 U.S. at 304. In prohibiting overly vague laws, the doctrine seeks to ensure that persons of ordinary intelligence have “fair warning” of what a law prohibits, prevent “arbitrary and discriminatory enforcement” of laws by requiring that they “provide explicit standards for those who apply them,” and, in cases where the “statute ‘abut(s) upon sensitive areas of basic First Amendment freedoms,’ “ avoid chilling the exercise of First Amendment rights. Grayned, 408 U.S. at 108–09 (alteration in original) (quoting Baggett v. Bullitt, 377 U.S. 360, 372 (1964)). In view of this last interest, the Constitution requires a “ ‘greater degree of specificity’ “ in cases involving First Amendment rights. Buckley, 424 U.S. at 77 (quoting Smith v. Goguen, 415 U.S. 566, 573 (1974)).
Even under the heightened standard for First Amendment cases, though, not all vagueness rises to the level of constitutional concern. “Many statutes will have some inherent vagueness, for ‘[i]n most English words and phrases there lurk uncertainties.’ “ Rose v. Locke, 423 U.S. 48, 49–50 (1975) (per curiam) (quoting Robinson v. United States, 324 U.S. 282, 286 (1945)); see also Ward v. Rock Against Racism, 491 U.S. 781, 794 (1989) (“[P]erfect clarity and precise guidance have never been required even of regulations that restrict expressive activity.”). Moreover, “[t]he mere fact that a regulation requires interpretation does not make it vague.” Ridley v. Mass. Bay Transp. Auth., 390 F.3d 65, 93 (1st Cir.2004). We have thus said that “a statute is unconstitutionally vague only if it ‘prohibits ․ an act in terms so uncertain that persons of average intelligence would have no choice but to guess at its meaning and modes of application.’ “ United States v. Councilman, 418 F.3d 67, 84 (1st Cir.2005) (en banc) (quoting United States v. Hussein, 351 F.3d 9, 14 (1st Cir .2003)).
With these standards in mind, we review NOM's vagueness challenges de novo. Hussein, 351 F.3d at 14. NOM poses challenges to three sets of terms, and variations thereof: (1) “promoting,” “support,” and “opposition”; (2) “influencing”; and (3) “initiation.” In addition, NOM claims that the definition of “expressly advocate” is unconstitutionally vague because it invites the use of context to determine the purpose of a communication.
A. “Promoting,” “Support,” and “Opposition”
We begin with NOM's vagueness challenge to variations of the terms “promoting,” “support,” and “opposition,” which appear in three separate provisions:38 the definition of a non-major-purpose PAC,39 the PAC statute's definition of an “expenditure,”40 and the independent expenditure provision.41 The district court held each of these terms to be sufficiently clear to evade due process concerns. We agree.
The Supreme Court rejected a vagueness challenge to substantially similar statutory language in McConnell, 540 U.S. 93, overruled on other grounds by Citizens United, 130 S.Ct. 876. The language at issue was a provision of federal election law defining “Federal election activity” to include “a public communication that refers to a clearly identified candidate for Federal office ․ and that promotes or supports a candidate for that office, or attacks or opposes a candidate for that office (regardless of whether the communication expressly advocates a vote for or against a candidate).” 2 U.S.C. § 431(20)(A)(iii). Applying due process standards, the Court observed that “[t]he words ‘promote,’ ‘oppose,’ ‘attack,’ and ‘support’ clearly set forth the confines within which potential party speakers must act in order to avoid triggering the provision.” McConnell, 540 U.S. at 170 n.64. The Court concluded that “[t]hese words ‘provide explicit standards for those who apply them’ and ‘give the person of ordinary intelligence a reasonable opportunity to know what is prohibited,’ “ and thus held that the provision was not unconstitutionally vague. Id. (quoting Grayned, 408 U.S. at 108–09).
NOM acknowledges McConnell's relevance, but argues that the opinion's holding is limited to the context of the federal law at issue there, citing several authorities that purportedly held similar statutory language to be “vague and overbroad vis-à-vis other speech or other speakers.” NOM's argument is misguided. The authorities NOM cites—circuit court opinions and a partial concurrence to the Court's 2007 decision in Wisconsin Right to Life—address the conceptually distinct question of whether terms such as “promote,” “oppose,” “attack,” and “support” maintain an acceptably clear distinction between express campaign advocacy and issue advocacy. See Wis. Right to Life, 551 U.S. at 492–93 (Scalia, J., concurring); Ctr. for Individual Freedom v. Carmouche, 449 F.3d 655, 662–66 (5th Cir.2006); N.C. Right to Life, Inc. v. Bartlett, 168 F.3d 705, 712–13 (4th Cir.1999). This is, as we have discussed, primarily an overbreadth issue, and we have already rejected NOM's arguments that the statutes here are unconstitutionally overbroad.
If, on the other hand, NOM offers these authorities solely for the purpose of countering McConnell's vagueness holding-which is the relevant point here-they also fall short of the mark. None of the cited cases is a majority Supreme Court opinion issued after McConnell, so McConnell remains the leading authority relevant to interpretation of the terms before us. Of course, the statutes here are distinct from the provision that McConnell construed, and thus the Court's reading is not dispositive. However, contrary to NOM's assertion, the statutory context here is close enough to McConnell to make the Court's conclusion that the terms are not vague particularly persuasive. In each of the provisions, the terms “promote”/”promoting,” “support,” and “oppose”/”opposition” have an election-related object: “candidate” in the federal law, 2 U.S.C. § 431(20)(A)(iii), and “candidate,” “nomination or election of any candidate” and “campaign, referendum or initiative” in the Maine provisions, Me.Rev.Stat. tit. 21–A, §§ 1019–B(3)(B), 1052(4)(A)(1), (5)(A)(5). If anything, the terms of Maine's statutes provide slightly more clarity: for example, § 1052(5)(A)(5)'s reference to “promoting ․ the nomination or election of any candidate” is more precise than the federal law's reference to “promot[ing] ․ a candidate,” 2 U.S.C. § 431(20)(A)(iii). We thus find the use of “promoting,” “support,” and “opposition” in §§ 1019–B and 1052 clear enough to “give the person of ordinary intelligence a reasonable opportunity to know what is prohibited.” Grayned, 408 U.S. at 108.
The term “influencing” (appearing also as “influence”) presents a closer question.42 The district court held “influencing” to be unconstitutionally vague and severed it from the various statutes challenged here. On appeal, defendants urge that we find “influencing” sufficiently clear to avoid due process concerns. NOM, in turn, contends that the district court erred in severing “influencing,” suggesting that we should find the statutes unconstitutional in their entirety. Because we agree with defendants that the use of “influencing” in the statutes before us is, when given a properly limited meaning, not unconstitutionally vague, we need not reach the question of severance.
In arguing that “influencing” is unconstitutionally vague, NOM relies on the Supreme Court's construction of similar language in Buckley v. Valeo. The relevant portion of Buckley concerned a disclosure requirement applicable to anyone “ ‘who makes contributions or expenditures' aggregating over $100 in a calendar year.” Buckley, 424 U.S. at 74–75 (quoting 2 U.S.C. § 434(e)). The statute defined “expenditures” to include “the use of money or other valuable assets ‘for the purpose of ․ influencing’ the nomination or election of candidates for federal office.” Id . at 77 (quoting 2 U.S.C. § 431(f)). The Court noted that the “ambiguity” of the phrase “for the purpose of influencing” “poses constitutional problems” and “raises serious problems of vagueness,” id. at 76–77, in that it had the “potential for encompassing both issue discussion and advocacy of a political result,” id. at 79. This was, of course, the same concern the Court raised with respect to the phrase “relative to a candidate” in FECA's independent expenditures cap, and the Court reached an identical solution. The Court skirted its constitutional concerns by imposing a limiting construction on the definition of expenditure “to reach only funds used for communications that expressly advocate the election or defeat of a clearly identified candidate.” Id. at 80 (footnote omitted).
NOM's argument that Buckley dictates a finding of vagueness here is flawed on several counts. First, as more recent Supreme Court precedents have made clear, Buckley's narrowing interpretation of the phrase “for the purpose of influencing” “was the product of statutory interpretation rather than a constitutional command.” McConnell, 540 U.S. at 192. The Court never squarely held in Buckley that the term “influencing” was unconstitutionally vague under due process standards, and the constitutional concern that prompted the Court to narrow the term—the fear that the statute might be read to reach issue discussion—is, as we have said, not a relevant one for review of disclosure laws. Second, even if Buckley were to have found “influencing” unconstitutionally vague in FECA, it would not be dispositive of the question here. Terms claimed to be vague must be interpreted in light of their precise statutory context, see URI Student Senate v. Town of Narragansett, 631 F.3d 1, 14 (1st Cir.2011); Welch v. United States, 750 F.2d 1101, 1112 (1st Cir.1985), and thus a phrase deemed problematic in federal election statutes might not run afoul of vagueness standards in Maine's statutes.
Nonetheless, Buckley's concerns aside, the term “influencing” does present some vagueness problems. The other candidate-related terms employed by the statutes here—such as “promoting,” “opposition,” “defeat,” and “support,” Me.Rev.Stat. tit. 21–A, §§ 1019–B(3)(B), 1052(4)(A)(1), (5)(A)(5)—are more plainly result-oriented, focusing on advocacy for or against a particular candidacy. Influence, on the other hand, covers a wider range of objectives. Conceivably falling within the meaning of “influence” are objectives as varied as advocacy for or against a candidate's election; championing an issue for inclusion in a candidate's platform; and encouraging all candidates to embrace public funding. Without more context, we believe the intended meaning of “influence” to be uncertain enough that a person of average intelligence would be forced to “ ‘guess at its meaning and modes of application.’ “ Councilman, 418 F.3d at 84 (quoting Hussein, 351 F.3d at 14).
Arguing that the statutes' use of “influencing” is adequately clear, defendants point us to the interpretive canon of noscitur a sociis, which provides that an ambiguous statutory term may be “given more precise content by the neighboring words with which it is associated.” Williams, 553 U.S. at 294. For example, in the non-major-purpose PAC definition, defendants suggest that “influencing” should be given a meaning similar to or consistent with “promoting” and “defeating.” See Me.Rev.Stat. tit. 21–A, § 1052(5)(A)(5) ( “for the purpose of promoting, defeating or influencing in any way”). This argument fails for two reasons.
First, “influencing” appears on its own in some of the statutes before us, thus defeating the noscitur a sociis exercise for those provisions. See, e.g., id. § 1014(2–A) (“The disclosure is not required if the communication was not made for the purpose of influencing the candidate's nomination for election or election.”). Second, in those statutes where “influencing” is paired with other terms, we find more persuasive the countervailing interpretive canon counseling that a statute should “ ‘be so construed that, if it can be prevented, no clause, sentence, or word shall be superfluous, void, or insignificant.’ “ TRW Inc. v. Andrews, 534 U.S. 19, 31 (2001) (quoting Duncan v. Walker, 533 U.S. 167, 174 (2001)). Our interpretation is guided by the fact that, even where it appears with other terms, “influencing” appears to have been intentionally set apart. For example, the relevant language of § 1052(5)(A)(5) reads “for the purpose of promoting, defeating or influencing in any way;” the addition of “in any way” logically gives “influencing” a broader sweep than the foregoing terms. The differentiation is even more apparent in the PAC statute's definition of expenditure, where the term “influencing” appears in a different clause and is given a different (though related) object from the other terms. See Me.Rev.Stat. tit. 21–A, § 1052(4)(A)(1) (defining expenditure to cover the transfer of anything of value “for the purpose of influencing the nomination or election of any person to political office; or for the initiation, support or defeat of a campaign, referendum or initiative.”). The natural inference from this separation is that the drafters intended “influencing” to carry a different meaning from the words with which it appears.
Despite their continued insistence that the use of “influencing” in the statutes here is not vague, defendants recognize that we, like the district court, might find “influencing” insufficiently clear on its face to satisfy due process standards. Therefore, as a fallback position, defendants offer a narrowing construction that has been adopted by the Commission with respect to a separate statute regulating ballot question committees. In written guidance, the Commission has clarified that it interprets the phrase “for the purpose of initiating, promoting, defeating or influencing in any way a campaign,” Me.Rev.Stat. tit. 21–A, § 1056–B, in the context of ballot-question campaigns, to “include communications and activities which expressly advocate for or against a ballot question or which clearly identify a ballot question by apparent and unambiguous reference and are susceptible of no reasonable interpretation other than to promote or oppose the ballot question,” Me. Comm'n on Governmental Ethics & Elections Practices, Guidance on Reporting as a Ballot Question Committee, available at http://www.maine.gov/ethics/bqcs/guidance.htm (last visited July 25, 2011). This narrowing construction was not offered to the district court.43 However, there is no barrier to our considering it here—and, indeed, we are required to do so. See Hoffman Estates, 455 U.S. at 494 n.5 (“In evaluating a facial challenge to a state law, a federal court must ․ consider any limiting construction that a state court or enforcement agency has proffered.”).
As narrowed, the terms “influencing” and “influence,” as used in the statutes at issue here, would include only “communications and activities that expressly advocate for or against [a candidate] or that clearly identify a candidate by apparent and unambiguous reference and are susceptible of no reasonable interpretation other than to promote or oppose the candidate.” This narrowed formulation is considerably more precise than the original, and succeeds both in “provid[ing] explicit standards for those who apply” the provisions at issue here and in ensuring that persons of average intelligence will have reasonable notice of the provisions' coverage. Grayned, 408 U.S. at 108. We thus conclude that the provisions' use of the terms “influencing” and “influence,” so limited, is not so vague as to offend due process.
Among the statutes at issue in this appeal, the term “initiation” appears only in the PAC statute's definition of “expenditure.”44 NOM offers no support for its argument that “initiation” is vague, contending only that “initiation” “fare[s] no better” than the other terms challenged on vagueness grounds. We find “initiation” to be adequately clear. The context-defining “expenditure” to include giving something of value “for the initiation ․ of a campaign”—makes plain that “initiation” is being used as the noun form of the verb “initiate,” the primary definition of which is “to begin, set going, or originate.” The Random House Dictionary of the English Language 982 (2d ed. unabr.1987). Used in this way, the language is unequivocal, and easily would put an individual of average aptitude on notice that the act of incurring an expense for the purpose of beginning an electoral campaign will constitute an “expenditure” within the meaning of § 1052(4)(A)(1). See Grayned, 408 U.S. at 108. NOM's argument that “initiation” is unconstitutionally vague thus fails.
D. Use of Context in Definition of “Expressly Advocate”
NOM's final vagueness argument is somewhat distinct from the preceding ones. While NOM's claim focuses on the phrase “expressly advocate” in the independent expenditure statute,45 NOM does not contend that the phrase itself is unconstitutionally vague. Instead, NOM argues that Maine's definition of “expressly advocate,” set forth in regulations promulgated by the Commission, renders the term vague because it invites reliance on a communication's context and employs a purportedly unconstitutional “appeal-to-vote” formulation for determining what qualifies as express advocacy. Specifically, the regulations provide that a communication will be considered to “expressly advocate” when it employs phrases that “in context can have no other reasonable meaning than to urge the election or defeat of one or more clearly identified candidate(s), such as posters, bumper stickers, advertisements, etc. which say ‘Pick Berry,’ ‘Harris in 2000,’ ‘Murphy/Stevens' or ‘Canavan!’ “ 94–270–001 Me.Code R. § 10(2)(B).46 As we explain, NOM's arguments read far too much into a limited line of Supreme Court precedents, and provide no basis for concluding that Maine's regulations are unconstitutionally vague.
NOM's arguments have their roots in the recent trio of Supreme Court cases addressing the constitutionality of the federal prohibition of independent expenditures by corporations and unions for “electioneering” communications-those made shortly before a primary or general election that clearly identify a candidate for federal office. The trio began with McConnell, in which the Court upheld the electioneering provision against a facial overbreadth challenge. In so doing, the Court found unavailing the contention that the provision would regulate a substantial amount of issue advocacy, noting that the argument “fail[ed] to the extent that the issue ads broadcast during the [relevant period] are the functional equivalent of express advocacy.” McConnell, 540 U.S. at 206.
This conditional assertion was put to the test several years later in Wisconsin Right to Life, when the Court, entertaining an as-applied challenge to the electioneering provision, considered whether several specific advertisements qualified as the “functional equivalent of express advocacy.” The Court concluded they did not, and accordingly held the provision unconstitutional in its application. Along the way, the principal opinion made two points relevant to NOM's arguments here. First, it suggested that an advertisement would qualify as the “functional equivalent of express advocacy,” and thereby could be regulated without triggering overbreadth concerns, only when it “is susceptible of no reasonable interpretation other than as an appeal to vote for or against a specific candidate.”47 Wis. Right to Life, 551 U.S. at 469–70. NOM refers to this formulation as the “appeal-to-vote test.” Second, the Court criticized efforts to use the advertisements' context to determine whether they qualified as the “ ‘functional equivalent’ of express advocacy,” noting that “contextual factors of the sort invoked [there] should seldom play a significant role in the inquiry.” Id. at 473–74.
Most recently, the Court concluded in Citizens United that Congress could not limit the campaign-related speech of corporations and unions and thus held the electioneering provision unconstitutional, overturning McConnell. Citizens United provides the launching point for NOM's first argument that Maine's definition of “expressly advocate” is vague. NOM contends that Citizens United eliminated “the appeal-to-vote test as a constitutional limit on government power,” and reads into this an implicit holding that the test was unconstitutionally vague.
NOM's reading finds no support in the text of Citizens United, though we agree with NOM that, in striking down the federal electioneering expenditure statute, Citizens United eliminated the context in which the appeal-to-vote test has had any significance.48 It is a large and unsubstantiated jump, however, to read Citizens United as casting doubt on the constitutionality of any statute or regulation using language similar to the appeal-to-vote test to define the scope of its coverage. The basis for Citizens United's holding on the constitutionality of the electioneering expenditure statute had nothing to do with the appeal-to-vote test or the divide between express and issue advocacy. Instead, the decision turned on a reconsideration of prior case law holding that a corporation's political speech may be subjected to greater regulation than an individual's. See Citizens United, 130 S.Ct. at 886. The opinion offered no view on the clarity of the appeal-to-vote test. In fact, the Court itself relied on the appeal-to-vote test in disposing of a threshold argument that the appeal should be resolved on narrower, as-applied grounds. See id. at 889–90 (applying appeal-to-vote test in determining that advertisements at issue were the functional equivalent of express advocacy).
We find similarly misguided NOM's argument that the definition of “expressly advocate” is vague due to the regulation's reference to consideration of an advertisement's words “in context.” NOM misinterprets Wisconsin Right to Life in suggesting that the principal opinion barred all consideration of context to determine whether an advertisement was the functional equivalent of express advocacy. To the contrary, the opinion explicitly acknowledges that “[c]ourts need not ignore basic background information that may be necessary to put an ad in context.” Wis. Right to Life, 551 U.S. at 474.49 It is apparent from the examples provided by the regulation here—“ ‘Pick Berry,’ ‘Harris in 2000,’ ‘Murphy/Stevens' or ‘Canavan!’ “—that “knowing that Berry is a candidate to be picked on the ballot, that 2000 is an election where Harris should win, etc.,” Nat'l Org. for Marriage, 723 F.Supp.2d at 266, is precisely the sort of basic background information that may be consulted in the express advocacy determination.
In any event, we find the regulation's definition of “expressly advocate,” as a whole, to be sufficiently clear to satisfy due process. The definition offers abundant examples (fourteen in all) of the sorts of language that will constitute express advocacy, and, as we have noted before, “[t]he existence of clear examples of conduct covered by a law may ․ help to insulate the law against an accusation of vagueness.” URI Student Senate, 631 F.3d at 14; see also Hotel & Motel Ass'n v. City of Oakland, 344 F.3d 959, 972–73 (9th Cir.2003) (finding ordinance provided sufficient notice where it listed “no less than nineteen specific examples of the types of conduct to which th[e] provision applie[d]”). Moreover, the phrase set forth in the regulation—“can have no other reasonable meaning than to urge the election or defeat of one or more clearly identified candidate(s)”-is certainly as clear, if not more so, as words such as “support” and “promote” that the Supreme Court has held non-vague. See McConnell, 540 U.S. at 170 n.64; see also Wis. Right to Life, 551 U.S. at 474 n.7 (explaining that the appeal-to-vote formulation meets the “imperative for clarity” in regulation of political speech). We therefore reject NOM's arguments that the regulation's definition of the phrase “expressly advocate” is unconstitutionally vague.
V. Challenge to the District Court's Unsealing Order
The remaining issue in this appeal is whether the district court erred in ruling that the trial record must be unsealed. Reviewing the court's unsealing order under a deferential standard, see Siedle v. Putnam Invs., Inc., 147 F.3d 7, 10 (1st Cir.1998) (unsealing orders are reviewed “only for mistake of law or abuse of discretion”), we find no abuse of discretion.
Decisions on the sealing of judicial documents require a balancing of interests, although the scales tilt decidedly toward transparency. The starting point must always be the common-law presumption in favor of public access to judicial records. See Nixon v. Warner Commc'ns, Inc., 435 U.S. 589, 597 (1978); Siedle, 147 F.3d at 9. As we have noted in prior cases, “[p]ublic access to judicial records and documents allows the citizenry to ‘monitor the functioning of our courts, thereby insuring quality, honesty and respect for our legal system.’ “ FTC v. Standard Fin. Mgmt. Corp., 830 F.2d 404, 410 (1st Cir.1987) (quoting In re Cont'l Ill. Secs. Litig., 732 F.2d 1302, 1308 (7th Cir.1984)). The presumption favoring public access, which extends to both civil and criminal trials, is not inviolate, and may on some occasions be overcome by competing interests. Siedle, 147 F.3d at 10; see also id. at 10–12 (finding abuse of discretion where unsealing order would make public information that was likely subject to the attorney-client privilege and a confidentiality agreement). That said, “the presumption is nonetheless strong and sturdy,” and thus “ ‘[o]nly the most compelling reasons can justify non-disclosure of judicial records.’ “ Standard Fin. Mgmt. Corp., 830 F.2d at 410 (alteration in original) (quoting In re Knoxville News–Sentinel Co., 723 F.2d 470, 476 (6th Cir.1983)).
Portions of the trial record here were initially filed in sealed form, albeit by the parties' stipulation rather than court order.50 Following trial, the district court issued sua sponte an order to show cause why the entire record should not be filed in publicly accessible form. NOM responded with a brief arguing, inter alia, that disclosure of certain of its strategic documents included in the record would severely burden NOM's ability to effectively engage in protected political activities, and would invade the privacy of NOM's third-party service providers and contractors identified in the record and risk subjecting them to harassment. The district court found NOM's arguments unavailing and ordered the record unsealed such that it would be “public in precisely the way that it would have been had live witnesses been called to testify.” Nat'l Org. for Marriage, 723 F.Supp.2d at 249 n.4. We granted an emergency motion to stay the unsealing order during the pendency of this appeal.
On appeal, NOM fields two arguments for abuse of discretion. It first argues that the district court erred in unsealing the documents without a “finding of true necessity.” NOM's argument flips the proper analysis on its head. The presumption here favors openness, and a court need make no finding, let alone one of “true necessity,” in order to make the proceedings and documents in a civil trial public. Instead, it is the party seeking to keep documents sealed who must make a showing sufficient to overcome the presumption of public access. See Standard Fin. Mgmt. Corp., 830 F.2d at 411. Second, NOM suggests that the district court erred in failing to consider a number of controlling legal principles. On examination, the authorities it cites are, without exception, inapposite.51
On the record before us, we cannot conclude that the district court abused its discretion in ordering the trial record unsealed. While NOM claims harm from disclosure of certain strategic documents, neither before the district court nor in this appeal has NOM identified any specific information that, if made public, would damage or chill its political advocacy efforts. Indeed, the documents it identifies as particularly sensitive, including a strategic planning document it terms its “playbook,” disclose primarily advocacy priorities and expenditures in past election cycles, and we see little among them that could advantage NOM's opponents going forward. NOM's claims that its contractors and service-providers could be subject to harassment also lack support, resting upon allegations of harassment against a vendor that performed work for supporters of California's Proposition 8. While “ ‘privacy rights of participants and third parties[ ] are among those interests which, in appropriate cases, can limit the presumptive right of access to judicial records,’ “ Standard Fin. Mgmt. Corp., 830 F.2d at 411 (internal quotation marks omitted) (quoting In re Knoxville News–Sentinel Co., 723 F.2d at 478), NOM failed to make a compelling case that the specific vendors referenced in the documents here have any reasonable privacy concerns relating to the disclosure of their business relationship with NOM.
For the reasons set forth above, we vacate the portion of the district court's judgment finding the terms “influencing” and “influence” unconstitutionally vague, remand for entry of judgment in defendants' favor in full on those claims, and affirm the judgment in all other respects. We also vacate our stay of the district court's unsealing order. Costs shall be awarded to the appellees.
1. In a companion opinion filed today, we consider appellant's challenges to a related provision of Rhode Island's election laws. See Nat'l Org. for Marriage v. Daluz, No. 10–2304 (1st Cir.2011).
2. The court also held unconstitutional a regulation governing the timing of disclosures. That holding is not at issue in this appeal.
3. A candidate will be considered “clearly identified” where the name or a depiction of the candidate appears, or where the “identity of the candidate is apparent by unambiguous reference.” Me.Rev.Stat. tit. 21–A, § 1012(1).
4. As with the independent expenditure reporting requirements, the relevant period is twenty-one days before a primary and thirty-five days before a general election. Me.Rev.Stat. tit. 21–A, § 1014(2–A).
5. NOM was joined in its initial complaint by co-plaintiff American Principles in Action. The specific claims at issue in this appeal were only pursued by NOM, and, accordingly, American Principles in Action does not appear as an appellant.
6. The constitutionality of § 1056–B is not at issue in this appeal.
7. The hearing addressed solely the specific claims at issue in this appeal, leaving the challenges to § 1056–B for a later summary judgment hearing.
8. The court declined, however, to reach the constitutionality of the “major-purpose” PAC provision, § 1052(5)(A)(4), noting that the record showed no basis for concluding that NOM would be subject to that regulation. See Nat'l Org. for Marriage, 723 F.Supp.2d at 254.
9. See Me.Rev.Stat. tit. 21–A, § 1052(5)(A)(5) (defining non-major-purpose PAC to include “[a]ny organization that ․ receives contributions or makes expenditures aggregating more than $5,000 ․ for the purpose of promoting, defeating or influencing in any way the nomination or election of any candidate to political office”).
10. See Me.Rev.Stat. tit. 21–A, § 1019–B(2) (“A person presumed ․ to have made an independent expenditure may rebut the presumption by filing a signed written statement ․ stating that the cost was not incurred with the intent to influence the nomination, election or defeat of a candidate․”).
11. See Me.Rev.Stat. tit. 21–A, § 1014(2–A) (“[D]isclosure is not required if the communication was not made for the purpose of influencing the candidate's ․ election.”).
12. The other statutes contained additional advocacy-related terms that remained after “influencing” was stricken-e.g., “promoting” and “defeating,” Me.Rev.Stat. tit. 21–A, § 1052(5)(A)(5)-whereas the independent expenditure rebuttal procedure turned solely on the word “influence,” see id. § 1019–B(2) (applying to expenditures made “with the intent to influence” candidate elections).
13. The district court reached this conclusion primarily on the basis of Citizens United, which upheld a similar federal-law disclosure provision for advertisements made shortly before an election. The Court ruled that the statute, which contained no rebuttal provision analogous to Maine's, passed First Amendment muster in light of a general “public ․ interest in knowing who is speaking about a candidate shortly before an election.”Citizens United, 130 S.Ct. at 915.
14. We subsequently stayed the unsealing order upon NOM's emergency motion for a stay pending appeal.
15. Such a targeted approach is consistent with the claim-specific nature of standing; “standing is not dispensed in gross,” Lewis v. Casey, 518 U.S. 343, 358 n.6 (1996), and thus a plaintiff must separately prove standing “for each claim he seeks to press.” DaimlerChrysler Corp. v. Cuno, 547 U.S. 332, 352 (2006).
16. Defendants do not appear to contest that the second and third prongs of the constitutional standing analysis would be met here; to the extent that the plaintiff has suffered a cognizable injury at all, the injury can “be traced to the existence and threatened enforcement of the challenged statutes,” and is redressable in an action against the defendants here, who are charged with enforcing Maine's election laws. N.H. Right to Life Political Action Comm. v. Gardner, 99 F.3d 8, 13 (1st Cir.1996).
17. At the time the district court issued its merits decision in August 2010, “NOM ha[d] endorsed no one, d[id] not ․ plan to make expenditures, and did not even budget for expenditures in [the 2010] Maine election cycle.” Nat'l Org. for Marriage, 723 F.Supp.2d at 258.
18. As defendants note, NOM did expend some resources in two legislative races in September 2010 (after the district court issued its opinion). Specifically, NOM sent out postcards that read: “In May 2009, the Maine Legislature approved homosexual ‘marriage.’ Rep. Linda Valentino and Rep. Donald Pilon voted to support same-sex ‘marriage.’ Now it's time to let Don Pilon and Linda Valentino know we don't agree with their decision to back same-sex ‘marriage.’ Email [them] ․ and tell them they stand on the wrong side of House District[s] 133 and 134.” The other side of the postcards contained pictures of the candidates' opponents, identified them as “stand[ing] for marriage as between one man and one woman,” and urged voters to email them to “thank them for standing for traditional marriage.” However, because these mailings took place after NOM had filed its notice of appeal, they are not properly part of the record here and we do not consider them in our standing analysis.
19. As described above, the provision also establishes a narrow exemption from registration and reporting for an out-of-state PAC if its “only financial activity within the State is to make contributions to candidates, party committees, political action committees or ballot question committees.” Me.Rev.Stat. tit. 21–A, § 1053–B. As NOM indicated that it intended to make independent expenditures for political advertising, it would not fall within this exemption.
20. “Expenditure,” which appears in the definition of a non-major-purpose PAC, is defined by another portion of the statute to include a “purchase, payment, distribution, loan, advance, deposit or gift of money or anything of value, made for the purpose of influencing the nomination or election of any person to political office; or for the initiation, support or defeat of a campaign, referendum or initiative, including the collection of signatures for a direct initiative, in this State.” Me.Rev.Stat. tit 21–A, § 1052(4)(A)(1).
21. While Supreme Court precedent does not explicitly brand this an issue of standing, see, e.g., Humanitarian Law Project, 130 S.Ct. at 2719, it is conceptually related to standing doctrine and has been so treated by a number of circuit courts. See, e.g., United States v. Tyler, 281 F.3d 84, 91 n.6 (3d Cir.2002); United States v. Hill, 167 F.3d 1055, 1064 (6th Cir.1999); see also Hunt v. City of Los Angeles, 638 F.3d 703, 710 (9th Cir.2011) (referring to the bar on vagueness challenges by those whose conduct is clearly covered as a “special standing principle[ ]”).
22. See supra note 18.
23. Humanitarian Law Project is not to the contrary. Humanitarian Law Project simply noted, in the context of an as-applied vagueness challenge to a federal criminal statute, that the general rule prohibiting such challenges where the petitioner's speech is clearly proscribed applies in the First Amendment arena. Humanitarian Law Project, 130 S.Ct. at 2719 (“Th[e] rule makes no exception for conduct in the form of speech.”). Consistent with Williams, however, Humanitarian Law Project noted that the petitioner's vagueness arguments might make out a valid claim if framed as an “overbreadth claim under the First Amendment.” Id.
24. The First Amendment is incorporated through the Fourteenth Amendment and thus applies to Maine's laws. Vote Choice, Inc. v. DiStefano, 4 F.3d 26, 31 (1st Cir.1993) (citing N.Y. Times Co. v.. Sullivan, 376 U.S. 254, 276–77 (1964)).
25. The Court provided specific examples of such “express terms,” including “ ‘vote for,’ ‘elect,’ ‘support,’ ‘cast your ballot for,’ ‘Smith for Congress,’ ‘vote against,’ ‘defeat,’ [and] ‘reject.’ “ Buckley, 424 U.S. at 44 n.52.
26. The Court grafted a similar limiting construction onto the language of a disclosure statute, 2 U.S.C. § 434(e), to address perceived problems with its use of the phrase “for the purpose of influencing [a candidate election].” Buckley, 424 U.S. at 80.
27. This is so to the extent that the line between issue advocacy and candidate advocacy was considered a valid distinction at all. Indeed, the majority in McConnell indicated that it was not “persuaded, independent of our precedents, that the First Amendment erects a rigid barrier between express advocacy and so-called issue advocacy,” noting that the “notion cannot be squared with our longstanding recognition that the presence or absence of magic words cannot meaningfully distinguish electioneering speech from a true issue ad.” McConnell, 540 U.S. at 193.
28. In FEC v. Akins, 524 U.S. 11, 27 (1998), the Court explicitly entertained the possibility, but did not decide, that Buckley's “express advocacy” narrowing construction was limited to addressing “the First Amendment problems presented by regulation of ‘independent expenditures.’ “
29. Of the provisions at issue here, Maine's requirement that non-major-purpose PACs register with the Commission is, on its face, the furthest from a traditional disclosure law. In function, however, it too is first and foremost a disclosure provision. The registration requirement does not obligate the PAC to form a separate entity, create a segregated fund, or make any substantive change to its operation or form; the law merely requires the reporting of certain information about the PAC after it crosses the applicable contribution/expenditure threshold (along with certain other de minimis requirements, such as ongoing maintenance of records). See Me.Rev.Stat. tit. 21–A, § 1053.Moreover, Citizens United may be read to suggest that the Court views this type of information-gathering registration requirement as akin to a disclosure requirement. In explaining why it was not importing the express advocacy limitation into its analysis of the disclosure law before it, the Court cited a case upholding against First Amendment challenge a federal law that imposed both disclosure and registration requirements on lobbyists, noting that such requirements were found permissible “even though Congress has no power to ban lobbying itself.” Citizens United, 130 S.Ct. at 915 (citing United States v. Harriss, 347 U.S. 612, 625 (1954)).
30. Additionally, the application of a disclosure requirement may be held to violate the First Amendment where the challengers can show “ ‘a reasonable probability that the compelled disclosure [of personal information] will subject them to threats, harassment, or reprisals from either Government officials or private parties.’ “ Reed, 130 S.Ct. at 2820 (alteration in original) (quoting Buckley, 424 U.S. at 74). NOM has not attempted to make such a showing here with respect to the disclosures required by Maine law.
31. Partly for this reason, the Court refused to consider the option of financing speech through a PAC to be a factor mitigating the corporate and union independent expenditure ban's burden on speech. See Citizens United, 130 S.Ct. at 897 (“Section 441b is a ban on corporate speech notwithstanding the fact that a PAC created by a corporation can still speak.”).
32. The only restriction on a PAC's expenditures is for direct contributions to candidates; PACs are subject to the same per-candidate contribution limits as any other entity or individual.
33. As the Court observed in First National Bank v. Bellotti, 435 U.S. 765 (1978), “the people in our democracy are entrusted with the responsibility for judging and evaluating the relative merits of conflicting arguments. They may consider, in making their judgment, the source and credibility of the advocate.” Id. at 791–92 (footnote omitted).
34. Defendants also cite an interest in “gathering data necessary to enforce substantive election law restrictions.” Though we note that Buckley recognized a similar interest in “gathering the data necessary to detect violations of [FECA's] contribution limitations,” 424 U.S. at 68, we find the informational interest sufficient to support Maine's laws and thus do not reach this second class of interests.
35. For instance, we have held that the First Amendment would permit, in some cases, a first-dollar reporting requirement, see Vote Choice, 4 F.3d at 33, whereas the First Amendment clearly sets a “lower bound” for contribution limits. Sorrell, 548 U.S. at 248.
36. In fact, the statute at issue in Citizens United was slightly more prescriptive, specifying that, for video messages, “[t]he required statement must be made in a ‘clearly spoken manner,’ and displayed on the screen in a ‘clearly readable manner’ for at least four seconds.” 130 S.Ct. at 914 (quoting 2 U.S.C. § 441d(d)(2)).
37. NOM argues that, though Citizens United and other courts may have approved disclaimer and attribution limitations in the precise circumstances before them, such measures have never been approved per se. That may be so, but the mere fact that disclaimer and attribution requirements have not been considered in the environment in which we review them now does not weaken our conclusion that the requirements withstand exacting scrutiny.
38. The term “promoting” also appears in the definition of a major-purpose PAC, Me.Rev.Stat. tit. 21–A, § 1052(5)(A)(4), which we do not address here in light of our holding that NOM lacks standing to challenge the major-purpose PAC provision.
39. The statute defines a non-major-purpose PAC to mean an entity that crosses the requisite threshold of contributions or expenditures “for the purpose of promoting, defeating or influencing in any way” a candidate election. Me.Rev.Stat. tit 21–A, § 1052(5)(A)(5) (emphasis added).
40. The statute defines “expenditure” to include giving something of value “for the initiation, support, or defeat” of a campaign or initiative. Me.Rev.Stat. tit. 21–A, § 1052(4)(A)(1) (emphasis added).
41. The independent expenditure statute instructs that reports submitted pursuant to the provision “must state whether the expenditure is in support of or in opposition to the candidate.” Me.Rev.Stat. tit. 21–A, § 1019–B(3)(B) (emphasis added).
42. “Influencing” appears in a number of places throughout the statutes challenged by NOM. These include: the definition of a non-major-purpose PAC, Me.Rev.Stat. 21–A, § 1052(5)(A)(5) (defining non-major-purpose PAC to mean an entity that crosses the requisite threshold of contributions or expenditures “for the purpose of promoting, defeating or influencing in any way ” a candidate election (emphasis added)); the out-of-state PAC provision, id. § 1053–B (providing that an out-of-state PAC need not register if, among other things, it “has not raised and accepted any contributions during the calendar year to influence an election or campaign in [Maine]” (emphasis added)); the attribution and disclaimer provision, id. § 1014(2–A) (clarifying that disclaimer and attribution requirements do not apply to communications made shortly before an election that name or depict a candidate “if the communication was not made for the purpose of influencing the candidate's ․ election” (emphasis added)); the independent expenditure provision, id. § 1019–B(2) (providing that a person presumed to have made an independent expenditure may rebut the presumption by filing a statement that “the cost was not incurred with the intent to influence the nomination, election or defeat of a candidate” (emphasis added)); and the definitions of “expenditure” used in connection with the attribution and disclaimer, independent expenditure, and PAC provisions, id. §§ 1012(3) (defining “expenditure” to include giving something of value “for the purpose of influencing the nomination or election of any person to political office” (emphasis added)), 1052(4)(A)(1) (same).
43. The Maine Attorney General had offered in prior litigation before the district court a narrowing construction limiting the term “influencing” to express advocacy. See Volle v. Webster, 69 F.Supp.2d 171, 175 (D.Me.1999). In light of the fact that Supreme Court precedent subsequent to that litigation “made clear that the state may regulate speech other than express advocacy,” the Attorney General believed that the narrowing construction was “no longer required” and therefore did not offer it in the proceedings below.
44. The statute defines “expenditure” to include the transfer of something of value “for the initiation, support or defeat of a campaign, referendum or initiative.” Me.Rev.Stat. tit. 21–A, § 1052(4)(A)(1) (emphasis added). NOM also challenges the use of “initiating” in the definition of a major-purpose PAC, but we have held that it lacks standing to pursue that claim.
45. The statute defines an “independent expenditure” to be one “that expressly advocates the election or defeat of a clearly identified candidate.” Me.Rev.Stat. tit. 21–A, § 1019–B(1)(A). Though the attribution and disclaimer provisions also use the phrase “expressly advocating,” see id. § 1014(1), the regulations challenged here relate solely to the definition of an “independent expenditure.”
46. The full text of the definition is as follows:“Expressly advocate” means any communication that uses phrases such as “vote for the Governor,” “reelect your Representative,” “support the Democratic nominee,” “cast your ballot for the Republican challenger for Senate District 1,” “Jones for House of Representatives,” “Jean Smith in 2002,” “vote Pro–Life” or “vote Pro–Choice” accompanied by a listing of clearly identified candidates described as Pro–Life or Pro–Choice, “vote against Old Woody,” “defeat” accompanied by a picture of one or more candidate(s), “reject the incumbent,” or communications of campaign slogan(s) or individual word(s), which in context can have no other reasonable meaning than to urge the election or defeat of one or more clearly identified candidate(s), such as posters, bumper stickers, advertisements, etc. which say “Pick Berry,” “Harris in 2000,” “Murphy/Stevens” or “Canavan!”.94–270–001 Me.Code R. § 10(2)(B).
47. En route to this test, the principal opinion rejected proposed intent- and effect-based standards, i.e. frameworks that would have required inquiry into the intent of the speaker to affect an election or an examination of the actual effect the speech would have on an election or on its target audience. Wis. Right to Life, 551 U.S. at 467–69.
48. We do not agree, however, with NOM's characterization of the appeal-to-vote test, or any of the other tests proposed by the Court for distinguishing between express and issue advocacy, as a “constitutional limit on government power.” Citizens United made clear that at least some forms of regulation may reach issue advocacy, see 130 S.Ct. at 915, and there are substantial questions as to whether the line between issue advocacy and express advocacy is constitutionally rooted, see McConnell, 540 U.S. at 193 (noting doubts that “the First Amendment erects a rigid barrier between express advocacy and so-called issue advocacy”).
49. In Citizens United, the Court also relied on contextual factors in determining that the communication at issue-a ninety-minute documentary about Hillary Clinton-constituted the functional equivalent of express advocacy. See 130 S.Ct. at 890 (“In light of historical footage, interviews with persons critical of her, and voiceover narration, the film would be understood by most viewers as an extended criticism of Senator Clinton's character and her fitness for the office of the Presidency.”).
50. In the course of discovery, the parties entered into a confidentiality agreement, which was entered as a consent order by the magistrate judge overseeing discovery matters. That consent order required, among other things, that any documents designated confidential that were filed with the court be submitted under seal. In subsequently stipulating to a joint trial record, the parties included a number of documents that had previously been filed under seal pursuant to the consent order.
51. For example, NOM cites cases involving the validity of a federal regulation that required release of materials compiled by the FEC during investigations into alleged election law violations, see Am. Fed'n of Labor & Cong. of Indus. Orgs. v. FEC, 333 F.3d 168 (D.C.Cir.2003), and a claim of First Amendment privilege against the compelled disclosure of internal documents pursuant to a court order enforcing a discovery request, see Perry v. Schwarzenegger, 591 F.3d 1147 (9th Cir.2010). Those cases involved the possible compelled disclosure of information to which there was no presumptive right of public access; here, in contrast, the documents at issue were voluntarily included in the record filed with the district court, and thus subject to a presumption of public access.NOM also contends that the two-step inquiry set forth in Press–Enterprise Co. v. Superior Court, 478 U.S. 1, 8 (1986), must be applied to determine whether the documents here are within the public's presumptive right of access. That inquiry relates to the categorical determination of whether a particular type of proceeding or class of court documents falls within the public's right of access, see In re Bos. Herald, Inc., 321 F.3d 174, 182–83 (1st Cir.2003); it does not govern whether individual documents filed with a court should be made public. NOM does not contest that the right of public access extends to the trial record in a civil matter, Siedle, 147 F.3d at 10, and thus Press–Enterprise's two-step inquiry is inapplicable.
LIPEZ, Circuit Judge.