Harvard Law Review Harvard Law Review Harvard Law Review

First Amendment

A Shield for David and a Sword Against Goliath: Protecting Association While Combatting Dark Money Through Proportionality

Election disclosure laws are traditionally used to combat corruption in the political process.1×1. See, e.g., David A. Strauss, Corruption, Equality, and Campaign Finance Reform, 94 Colum. L. Rev. 1369, 1388 (1994). These laws typically require candidates, parties, or political committees to disclose “the identity of contributors and the dollar amount of their contributions.”2×2. Trevor Potter, Buckley v. Valeo, Political Disclosure and the First Amendment, 33 Akron L. Rev. 71, 73 (1999). In the campaign finance context, disclosure can refer to reporting the information, publicly disseminating the information, or using disclaimers that identify “the proponents of particular political messages.” Jennifer A. Heerwig & Katherine Shaw, Through a Glass, Darkly: The Rhetoric and Reality of Campaign Finance Disclosure, 102 Geo. L.J. 1443, 1449 (2014). Shining a light on who is financing a candidate is one of the last remaining tools for regulating campaign finance3×3. Ciara Torres-Spelliscy, Has the Tide Turned in Favor of Disclosure? Revealing Money in Politics After Citizens United and Doe v. Reed, 27 Ga. St. U. L. Rev. 1057, 1102–03 (2011). in a legal landscape that allows near-unlimited spending for certain groups.4×4. See Lear Jiang, Disclosure’s Last Stand? The Need to Clarify the “Informational Interest” Advanced by Campaign Finance Disclosure, 119 Colum. L. Rev. 487, 494 (2019). But dark money threatens this tool.

Dark money refers to financial influences that affect the outcome of elections — through uncoordinated advocacy of a candidate — without being subject to any campaign finance disclosure requirements.5×5. See Frances R. Hill, Dark Money in Motion: Mapping Issues Along the Money Trail, 49 Val. U. L. Rev. 505, 505–07 (2015) (describing the phenomenon of dark money). This is done in part through donating to organizations that are not categorized as political committees6×6. Trevor Potter & Bryson B. Morgan, The History of Undisclosed Spending in U.S. Elections & How 2012 Became the “Dark Money” Election, 27 Notre Dame J.L. Ethics & Pub. Pol’y 383, 463–65 (2013) (describing the “major purpose” test for political committees and the use of 501(c)(4) organizations to avoid such requirements). such as 501(c)(4) groups: a “social welfare” category that includes any group broadly focused on promoting the general welfare of the community.7×7. I.R.C. § 501(c)(4) (2012). 501(c)(4)s span from the National Rifle Association to volunteer firefighter companies8×8. See IRS, Types of Organizations Exempt Under Section 501(c)(4) (2018), https://www.irs.gov/charities-non-profits/other-non-profits/types-of-organizations-exempt-under-section-501c4 [https://perma.cc/859Q-3RQ6]. and have virtually no limit on the legislative drafting and lobbying, advocacy of candidates, and donations to political committees they can engage in.9×9. This lobbying includes “drafting legislation,” “persuading legislators to introduce legislation,” and circulating materials advocating for passage or defeat of a bill, as long as the issues relate to the organization’s social welfare purpose. B. Holly Schadler, The Connection: Strategies for Creating and Operating 501(c)(3)s, 501(c)(4)s and Political Organizations 11 (2012). The ability for donors working through these groups to avoid revealing their identities provides a substantial loophole in a disclosure regime aimed at answering those questions for the polity.10×10. See Richard Briffault, Campaign Finance Disclosure 2.0, 9 Election L.J. 273, 273–74 (2010).

Some advocates have accordingly called for broader disclosure laws that cover organizations like 501(c)(4) groups whose primary social purpose is not political.11×11. See, e.g., Chisun Lee et al., Brennan Ctr. for Justice, Secret Spending in the States 23–26 (2016). Indeed, four states now have some sort of disclosure regulation that includes some 501(c)-class organizations.12×12. Colorado, Washington State, California, and most recently this summer, New Jersey have all passed donor disclosure laws. Andrew Garrahan, New Jersey, Colorado Join Growing List of States Regulating “Dark Money,” Inside Pol. L. (July 2, 2019), https://www.insidepoliticallaw.com/2019/07/02/new-jersey-colorado-join-growing-list-of-states-regulating-dark-money/ [https://perma.cc/3MU9-CR8K]. The most recent of such laws — passed this year in New Jersey — suggests that its goal is to “curb the influence of so-called dark-money groups.”13×13. Nick Corasaniti, Ending Secret “Dark Money” Political Donations in New Jersey, N.Y. Times (June 11, 2019), https://nyti.ms/2F0J6fR [https://perma.cc/WG22-JBQ3].

However, these laws — though admirable in their transparency-related goals — run into significant First Amendment issues. The Supreme Court’s freedom of association jurisprudence, beginning with NAACP v. Alabama ex rel. Patterson,14×14. 357 U.S. 449 (1958). protects the privacy of one’s associations, including membership in organizations engaged in issue advocacy. Indeed, this “right of association . . . was the lifeline that prevented Southern states from using public disclosure, registration, or tax laws to destroy either the NAACP or the civil rights movement.”15×15. L. Darnell Weeden, Fifty Plus Years After the Start of the Civil Rights Movement: A Contextual Analysis of the Freedom of Association for the National Association for the Advancement of Colored People’s Pursuit of Reforming the Law, 12 Fla. Coastal L. Rev. 337, 344–45 (2011).

This Note argues that recent cases upholding broader donor disclosures for nonprofit organizations represent a departure from established First Amendment jurisprudence. The outcomes of these cases are mistaken, and these mistakes have resulted from gradual changes in the “exacting scrutiny” standard required in freedom of association cases outside of the election context. This Note further argues that rather than return to the original standard, which does little to combat the challenge of dark money, courts should employ a proportionality-based approach. Such an approach would allow the First Amendment, in the context of association, to serve as both a shield for David — protecting smaller, more fragile movements’ and organizations’ associational interests in anonymity — while also providing a sword against Goliath — upholding disclosure regulations targeted at larger organizations funneling dark money into elections and harming the democratic process.

Part I describes the development of freedom of association under the First Amendment. After the Supreme Court first recognized the right to association in Patterson, a wave of cases in the late twentieth century upheld its strict scrutiny standard for analyzing government actions that infringed on association. Part II describes the subsequent split in jurisprudence developed by Buckley v. Valeo.16×16. 424 U.S. 1 (1976). The issue of direct campaign contributions created a fork in the road of association jurisprudence; restrictions on someone’s right to freely and privately associate were upheld in light of the compelling government interest to promote transparency in election financing for voters. But Buckley’s universe of election finance cases was never meant to “displace[] the stringent standard set out in [Patterson].”17×17. Ams. for Prosperity Found. v. Becerra, 919 F.3d 1177, 1181 (9th Cir. 2019). Nevertheless, some circuits have recently departed from the traditional doctrinal separation of the Buckley line of cases and the Patterson line of cases, applying Buckley’s lower scrutiny to nonpolitical committees. Part III presents proportionality as a way forward for managing the challenges that dark money legislation aims to combat while preserving the associational freedoms of Patterson.18×18. For a discussion of the major interests at play in disclosure legislation, see Archon Fung, Mary Graham & David Weil, Full Disclosure: The Perils and Promise of Transparency 1–7 (2007); and Heerwig & Shaw, supra note 2, at 1465–70.


A. At First Just a Shield — The Early Association Cases

The right to “freely associate” is not found in the text of the First Amendment. It was first introduced in Patterson,19×19. See Robert M. O’Neil, The Neglected First Amendment Jurisprudence of the Second Justice Harlan, 58 N.Y.U. Ann. Surv. Am. L. 57, 57 (2001). which concerned a state law that required foreign corporations to register a corporate charter before conducting business in the state.20×20. See NAACP v. Alabama ex rel. Patterson, 357 U.S. 449, 451 (1958). When Alabama demanded a list of the National Association for the Advancement of Colored People (NAACP)’s leaders and membership, the government asserted that its interest lay in combatting organizational violations of this foreign corporation registration statute.21×21. Id. at 464.

The NAACP challenged the statute. In light of the threats members faced in being a part of the NAACP’s work, compelled disclosure would directly “induce members to withdraw from the Association and dissuade others from joining it because of fear of exposure of their beliefs shown through their associations.”22×22. Id. at 463. The Court found “privacy in group association may . . . be indispensable to preservation of freedom of association, particularly where a group espouses dissident beliefs.”23×23. Id. at 462 (emphasis added). As such, violations of this associative interest were suspect even when the regulation had an independent, compelling intention like combatting business registration fraud.24×24. Id. at 461 (“[Infringement], even though unintended, may inevitably follow from varied forms of governmental action.”).

The opinion’s discussion of the standard never used the term “exacting scrutiny.” It instead promulgated that “state action which may have the effect of curtailing the freedom to associate is subject to the closest scrutiny25×25. Id. at 460–61 (emphasis added). and that the “interest of the State must be compelling.”26×26. Id. at 463 (emphasis added) (quoting Sweezy v. New Hampshire, 354 U.S. 234, 265 (1957) (Frankfurter, J., concurring)). In Patterson, no such interest could be found.27×27. Weeden, supra note 15, at 346. Commenters have “equated the failure of the State of Alabama in Patterson to show a controlling justification for requiring a disclosure of a membership list as the functional equivalent of a failure to demonstrate a compelling state interest under the strict scrutiny test.” Id.

A year later, another NAACP chapter also refused a demand to produce a list of members in Bates v. City of Little Rock.28×28. 361 U.S. 516, 517 (1960). Citing Patterson, the Court affirmed that “it is now beyond dispute that freedom of association for the purpose of advancing ideas and airing grievances is protected.”29×29. Id. at 523. The Court required the government to demonstrate a “compelling” and “cogent . . . interest in obtaining and making public the membership lists.”30×30. Id. at 524.

Fairly quickly, the Court extended these principles from organizations that were being compelled to produce lists to individuals who were compelled to disclose their associations and organizational memberships. The Court confronted an Arkansas statute that required public school teachers to disclose any organization they were members of or regularly contributed to.31×31. See Shelton v. Tucker, 364 U.S. 479, 480 (1960). The Court recognized the state’s interest in investigating teacher fitness,32×32. Id. at 485. but the law’s burden was impermissibly limitless, requiring teachers to reveal “every conceivable kind of associational tie — social, professional, political, avocational, or religious.”33×33. Id. at 488.

In this discussion, the Court explicitly reaffirmed that narrow tailoring was required for compelled disclosures, saying even a compelling interest “cannot be pursued by means that broadly stifle fundamental personal liberties when the end can be more narrowly achieved.”34×34. Id. In contrast, the dissent notably advocated for a Buckley-like standard. Justice Harlan suggested that a “substantial relevance to a legitimate state interest” test was sufficient. Id. at 498 (Harlan, J., dissenting). Critically, the opinion also clarified that immediate public disclosure of the associations was not necessary for the disclosure to be impermissible; there was still an underlying pressure to stay away from associations “which might displease” the administrators receiving the report.35×35. Id. at 486 (majority opinion).

B. Remaining Faithful to Patterson’s Strict Spirit

Circuit courts addressing compelled disclosure of novel associations in a non-election-finance context continued to remain faithful to Patterson as they expanded and applied the rule. The D.C. Circuit considered one statute that required members of the Communist Party to disclose their affiliation.36×36. See Boorda v. Subversive Activities Control Bd., 421 F.2d 1142, 1145 (D.C. Cir. 1969). The court applied Patterson and found that compelling disclosure could be a “substantial burden” on free association.37×37. Id. at 1148. In the Fourth Circuit, Master Printers of America v. Donovan38×38. 751 F.2d 700 (4th Cir. 1984). concerned reporting requirements under the Labor-Management Reporting and Disclosure Act (LMRDA); the requirements were challenged on grounds that they would chill membership.39×39. Id. at 703. Specifically, the Master Printers of America (MPA) argued this was an impermissible burden on association because “[f]orced disclosure of membership and contribution lists . . . would deter current members from participating in MPA programs, jeopardize efforts to recruit additional members, and thereby unconstitutionally burden its speech activity.” Id. Though labor unions are treated differently from other organizations when it comes to First Amendment jurisprudence in some respects, see, e.g., Marion Crain & John Inazu, Re-Assembling Labor, 2015 U. Ill. L. Rev. 1791, 1796–99, the line of cases related to Master Printers is still useful and doctrinally relevant to this discussion because it addresses the particular issue of compelled disclosure of membership lists. Like in the teacher case, the court stated that “a mere showing of some legitimate governmental interest” was insufficient to survive “‘exacting’ scrutiny.”40×40. Master Printers, 751 F.2d at 704 (quoting Buckley v. Valeo, 424 U.S. 1, 65 (1976)). The court’s opinion also clarified that exacting scrutiny applied regardless of whether the infringement concern arose “not through direct government action, but indirectly as an unintended but inevitable result of the government’s conduct in requiring disclosure.” Id. (quoting Buckley, 424 U.S. at 65). As described by the court, this took form in a four-factor test and still required tailoring:

The degree of infringement on [F]irst [A]mendment rights; the importance of the governmental interest protected by the Act; whether a “substantial relation” exists between the governmental interest and the information required to be disclosed; and the closeness of the “fit” between the Act and the governmental interest it purports to further. After weighing these factors, it will be possible to determine whether the burden imposed . . . is sufficiently justified by the interests the Act seeks to protect.

41×41. Id. (emphasis added).

The court also interpreted Patterson as requiring an additional “finding of a substantial ‘chill,’” by demonstrating the statute would lead to “threats, harassment, or reprisals to specific individuals.”42×42. Id.

The Sixth Circuit followed suit43×43. See Humphreys, Hutcheson & Moseley v. Donovan, 755 F.2d 1211 (6th Cir. 1985). by explicitly referencing Master Printers’s analysis, reiterating its four factors, and echoing that compelled disclosures were subject to exacting scrutiny.44×44. Id. at 1220 (“The Master Printers court recognized that the Supreme Court had declared unconstitutional compelled disclosure of membership in groups involved in advocacy . . . .”). The court also restated Master Printers’s requirement of a threshold “showing that the statutory scheme will result in threats, harassment, or reprisals.”45×45. Id. (quoting Master Printers, 751 F.2d at 704). As a result, no substantial burden on associational rights was found here because the organization “failed to establish that the report requirements . . . produce[d] a ‘deterrent effect’ akin to that present in [Patterson].” Id. at 1221. Here, allegations of the chilling effect were consequential enough to inquire whether the disclosure requirement was “narrowly tailored to serve a compelling governmental interest.”46×46. Id. (emphasis added). The court found the interest in labor peace was sufficiently compelling, “outweigh[ed] the chill,” and the compelled disclosure was “carefully tailored.”47×47. Id. at 1222. In this way, the Sixth Circuit used the language of “exacting scrutiny” right alongside a strict scrutiny analysis — searching for a compelling interest and narrow tailoring. And it indicated that it read the Fourth Circuit to be doing the same.48×48. Id. at 1220.

The Ninth Circuit also agreed that compelled disclosure of associations “must survive exacting scrutiny.”49×49. Perry v. Schwarzenegger, 591 F.3d 1147, 1160 (9th Cir. 2010) (quoting Buckley v. Valeo, 424 U.S. 1, 64 (1976)). The court followed Ninth Circuit precedent from Dole v. Service Employees Union, Local 280, 950 F.2d 1456, 1461 (9th Cir. 1991). Other Ninth Circuit cases discussing associational interest outside of the election-financing context do not always include these thresholds, but have consistently indicated that narrow tailoring is required. See Brock v. Local 375, Plumbers Int’l Union, 860 F.2d 346, 350 (1988). Similar in some ways to the Master Printers approach, the court read Patterson as requiring a threshold finding of chilling effects to trigger strict scrutiny and used a two-part framework. First, a prima facie showing of a First Amendment infringement through harassment, membership withdrawal, or a chilling effect must be demonstrated, and then the burden is upon the government to demonstrate “that the information sought . . . is rationally related” to a compelling government interest that uses “least restrictive means.”50×50. Perry, 591 F.3d at 1161 (emphasis added) (quoting Brock, 860 F.2d at 350). The court characterized this analysis as a balancing test to determine “whether the interest in disclosure . . . outweighs the harm,” id. at 1161 (alteration in original) (quoting Buckley, 424 U.S. at 72), and requiring that “[t]he request must also be carefully tailored,” id. (emphasis added).

In sum, until recently the development of circuit case law continued to stay faithful to the strict requirements of Patterson. Though some of these decisions began to use the language of exacting scrutiny or expanded their tests to include certain thresholds, their discussions always made clear that a compelling government interest and close tailoring were still required. This suggests “exacting scrutiny” was not meant to introduce a new, lower standard, but was seen as interchangeable with strict scrutiny. As the next Part will show, however, the intervening proliferation of election cases concerning compelled disclosure became mistaken for a change in the standard that applies to nonelection cases.


A. Buckley and the Development of Election-Related Association

The language of “exacting scrutiny” first appeared in the associational context in Buckley v. Valeo. The opinion extends over a hundred pages before reaching the issue of compelled disclosure and associational freedoms. But, at the outset, Buckley took a moment to recognize the “freedom to associate with others for the common advancement of political beliefs and ideas.”51×51. Buckley, 424 U.S. at 15 (quoting Kusper v. Pontikes, 414 U.S. 51, 56 (1973)). Critically for modern-day dark money legislation, the Court also indicated that its past decisions have not drawn fine lines between contributors and members but have treated them interchangeably.” Id. at 66. The Court expressed that state action threatening this freedom “is subject to the closest scrutiny,”52×52. Id. at 25 (emphasis added) (quoting NAACP v. Alabama ex rel Patterson, 357 U.S. 449, 460–61 (1958)). requiring “a sufficiently important interest and employ[ing] means closely drawn to avoid unnecessary abridgement of associational freedoms.”53×53. Id. (emphasis added).

When discussing the disclosure requirements specifically, the Court confirmed that the “strict test established by [Patterson]”54×54. Id. at 66. applied, and that it was one of “exacting scrutiny.”55×55. Id. at 64. As such, the compelled disclosure could not “be justified by a mere showing of some legitimate governmental interest.”56×56. Id. Unlike the cases before it, Buckley dove into the merit of government interests where the “‘free functioning of our national institutions’ is involved.”57×57. Id. at 66 (quoting Communist Party v. Subversive Activities Control Bd., 367 U.S. 1, 97 (1961)). The interests in “provid[ing] the electorate with information ‘as to where political campaign money comes from and how it is spent by the candidate’”58×58. Id. (quoting H.R. Rep. No. 92-564, at 4 (1971)). and in deterring corruption and its appearance were found to be substantial.59×59. Id. at 67–68 (“[D]isclosure requirements deter actual corruption and avoid the appearance of corruption by exposing large contributions and expenditures to the light of publicity.” Id. at 67.). Notably, this Patterson standard — which Buckley explicitly referred to, upheld, and applied in the portions of its opinion that addressed disclosure and association — gave no mention of “exacting” language. This suggests that Buckley’s use of this term in the context of upholding Patterson’s strict test did not mean to set a new standard. Indeed, even when the Court’s opinion turned to discussing the personal disclosure requirement of Section 434(e),60×60. As the Court explained: “Section 434(e) requires ‘[e]very person (other than a political committee or candidate) who makes contributions or expenditures aggregating over $100 in a calendar year ‘other than by contribution to a political committee or candidate’ to file a statement with the Commission.” Id. at 74–75 (quoting id. app. at 160) (listing out the text of the statutes discussed within). it affirmed that “in considering this provision we must apply the same strict standard of scrutiny, for the right of associational privacy developed in [Patterson] derives from the rights of the organization’s members.”61×61. Id. at 75 (emphasis added).

The Buckley Court also referred to a need for narrow tailoring.62×62. Id. at 68. The Court made clear that finding a compelling interest was not enough, even if it was substantially related to the regulation employed.63×63. Id. Notably, the opinion recognized that a chilling effect was expected and that “public disclosure . . . may even expose contributors to harassment or retaliation,”64×64. Id. but that in the context of campaign finance, disclosure was still “the least restrictive means of curbing the evils of campaign ignorance and corruption that Congress found to exist.”65×65. Id. (emphasis added). The Court went on to analyze why this was also the case for minor parties and independents because there was an insufficient showing of “focused and insistent harassment of contributors and members that existed in the NAACP cases.” Id. at 72 (quoting Brief for Appellants at 173, Buckley, 424 U.S. 1 (Nos. 75-436 and 75-437)). It was not that campaign disclosure merited a new standard, but just that in this context, the strict standard was met.

However, as election financing jurisprudence continued to develop, this standard relaxed and became distinct from Patterson. Courts began to cite and apply Buckley in the context of campaign finance disclosures without requiring any narrow tailoring in their analysis.66×66. The Court first used “exacting scrutiny” to mean something other than strict scrutiny in this context in Nixon v. Shrink Missouri Government PAC, 528 U.S. 377, 386 (2000). For example, in Citizens United v. FEC,67×67. 558 U.S. 310 (2010). an organization challenged parts of the Bipartisan Campaign Reform Act of 200268×68. Pub. L. No. 107-155, 116 Stat. 81 (codified as amended in scattered sections of 2, 8, 18, 28, 36, 47, and 52 U.S.C.). (BCRA). One portion in particular required individuals to file disclosure statements for spending on electioneering communications.69×69. 52 U.S.C. § 30104(f) (2012). Though the Court stated that this requirement was subjected to “exacting scrutiny,” it only demanded “a ‘substantial relation’ between the disclosure requirement and a ‘sufficiently important’ governmental interest.”70×70. Citizens United, 558 U.S. at 366–67 (quoting Buckley, 424 U.S. at 64, 66). The Court had already held that this portion of the BCRA met this standard71×71. See McConnell v. FEC, 540 U.S. 93, 196–97 (2003). because it served in “help[ing] citizens ‘make informed choices in the political marketplace,’”72×72. Citizens United, 558 U.S. at 367 (quoting McConnell, 540 U.S. at 197). and the Court found that the statute was still valid as applied to requiring disclosure around commercial advertisements at issue in the case.73×73. Id. The organization had not demonstrated its members faced “threats or reprisals.” Id. at 370. Accordingly, “disclosure requirements could reach beyond express advocacy”74×74. Richard Briffault, Two Challenges for Campaign Finance Disclosure After Citizens United and Doe v. Reed, 19 Wm. & Mary Bill Rts. J. 983, 995 (2011). because the informational interest still held up as compelling: “This transparency enables the electorate to make informed decisions and give proper weight to different speakers and messages.”75×75. Citizens United, 558 U.S. at 371. However, while broadening the universe of what could mandate disclosure beyond express advocacy, the majority also narrowed the government interest around combatting corruption to combatting only quid pro quo corruption or its appearance.76×76. See id. at 448 (Stevens, J., concurring in part and dissenting in part). Though Citizens United also indicated that an organization did not have to be a “political committee” as previously defined to be subject to disclosure requirements, subsequent federal actions still “make[] it extremely easy for such organizations to avoid having to disclose their donors.” Briffault, supra note 74, at 1008. For a more complete discussion of the intervening case law and administrative action that functionally led to the present landscape, see id. at 1008–13.

The departure from narrow tailoring has become more entrenched, with one Third Circuit case explicitly indicating that exacting scrutiny in the associational context “differs from ‘strict scrutiny’. . . in that it does not engage in a ‘least-restrictive-alternative analysis,’”77×77. Del. Strong Families v. Att’y Gen. of Del., 793 F.3d 304, 309 n.4 (3d Cir. 2015). in direct opposition to Buckley’s language.

B. Association as a Sword: Recent Misapplications of Buckley

Two circuits have addressed recent state dark money regulations. They have found that the weakened form of exacting scrutiny described above applies, and that dark money disclosure requirements survive this scrutiny.

In California, Americans for Prosperity Foundation v. Becerra78×78. 903 F.3d 1000 (9th Cir. 2018). most recently addressed this issue. The California Attorney General keeps a registry of charitable corporations and is authorized to “conduct whatever investigation is necessary” to maintain it.79×79. Cal. Gov’t Code § 12584 (2005). Organizations in the registry must submit a copy of the IRS Form 990.80×80. Becerra, 903 F.3d at 1004. This form typically includes a Schedule B, which requires nonprofits to “report the name and addresses of their largest contributors.”81×81. Id. at 1004–05. Though the Schedule B information is never made public under this law, this law is fit to be discussed among other compelled public disclosure laws because the impact may functionally be the same, because states often lack the capacity to keep information truly private on their servers. See Ams. for Prosperity Found. v. Harris, 182 F. Supp. 3d 1049, 1056–57 (C.D. Cal. 2016). As a result, much of the briefing and litigation argued in the alternative and reached the merits of the associational interest as if this information were to become public. See id.; Becerra, 903 F.3d at 1005. Additionally, “non-public disclosures can still chill protected activity where a plaintiff fears the reprisals of a government entity.” Ctr. for Competitive Politics v. Harris, 784 F.3d 1307, 1316 (9th Cir. 2015). In effect, the state compels organizations to disclose their donors by requiring that they submit the full Schedule B. When the Americans for Prosperity Foundation challenged this forced disclosure, the Ninth Circuit found it survived “exacting scrutiny.”82×82. Becerra, 903 F.3d at 1008 (quoting Doe v. Reed, 561 U.S. 186, 196 (2010)). But it applied the lower exacting scrutiny; no tailoring was required.83×83. See id. This followed in the footsteps of an earlier Ninth Circuit case, Center for Competitive Politics v. Harris,84×84. 784 F.3d 1307. which found there was no narrow tailoring requirement because the chilling burden did not reach a threshold triggering stricter scrutiny.85×85. Id. at 1313–14 (citing Buckley v. Valeo, 424 U.S. 1, 66 (1976)). This misinterpretation carried over into Becerra.86×86. See Becerra, 903 F.3d at 1008.

The Second Circuit also followed in this trend in the recent case Citizens United v. Schneiderman.87×87. 882 F.3d 374 (2d Cir. 2018). A similar charities regulation compelling organizations to submit the Schedule B was at issue.88×88. Id. at 379. The Second Circuit found that the exacting scrutiny standard applied to both the Buckley and Patterson lines of cases, which it defined similarly to the Ninth Circuit. Under this framework, the court found that its “task [was] easy,”89×89. Id. at 385. but it made its job “easy” by slicing off the narrow tailoring analysis Patterson required.

The alternative standard that misapplications of Buckley created does not represent an overall shift down, but rather a circuit split; other recent cases stand faithful to the Patterson standard.90×90. See, e.g., Coal. for Secular Gov’t v. Williams, 815 F.3d 1267, 1275 (10th Cir. 2016). For example, one Fifth Circuit case concerned Familias Unidas, an unincorporated organization that organized for improvements to its school district.91×91. Familias Unidas v. Briscoe, 544 F.2d 182, 183–84 (5th Cir. 1976). Following a nonviolent boycott by Mexican-American students in protest of their school board’s actions, a county judge requested a list of the organization’s members per a local law empowering him to compel such disclosure.92×92. Id. at 183–85. The judge’s authority was § 4.28 of the Texas Educational Code, which authorized the county judge to require any group “engaged in activities designed to hinder, harass, or interfere” with the operation of the school system to disclose its members, purposes, and parent affiliations. Id. at 184 n.1 The Fifth Circuit emphasized that privacy in association is critical “with respect to groups championing unpopular causes”93×93. This was indicated later in the proceedings, upon a second appeal of the case after remand. Familias Unidas v. Briscoe, 619 F.2d 391, 399 (5th Cir. 1980). and that “the Supreme Court has applied an exacting scrutiny to compulsory disclosures of membership lists or associational ties.”94×94. Id. at 395–96. Unlike the Ninth or Second Circuits’ exacting scrutiny, this test required an “overriding, legitimate state purpose” and tailoring “drawn with sufficiently narrow specificity.”95×95. Id. The regulation swept “too broadly” and failed this tailoring.96×96. Id. at 400.

C. Most Recent and Ongoing Litigation

After the opinion in Americans for Prosperity Foundation v. Becerra, Americans for Prosperity petitioned for rehearing en banc and was denied.97×97. See Ams. for Prosperity Found. v. Becerra, 919 F.3d 1177 (9th Cir. 2019). The opinion denying rehearing affirmed that exacting scrutiny was a lower standard than strict scrutiny and was met here.98×98. Id. at 1189 (Fisher, Paez, Nguyen, JJ., responding to the dissent from the denial of rehearing en banc). The dissent from this denial of rehearing en banc rejected this distinction and insisted that exacting scrutiny referred to one single standard in election and non-election cases that never lost its tailoring requirement. Rather, the requirement was simply always met in the election context because “Buckley fashioned a per se rule” which categorically established “that disclosure is the least restrictive means of reaching Congress’s goals.”99×99. Id. at 1180 (Ikuta, J., dissenting from denial of rehearing en banc). Under this logic, in the election context the narrow tailoring analysis was occasionally dropped from opinions because it did not require any new examination into less restrictive means.100×100. See id.

Most recently, Americans for Prosperity challenged the current New Jersey law in Americans for Prosperity v. Grewal.101×101. A preliminary injunction against enforcement of the law was granted on October 2, 2019. See Order Granting Preliminary Injunction, Ams. for Prosperity Found. v. Grewal, No. 19-cv-14228 (D.N.J. Oct. 27, 2019). The law regulates what it defines as independent expenditure committees: any 501(c)(4) or 527 organizations that spend a certain amount on “(1) influencing elections; (2) influencing legislation; or (3) providing political information.”102×102. Petitioner’s Brief at 4, Ams. for Prosperity Found. v. Grewal, No. 19-cv-14228; see also N.J. Stat. Ann. § 19:44A-3(h) (2019).

In the face of ongoing litigation and circuit disagreement, the question becomes: What standard should apply? Part III discusses using proportionality to reconcile this circuit split.


As described above, some circuits have departed from the strict scrutiny required by Patterson, while other circuits have remained faithful to its narrow tailoring requirement, creating a circuit split in need of resolution. This could be easily resolved by a reaffirmation of the Patterson standard. However, the departure also provides an opportunity to reset. The Court could use a third approach — proportionality — because neither current test is equipped to address the challenges in freedom of association jurisprudence.

Shortcomings of Current Approaches

Neither Patterson’s strict scrutiny test nor the apparently lower Buckley test is capable of satisfactorily solving the tension between needing to address the modern dark money phenomenon and ensuring that legitimate social movements are not chilled. This challenge accordingly calls for a proportionality approach.

First, Patterson’s strict scrutiny test is ill suited for the unique confluence of challenges that come with effectively tackling dark money through legislation. For any disclosure law aimed at combatting dark money, the state would first need to show a compelling interest. Under a Court that has narrowed the universe of what constitutes a permissible compelling interest, there is no incentive for an honest discussion about the “real interest” — that is, a more honest discussion of the complex and varied interests at play — if surviving challenge requires judges to rationalize in the language of a few permissible compelling interests.103×103. This hypothesis draws in part from strands of legal realism. See, e.g., Elizabeth Mertz & Cynthia Grant Bowman, Balanced Judicial Realism in the Service of Justice: Judge Richard D. Cudahy, 67 DePaul L. Rev. 655, 671–75 (2018). It might be that under a realist theory, if judges are already manipulating tests to achieve desired outcomes, proportionality allows a more upfront balancing of the judge’s considerations and allows a more transparent vehicle to consider a bigger variety of interests. See, e.g., Victoria Nourse & Gregory Shaffer, Varieties of New Legal Realism: Can a New World Order Prompt a New Legal Theory?, 95 Cornell L. Rev. 61, 106 n.204 (2009); see also Richard H. Fallon, Jr., Strict Judicial Scrutiny, 54 UCLA L. Rev. 1267, 1271 (2007) (“[I]ndividual Justices tend to vary their applications of strict scrutiny based on their personal assessments of the importance of the right in question.”). The primary interests currently available104×104. See Heerwig & Shaw, supra note 2, at 1451–52 (describing the three successful compelling interests in this context, one of which no longer applies). are the informational interest in “alert[ing] the voter to the interests to which a candidate is most likely to be responsive” and combatting corruption and its appearance.105×105. Buckley v. Valeo, 424 U.S. 1, 67 (1976). The latter is largely unavailable; the Court has said that an interest in combatting corruption or the appearance of corruption in the electoral context is insufficient for nonpolitical committees such as 501(c)(4)s.106×106. See SpeechNow.org v. FEC, 599 F.3d 686, 692–94 (D.C. Cir. 2010) (citing Citizens United v. FEC, 558 U.S. 310, 357 (2010)) (describing the Court’s view “that independent expenditures do not corrupt or create the appearance of . . . corruption,” id. at 694). And the use of the “informational interest” has been checkered.107×107. See Jiang, supra note 4, at 507–15 (describing different conceptions of the broadly recognized “informational interest” in disclosure). The inconsistency of its application, “compounded with growing theoretical pressures arguing that disclosure’s ability to educate the public is greatly overstated,”108×108. Id. at 487. puts it on shaky ground. Even if a compelling interest were found, many laws aimed at dark money would fail narrow tailoring analysis. The primary ways states have attempted to regulate dark money so far are sweeping and have survived only because both circuits addressing dark money laws mistakenly applied Buckley’s standard.109×109. See supra section II.B, pp. 651–52. The current New Jersey law, for example, would likely fail narrow tailoring because it sweeps in too many nonprofit groups that are not the target of the government’s anti-corruption interest.110×110. See Brief for Petitioner at 26, ACLU of N.J. v. Grewal, No. 19-CV-17807 (D.N.J. Sept. 10, 2019). Put simply, it is unclear if any legislative approach to dark money could actually survive the strict standard on its face, leaving governments with no solutions.

On the other hand, fully adopting the Buckley test is also a poor solution. If narrow tailoring is not required for disclosures, then legislation like the New Jersey law that sweeps in benign organizations will survive. But this chills membership in smaller organizations facing violence or other retaliation. If NAACP leaders in Alabama had been subject to Buckley’s standard in 1958, their organizing efforts to support bus boycotts and provide legal assistance to black students seeking university admission would have shut down because of the risks members would face if their identities become public.111×111. See NAACP v. Alabama ex rel. Patterson, 357 U.S. 449, 452 (1958). More recently, because Familias Unidas was unpopular among school board leadership and the broader community, the Fifth Circuit recognized that membership would be impermissibly chilled because “public recrimination [was] certain to follow from public disclosure.”112×112. Familias Unidas v. Briscoe, 619 F.2d 391, 401 (5th Cir. 1980). Critically, the court found a distinction between Communist cases where exposure was the only deterrent from illicit behavior, and the punitive disclosure here that would happen only after an organization was found engaging in proscribed conduct. In this way, requiring after-the-fact disclosure from Familias Unidas would not have deterred conduct, but would have chilled activities and action that may have been permissible. See id. at 401–02. Small, unpopular movements like this are at high risk of shuttering if they are subject to disclosure.113×113. See Seth F. Kreimer, Sunlight, Secrets, and Scarlet Letters: The Tension Between Privacy and Disclosure in Constitutional Law, 140 U. Pa. L. Rev. 1, 34–40, 54 (1991) (describing “the mechanisms by which exposure can achieve a deterrent effect and the situations in which the effect is most likely” to affect movements, id. at 34). At the same time, the larger organizations involved in political spending have little risk of their message being chilled or their movement being stomped out by disclosure alone, but their harm to democratic elections is high.114×114. See Jane Mayer, Dark Money: The Hidden History of the Billionaires Behind the Rise of the Radical Right 240–43, 250, 272–73 (2016) (describing the impact of undisclosed 501(c)(4) funds on swinging a reliably Democratic Senate seat for a Republican); Kenneth P. Vogel, Big Money: 2.5 Billion Dollars, One Suspicious Vehicle, and a Pimp –– On the Trail of the Ultra-Rich Hijacking American Politics 15–17 (2014); Nicholas O. Stephanopoulos, Aligning Campaign Finance Law, 101 Va. L. Rev. 1425, 1427 (2015) (noting that “there is evidence that politicians’ positions reflect the preferences of their donors” rather than the general public). It is plausible that owing to their numbers, even if some donations were deterred, enough engagement would remain for these larger movements to exert undesirable influence.115×115. See Richard L. Hasen, Plutocrats United 9 (2016) (describing the challenge of regulating larger organizations by protecting their speech but guarding against “the rise of a plutocratic class that has too much influence” over elections and legislative agendas).

In short, neither the Patterson nor the Buckley standard currently protects the anonymity interests of fragile or smaller movements while addressing the campaign abuse of larger groups that can remain anonymous. These inadequacies make the case for proportionality.

The Benefits of a Proportionality Approach

The best way forward from this jurisprudential checkerboard is to employ a proportionality approach to manage compelled disclosure laws targeting dark money. A few reasons favor this approach. First, the flexibility it affords courts in addressing the unique challenges of dark money makes it more effective. Second, a proportionality approach is consistent with the spirit of First Amendment jurisprudence. Third, a proportionality approach best promotes the theoretical and philosophical justifications of freedom of association in particular.

Professor Jamal Greene’s “typical proportionality formulation” is a useful model for how this would look in practice:

(1) [S]ome discernment of the nature of the claimed right; (2) an assessment of the means-ends fit between the law or act and some legitimate governmental objective; (3) a “least-restrictive means” or “minimal impairment” test that asks whether the government has less rights-impairing alternatives it could have pursued; and (4) balancing in the strict sense, which requires the adjudicator to assess whether there is significant disproportionality between the marginal benefit to the government and the marginal cost to the rights-bearer.

116×116. Jamal Greene, The Supreme Court, 2017 Term –– Foreword: Rights as Trumps?, 132 Harv. L. Rev. 28, 59 (2018). Greene notes that “[d]ifferent jurisdictions place different weights on the various steps in the analysis . . . [and it] can take different shape for different kinds of rights.” Id.

Applying this to a hypothetical regulation similar to the California regulation at issue in Harris demonstrates how this might work in practice. If a state attorney general or charity bureau began to compel the Schedule B form for all 501(c)(4) organizations, the nature of the claimed right would be the right to privacy in group associations. The means-ends fit would be substantial — there is a connection between promoting the informational interest in educating voters117×117. See Jiang, supra note 4, at 495, 507–15. and requiring the organizations to disclose the donors’ funding, lobbying, and candidate advocacy.118×118. The Fifth Circuit in Familias Unidas discussed in more detail when there is an adequate means-ends fit from disclosure and when there is not. The court found that when trying to identify and deter organized disruptions in the school day, “disclosure . . . does bear a relevant correlation to the legitimate object of peaceful operation of the schools,” and that this was unlike Patterson, “in which no justifiable relationship could be found.” Familias Unidas v. Briscoe, 619 F.2d 391, 400 (5th Cir. 1980). The minimal impairment analysis would consider whether limiting the regulation to 501(c)(4) organizations (leaving out 501(c)(3) groups119×119. 501(c)(3) groups are groups “organized and operated exclusively for religious, charitable, scientific, . . . literary, or educational purposes.” I.R.C. § 501(c)(3) (2012). ) was the narrowest way to effectively tackle this problem. One argument to the contrary would be that not all 501(c)(4)s in a particular state perform functions similar to a political committee.120×120. Indeed, this is the ACLU of New Jersey’s argument in their brief challenging the New Jersey law. See Brief for Petitioner, supra note 110, at 10, 27–29. Consequently, this regulation might still chill the donations and activity of several organizations that focus more on general community advocacy or other social welfare purposes. A more appropriately narrow regulation may not rely at all on a tax-code-based status121×121. For a discussion of why addressing dark money should be decoupled from tax exemption, see Ciara Torres-Spelliscy, Corporate Citizen? An Argument for the Separation of Corporation and State 52–55 (2019). and instead on an affirmative definition of politically involved organizations that can more accurately circumscribe just those organizations that engage in high levels of lobbying or advocacy for candidates. On the other hand, a more sweeping formulation might fail this prong but still be found permissible. This could happen if, in the fourth prong, the state could demonstrate that the sheer magnitude of influence by such organizations is much higher than the organizations or movements it may chill — such that it “justif[ies] the harm to constitutionally protected interests.”122×122. Vicki C. Jackson, Constitutional Law in an Age of Proportionality, 124 Yale L.J. 3094, 3116 (2015).

In this way, a disclosure regulation that would fail strict scrutiny under narrow tailoring could survive proportionality because of the more complex considerations this four-prong framework accommodates. More specifically, proportionality offers three benefits.

1. Flexible Functionality. — First, this approach is the most impactful in directly addressing the unique universe of dark money organizations.123×123. See Linda Sugin, Politics, Disclosure, and State Law Solutions for 501(c)(4) Organizations, 91 Chi.-Kent L. Rev. 895, 927–28, 932–35 (2016) (describing the limited effectiveness of more rigid state law disclosure regulations). As the hypothetical shows, the challenge with legislative solutions is that they may choose to circumscribe based on broad descriptions of “political information” or simply apply to whole categories of nonprofit organizations, which could include anything from local soup kitchen operations to public radio. As a result, the state is left with asserting interests such as combatting charity corruption, which conceal the “true” motivating purpose of the legislation,124×124. For example, one organization’s coverage of the California law recognizes that the government’s asserted interest was “to police fraud and self-dealing by charities,” but immediately then discusses a “larger litigation effort to weaken political disclosure laws, which secure public transparency in the electoral context.” Ninth Circuit Decision Ensures Proper Oversight of State Charities and Prevents Broader Assault on Political Transparency Laws, Campaign Legal Ctr. (Sept. 17, 2018), https://campaignlegal.org/update/ninth-circuit-decision-ensures-proper-oversight-state-charities-and-prevents-broader-assault [https://perma.cc/22QF-NJHU]; see also Steven Greenhut, Is AG Ignoring Lessons from IRS Scandal?, San Diego Union-Trib. (Dec. 10, 2014), https://www.sandiegouniontribune.com/news/politics/sdut-is-attorney-general-harris-targeting-conservatives-2014dec10-story.html [https://perma.cc/86DZ-H5VU] (suggesting that though the interest asserted was in combatting charity fraud, the California Attorney General was targeting those nonprofits engaged in political activity). because the means-ends fit may be more defensible and lead to fewer problems around narrow tailoring.125×125. See Fallon, supra note 103, at 1296–97 (observing the lack of clarity around what becomes a sufficiently compelling government interest). When the interests found to be sufficiently compelling are sporadic and far between, they incentivize putting a round peg into a square hole.

Proportionality addresses these challenges by encouraging a more open and honest discussion by government when developing legislation,126×126. See Jackson, supra note 122, at 3146–47 (“Legislators who understand that statutes will be evaluated under proportionality standards . . . will have reason to give attention to the rationality of the means. . . .” Id. at 3146.). while a categorical approach hides the ball.127×127. Cf. id. at 3142 n.226 (discussing how proportionality may encourage “judges to articulate the actual reasons for their opinions”). It seems to follow that there would be an analogous benefit in the drafting of legislation itself if a government knows that a holistic view of the competing interests involved will be considered. This discussion could include things that (1) on their own have not been found to be appropriate compelling interests in current jurisprudence but could be sufficiently compelling taken together128×128. In addition to building trust, another example is “political equality.” Hasen, supra note 115, at 11. This seems unlikely to be adopted as a compelling interest, but it could be more transparently considered and weighed under a proportionality framework where the ability to overturn an overall harmful law does not rely so singularly on finding one of the existing compelling interests is at play. or (2) do not lend themselves to legislative solutions that can be narrowly tailored but are better for the democratic process on balance. For example, there may be a slightly weaker correlation between disclosure and fighting quid pro quo corruption in a dark money context,129×129. Cf. Mayer, supra note 114, at 250–53 (describing the secretiveness and difficulty in tracking the origins or funding streams of several dark money organizations). It may be more difficult to indicate that money or ads were traded for a specific policy when the organization is not directly funding the candidate. but that weaker interest could be considered in combination with an interest in increasing minority democratic participation through building trust.

One concern is that a framework that always allows for a variety of factors on either side could “creat[e] an unacceptable level of indeterminacy.”130×130. Jackson, supra note 122, at 3153. However, courts are able to make rational judgments about the relative weight of interests when the downsides associated with impinging on the associational right are sufficiently strong.131×131. See Frederick Schauer, Balancing, Subsumption, and the Constraining Role of Legal Text, 4 Law & Ethics Hum. Rts. 33, 35–40 (2010) (suggesting that outcomes will be largely predictable and consistent when considering differently weighted factors); see also T. Alexander Aleinikoff, Constitutional Law in the Age of Balancing, 96 Yale L.J. 943, 972 (1987) (suggesting that similar types of reliable weighting decisions must be made when developing common law doctrine generally). The impact on the rights of a smaller, less mainstream organization would be outsized in comparison to the democratic legitimacy interest achieved from tracking a relatively small amount of advocacy funds.132×132. See Jackson, supra note 122, at 3157 (discussing “large-small trade-offs”). Though in hard cases the relative weights are closer and two judges might come out differently, that risk comes with applying any test. See id. at 3156. Though there will always be a concern that departures from the ideal in a proportionality framework might even be greater than the departures that have already occurred, it seems more appropriate to trust judicial discretion in the latter because it structurally allows for greater discussion and weighing of interests — more transparency in how the discretion is employed.

2. Comports with Jurisprudence. — Second, Justices and scholars have indicated that a proportionality approach is appropriate in First Amendment jurisprudence.133×133. See, e.g., Stephen Breyer, Active Liberty: Interpreting Our Democratic Constitution 49 (2006); Jackson, supra note 122, at 3096. Professor Richard Fallon has suggested that because the Court never “reach[ed] agreement about the precise nature of the inquiry”134×134. Fallon, supra note 103, at 1271. required by strict scrutiny, in its application the Supreme Court has at times engaged in “relatively ad hoc, weighted balancing of public and private interests in freedom of association cases.”135×135. Id. at 1306. In fact, Professor Fallon states that “strict scrutiny was capable of adaptation on an almost generic basis to protect relatively broadly defined rights that the Court thought merited strong, but less than absolute, protection.” Id. at 1292. Indeed, Justice Scalia explicitly referred to the strict scrutiny applied in one First Amendment case as a “balancing test” in practice.136×136. Emp’t Div. v. Smith, 494 U.S. 872, 883 (1990). Justice Marshall’s formulation of a “compelling interest” itself was a proportionate inquiry when applied in First Amendment contexts.137×137. See Richardson v. Ramirez, 418 U.S. 24, 78–80 (1974) (Marshall, J., dissenting); San Antonio Indep. Sch. Dist. v. Rodriguez, 411 U.S. 1, 98–99 (1973) (Marshall, J., dissenting). And just last Term, Justice Breyer advocated to reject the categorical tiers of scrutiny in a case that put governmental interests in promoting decency in trademark speech against the expressive interest of an individual, noting that “[t]he First Amendment is not the Tax Code.”138×138. Iancu v. Brunetti, 139 S. Ct. 2294, 2304 (2019) (Breyer, J., concurring in part and dissenting in part).

Proportionality-based balancing also already occurs in First Amendment jurisprudence to the more fundamental extent that different types of speech are protected to different degrees. The proscription of libel or obscenity, for example, is appropriate because free speech protection “was fashioned to assure unfettered interchange of ideas for the bringing about of political and social changes desired by the people,”139×139. Roth v. United States, 354 U.S. 476, 484 (1957). rather than to “protect every utterance” that strayed from these civic engagement purposes.140×140. Id. at 483; see also Charlotte Taylor, Free Expression and Expressness, 33 N.Y.U. Rev. L. & Soc. Change 375, 376 (2009) (“[C]ertain categories of speech are subject to regulation or suppression because they do not further central First Amendment values.”). Indeed, Chaplinsky v. New Hampshire141×141. 315 U.S. 568 (1942). explicitly applied these values in a balancing context to so-called “‘fighting’ words,” stating that “such utterances are no essential part of any exposition of ideas, and are of such slight social value . . . that any benefit that may be derived from them is clearly outweighed.”142×142. Id. at 572 (emphasis added). Consequently, adopting something like Greene’s framework may not be such a radical departure from First Amendment jurisprudence, as the Court already engages in proportionality analysis.

3. Theoretical and Philosophical Justifications. — Proportionality best serves the civic republican ends that underlie First Amendment rights of speech, and by extension, association, because it is “designed to discipline the process of rights adjudication on the assumption that rights are both important and, in a democratic society, limitable.”143×143. Greene, supra note 116, at 58. Civic republicanism is a major theoretical justification for the First Amendment144×144. The two other major justifications offered for the First Amendment are the marketplace of ideas and self-expression. Marketplace theory asserts that the First Amendment should be as speech permissive as possible so that ideas can compete and the “truth” can emerge. See Stanley Ingber, The Marketplace of Ideas: A Legitimizing Myth, 1984 Duke L.J. 1, 2–3. The other holds that self-expression is valuable in itself and should therefore be protected. See, e.g., Thomas I. Emerson, First Amendment Doctrine and the Burger Court, 68 Calif. L. Rev. 422, 424–25 (1980). and holds that the First Amendment should focus on creating conditions ideal for people to participate in governance and democratic discourse.145×145. See William V. Luneburg, Civic Republicanism, the First Amendment, and Executive Branch Policymaking, 43 Admin. L. Rev. 367, 371–74 (1991). The civic republican theory views public discussion as a political duty.146×146. See Mark A. Graber, Transforming Free Speech 92–93 (1991). The “final end of the state [is] to make men free to develop their faculties,”147×147. Whitney v. California, 274 U.S. 357, 375 (1927) (Brandeis, J., concurring); see also Vincent Blasi, The First Amendment and the Ideal of Civic Courage: The Brandeis Opinion in Whitney v. California, 29 Wm. & Mary L. Rev. 653, 672 (1988). which occurs through political deliberation and the exercise of governing.

This theory and its relation to the First Amendment originate from Alexander Meiklejohn. Evoking the image of a town hall, Meiklejohn explains that the “welfare of the community requires that those who decide issues shall understand them.”148×148. Alexander Meiklejohn, Political Freedom: The Constitutional Powers of the People 26 (1948). “[T]he public must have access to the greatest possible breadth and quantity of information in order to facilitate self-government.”149×149. See Gregory P. Magarian, Regulating Political Parties Under a “Public Rights” First Amendment, 44 Wm. & Mary L. Rev. 1939, 1975 (2003).

However, in the context of participation in the modern polity, the speech of unpopular movements or opinions risks being drowned out by louder voices or by hostile forces when those associations cannot remain private. In some instances, fear of facing economic or physical retaliation for voicing an unpopular opinion may chill speakers from speaking altogether.150×150. See, e.g., NAACP v. Alabama ex rel. Patterson, 357 U.S. 449, 462–63 (1958). This is bad for democracy more broadly because it stifles new, unpopular, or radical ideas — ideas that are necessary to help refine and distill views about government so that “wise decisions” can be made in participatory forums.151×151. Meiklejohn, supra note 148, at 26–27. Under a civic republican theory of association,152×152. For a broader discussion of how civic republican theory operates in the First Amendment expression and association contexts, see Luneburg, supra note 145. the inclusion of those views outside of the status quo helps sharpen and inform the ideas on how the polity should operate; this is made possible when groups can come together to develop those ideas.153×153. See, e.g., Meiklejohn, supra note 148, at 26–27. That pursuit is materially harmed when not all ideas can be freely developed or advocated for.

The proportionality approach does a better job of capturing the civic republican interests necessary for creating good conditions for political participation because it can accommodate more democratic process interests beyond the confused informational interest already found to be compelling by the courts.154×154. See Scot J. Zentner, Revisiting McConnell: Campaign Finance and the Problem of Democracy, 23 J.L. & Pol. 475, 475–79 (2007). A framework that aims to further civic republican ends acknowledges that effective self-governance does not just require combatting direct corruption. Though combatting undue influence is not a legitimate state interest,155×155. Regardless of whether proportionality or strict scrutiny is applied, the Court has indicated that combatting undue access and undue influence is not a permissible government interest. See McCutcheon v. FEC, 572 U.S. 185, 239–40 (2014) (Breyer, J., dissenting) (collecting cases and explaining how the Court has discussed these potential state interests). proportionality’s more holistic weighing could still bring in other permissible process interests like increasing the engagement of minority communities in the democratic process, which on its own would not be sufficiently compelling to carry a regulation.156×156. See Magarian, supra note 149, at 2027 (discussing how a “public rights” conception of the First Amendment that embodied parts of Meiklejohn’s theory of expression would allow for the consideration of values such as “engaging marginalized members of the political community in collective self-determination”).

On the other end, current disclosure jurisprudence damages civic republican ends. The Court’s “political engineering” has discouraged individuals from participating in governance; “political debate contributes far less to the resolution of important issues than our democratic culture requires.”157×157. Id. at 1944 (emphasis added). Consequently, disclosure laws that are carefully delineated to address process interests and strike a balance would be more favorably treated under a proportionality standard that can consider and weigh each of them (and their effects on civic republicanism) in turn.

C. Potential Challenges with a Proportionality Approach

A major counterargument to this proposal is that a proportionality approach is simply not how the Fourteenth and First Amendments work. That is, strict scrutiny is employed specifically because currently, constitutional jurisprudence in the United States is not proportional — and as a result this approach does not solve the problem. However, “doctrinal structures that approximate proportionality are a recurrent feature of our constitutional practice, dating back more than a century.”158×158. Jud Mathews & Alec Stone Sweet, All Things in Proportion? American Rights Review and the Problem of Balancing, 60 Emory L.J. 797, 813 (2011). Proportionality is used under other names,159×159. See Yun-Chien Chang & Xin Dai, How to Use, and Not Use, Proportionality Principle 1 (N.Y.U. Sch. of Law Pub. Law & Legal Theory Research Paper Series, Working Paper No. 19-32, 2019), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3345798 [https://perma.cc/FPT7-Z8YS] (“While the proportionality principle is not in the everyday legal parlance in the U.S., a similar mode of thinking has been inherent in U.S. Supreme Court jurisprudence since the nineteenth century.”). and it is simplistic to think that it is never employed in current jurisprudence. As Justice Breyer explained most recently, even when the Court “consider[s] a regulation that is . . . subject to ‘strict scrutiny,’ [it] sometimes find[s] the regulation to be constitutional after weighing the competing interests involved.”160×160. Iancu v. Brunetti, 139 S. Ct. 2294, 2304 (2019) (Breyer, J., concurring in part and dissenting in part). The Court has also employed this proportionality-type approach in Morse v. Frederick, 551 U.S. 393, 406–08 (2007), and Williams-Yulee v. Florida Bar, 135 S. Ct. 1656, 1670 (2015). Indeed, scholars have traced something akin to “proto-proportionality” to the early Commerce Clause cases, when the Court balanced the legitimacy of a state purpose under the Commerce Clause against the reasonableness of the burden.161×161. Mathews & Stone Sweet, supra note 158, at 815; see also id. at 801 (arguing that incorporating proportionality “would reclaim and build on the foundations that already exist in our own history and doctrine”). Empirical analyses of strict scrutiny also suggest that courts have used threshold inquiries and other tools to adjust strict scrutiny such that it is a “context-sensitive tool” the way that proportionality analysis might function.162×162. Adam Winkler, Fatal in Theory and Strict in Fact: An Empirical Analysis of Strict Scrutiny in the Federal Courts, 59 Vand. L. Rev. 793, 795 (2006). Professor Adam Winkler’s study finds that “[w]ithin discrete areas of law, strict scrutiny varies even more with the underlying context and courts have identifiable tendencies to uphold or reject particular types of laws.” Id. at 870. This seems analogous to what would happen if a full-fledged proportionality approach were used — some laws would be found permissible, whereas other types would typically fail to pass muster when, on balance, the burdens posed were too high.

Second, more foundationally, the circumstances that trigger the nonproportional or “rights-as-trumps” frameworks are “not themselves governed by traditional doctrinal machinery or constitutional interpretation.”163×163. Joseph Blocher, Rights as Trumps of What?, 132 Harv. L. Rev. F. 120, 127 (2019). Determinations like that required by the Master Printers test employed by two circuits, asking if there has been a burden on associational rights, “can trigger nondiscretionary, categorical rules, but the determinations themselves will almost inevitably involve the kinds of empirical questions that are more suited to proportionality review.”164×164. Id. at 128. These additional questions, like requiring a threshold finding that the disclosure risks a chilling through threats or harassment, actually add considerations to the strict scrutiny framework. Such glosses may have maintained the appearance of strict scrutiny,165×165. Matthew D. Bunker et al., Strict in Theory, but Feeble in Fact? First Amendment Strict Scrutiny and the Protection of Speech, 16 Comm. L. & Pol’y 349, 380 (2011) (indicating the current process is a “disturbingly mysterious” one “in which courts seemingly anoint interests with the label ‘compelling’ at will”). but in practice their application creates a framework more akin to proportionality.

Another counterargument may be that as-applied suits already provide an adequate solution to the dark-money problem because if the lower Buckley standard is used universally to tackle transparency challenges, fragile organizations fearing risk of threats and reprisals could simply challenge the law as applied to them.166×166. See Richard H. Fallon, Jr., As-Applied and Facial Challenges and Third-Party Standing, 113 Harv. L. Rev. 1321, 1321–22 (2000) (describing the use of as-applied challenges). Though some smaller organizations and movements might have the resources to bring such a lawsuit, those that are the smallest in number or financial resources will still be subject to disclosure.167×167. See Joshua A. Douglas, The Significance of the Shift Toward As-Applied Challenges in Election Law, 37 Hofstra L. Rev. 635, 635–36 (2009). In this way, the smallest of these associations, which are arguably most in need of protection so their ideas survive in the civic republican space against chilling, will have no recourse. Consequently, as-applied suits cannot truly address the broader concern of chilling or harassment for the most vulnerable movements.


The exacting scrutiny standard required in nonprofit freedom of association cases has become muddled in recent years from resembling something close to strict scrutiny to a much lower standard. This has been exacerbated most recently by decisions in the Second and Ninth Circuits that apply Buckley’s standard for election disclosure to cases falling squarely in the Patterson framework. Rather than attempt to return to Patterson’s approach or reconcile the two frameworks in the context of dark money, the most effective way forward is to embrace the use of proportionality in freedom of association jurisprudence.