Ted Cruz for Senate v. Federal Election Commission

• Court: District Court, District of Columbia
• Decided: June 3, 2021
• Docket: Civil Action No. 2019-0908
• Status: Published

Opinion

TED CRUZ FOR SENATE, et al., Plaintiffs v. FEDERAL ELECTION COMMISSION, et al., Defendants.

Civil No. 19-cv-908 (NJR) (APM) (TJK).

United States District Court, District of Columbia

June 3, 2021.

Charles J. Cooper, John D. Ohlendorf, J. Joel Alicea, Cooper & Kirk, PLLC, and Chris Gober, The Gober Group PLLC, for Plaintiffs.

Lisa J. Stevenson, Kevin Deeley, Harry J. Summers, Seth Nesin, and Tanya Senanayake, Federal Election Commission, Washington, DC, for Defendants.

Before: RAO, Circuit Judge, MEHTA and KELLY, District Judges.

MEMORANDUM OPINION

RAO, Circuit Judge:

In our constitutional democracy, elections are the primary way for the people to express their political will. Political speech promotes the free exchange of ideas about principles of government, pressing policy matters, and the relative merits of candidates for office. In recognition of the centrality of free speech to our democracy, the Supreme Court has consistently held that “the First Amendment ‘has its fullest and most urgent application’ to speech uttered during a campaign for political office.” Eu v. San Fran. Cnty. Dem. Cent. Comm., 489 U.S. 214, 223 (1989) (quoting Monitor Patriot Co. v. Roy, 401 U.S. 265, 272 (1971)). Protections for political speech extend to 2

campaign financing because effective speech requires spending money. See Buckley v. Valeo, 424 U.S. 1, 19–23 (1976) (per curiam).

This case raises a constitutional challenge to a somewhat obscure campaign finance restriction that limits the amount of post-election contributions that may be used to repay a candidate’s pre-election loans. Section 304 of the Bipartisan Campaign Reform Act of 2002 prohibits candidates from using post-election contributions to repay personal loans over $250,000. See 52 U.S.C. § 30116(j) (the “loan-repayment limit”). Senator Rafael Edward “Ted” Cruz and his campaign committee Ted Cruz for Senate brought this suit to invalidate and enjoin the enforcement of Section 304 and its implementing regulation. We find that the loan-repayment limit burdens political speech and thus implicates the protection of the First Amendment. Because the government has failed to demonstrate that the loan-repayment limit serves an interest in preventing quid pro quo corruption, or that the limit is sufficiently tailored to serve this purpose, the loan- repayment limit runs afoul of the First Amendment. We therefore grant summary judgment for Senator Cruz and his campaign.

I.

A.

Candidates for federal office require substantial funds to support their campaigns. Funding may come from individual contributions, which are subject to a per-election cap. 1 See

1 The current base limit is set at $2,900 per election. See 86 Fed. Reg. 7,867, 7,869 (Feb. 2, 2021). A primary election, general election, 3

Federal Election Campaign Act of 1971 (“FECA”), Pub. L. No. 92-225, 86 Stat. 3 (codified as amended at 52 U.S.C. § 30116); see also 52 U.S.C. § 30116(a)(1)(A) & (c). Candidates may also self-finance their campaigns without monetary limits. See 11 C.F.R. § 110.10; see also Buckley, 424 U.S. at 51–54. Self- financing often takes the form of loans, either from a candidate’s personal funds or through a third-party lender. A campaign may repay a candidate’s loans using contributions received both before and after the election. Under Section 304 of the Bipartisan Campaign Reform Act of 2002 (“BCRA”), however, a campaign may repay only $250,000 of a candidate’s pre-election loans with post-election contributions. See Pub. L. No. 107-155, § 304, 116 Stat. 81 (codified at 52 U.S.C. § 30116(j)).

The loan-repayment limit intersects with other restrictions on the use of campaign contributions promulgated by the Federal Election Commission (“FEC” or “Commission”). For instance, an individual may designate a contribution for a particular election, including a previous election. See 11 C.F.R. § 110.1(b)(2)(i). If designated for a previous election, a contribution may be accepted “only to the extent that [it] does not exceed net debts outstanding” from that election. See id. § 110.1(b)(3)(i). A campaign’s “net debts outstanding” for an election equals the “total amount of unpaid debts and obligations” minus its total available resources. Id. § 110.1(b)(3)(ii)(A)–(C). A campaign may accept post- election contributions only to the extent necessary to pay down a net shortfall. To effectuate the loan-repayment limit in Section 304, the calculation of “net debts outstanding” excludes the amount of any candidate loans “that in the

runoff election, and special election are treated as separate elections. See 52 U.S.C. § 30101(1)(A). 4

aggregate exceed $250,000 per election.” Id. § 110.1(b)(3)(ii)(C). The $250,000 limit applies to third-party loans secured by the candidate and also to loans from the candidate’s personal funds. See id. § 116.11(a).

A campaign has two options to pay back a candidate’s personal loans. First, a campaign “[m]ay repay the entire amount of the personal loans using contributions” made before the election. Id. § 116.11(b)(1). If the campaign chooses to use pre-election contributions, “it must do so within 20 days of the election.” Id. § 116.11(c)(1). Second, pursuant to Section 304, a campaign may repay up to $250,000 of the personal loans with post-election contributions. After the election, any balance of the personal loan that exceeds $250,000 will be treated “as a contribution by the candidate.” Id. § 116.11(c)(2).

B.

This case arose from Senator Cruz’s 2018 campaign for reelection to the United States Senate. The day before the general election, Senator Cruz made two loans totaling $260,000 to his campaign: $5,000 from his personal bank account and $255,000 from a third-party lender secured with his personal assets. Senator Cruz won reelection.

After the election, Senator Cruz’s campaign had almost $2.5 million in debt against approximately $2.2 million in cash on hand. The campaign “used the funds it had on hand to pay vendors and meet other obligations instead of repaying [Senator Cruz’s] loans.” Compl. ¶ 29, ECF No. 1. The campaign did not use any pre-election funds within twenty days of the election to repay the Senator’s loans, as Section 304’s implementing regulation would have permitted. Instead, the campaign repaid Senator Cruz the maximum $250,000 with post-election contributions but Section 304 prevented the campaign from 5

paying back the final $10,000. The $10,000 balance of those loans was subsequently deemed a campaign contribution from Senator Cruz.

Senator Cruz and his campaign (collectively, the “Cruz campaign”) brought suit against the FEC, alleging that Section 304 of BCRA and its implementing regulation, 11 C.F.R. § 116.11, violate the First Amendment. The complaint contends that the loan-repayment limit unconstitutionally infringes the First Amendment rights of Senator Cruz, his campaign, other candidates, and any individuals who might seek to make post-election contributions. Because the complaint concerned a constitutional challenge to a provision of BCRA, the Cruz campaign also applied for a three-judge district court pursuant to Section 403 of BCRA and 28 U.S.C. § 2284. The FEC moved to dismiss for lack of standing and also argued that a three-judge court would not have subject matter jurisdiction. The one-judge district court denied the FEC’s motion to dismiss, held the Cruz campaign had standing to challenge the loan-repayment limit, and granted the Cruz campaign’s application for a three-judge district court. See Ted Cruz for Senate v. FEC, 2019 WL 8272774, at *5–8 (D.D.C. Dec. 24, 2019). We convened to hear and decide the case.

Following additional preliminary proceedings, 2 the Cruz campaign and the FEC both moved for summary judgment.

2 We assumed supplemental jurisdiction over the Cruz campaign’s constitutional and Administrative Procedure Act claims against the implementing regulation. See Ted Cruz for Senate v. FEC, 451 F. Supp. 3d 92, 100 (D.D.C. 2020). We held these claims in abeyance pending resolution of the constitutional challenge to Section 304. Order, Ted Cruz for Senate v. FEC, No. 1:19-cv-00908 (D.D.C. Apr. 15, 2020), ECF No. 49. Our holding that Section 304 cannot pass 6

Summary judgment is warranted if “there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” FED. R. CIV. P. 56(a). “[I]n ruling on cross- motions for summary judgment, the court shall grant summary judgment only if one of the moving parties is entitled to judgment as a matter of law upon material facts that are not genuinely disputed.” Shays v. FEC, 424 F. Supp. 2d 100, 109 (D.D.C. 2006). Because the Cruz campaign and the FEC agree that there is no genuine dispute of material fact, we resolve this case by summary judgment.

II.

To determine whether the loan-repayment limit abridges First Amendment rights we follow the approach taken in McCutcheon v. FEC, the Supreme Court’s most recent foray into the constitutionality of a campaign finance regulation. 572 U.S. 185 (2014) (plurality opinion). First, we assess whether the loan-repayment limit burdens political speech and thus implicates the protection of the First Amendment. Second, because we conclude that the limit burdens political speech, we must carefully scrutinize the government’s interests and the fit between that interest and the regulatory means chosen to effectuate it. Even under the less exacting test of closely drawn scrutiny, we find the government fails to demonstrate that the loan-repayment limit serves an interest in preventing quid pro quo corruption or its appearance. Moreover, the loan- repayment limit has only a tenuous connection to the asserted government interest in preventing corruption and thus lacks the close tailoring necessary under the First Amendment.

constitutional muster moots the Cruz campaign’s regulatory challenges. 7

When presented with a less familiar type of campaign finance regulation, we must determine at the outset whether the restriction burdens the exercise of political speech. See id. at 203–06; Ariz. Free Enter. Club’s Freedom Club PAC v. Bennett, 564 U.S. 721, 736–47 (2011); Davis v. FEC, 554 U.S. 724, 738–40 (2008). The Cruz campaign argues the loan- repayment limit burdens speech by limiting campaign expenditures and contributions. The FEC maintains the limit does not burden speech at all. We find the loan-repayment limit burdens political speech and thus implicates the protection of the First Amendment.

The First Amendment provides that “Congress shall make no law … abridging the freedom of speech.” U.S. CONST. amend. I. This Amendment “is designed and intended to remove governmental restraints from the arena of public discussion, putting the decision as to what views shall be voiced largely into the hands of each of us, in the belief that no other approach would comport with the premise of individual dignity and choice upon which our political system rests.” McCutcheon, 572 U.S. at 203 (cleaned up). Robust and free political discussion is essential to the republican form of government established by our Constitution. Given the fundamental interests at stake, the First Amendment “safeguards an individual’s right to participate in the public debate through political expression and political association.” Id. Because financing for political campaigns implicates the freedom to speak and to associate, the Supreme Court has long recognized that limitations on campaign spending “necessarily reduce[] the quantity of expression by restricting the number of issues discussed, the depth of their exploration, and the size of the audience reached.” Buckley, 424 U.S. at 19. 8

Since Buckley, the Court’s decisions have focused on identifying whether a restriction on campaign finance burdens expenditures or contributions, in part because the distinction can affect the standard of review. 3 See id. at 25, 44–45. But it is well established that both expenditures and contributions implicate “fundamental First Amendment activities.” Id. at 14. When a candidate makes expenditures on behalf of her campaign, she exercises her right to speak; and when a contributor donates to that campaign, he exercises the right to associate with the candidate and to express his support. The contributions to a campaign in turn promote more expenditures and political speech by the candidate.

In recent decisions, the Court has declined to eliminate the distinction between expenditures and contributions even as it has focused on speech interests more generally. See, e.g., McCutcheon, 572 U.S. at 199; id. at 228 (Thomas, J., concurring in the judgment) (suggesting that the distinction between expenditures and contributions “has only continued to erode in the intervening years”) (cleaned up). The Court has emphasized the central question of whether and how a challenged regulation burdens political speech. For example, in McCutcheon, the Court explained that Buckley’s distinction between expenditure and contribution limits stemmed from the “the degree to which each encroaches upon protected First Amendment interests.” Id. at 197. The Court assessed the burden on expressive and associational rights imposed by the aggregate contribution limits challenged in that case. See id. at 204–05. In Davis, the Court found that a regulation burdened a

3 While burdens on expenditures must withstand strict scrutiny, the Court has assessed burdens on contributions under a less demanding, “but still ‘rigorous standard of review.’” McCutcheon, 572 U.S. at 197 (quoting Buckley, 424 U.S. at 29). 9

candidate’s expenditures because it raised contribution limits asymmetrically, that is, only for the opponents of a candidate who spent over a certain amount of his own money. See 554 U.S. at 738–40. The Court focused on how the regulation functioned to analyze the burden that it imposed. See id.; see also Bennett, 564 U.S. at 736–47 (evaluating the specific operation of Arizona’s matching funds provision and holding that it substantially burdened speech). In a political campaign, expenditures and contributions are part of a connected cycle of speech and association protected by the First Amendment.

We find that the loan-repayment limit restricts political expression and association for candidates and their contributors. To begin with, the loan-repayment limit burdens candidates who wish to make expenditures through personal loans because the limit constrains the repayment options available to the candidate. 4 Whereas other campaign debts may be repaid by post-election contributions, candidate loans above $250,000 do not receive the same treatment. That the candidate makes a choice to finance his campaign with personal loans, rather than through other forms of debt, does not minimize the First Amendment harm. Cf. Davis, 554 U.S. at 739 (“The

4 In general, a loan from a candidate to his campaign is treated as an expenditure. Both FECA and its regulations define the term “expenditure” to include loans. See 52 U.S.C. § 30101(9)(A)(i) (“The term ‘expenditure’ includes … any … loan, … made by any person for the purpose of influencing any election for Federal office[.]”); 11 C.F.R. § 100.111(a) (“A … loan … made by any person for the purpose of influencing any election for Federal office is an expenditure.”); 11 C.F.R. § 100.111(b) (“For purposes of this section, the term payment includes … any guarantee or endorsement of a loan by a candidate or a political committee.”); see also Anderson v. Spear, 356 F.3d 651, 672 (6th Cir. 2004) (“[L]oans are candidate expenditures, unless and until they are repaid.”). 10

resulting drag on First Amendment rights is not constitutional simply because it attaches as a consequence of a statutorily imposed choice.”). Candidate loans comprise the majority of campaign debt, and personal loans will sometimes be the only way for a candidate to raise enough money for an effective campaign in the short term. The limit places a particular burden on relatively unknown challengers who may require more financing up front in order to wage an effective campaign against a better funded incumbent. See Anderson v. Spear, 356 F.3d 651, 673 (6th Cir. 2004) (“[A] candidate may need to speak early in order to establish her position and garner contributions.”).

We also note that since the enactment of BCRA and the loan-repayment limit, “there is a clear clustering of loans right at the $250,000 threshold.” Alexei Ovtchinnikov & Philip Valta, Debt in Political Campaigns 24 (HEC Paris Research Paper No. FIN-2016-1165, May 2020). During this same time period, the percentage of candidate loans above $250,000 has remained roughly the same while spending on Senate and House campaigns has more than doubled, indicating that the loan-repayment limit constricts candidate lending.

We find the burden imposed by Section 304 “is evident and inherent in the choice that confronts” candidates who wish to use personal loans to finance their campaigns. Bennett, 564 U.S. at 745 (citing Davis, 554 U.S. at 738–40). The limit imposes a “drag” on the candidate’s First Amendment activity by discouraging the personal financing of campaign speech. Davis, 554 U.S. at 739.

The FEC defends the constitutionality of the loan- repayment limit by maintaining that it does not burden political speech at all, because “[m]oney that repays a candidate’s 11

personal loan after an election effectively goes into the candidate’s pocket, and not to fund speech or speech-related activities.” FEC Mem. in Supp. of Mot. for Summ. J. (“FEC Mot.”) 20, ECF No. 65. The Commission highlights that the loan-repayment limit does not cap the amount of candidate financing or prohibit a candidate from loaning his campaign more than $250,000, and the candidate remains free to repay the full amount of the loan with pre-election contributions.

While it is true that the loan-repayment limit is not a ban on personal financing, the First Amendment’s protection has never been limited to direct restrictions on expenditures, because “[t]he First Amendment would … be a hollow promise if it left government free to destroy or erode its guarantees by indirect restraints.” United Mine Workers of Am. v. Ill. State Bar Ass’n, 389 U.S. 217, 222 (1967). Laws that regulate in the First Amendment arena must be scrutinized even when the “deterrent effect on [speech] arises, not through direct government action, but indirectly as an unintended but inevitable result of the government’s conduct.” Buckley, 424 U.S. at 65.

Even indirect regulations of speech may run afoul of the First Amendment, because they can “abridg[e] the freedom of speech.” U.S. CONST. amend. I. The word “abridge” means “to contract, to diminish, to cut short.” 1 SAMUEL JOHNSON, A DICTIONARY OF THE ENGLISH LANGUAGE (6th ed. 1785); see also OXFORD ENGLISH DICTIONARY 43 (2d ed. 1989) (“abridge”: “To curtail, to lessen, to diminish (rights, privileges, advantages, or authority)”). At the time of the enactment of the First Amendment, as well as today, the plain meaning of “abridge” is to diminish or to curtail the freedom of speech. Consistent with this meaning, the First Amendment protects individuals not only from direct and outright bans on 12

speech, but also indirect actions the government might take to “abridge” the central freedom to speak freely in the democratic process.

Following these general principles, the Supreme Court has found a First Amendment burden even absent an outright ban or cap, when the regulation acted as a “drag” on speech— which is to say an “abridgment” of speech. Davis, 554 U.S. at 739–40. In Davis, the Supreme Court held unconstitutional a provision of BCRA that relaxed the base contribution limits for a candidate’s opponents if the candidate spent more than $350,000 of his own funds. The provision burdened free speech rights even though it “d[id] not impose a cap on a candidate’s expenditure of personal funds.” Id. at 738–39. Instead, the challenged provision “impose[d] an unprecedented penalty” on candidates who chose to “robustly exercise[] [their] First Amendment right[s].” Id. at 739. Similarly, in Bennett, the Court held unconstitutional an Arizona law that gave matching funds to publicly financed candidates if privately financed candidates—or independent expenditure groups—spent over a set amount. See 564 U.S. at 728. The Court concluded that the Arizona law “plainly force[d] the privately financed candidate to ‘shoulder a special and potentially significant burden’ when choosing to exercise his First Amendment right to spend funds on behalf of his candidacy.” Id. at 737 (quoting Davis, 554 U.S. at 739). If the law curtails a candidate’s ability to speak on his behalf, it runs afoul of the First Amendment even when the law is not an outright ban.

The FEC seeks to distinguish Davis and Bennett because those cases involved a penalty for candidate speech above a certain threshold, whereas the loan-repayment limit has no similar penalty—by loaning his campaign more than $250,000 13

a candidate does not indirectly fund his opponent through either liberalized, asymmetrical contribution limits (Davis) or matching funds (Bennett). First Amendment burdens, however, are not limited to prescribed forms. Our review must scrutinize regulatory burdens in order to vigorously protect the freedom of speech. While not identical to previously challenged regulations, the loan-repayment limit restricts a candidate’s campaign expenditures by circumscribing the repayment options for candidate loans over $250,000. 5

The FEC’s insistence that the loan-repayment limit does not burden political speech overlooks the reality of how the limit functions. The FEC narrowly focuses on the repayment of the loan and through this lens notes that the loan-repayment limit does not restrict expenditures because the candidate remains free to loan or contribute as much money as he wishes to his campaign. 6 The FEC’s cramped understanding of the

5 On the flip side, the loan-repayment limit may also impact contributors. Candidate loans over $250,000 are singled out and excluded from the “net debts outstanding” that a campaign may pay off with post-election contributions. The FEC’s regulations permit contributors to designate their contributions for a prior election. An individual who wanted to contribute to Senator Cruz after the 2018 election could not have contributed to—and thus expressed his support for—Senator Cruz’s 2018 election campaign if the only debt remaining was the Senator’s loan in excess of $250,000. 6 The FEC suggests there is no restriction on political speech in this case, relying on FEC v. O’Donnell, 209 F. Supp. 3d 727 (D. Del. 2016). That case is inapposite, however, because it concerned FECA’s ban on the use of contributions to pay a candidate’s personal expenses. The court held such contributions did not “facilitate political expression.” Id. at 739. By contrast, the loan-repayment limit restricts political expression and implicates the First Amendment in a way that personal expenses for a new outfit and a 14

First Amendment fails to provide adequate protection to the important free speech interests at stake. The FEC would isolate the transactions at issue until they no longer resemble campaign expenditures or contributions.

In determining whether First Amendment interests are implicated, however, we must focus on whether a statute burdens political speech, not whether a particular regulatory label is a perfect fit. The relative novelty of a campaign finance regulation cannot insulate it from judicial scrutiny because “political speech must prevail against laws that would suppress it, whether by design or inadvertence.” Citizens United v. FEC, 558 U.S. 310, 340 (2010). Legislators may try different regulatory approaches to protect against quid pro quo corruption; however, any such regulation of campaigns must comport with the First Amendment.

The loan-repayment limit implicates First Amendment interests. A candidate’s loan to his campaign is an expenditure that may be used for expressive acts. Such expressive acts are burdened when a candidate is inhibited from making a personal loan, or incurring one, out of concern that she will be left holding the bag on any unpaid campaign debt.

This case illustrates the reality that contributions and expenditures are often “two sides of the same First Amendment coin.” Buckley, 424 U.S. at 241 (Burger, C.J., concurring in part and dissenting in part). Contributions allow a candidate to make further expenditures, reflecting the practical link between the associational and expressive activity of the candidate and contributor. By limiting the amount of post-election contributions that can be used to retire candidate loans, the

gym membership arguably do not. See 52 U.S.C. § 30114(b)(2)(B) & (I). 15

loan-repayment limit abridges political speech and implicates the protection of the First Amendment.

Because the loan-repayment limit encumbers political speech, the government has “the burden of proving the constitutionality of its actions.” McCutcheon, 572 U.S. at 210 (cleaned up). The parties dispute the relevant standard of review. The Cruz campaign maintains we should apply either the strict scrutiny applicable to expenditure limits or the closely drawn scrutiny applied to contribution limits. By contrast, the FEC suggests the loan-repayment limit must be analyzed under deferential rational basis review because the limit burdens no First Amendment interests. Because we find the loan- repayment limit restricts expressive and associational interests in political campaigns, we must apply a form of heightened scrutiny, either strict or closely drawn.

Under either form of heightened scrutiny, we assess the government’s asserted interest in restricting speech and the fit between that interest and the means the government has chosen to fulfill it. See id. at 199. Applying strict scrutiny, a regulation will be upheld only if it furthers a compelling government interest and the government uses the least restrictive means of furthering that interest; whereas under closely drawn scrutiny a regulation will be upheld “if the State demonstrates a sufficiently important interest and employs means closely drawn to avoid” abridging First Amendment freedoms. See id. at 197.

The loan-repayment limit fails under even the less exacting test of closely drawn scrutiny and so, as in McCutcheon, we have no need to “parse the differences” between the standards of scrutiny. Id. at 199. The government 16

fails to demonstrate that the loan-repayment limit serves an interest in addressing quid pro quo corruption. In addition, we find “a substantial mismatch,” id., between the government’s asserted interest and the loan-repayment limit.

1.

The government bears the burden of demonstrating that the loan-repayment limit serves a sufficiently important interest that justifies the burden on political speech. The Supreme Court has made clear that the only recognized government interest in restraining political speech is “preventing corruption or the appearance of corruption.” Id. at 206–07. The Court has considered—and rejected—other government justifications such as “reduc[ing] the amount of money in politics,” id. at 191; “level[ing] electoral opportunities by equalizing candidate resources and influence,” Bennett, 564 U.S. at 748 (cleaned up); reducing “[i]ngratiation and access,” Citizens United, 558 U.S. at 360; or equalizing viewpoints among individuals and groups, Buckley, 424 U.S. at 48–49. The government’s interest in eliminating corruption is limited to quid pro quo corruption, in other words, “dollars for political favors.” McCutcheon, 572 U.S. at 192 (quoting FEC v. Nat’l Conserv. PAC, 470 U.S. 480, 497 (1985)). To comport with the First Amendment, a regulation of political speech must target only this particular form of corruption, which means “the Government may not seek to limit the appearance of mere influence or access.” Id. at 208.

In addition, it is not sufficient for the FEC merely to assert an interest in preventing quid pro quo corruption. The government must demonstrate the validity of its interest by more than “mere conjecture.” Nixon v. Shrink Mo. Gov’t PAC, 17

528 U.S. 377, 392 (2000). “When the Government defends a regulation on speech as a means to … prevent anticipated harms, it must do more than simply posit the existence of the disease sought to be cured.” Colo. Repub. Fed. Campaign Comm. v. FEC, 518 U.S. 604, 618 (1996) (cleaned up). Moreover, “[t]he quantum of empirical evidence needed to satisfy heightened judicial scrutiny of legislative judgments will vary up or down with the novelty and plausibility of the justification raised.” Shrink Mo., 528 U.S. at 391; see also Zimmerman v. City of Austin, 881 F.3d 378, 392–93 (5th Cir. 2018) (discussing cases). We assess the FEC’s asserted interests in light of these standards.

The FEC maintains that the loan-repayment limit addresses the heightened risk and appearance of quid pro quo corruption that results from elected officeholders soliciting contributions that will be used to repay their personal loans. The Commission posits that “[m]oney given after the election … provides the contributor with even more influence over the candidate since the candidate is benefiting personally from the contribution.” FEC Statement of Material Facts (“FEC SMF”) ¶ 73, ECF No. 65 (cleaned up). The Commission repeatedly characterizes post-election contributions used to repay candidate loans as going into the candidate’s pocket. The FEC also points to media reports of debt retirement parties as giving rise “to at least the appearance of federal candidates trading dollars for favors in the context of repayment of candidate loans.” FEC Mot. 33. The Commission maintains there is a public perception that individuals who contribute to candidates after an election are likely to expect a political favor in return.

Despite these assertions, the Commission fails to demonstrate that quid pro quo corruption or its appearance 18

arises from post-election contributions to retire a candidate’s personal debt. We first observe that the FEC has not identified a single case of actual quid pro quo corruption in this context. This is particularly notable given that many states impose no restriction on using post-election contributions to repay candidate loans, 7 and the Commission fails to identify any problems with quid pro corruption or its appearance in these states. Cf. Citizens United, 558 U.S. at 357 (finding it significant that the government failed to claim that “independent expenditures by for-profit corporations … corrupted the political process” in the twenty- six states that did not restrict such expenditures). Here the FEC’s few state examples involve only concerns that candidates will be too responsive to the influence of special

7 The Cruz campaign identifies ten states that cap candidate loans or restrict candidate loan repayment in some fashion. See Cruz Mem. in Supp. of Mot. for Summ. J. 28 & n.4, ECF No. 61-1. Georgia and South Carolina cap the repayment of candidate loans with post- election contributions, similar to BCRA’s loan-repayment limit. See GA. CODE ANN. § 21-5-41(h); S.C. CODE ANN. § 8-13-1328. Although Florida permits candidate loans and their repayment with pre-election contributions, it bans all post-election contributions. See FLA. STAT. § 106.08(3)(b). Alaska, Rhode Island, Texas, and Washington cap the repayment of candidate loans with either pre- or post-election contributions. See ALASKA STAT. § 15.13.078(b)(1); 17 R.I. GEN. LAWS § 17-25-7.4; TEX. ELEC. CODE § 253.042(a); WASH. REV. CODE § 42.17A.445(3). California, Massachusetts, and Nebraska place no limit on the repayment of candidate loans but instead cap the amount that candidates may loan their campaigns. See CAL. GOV’T CODE § 85307(b); MASS. GEN. LAWS ch. 55, § 7; NEB. REV. STAT. § 49-1446.04; 4 NEB. ADMIN. CODE ch. 10, § 004(02). The Commission does not contest that “only a minority of states” restrict candidate campaign loans in some way. FEC Mot. 34. 19

interests or concerns about contributions unrelated to the repayment of candidate loans. See, e.g., FEC SMF ¶¶ 76, 79.

By contrast, in cases that have found a sufficient anticorruption interest, the record has been robust. In Buckley, the Court cited “the deeply disturbing examples surfacing after the 1972 election” as demonstrating that the problem of quid pro quo corruption was “not an illusory one.” 424 U.S. at 27 & n.28; see also Buckley v. Valeo, 519 F.2d 821, 838–40 & nn.26–38 (D.C. Cir. 1975) (en banc) (describing extensive factual record before Congress). In McConnell v. FEC, the omnibus challenge to BCRA, the record before the court consisted of more than 100,000 pages, including “576 pages of proposed findings of fact” and “the testimony and declarations of over 200 fact and expert witnesses.” 251 F. Supp. 2d 176, 208–09 (D.D.C. 2003). In Bluman v. FEC, the court pointed to “public controversy and an extensive investigation by the Senate Committee on Governmental Affairs,” including specific examples of foreign governments attempting “to ‘influence U.S. policies and elections through, among other means, financing election campaigns,’” as justification for BCRA’s ban on expenditures and contributions by foreign nationals. 800 F. Supp. 2d 281, 283 (D.D.C. 2011) (quoting S. REP. NO. 105–67, at 47 (1998)).

A lengthy record may not be sufficient to demonstrate corruption, but the absence of any record of such corruption undermines the government’s proffered interest. The FEC cannot carry its substantial burden by simply asserting that post-election contributions to repay a candidate’s loans may come with expectations of a political favor.

In the absence of any evidence of actual corruption, the FEC turns elsewhere. For instance, the Commission relies 20

heavily on an academic article that concluded “[i]ndebted politicians … exhibit a heightened sensitivity in their voting decisions to political contributions received from special interest groups.” Ovtchinnikov & Valta, Debt in Political Campaigns 29. The article, however, does not distinguish between voting pattern changes as a consequence of donor influence or access and voting pattern changes as part of quid pro quo corruption. In a representative democracy, mere influence or access is not the type of quid pro quo corruption that justifies infringements on political speech. A “generic favoritism or influence theory … is at odds with standard First Amendment analyses because it is unbounded and susceptible to no limiting principle.” McConnell v. FEC, 540 U.S. 93, 296 (2003) (Kennedy, J., concurring in the judgment in part and dissenting in part). “The line between quid pro quo corruption and general influence may seem vague at times, but the distinction must be respected in order to safeguard basic First Amendment rights.” McCutcheon, 572 U.S. at 209.

The Commission also places great weight on a selective legislative history of the loan-repayment limit, arguing that lawmakers intended to “mitigate the heightened risk of quid pro quo corruption and its appearance resulting from already- elected officeholders soliciting contributions for their own personal benefit.” 8 FEC Mot. 6. Even on the doubtful 8 See, e.g., 147 CONG. REC. S2,462 (Mar. 19, 2001) (statement of Sen. Domenici) (“In fact, it should be a condition to your putting up your own money, knowing right up front you are not going to get it back from your constituents under fundraising events that you would hold and then ask them: How would you like me to vote now that I am a Senator?”); 147 CONG. REC. S2,541 (Mar. 20, 2001) (statement of Sen. Hutchison) (“[Candidates] have a constitutional right to try to buy the office, but they do not have a constitutional right to resell it.”). 21

proposition that assertions in legislative debates could carry the government’s burden, these statements from the legislative history amount to mere suppositions about the appearance of corruption. Moreover, the Cruz campaign proffers other tidbits of legislative history, including numerous statements suggesting that some legislators thought the loan-repayment limit would protect incumbents from wealthy challengers. 9 The competing statements in the legislative history of BCRA establish no clear emphasis on eradicating quid pro quo corruption as opposed to the impermissible purpose of leveling the playing field.

In addition, the loan-repayment limit, Section 304 of BCRA, was enacted at the same time as Section 319, the so- called “Millionaire’s Amendment,” which the Supreme Court held unconstitutional in part because it was intended to “level electoral opportunities for candidates of different personal wealth.” Davis, 554 U.S. at 741 (cleaned up). While

9 See, e.g., 147 CONG. REC. S2,541 (Mar. 20, 2001) (statement of Sen. Hutchison) (“Our purpose is to level the playing field so that one candidate who has millions, if not billions, of dollars to spend on a campaign will not be at such a significant advantage over another candidate who does not have such means as to create an unlevel playing field.”); 147 CONG. REC. S2,465 (Mar. 19, 2001) (statement of Sen. Sessions) (“It also prohibits wealthy candidates, who incur personal loans in connection with their campaign that exceed $250,000, from repaying those loans from any contributions made to the candidate. … I know there were large contributions in this last Senate campaign from candidates of $10 million, $60 million, and other amounts of money that the winning candidates in this body contributed from their own funds. I tell you, I am glad I didn’t face a person who could write a check for $60 million, $10 million—or $5 million, for that matter. If so, I would like to be able to have a level playing field so I could stay in the ball game.”). 22

Section 304 may serve a different purpose from Section 319, the text of BCRA, as well as the legislative debates, linked the two provisions, which suggests that the loan-repayment limit may also “further the impermissible objective of simply limiting the amount of money in political campaigns.” McCutcheon, 572 U.S. at 218. At a minimum, the connection between the provisions casts further doubt on the government’s asserted anticorruption interest.

Finally, the FEC relies on media reports and a YouGov poll, but these similarly fail to establish that restrictions like the loan-repayment limit serve the purpose of preventing quid pro quo corruption. The media reports merely hypothesize that individuals who contribute after the election to help retire a candidate’s debt might have greater influence with or access to the candidate. Yet this is not evidence of quid pro quo corruption, and minimizing influence and access is not a proper goal for campaign finance regulation. The YouGov poll was conducted at the FEC’s behest for this litigation to demonstrate that the loan-repayment limit addresses the appearance of corruption. The poll first asked respondents whether they were aware that candidates could loan their campaigns money and then be paid back with post-election contributions. FEC Mot. Ex. 16, ECF No. 65-16 (Decl. of Ashley Grosse, Ex. A). In the poll’s only two follow-up questions, 81 percent of respondents thought it “very likely” or “likely” that individuals who donate money to a federal candidate’s campaign after an election “expect a political favor in return,” and 67 percent of respondents thought donors would “be more likely to expect political favors” if there were no limit on repaying a candidate loan with post-election contributions. Id. The FEC relies on these responses as evidence that the loan-repayment limit addresses “at least the appearance of quid pro quo corruption.” FEC Mot. 32. 23

We disagree. Such generic questions do not get at the specific problem of quid pro quo corruption the government asserts this statute combats. On the government’s reasoning, the poll answers would raise doubts about any contributions to incumbents (i.e. winning candidates) who use post-election contributions to retire any type of campaign debt. Even if contributors who donate to retire a candidate’s debt expect political favors, that hardly demonstrates that the (now elected) official is more likely to grant such political favors. Moreover, the poll did not define the term “political favor,” so the poll’s responses are not evidence that the public associates such contributions with quid pro quo corruption, which Congress may regulate, or simply increased influence and access, which Congress may not. See McCutcheon, 572 U.S. at 208. Finally, the poll failed to mention that the individual contribution limit applies to post-election contributions just as it does to pre- election contributions. That omission renders the poll an ineffective measure of public perception of possible corruption in this context. At most, the poll suggests that some members of the public distrust or are skeptical about using contributions to repay candidate loans, but the “tendency to demonstrate distrust” is insufficient to establish corruption or its appearance. Nat’l Conserv. PAC, 470 U.S. at 499. We conclude the FEC fails to demonstrate that the loan-repayment limit serves an interest in preventing quid pro quo corruption.

The FEC also maintains that the loan-repayment limit prevents the circumvention of base contribution limits because without the limit a candidate could keep outstanding loans from past campaigns, which would allow individuals to stack up maximum contributions for each election for which the candidate had open loans. The problem with the FEC’s position, however, is that contributors are permitted to make multiple contributions at a single time—they can contribute to 24

retire debt from a previous election (subject to the loan- repayment limit) and they can contribute to any ongoing campaign for a future election. Each of these separate per- election contributions, however, is limited by the base contribution limit. Nothing about the potential for stacking circumvents the base limits. What the FEC terms “circumvention” is in fact a lawful contribution under existing campaign finance laws.

The government suggests it is dissatisfied with the possibility of large one-time contributions, which the FEC treats as a kind of legal loophole. Yet the loan-repayment limit does little to close the ostensible loophole, because the limit applies only to a candidate’s personal loans, not to other campaign debt. Also, the FEC fails to identify a plausible financial incentive for a candidate to carry significant personal campaign debt over many years simply to keep open the possibility of soliciting larger stacked donations in the future.

In sum, the FEC’s position amounts to speculation that contributions to pay off a candidate’s personal loans carry a danger of quid pro quo corruption, but the Supreme Court has “never accepted mere conjecture as adequate to carry a First Amendment burden.” Shrink Mo., 528 U.S. at 392. The government has failed to demonstrate that its interest in the loan-repayment limit is sufficiently important, because the limit serves no additional purpose in preventing quid pro quo corruption or the circumvention of base contribution limits. With little connection to any actual or perceived quid pro quo corruption interest, the FEC’s asserted rationale boils down to a general concern about money in politics and campaign contributions to incumbents—but such general concerns about influence or access cannot justify government regulation in the vital area of political speech. 25

2.

Even if the government had shown that the limit was justified by an important government interest, the loan- repayment limit is not “closely drawn” to protect expressive and associational freedoms. McCutcheon, 572 U.S. at 218 (quoting Buckley, 424 U.S. at 25). “In the First Amendment context, fit matters.” Id. The government’s rationale for the loan-repayment limit fits about as well as a pair of pandemic sweatpants. The First Amendment requires a better fit than that.

When assessing fit even under standards short of strict scrutiny, we “require a fit that is not necessarily perfect, but reasonable; that represents not necessarily the single best disposition but one whose scope is in proportion to the interest served, that employs not necessarily the least restrictive means but a means narrowly tailored to achieve the desired objective.” Id. (cleaned up). As part of the inquiry we consider “whether experience under the present law confirms a serious threat of abuse,” and whether there are less burdensome alternatives available to the government in securing its interests. Id. at 219 (quoting FEC v. Colo. Repub. Fed. Campaign Comm., 533 U.S. 431, 457 (2001)).

In arguing for a close fit, the FEC maintains “[t]he Loan Repayment Limit is tailored to apply in situations when the strength of the government’s important anti-corruption interests are at their peak,” because “the candidate will be in a position to grant political favors to [post-election] contributors.” FEC Mot. 40. Moreover, the FEC asserts, the limit is well tailored because it applies only to situations in which “the candidate or officeholder is directly, personally benefiting from the contributions,” and it does not prevent 26

campaigns from repaying the loans in full with pre-election funds. Id. at 41.

Contrary to the government’s assertions, the loan- repayment limit is not sufficiently tailored to achieve the objective of preventing quid pro quo corruption or its appearance. To begin with, the loan-repayment limit is over inclusive. It applies across the board to winning and losing candidates, although any purported anticorruption rationale applies only to winning candidates. The FEC’s primary defense of the regulation is that post-election contributions used to retire a candidate’s personal campaign loans are particularly susceptible to quid pro quo corruption or its appearance. This justification, however, does not apply to candidates who lose an election and therefore have no way to provide improper benefits to contributors who donate to retire election debt. Losing candidates are less likely to receive post- election contributions and, in any event, contributions made to a losing candidate pose essentially no risk of corruption or its appearance. See Anderson, 356 F.3d at 673 (invalidating a state cap on candidate loans and explaining that “the risk of quid pro quo is virtually non-existent where the contribution is made to a losing candidate who seeks to recoup some of his debt”). When a campaign finance regulation sweeps in conduct well beyond the government’s asserted rationale, it does not provide the close fit required by the First Amendment.

The loan-repayment limit is also substantially underinclusive as to the government’s asserted interests. Although “the First Amendment imposes no freestanding underinclusiveness limitation,” a law’s underinclusiveness can indicate a poor fit and can raise doubts about whether the law advances the interests invoked by the government. Williams- Yulee v. Fla. Bar, 135 S. Ct. 1656, 1668 (2015) (cleaned up). 27

Here, aside from the loan-repayment and base contribution limits, there are no restrictions on post-election contributions made to retire other types of campaign debt. A person may contribute to retire any outstanding campaign debt, with the exception of a candidate’s personal loans over $250,000. The FEC argues that a candidate who makes a loan to his campaign that he expects will be repaid is more dependent on outside contributions than a candidate who simply gives the money to his campaign. Yet not all candidates can afford to just give money to their campaigns—and there is nothing inherently corrupting about receiving campaign contributions after an election.

The FEC’s concerns regarding post-election contributions to retire candidate loans seem to apply equally to any contribution made to an incumbent, because all incumbents are in a position to grant favors. But Congress does not restrict pre- election contributions to incumbents except through the base contribution limit. The government has advanced no reason why a contribution made to an incumbent before the election poses no risk of corruption, but the same contribution made after the election to a winning candidate (now incumbent) and applied to pre-election debt poses a unique and heightened concern of quid pro quo corruption.

The government’s fit rationale also cannot explain why post-election contributions to retire pre-election debt are permissible up to the $250,000 cap. This cap means that in the current election cycle, a campaign committee can accept just over eighty-six maximum contributions after the election to repay a candidate loan (eighty-six contributions of $2,900 aggregates to $249,400, just shy of the $250,000 ceiling). It is hardly clear why the eighty-seventh or eighty-eighth contributor poses a particular danger of quid pro corruption. Cf. 28

McCutcheon, 572 U.S. at 210. Instead, the $250,000 cap operates to limit or disincentivize the total amount of campaign expenditure a candidate makes through personal loans.

The loan-repayment limit also imposes an additional regulatory requirement on top of the existing base limits. The loan-repayment limit is exactly the sort of “prophylaxis-upon- prophylaxis approach” that demands “we be particularly diligent in scrutinizing the law’s fit.” Id. at 221 (cleaned up). As the D.C. Circuit has explained, “an additional constraint layered on top of the base limits … separately need[s] to serve the interest in preventing the appearance or actuality of corruption.” 10 Holmes v. FEC, 875 F.3d 1153, 1161 (D.C. Cir. 2017) (en banc) (cleaned up). Post-election contributions, like contributions made before an election, are subject to the base limits, which serve to prevent the dangers of quid pro quo corruption. Layered on top of the base limits, the loan- repayment limit places an additional restriction on pre-election expenditures and post-election contributions, but the government has failed to demonstrate that the limit provides additional protection against quid pro quo corruption or its appearance.

10 Other circuit courts have similarly interpreted McCutcheon as requiring the government to make an additional showing to justify campaign finance restrictions that operate on top of base limits. See, e.g., Jones v. Jegley, 947 F.3d 1100, 1106 (8th Cir. 2020) (“Just as in McCutcheon, Arkansas’s failure here to provide any evidence that its blackout period accomplishes anything more than the $2,700 base limits alone means that it cannot survive exacting scrutiny.”); Zimmerman v. City of Austin, 881 F.3d 378, 392 (5th Cir. 2018) (holding restrictions in addition to the base limit “must be justified by evidence that the additional limit serves a distinct interest in preventing corruption that is not already served by the base limit”). 29

The Commission next tries to demonstrate fit by minimizing the burden of the loan-repayment limit. For instance, the Commission maintains that the loan-repayment limit “increase[s] the funds available to campaign committees,” and so does not “prevent[] campaigns from ‘amassing the resources necessary for effective advocacy.’” FEC Mot. 41 (quoting Randall v. Sorrell, 548 U.S. 230, 247 (2006) (plurality opinion)). The FEC overreads Randall, which noted that if a contribution limit prevents a campaign from amassing the necessary resources, it cannot survive under the First Amendment. See Randall, 548 U.S. at 248. It does not logically follow, however, that if a campaign can manage to amass necessary resources, the regulation survives First Amendment scrutiny. Preventing candidates from amassing resources is only one of the reasons a regulation of political speech may fail under the First Amendment, and therefore it cannot serve as an independent basis for upholding a regulation. Cf. Libertarian Nat’l Comm. v. FEC, 924 F.3d 533, 558–59 (D.C. Cir. 2019) (en banc) (Katsas, J., concurring in part, concurring in the judgment, and dissenting in part). Moreover, the determination of what resources are “necessary” for effective speech must be left to individual speakers, not the FEC.

Finally, the Commission urges this court to defer to Congress’s judgment that the loan-repayment limit is necessary for combatting corruption. While we must respect the legislative choices of Congress acting within its constitutional sphere, we cannot defer on the question of whether a particular legislative choice is in fact constitutional. “We must give weight to attempts by Congress to seek to dispel either the appearance or the reality of [corruptive] influences. The remedies enacted by law, however, must comply with the First Amendment; and it is our law and our tradition that more 30

speech, not less, is the governing rule.” Citizens United, 558 U.S. at 361; see also Schneider v. State, 308 U.S. 147, 161 (1939) (explaining that legislative judgments may be “insufficient to justify” a restriction that “diminishes the exercise of rights so vital to the maintenance of democratic institutions”). Courts cannot rubber stamp congressional preferences when important First Amendment interests are at stake.

In sum, we hold that the government failed to meet its burden of demonstrating that the loan-repayment limit serves an interest in combatting quid pro quo corruption or its appearance and that in any event the loan-repayment limit is insufficiently tailored to meet this objective.

***

When it comes to campaign finance regulation, the foxes are effectively in charge of the political henhouse, because elected officials set the rules for future elections. The Constitution, however, does not leave our liberties to the foxes. Laws regulating political speech implicate First Amendment rights essential to a free democracy, and courts have an independent duty to scrutinize the government’s interest as well as the means chosen to realize it. To protect “the political responsiveness at the heart of the democratic process,” McCutcheon, 572 U.S. at 227, Congress may regulate political speech only to prevent the specific problem of quid pro quo corruption. The loan-repayment limit does not serve that interest, and the government’s arguments to the contrary boil down to hypothetical concerns about influence and access to incumbents. Such justifications are not sufficient under the First Amendment to uphold a statute that burdens political speech. The loan-repayment limit intrudes on fundamental 31

rights of speech and association without serving a substantial government interest.

For the foregoing reasons, we hold that the loan- repayment limit, Section 304 of BCRA, is unconstitutional because it violates the First Amendment. Thus, the court denies the Commission’s motion for summary judgment and grants the Cruz campaign’s motion for summary judgment. A separate order accompanies this memorandum opinion.