Thompson v. Hebdon

Full title: David Thompson; Aaron Downing; Jim Crawford, Plaintiffs-Appellants, v…

Court: United States Court of Appeals, Ninth Circuit

Date published: Jul 30, 2021

7 F.4th 811 (9th Cir. 2021)

Opinion

No. 17-35019

07-30-2021

David THOMPSON ; Aaron Downing; Jim Crawford, Plaintiffs-Appellants, v. Heather HEBDON, in Her Official Capacity as the Executive Director of the Alaska Public Offices Commission; Richard Stillie; Irene Catalone; Anne Helzer; Robert Clift; and Jim McDermott, in their official capacities as members of the Alaska Public Offices Commission, Defendants-Appellees.

Erin Murphy (argued), Paul D. Clement, and Lauren N. Beebe, Kirkland & Ellis LLP, Washington, D.C.; Kevin G. Clarkson, Matthew C. Clarkson, Robin O. Brena, and Laura S. Gould, Brena Bell & Walker P.C., Anchorage, Alaska; for Plaintiffs-Appellants. Laura Fox (argued), Senior Assistant Attorney General; Tregarrick R. Taylor, Deputy Attorney General; Department of Law, Anchorage, Alaska, for Defendants-Appellees. Brian A. Sutherland, Reed Smith LLP, San Francisco, California; M. Patrick Yingling, Reed Smith LLP, Chicago, Illinois; Brent Ferguson and Daniel I. Weiner, Brennan Center for Justice, New York, New York; for Amicus Curiae Brennan Center for Justice at NYU School of Law. Tara Malloy and Megan P. McAllen, Campaign Legal Center, Washington, D.C., for Amicus Curiae Campaign Legal Center. Ronald A. Fein and John C. Bonifaz, Free Speech for People, Newton, Massachusetts, for Amici Curiae Free Speech for People and Professor David Fontana. Tara Malloy, Noah B. Lindell, Megan P. McAllen, and Mark P. Gaber, Campaign Legal Center, Washington, D.C., for Amicus Curiae Campaign Legal Center.

CALLAHAN, Circuit Judge

Erin Murphy (argued), Paul D. Clement, and Lauren N. Beebe, Kirkland & Ellis LLP, Washington, D.C.; Kevin G. Clarkson, Matthew C. Clarkson, Robin O. Brena, and Laura S. Gould, Brena Bell & Walker P.C., Anchorage, Alaska; for Plaintiffs-Appellants.

Laura Fox (argued), Senior Assistant Attorney General; Tregarrick R. Taylor, Deputy Attorney General; Department of Law, Anchorage, Alaska, for Defendants-Appellees.

Brian A. Sutherland, Reed Smith LLP, San Francisco, California; M. Patrick Yingling, Reed Smith LLP, Chicago, Illinois; Brent Ferguson and Daniel I. Weiner, Brennan Center for Justice, New York, New York; for Amicus Curiae Brennan Center for Justice at NYU School of Law.

Tara Malloy and Megan P. McAllen, Campaign Legal Center, Washington, D.C., for Amicus Curiae Campaign Legal Center.

Ronald A. Fein and John C. Bonifaz, Free Speech for People, Newton, Massachusetts, for Amici Curiae Free Speech for People and Professor David Fontana.

Tara Malloy, Noah B. Lindell, Megan P. McAllen, and Mark P. Gaber, Campaign Legal Center, Washington, D.C., for Amicus Curiae Campaign Legal Center.

Before: Sidney R. Thomas, Chief Judge, and Consuelo M. Callahan and Carlos T. Bea, Circuit Judges.

Partial Concurrence and Partial Dissent by Chief Judge Thomas

ORDER

The opinion filed on November 27, 2018, and published at 909 F.3d 1027, is withdrawn, and replaced by the opinion filed concurrently with this order.

CALLAHAN, Circuit Judge:

We must decide whether an Alaska law regulating campaign contributions violates the First Amendment. At issue are Alaska's limits on contributions made by individuals to candidates, individuals to election-related groups, and political parties to candidates, and also its limit on the total funds a candidate may receive from out-of-state residents. The district court upheld all four provisions against a constitutional challenge by three individuals and a subdivision of the Alaska Republican Party. See Thompson v. Hebdon , 909 F.3d 1027, 1032–33 (9th Cir. 2018). In a prior opinion, we affirmed except as to the nonresident limit. Id . at 1031. Plaintiffs filed a petition for certiorari. See Thompson v. Hebdon , ––– U.S. ––––, 140 S. Ct. 348, 351, 205 L.Ed.2d 245 (2019). The Supreme Court issued a per curiam opinion granting the petition, vacating our judgment, and remanding the case for us to "revisit" whether the individual-to-candidate and individual-to-group limits "are consistent with [the Supreme Court's] First Amendment precedents," in particular Randall v. Sorrell , 548 U.S. 230, 126 S.Ct. 2479, 165 L.Ed.2d 482 (2006). Id . Following remand, we received supplemental briefs from the parties and an amicus, and we heard oral argument. We now issue a revised opinion. Our resolution of the challenges to the political-party-to-candidate and nonresident limits remains the same, affirming the district court's decision upholding the former but reversing the decision upholding the latter. But we now reverse the district court's decision upholding the individual-to-candidate and individual-to-group limits.

I

A

Alaska has long regulated campaign contributions to political candidates. In 1974, Alaska enacted a statute prohibiting individuals from contributing more than $1,000 annually to a candidate. See Alaska v. Alaska Civil Liberties Union , 978 P.2d 597, 601 (Alaska 1999). In 1996, the Alaska Legislature enacted a revised campaign finance law "to restore the public's trust in the electoral process and to foster good government." 1996 Alaska Sess. Laws ch. 48 § 1(b). Among other things, the law lowered the annual limit on contributions by individuals to a candidate from $1,000 to $500 and set a $500 limit on annual contributions by individuals to a group that is not a political party. Id . §§ 10–11. The law also set aggregate limits on the amount candidates could accept from nonresidents of Alaska. In 2003, the Alaska legislature revised the 1996 law by raising the individual-to-candidate and individual-to-group limits from $500 to $1,000. 2003 Alaska Sess. Laws ch. 108, §§ 8–10.

In 2006, a ballot initiative—Ballot Measure 1 (the "2006 Initiative")—proposed a further revision of the limits. 2006 Alaska Laws Initiative Meas. 1, § 1. The 2006 Initiative, which is the law at issue here, returned the individual-to-candidate and individual-to-group limits to their pre-2003 levels of $500 per year. Alaska Stat. § 15.13.070(b)(1). It also capped the amount a non-political party group could contribute to a candidate at $1,000, restricted the amount candidates could receive from nonresidents to $3,000 per year, and limited the amount a political party—including its subdivisions—could contribute to a candidate. Alaska Stat. §§ 15.13.070(c) & (d), 15.13.072(a)(2) & (e)(3), 15.13.400(15). The 2006 Initiative passed with 73% of the popular vote.

B

Plaintiffs are three individuals and a subdivision of the Alaska Republican Party. In 2015, Plaintiffs brought a First Amendment challenge against Defendants, Alaska public officials, targeting, as relevant to this appeal, (1) the $500 annual limit on an individual contribution to a political candidate, (2) the $500 limit on an individual contribution to a non-political party group, (3) annual limits on what a political party—including its subdivisions—may contribute to a candidate, and (4) the annual aggregate limit on contributions a candidate may accept from nonresidents of Alaska. Plaintiffs sought a declaratory judgment that each of the challenged provisions is unconstitutional, a permanent injunction prohibiting enforcement of the challenged provisions, and costs and attorney's fees under 42 U.S.C. § 1983. Thompson v. Dauphinais , 217 F. Supp. 3d 1023, 1027 (D. Alaska 2016).

Two of the Plaintiffs, Aaron Downing and Jim Crawford, are Alaska residents who wanted to, but legally could not, contribute more than $500 to individual candidates running for state or municipal office. Crawford also wanted to give more than $500 to a non-political party group. David Thompson is a Wisconsin resident whose brother-in-law is former Alaska State Representative Wes Keller. Thompson sent Keller a $100 check for his campaign in 2015, but Keller returned the check because the campaign had already hit the $3,000 nonresident limit. Finally, District 18 is a subdivision of the Alaska Republican Party that was limited in the amount it could give to Amy Demboski's mayoral campaign due to Alaska's aggregate limit on the amount a campaign can accept from a political party.

After granting Alaska's motion for partial summary judgment for lack of standing on certain of Plaintiffs’ claims, the district court held a seven-day bench trial. In November 2016, the district court issued a decision rejecting all of Thompson's remaining claims. Thompson , 217 F. Supp. 3d at 1027–40. Applying the intermediate scrutiny standard for evaluating contribution limitations set forth in Montana Right to Life Ass'n v. Eddleman , 343 F.3d 1085 (9th Cir. 2003), the district court determined that each of the four challenged provisions was aimed at the "important state interest" of combating quid pro quo corruption (or its appearance) and was "closely drawn" to meet that interest. Thompson , 217 F. Supp. 3d at 1040. Plaintiffs timely appealed.

On appeal, our prior opinion analyzed whether those limits furthered a "sufficiently important state interest" and were "closely drawn" to that end. Thompson , 909 F.3d at 1034 (quoting Eddleman , 343 F.3d at 1092 ) (internal quotation marks omitted). We recognized that the Supreme Court's decisions in Citizens United v. Fed. Election Comm'n , 558 U.S. 310, 130 S.Ct. 876, 175 L.Ed.2d 753 (2010), and McCutcheon v. Fed. Election Comm'n , 572 U.S. 185, 134 S.Ct. 1434, 188 L.Ed.2d 468 (2014), narrow "the type of state interest that justifies a First Amendment intrusion on political contributions" to combating "actual quid pro quo corruption or its appearance." Thompson , 909 F.3d at 1034. We concluded that the individual-to-candidate contribution limit " ‘focuses narrowly on the state's interest,’ ‘leaves the contributor free to affiliate with a candidate,’ and ‘allows the candidate to amass sufficient resources to wage an effective campaign,’ " and thus survived First Amendment scrutiny. Thompson , 909 F.3d at 1036–39 (quoting Eddleman , 343 F.3d at 1092 ) (alterations omitted). We then found the individual-to-group contribution limit valid as a tool for preventing circumvention of the individual-to-candidate limit. See id. at 1039–40. We also upheld the political party-to-candidate limit. Id . at 1040. However, we reversed as to the nonresident limit. While we found that the first three restrictions narrowly tailored to prevent quid pro quo corruption or its appearance and thus did not impermissibly infringe constitutional rights, we found that the nonresident limit did not target an "important state interest" and therefore violated the First Amendment. Id . at 1040–43.¹ C

1 Chief Judge Thomas concurred in part and dissented in part. Thompson , 909 F.3d 1027 at 1044 (Thomas, C.J., concurring in part and dissenting in part). He agreed that Alaska's limitations on contributions made by individuals to candidates, individuals to election-related groups, and political parties to candidates do not violate the First Amendment. Id . But he would hold that the nonresident aggregate contribution limit also does not violate the First Amendment. Id .

The Supreme Court remanded, taking issue with our failure to apply Randall to the two $500 limits on individuals to candidates and election-related groups.² Thompson , 140 S. Ct. at 350. In Randall , the Supreme Court "invalidated a Vermont law that limited individual contributions on a per-election basis to: $400 to a candidate for Governor, Lieutenant Governor, or other statewide office; $300 to a candidate for state senator; and $200 to a candidate for state representative." See id . Justice Breyer's opinion for the plurality observed that "contribution limits that are too low can ... harm the electoral process by preventing challengers from mounting effective campaigns against incumbent officeholders, thereby reducing democratic accountability." Randall , 548 U.S. at 248–49, 126 S.Ct. 2479 ; see also id . at 264–65, 126 S.Ct. 2479 (Kennedy, J., concurring in judgment) (agreeing that Vermont's limits were unconstitutional); id . at 265–73, 126 S.Ct. 2479 (Thomas, J., joined by Scalia, J., concurring in judgment) (agreeing that Vermont's limits were unconstitutional and arguing that such limits should be analyzed under strict scrutiny). Justice Breyer explained that a contribution limit that is too low can therefore "prove an obstacle to the very electoral fairness it seeks to promote." Id . at 249, 126 S.Ct. 2479 (plurality opinion).

2 We declined to apply Randall because we believed that it was "not binding authority because no opinion commanded a majority of the Court." Thompson , 909 F.3d at 1037 n.5.

Randall "identified several ‘danger signs’ about Vermont's law that warranted closer review." Thompson , 140 S. Ct. at 350. In remanding this matter, the Supreme Court found that "Alaska's limit on campaign contributions shares some of those characteristics" in three ways. Id. "First, Alaska's $500 individual-to-candidate contribution limit is ‘substantially lower than ... the limits [the Supreme Court has] previously upheld.’ " Id . (quoting Randall , 548 U.S. at 253, 126 S.Ct. 2479 ). "Second, Alaska's individual-to-candidate contribution limit is ‘substantially lower than ... comparable limits in other States.’ " Id. at 351 (quoting Randall , 548 U.S. at 253, 126 S.Ct. 2479 ). "Third, Alaska's contribution limit is not adjusted for inflation." Id.

The Randall Court, after finding several danger signs, considered "five sets of considerations" or "factors" to determine whether Vermont's limits were constitutional: (1) whether the limits would significantly restrict the amount of funding available for challengers to run competitive campaigns; (2) whether political parties must abide by the same low limits that apply to individual contributors; (3) whether volunteer services or expenses are considered contributions that would count toward the limit; (4) whether the limits are indexed for inflation; and (5) whether there is any "special justification" that might warrant such low limits. Randall , 548 U.S. at 244–62, 126 S.Ct. 2479. The remand here specifically noted the "special justification" factor for our consideration. Thompson , 140 S. Ct. at 351.

II

"When the Government restricts speech, the Government bears the burden of proving the constitutionality of its actions." McCutcheon , 572 U.S. at 210, 134 S.Ct. 1434 (citation omitted). "We review a district court's legal determinations, including constitutional rulings, de novo ." Berger v. City of Seattle , 569 F.3d 1029, 1035 (9th Cir. 2009) (en banc). "When the issue presented involves the First Amendment .... [h]istorical questions of fact (such as credibility determinations or ordinary weighing of conflicting evidence) are reviewed for clear error, while constitutional questions of fact (such as whether certain restrictions create a ‘severe burden’ on an individual's First Amendment rights) are reviewed de novo ." Prete v. Bradbury , 438 F.3d 949, 960 (9th Cir. 2006).

A

For the $500 individual-to-candidate limit on remand, we "examine the record independently," see Randall , 548 U.S. at 253, 126 S.Ct. 2479, and apply the five-factor test outlined in Randall with an emphasis on the "special justification" factor.

i

We begin with the first Randall factor and ask whether Alaska's individual-to-candidate contribution limit "significantly restrict[s] the amount of funding available for challengers to run competitive campaigns." See id . This first factor favors Plaintiffs.

Incumbent officeholders in Alaska have a number of advantages that challengers must overcome in order to be competitive. First, they have name-recognition. Challengers need to expend sufficient funds from the start of campaigns so voters even know who they are. See id . at 256, 126 S.Ct. 2479 (recognizing "the typically higher costs that a challenger must bear to overcome the name-recognition advantage enjoyed by an incumbent"). This can be especially difficult in Alaska's geographically large districts or in its state-wide races. Former Senator John Coghill testified that when he campaigned for his senate seat, he "had to drive from North Pole to Valdez to Glennallen to Delta, to as far as the Palmer area, and those kinds of things just make it more expensive," and that "[s]ome people have to fly from village to village." He remembered that he had "a rural area where [he] had to drive literally for two days to cover the district."

Second, Alaska's use of annual limits favors incumbents. Alaska Stat. § 15.13.070(b)(1) (2018). General elections occur in even-numbered years. The odd-numbered years are known as "off-years." Challengers tend to register to run in election years. Their tendency not to enter political races earlier is not necessarily conscious or negligent. Often it is just that they are not recruited to run until the year of the general election. It follows that challengers are often not registered as candidates and therefore cannot raise money in the off-years. Meanwhile, most incumbents are registered as candidates and raise money year in and year out. Thus, challengers are short the contributions from those who contributed to them during the election year but would have also given during the off-year. For example, in the election cycles from 2002 to 2014, challengers overwhelmingly did not raise money in off-years, while incumbents overwhelmingly did.

Third, challengers often have to run first in primary elections for which they need to spend money, while incumbents are less likely to face primary challenges, and hence they may save that money to use against their challengers in general elections.

One example of the convergence of these factors is the 2012 state senate race between incumbent Hollis French and challenger Bob Bell. French raised money in the off-year (228 contributions compared to Bell's 0 contributions). Then Bell entered the race the year of the general election. He first had to spend money defeating a primary opponent, which French did not have. In the end, French raised about $172,000, and Bell $126,000. French won.

These advantages to incumbents raise questions as to whether the $500 individual-to-candidate contribution limit establishes too low a ceiling to allow challengers to launch campaigns and continue to run against incumbents competitively. The Supreme Court noted that this "contribution limit is substantially lower than the limits we have previously upheld," and "substantially lower than comparable limits in other States." Thompson , 140 S. Ct. at 351 (citations and alterations omitted). In particular, "[o]nly five other States have any individual-to-candidate contribution limit of $500 or less per election." Id . (noting that "the per-election contribution limit is comparable to Alaska's annual limit" because in most states "primary and general elections count[ ] as separate elections"). "Moreover, Alaska's $500 contribution limit applies uniformly to all offices, including Governor and Lieutenant Governor ..., making Alaska's law the most restrictive in the country in this regard." Id . (citations omitted). Such a low limit "magnif[ies] the advantages of incumbency to the point where they put challengers to a significant disadvantage." See Randall , 548 U.S. at 248, 126 S.Ct. 2479.

The record³ and logic indicate that a low limit restricts contributions from those who would donate more. In former Alaska governor Tony Knowles's first gubernatorial race in 1994, he raised $1.7 million under a $1,000 limit, but only $1.0 million in 1998 under a $500 limit. When he ran again in 2006 under a $1,000 limit, he again raised $1.7 million. When the limit was $1,000, there was a 60% increase in funds among gubernatorial candidates generally. In races for the state House of Representatives, the average total annual contributions to all registered candidates in the $1,000-limit years (2003–2006) was $3.15 million, while in the $500-limit years (2002, 2008–2014) it was around $2.5 million, a difference of over $600,000. Applied to challengers, the low limit here significantly restricts the amount of funds they have to mount successful challenges against advantaged incumbents. This is especially so considering that in competitive campaigns in Alaska, candidates who raise more money generally win, incumbents regularly raise more than challengers, and indeed incumbents win almost all elections.

3 One of Plaintiffs’ experts attempted to conduct a study to measure lost revenue to candidates by estimating how many contributors who donated the maximum amount would give more if they could. But by his own admission, his analysis was flawed in certain respects and thus both we and the district court were unable to rely on it. Thompson , 217 F. Supp. 3d at 1035 ; Thompson , 909 F.3d at 1038.

The dissent recycles the district court's factual findings to support a contrary conclusion and claims that we "do not pay adequate deference" to them. Dissent at 34. However, that those findings supported a conclusion that the limit allows candidates "to amass sufficient resources to run effective campaigns," see Eddleman , 343 F.3d at 1092, does not necessarily answer the different question the Supreme Court posed to us by asking us to apply Randall —whether the limit "significantly restrict[s] the amount of funding available for challengers to run competitive campaigns," Randall , 548 U.S. at 253, 126 S.Ct. 2479. To answer that question, we must "review the record independently ." Id . at 249, 126 S.Ct. 2479 (emphasis added). In Randall , the Supreme Court stated that "the record ... does not conclusively prove ... that [Vermont's] contribution limits will significantly restrict the amount of funding available for challengers to run competitive campaigns," yet found that "the inference amounts to one factor (among others) that here counts against the constitutional validity of the contribution limits." Randall , 548 U.S. at 256, 126 S.Ct. 2479. It was enough that "the record suggest[ed]" that the limits were a significant restriction. Id . at 253, 126 S.Ct. 2479. Similarly, here, we find that the record at least "suggests" that Alaska's individual contribution limit "significantly restrict[s] the amount of funding available for challengers to run competitive campaigns" and thus "counts against the constitutional validity of the contribution limits." See id . at 253, 256, 126 S.Ct. 2479.

ii

The second Randall factor, whether political parties must abide by the same low limits that apply to individual contributors, and the third Randall factor, whether volunteer services or expenses are considered contributions that would count toward the limit, both favor Alaska. As Justice Ginsburg observed in her separate statement accompanying the remand, "political parties in Alaska are subject to much more lenient contribution limits than individual donors." Thompson , 140 S. Ct. at 351 (Ginsburg, J., concurring) (citing Alaska Stat. § 15.13.070(d) (2018) ). And Alaska does not count volunteer services toward the contribution limits. See Alaska Stat. § 15.13.400(4)(B)(i) ; Alaska Admin. Code tit. 2, § 50.250(d). As for volunteer expenses, we read Alaska's definition of "contribution" as generally excluding such expenses from being defined as contributions. For anything to count as a "contribution," it apparently needs to be directly "rendered to the candidate or political party" and spent "for the purpose of ... influencing the nomination or election of a candidate" or "influencing a ballot proposition or question." Alaska Stat. § 15.13.400(4)(A) (emphasis added). Thus, Justice Breyer's concern that Vermont's law would "count [volunteer] expenses against the volunteer's contribution limit" does not seem to be present here. See Randall , 548 U.S. at 259, 126 S.Ct. 2479 (noting how under Vermont's law, "anything of value, paid ... for the purpose of influencing an election" might count as a "contribution"). Applying Justice Breyer's hypothetical to Alaska, "a gubernatorial campaign volunteer who makes four or five round trips driving across the State performing volunteer activities coordinated with the campaign [should not] find that he or she is near, or has surpassed, the contribution limit." See id . at 260, 126 S.Ct. 2479.

iii

As for the fourth factor, the parties do not dispute that Alaska's limits are not indexed for inflation, and thus this factor favors Plaintiffs. The Supreme Court noted that "Alaska's $500 contribution limit is the same as it was 23 years ago, in 1996." Thompson , 140 S. Ct. at 351. Justice Breyer explained in Randall , "[a] failure to index limits means that limits which are already suspiciously low, will almost inevitably become too low over time." Randall , 548 U.S. at 261, 126 S.Ct. 2479 (citation omitted). He explained that the lack of indexing "means that future legislation will be necessary to stop that almost inevitable decline, and it thereby imposes the burden of preventing the decline upon incumbent legislators who may not diligently police the need for changes in limit levels to ensure the adequate financing of electoral challenges." Id . Here, $500 in 2021 dollars appears to have a real value of about $375 in 2006 dollars, 2006 being the year the Alaska contribution limits at issue were passed.

Alaska asserts that candidates are increasingly turning to social media to campaign, which costs less than traditional means. However, the bread-and-butter approach of meeting voters face-to-face, sending them mailings, and reaching them by television and radio remain, and the costs of those methods inevitably rise over time. On top of that, the costs of hiring staff and renting space is ever increasing. As the cost of living rises so does the cost of campaigning.

iv

We finally turn to the fifth factor, whether there is any "special justification" for Alaska's low limits, which the Supreme Court specifically mentioned in its remand. Thompson , 140 S. Ct. at 351. Justice Ginsburg noted in her separate statement that "Alaska has the second smallest legislature in the country and derives approximately 90 percent of its revenues from one economic sector—the oil and gas industry. ... [T]he [d]istrict [c]ourt suggested [that] these characteristics make Alaska ‘highly, if not uniquely, vulnerable to corruption in politics and government.’ " Id . at 351–52 (citation omitted).

The district court recognized that "the prevention of quid pro quo corruption, or its appearance, is the only state interest that can support limits on campaign contributions." Thompson , 217 F. Supp. 3d at 1028 (citing Citizens United and McCutcheon ). It reasonably observed that "[l]ower limits often increase the donor base and decrease the impact of an individual contribution, thus making it easier for a candidate to decline a contribution contingent upon the performance of a political favor." See id . at 1033. However, the Supreme Court found that Alaska's individual contribution limits are so low as to exhibit "danger signs," which requires that we determine whether Alaska has a "special justification" indicating that "corruption (or its appearance) in [Alaska] is significantly more serious a matter than elsewhere." Randall , 548 U.S. at 261, 126 S.Ct. 2479. Trial witnesses certainly testified to a number of incidents where legislators were "pressure[d] to vote in a particular way or support a certain cause in exchange for past or future campaign contributions while in office." Thompson , 217 F. Supp. 3d at 1029. However, the record contains no indication that corruption or its appearance is more serious in Alaska than in other states. The small size of the legislature and the influence of the oil industry are risk factors, but Alaska's anecdotal evidence is insufficient to establish that "corruption (or its appearance) in [Alaska] is significantly more serious a matter than elsewhere ." Randall , 548 U.S. at 261, 126 S.Ct. 2479 (emphasis added). Alaska continually draws our attention to the VECO scandal. But a scandal that occurred and was widely publicized some fifteen years ago involving six to seven legislators who engaged largely in criminal bribery by putting money into their own pockets instead of their campaigns, though disturbing, is insufficient for us to conclude that there is a present special justification for Alaska's low individual contribution limit.

v

On balance, our consideration of the five factors leads us to hold that Alaska has failed to meet its burden of showing that its individual contribution limit is "closely drawn to meet its objectives." See Randall , 548 U.S. at 253, 126 S.Ct. 2479. On top of its danger signs, the limit significantly restricts the amount of funds available to challengers to run competitively against incumbents, and the already-low limit is not indexed for inflation. Moreover, Alaska has not established a special justification for such a low limit.

B

Similarly, Alaska has not met its burden of showing that the $500 individual-to-group limit is closely drawn to restrict contributors from circumventing the individual-to-candidate limit. "This ‘prophylaxis-upon-prophylaxis approach’ requires that we be particularly diligent in scrutinizing the law's fit." McCutcheon , 572 U.S. at 221, 134 S.Ct. 1434 (citation omitted). The individual-to-group limit here exhibits danger signs in its own right: like the individual-to-candidate limit, it is not adjusted for inflation, and it is lower than limits in other states. It appears that only two other states, Colorado and Massachusetts, impose comparably low individual-to-group limits. See Colo. Const. art. XXVIII, § 3 (5) (individual-to-group limit of $625 per two-year election period); Mass. Gen. Laws ch. 55 § 7A(a)(3) (individual-to-group limit of $500 per year).

The Supreme Court has observed that "there is not the same risk of quid pro quo corruption or its appearance when money flows through independent actors to a candidate, as when a donor contributes to a candidate directly." McCutcheon , 572 U.S. at 210–11, 134 S.Ct. 1434. Nonetheless, there is concern that the donor may channel money to the candidate through a series of contributions to groups that have stated their intention to support the candidate. Id . at 211–12, 134 S.Ct. 1434. However, that concern is already addressed by Alaska's also-low $1,000 group-to-candidate contribution limit. Alaska Stat. § 15.13.070(c). "But what if this donor does the same thing via, say, 100 different PACs?" as McCutcheon asks. 572 U.S. at 212, 134 S.Ct. 1434. We respond that we see no indication in the record that "the individual donor will necessarily have access to a sufficient number of PACs to effectuate such a scheme." See id . at 213, 134 S.Ct. 1434. It is more likely that the donor would opt to spend unlimited funds on independent expenditures on behalf of his or her favored candidate. See id . at 213–14, 134 S.Ct. 1434. Moreover, as the Supreme Court noted in McCutcheon , there are potential alternative solutions that are less likely to abridge constitutional speech, without "opin[ing] on the[ir] validity." See id . at 223, 134 S.Ct. 1434. For example, Alaska could pass a law similar to the federal law treating "all contributions made by a person, either directly or indirectly, on behalf of a particular candidate, including contributions which are in any way earmarked or otherwise directed through an intermediary or conduit to such candidate , ... as contributions from such person to such candidate." See 52 U.S.C. § 30116(a)(8) (emphasis added). In any event, we find that "because the statute is poorly tailored to the Government's interest in preventing circumvention of the base limits, it impermissibly restricts participation in the political process." See McCutcheon , 572 U.S. at 218, 134 S.Ct. 1434.

C

As we did in our prior opinion, we uphold the $5,000 limit on the amount a political party may contribute to a municipal candidate. Alaska Stat. §§ 15.13.070(d), 15.13.400(17). Plaintiffs do not challenge the dollar amount; they instead argue that the law's aggregation of political party sub-units is unconstitutional. They reason that limiting party sub-units to the $5,000 limit but not limiting multiple labor-union PACs to the same limit is discriminatory.

Plaintiffs’ discriminatory treatment argument fails because independent labor union PACs are not analogous to political party sub-units. Party sub-units, by definition, are subsidiaries of a parent entity—the umbrella political party. As such, they share the objectives and rules of the party. In the past, we have observed without remark that at least one other state similarly aggregates party sub-units for purposes of campaign contribution limits. See, e.g. , Lair v. Bullock , 798 F.3d 736, 741 (9th Cir. 2015) ("Montana treats all committees that are affiliated with a political party as one entity."). Different labor unions, by contrast, are different entities. Moreover, political parties may donate more than labor union PACs ($5,000 versus $1,000), which undercuts the basis for a direct comparison between the two disparate sets of organizations. Alaska Stat. § 15.13.070(c), (d). We therefore reject Plaintiffs’ inchoate disparate treatment argument and uphold the political party-to-candidate limit.⁴

4 Our holding should not be construed as foreclosing a constitutional challenge to the dollar amount of Alaska's (or some other state's) limit on political party-to-candidate contributions.

D

Finally, as we did in our prior opinion, we reverse on Alaska's nonresident aggregate limit, which bars a candidate from accepting more than $3,000 per year from individuals who are not residents of Alaska. Alaska Stat. § 15.13.072(a)(2), (e). This particular provision prevented Thompson from making a desired $100 contribution to a candidate for the Alaska House of Representatives—his brother-in-law—because his brother-in-law had already received $3,000 in out-of-state contributions.

The district court held that the nonresident aggregate limit serves an anti-corruption purpose. The court cited Alaska's unique vulnerability to "exploitation by outside industry and interests," and referenced trial testimony that those entities "can and do exert pressure on their employees to make contributions to state and municipal candidates." Thompson , 217 F. Supp. 3d at 1039. The court determined that the nonresident limit therefore

furthers Alaska's sufficiently important interest in preventing quid pro quo corruption or its appearance in two ways. First, [it] furthers the State's anticorruption interest directly by avoiding large amounts of out-of-state money from being contributed to a single candidate, thus reducing the appearance that the candidate feels obligated to outside interests over those of his constituents. Second, the nonresident aggregate limit discourages circumvention of the $500 base limit and other game-playing by outside interests, particularly given [the Alaska Public Offices Commission's] limited ability and jurisdiction to investigate and prosecute out-of-state violations of Alaska's campaign finance laws.

Id.

Taking the district court's evidentiary findings as true, on de novo review we cannot agree that the nonresident limit targets quid pro quo corruption or its appearance. At most, the law aims to curb perceived "undue influence" of out-of-state contributors—an interest that is no longer sufficient after Citizens United and McCutcheon . McCutcheon , 572 U.S. at 206–08, 134 S.Ct. 1434. Indeed, Alaska's argument that the nonresident limit "reduces the appearance that a candidate will be obligated to outside interests rather than constituents" says nothing about corruption .⁵ It is not enough to show that out-of-state firms—and particularly those wishing to exploit Alaska's natural resources—"can and do exert pressure on their employees to make contributions to state and municipal candidates." Thompson , 217 F. Supp. 3d at 1039.

5 In Landell v. Sorrell , the Second Circuit opined that the Alaska Supreme Court's upholding of the nonresident limit "is a sharp departure from the corruption analysis adopted by the Supreme Court in Buckley and Shrink. " 382 F.3d 91, 148 (2d Cir. 2004), rev'd on other grounds sub nom. Randall , 548 U.S. at 230, 126 S.Ct. 2479.

Moreover, even if we agreed with Alaska that limiting the inflow of contributions from out-of-state extractive industries served an anti-corruption interest, the nonresident aggregate limit is a poor fit. Out-of-state interests can still maximize their influence across a large number of candidates—they just need to be early players so that they can contribute the maximum $500 donation before each of those candidates reaches the $3,000 limit.

McCutcheon is instructive on this point. There, the Court invalidated aggregate contribution limits that allowed an individual to contribute the maximum to multiple candidates but not to any additional candidates once the contributor hit the aggregate limit. 572 U.S. at 210–18, 134 S.Ct. 1434. The Court held that the law was a poor fit for combating quid pro quo corruption or its appearance because contributions to a candidate before a contributor has reached the aggregate limit are not somehow less corrupting than contributions to another candidate after the aggregate limit is reached. See id .

Alaska's showing as to its nonresident limit is analogous. Alaska fails to show why an out-of-state individual's early contribution is not corrupting, whereas a later individual's contribution—i.e. , a contribution made after the candidate has already amassed $3,000 in out-of-state funds—is corrupting. Nor does Alaska show that an out-of-state contribution of $500 is inherently more corrupting than a like in-state contribution—only the former of which is curbed under Alaska's nonresident limit. Alaska fails to demonstrate that the risk of quid pro quo corruption turns on a donor's particular geography. Accordingly, while we do not foreclose the possibility that a state could limit out-of-state contributions in furtherance of an anti-corruption interest, Alaska's aggregate limit on what a candidate may receive is a poor fit.

As an alternative defense of the law, Alaska argues that the nonresident limit targets the important state interest of protecting its system of self-governance. We reject Alaska's proffered state interest for three reasons.

First, what Alaska calls "self-governance" is really a re-branding of the interest of combating influence and access that the Supreme Court has squarely rejected. To understand Alaska's proffered state interest, it is important to be clear on what the State does not mean by "self-governance." In the distinct context of a law restricting "who may exercise official, legislative powers," we recognized "self-government" as a legitimate state interest. Chula Vista Citizens for Jobs & Fair Competition v. Norris , 782 F.3d 520, 531 (9th Cir. 2015) (en banc). In Norris , we used the term "self-government" to mean a state's interest in controlling who governs.

Alaska's (and the dissent's) proffered state interest is materially different from what we called self-governance in Norris . Alaska's version of "self-governance" is concerned with limiting not who governs (as in Norris ) but who is allowed to contribute to the campaigns of those who would govern. Indeed, the dissent correctly characterizes Alaska's proffered interest as seeking "to ensure that its legislators are responsive to the individuals that they represent, not to out-of-state interests." Dissent at 47. The premise of Alaska's concern with "outside control" is that Alaska state officials will feel pressure to kowtow to out-of-state entities because of nonresident contributions.

The dissent makes a cogent case for the view that states should be able to limit who may "directly influence the outcome of an election" by making financial contributions. See Dissent at ––––. But that debate is over. The Supreme Court has expressly considered and rejected those arguments. See McCutcheon , 572 U.S. at 206–08, 134 S.Ct. 1434 (holding that states do not have a legitimate interest in curbing " ‘influence over or access to’ elected officials" by individuals "spend[ing] large sums" (quoting Citizens United , 558 U.S. at 359, 130 S.Ct. 876 )). In short, Alaska's proffered interest in "self-governance" is indistinguishable from the disavowed state interest in combating "influence over or access to" public officials.⁶

6 The Supreme Court has given no indication that the First Amendment interest in protecting political access waxes or wanes depending on the representative relationship between contributor and candidate. See Buckley v. Valeo , 424 U.S. 1, 48–49, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976). In fact, Buckley ’s language arguably compels the opposite conclusion:

[T]he concept that government may restrict the speech of some elements of our society in order to enhance the relative voice of others is wholly foreign to the First Amendment, which was designed to secure the widest possible dissemination of information from diverse and antagonistic sources, and to assure unfettered interchange of ideas for the bringing about of political and social changes desired by the people.

Id. (internal quotation marks omitted). Far from serving the goal of "secur[ing] the widest possible dissemination of information from diverse and antagonistic sources," the nonresident limit artificially suppresses the free exchange of political ideas.

Second, even if Alaska's "self-governance" interest could be construed as distinct from the interest in combating influence and access, the Supreme Court's recent campaign finance decisions leave no room for us to accept the State's proffered interest. The Supreme Court's opinions articulate "only one" narrowly defined legitimate state interest in capping campaign contributions: preventing quid pro quo corruption or its appearance. McCutcheon , 572 U.S. at 206–07, 134 S.Ct. 1434. In McCutcheon , its banner campaign contribution case, the Court explains that it has "consistently rejected attempts to suppress campaign speech based on other legislative objectives." Id. at 207, 134 S.Ct. 1434. McCutcheon resolved that "[a]ny regulation must instead target what we have called ‘quid pro quo ’ corruption or its appearance." Id. at 192, 134 S.Ct. 1434 (emphasis added) (citing Citizens United , 558 U.S. at 359, 130 S.Ct. 876 ). Indeed, "[c]ampaign finance restrictions that pursue other objectives ... impermissibly inject the Government ‘into the debate over who should govern.’ " Id. (quoting Ariz. Free Enter. Club's Freedom Club PAC v. Bennett , 564 U.S. 721, 750, 131 S.Ct. 2806, 180 L.Ed.2d 664 (2011) ); see also VanNatta v. Keisling , 151 F.3d 1215, 1217 (9th Cir. 1998) (noting "the lack of support for any claim based on the right to a republican form of government"). That unqualified directive leaves no room for Alaska's averred self-governance interest. Campaign contribution limits rise or fall on whether they target quid pro quo corruption or its appearance.

The dissent suggests we are free to accept "self-governance" as an important state interest in justifying limits on campaign contributions because the Supreme Court has not expressly considered and rejected that specific interest. Although a prior three-judge opinion of our court does not bind a later panel on an issue that was not before the prior panel, when it comes to Supreme Court precedent, our court is bound by more than just the express holding of a case. Our decisions must comport with the "reasoning or theory," not just the holding, of Supreme Court decisions (even in the face of prior contrary Ninth Circuit precedent). Miller v. Gammie , 335 F.3d 889, 893, 900 (9th Cir. 2003) (en banc) (adopting the view that lower courts are "bound not only by the holdings of higher courts’ decisions but also by their ‘mode of analysis’ " (quoting Antonin Scalia, The Rule of Law as a Law of Rules , 56 U. Chi. L. Rev. 1175, 1177 (1989) )); see id . at 900 ("[T]he issues decided by the higher court need not be identical in order to be controlling."). The dissent's conclusion that self-governance is an important state interest in this context is "clearly irreconcilable" with the Supreme Court's reasoning in McCutcheon . See id .

Third, even if McCutcheon did not shutter the possibility of alternative state interests, self-governance is not an important state interest in light of countervailing First Amendment concerns. Indeed, Alaska fails to prove that nonresident participation in a state's election infringes state sovereignty. Instead, it alleges in conclusory fashion that the "nonresident limit also furthers the important state interest in protecting Alaska's system of self-government from outside control."

Accordingly, we hold that Alaska's aggregate nonresident contribution limit violates the First Amendment, and we reverse the district court's judgment on this issue.⁷

7 The dissent relies on Bluman v. Fed. Election Comm'n , 800 F. Supp. 2d 281 (D.D.C. 2011), but that case is inapplicable. The plaintiffs in Bluman were foreign citizens who sought the right to participate in the United States campaign process by, among other things, making financial contributions to candidates. Id. at 282–83. They argued they should be treated the same as American citizens (such as minors and American corporations) who, though unable to vote, are permitted to make campaign contributions. Id. at 290. The court rejected that argument and based its holding on the conclusion that the plaintiffs, in contrast to American citizens who are unable to vote, were, by definition, outside "the American political community." Id. Thus, contrary to the dissent's statement that Bluman cannot "be distinguished on the grounds that it involved a distinction between United States citizens and foreign nationals," Dissent at ––––, that distinction was the very basis for the Bluman court's holding.

III

While the district court correctly held that the political party-to-candidate limit is constitutional, it erred under Supreme Court precedent in upholding the individual-to-candidate limit, the individual-to-group limit, and the nonresident aggregate limit. Accordingly, we reverse the district court on those three provisions and remand for entry of judgment consistent with this opinion.

AFFIRMED in part, REVERSED in part, and REMANDED .

The parties shall bear their own costs.

THOMAS, Chief Judge, concurring in part and dissenting in part:

I concur in Section II.C of the majority opinion because I agree that Alaska's $5,000 limit on a political party's contribution to a municipal candidate does not violate the First Amendment. However, I respectfully dissent from Sections II.A, II.B, and II.D.

Applying Randall v. Sorrell , 548 U.S. 230, 126 S.Ct. 2479, 165 L.Ed.2d 482 (2006), as is our task on remand, see Thompson v. Hebdon , ––– U.S. ––––, 140 S. Ct. 348, 350, 205 L.Ed.2d 245 (2019) (" Thompson III "), I would uphold Alaska's $500 limit on individual contributions to candidates and election-related groups. That limit, although not indexed for inflation, passes muster under Randall because Alaska permits political parties to donate significantly more than $500 to candidates; Alaska does not count volunteer services and at least some volunteer expenses toward the $500 limit; the record does not suggest that the $500 limit significantly restricts the amount of funding available for challengers to run competitive campaigns; and the record indicates that corruption (or its appearance) is significantly more serious a problem in Alaska than elsewhere. Moreover, I remain persuaded that the nonresident aggregate contribution limit, which furthers Alaska's important state interests in preventing quid pro quo corruption or its appearance and in preserving self-governance, does not violate the First Amendment either. See Thompson v. Hebdon , 909 F.3d 1027, 1044 (9th Cir. 2018) (" Thompson II ") (Thomas, J., concurring in part and dissenting in part). Accordingly, I would affirm the district court's decision in its entirety.

I

Assuming the $500 limit exhibits the "danger signs" that "warrant[ ] closer review" under Randall , Thompson III , 140 S. Ct. at 350, I would begin the tailoring analysis, as the majority does, with consideration of Randall ’s five factors. Applying those factors to the existing record, I would affirm the district court's decision upholding the $500 limit on individual-to-candidate and individual-to-group contributions.

A

Only one Randall factor favors Thompson: Alaska's limits are not adjusted for inflation. See Randall , 548 U.S. at 261, 126 S.Ct. 2479 ; Thompson II , 909 F.3d at 1037 & n.5 ; Thompson v. Dauphinais , 217 F. Supp. 3d 1023, 1027 (D. Alaska 2016) ("Thompson I "). But this factor is far from dispositive. See Nixon v. Shrink Mo. Gov't PAC , 528 U.S. 377, 397, 120 S.Ct. 897, 145 L.Ed.2d 886 (2000) ("[T]he [First Amendment] issue ... cannot be truncated to a narrow question about the power of the dollar, but must go to the power to mount a campaign with all the dollars likely to be forthcoming. ... [T]he dictates of the First Amendment are not mere functions of the Consumer Price Index."). And the remaining factors support the district court's judgment.

1

The majority acknowledges that two Randall factors favor Alaska: the $500 contribution limit does not apply to political parties, and volunteer services and expenses do not appear to constitute contributions under Alaska law. See Randall , 548 U.S. at 256–60, 126 S.Ct. 2479. I agree and will not repeat the majority's analysis. It bears mentioning, however, that Alaska law permits political parties to annually contribute up to $10,000 to candidates for the state house of representatives, $15,000 to state senate candidates, and $100,000 to candidates for governor or lieutenant governor. See Alaska Stat. § 15.13.070(d)(1)–(3). In stark contrast, under the Vermont campaign finance laws at issue in Randall , political parties, "taken together with all [their] local affiliates," could "make one contribution of at most $400 to [a] ... gubernatorial candidate, one contribution of at most $300 to a ... candidate for State Senate, and one contribution of at most $200 to a ... candidate for the State House of Representatives." 548 U.S. at 257, 126 S.Ct. 2479. Unlike Vermont, Alaska does not "reduce the voice of political parties" in Alaska "to a whisper" by "preventing a political party from using contributions by small donors to provide meaningful assistance to any individual candidate." Id. at 258–59, 126 S.Ct. 2479. Accordingly, there are no "special party-related harms" that "weigh[ ] against the constitutional validity of [Alaska's] contribution limits." Id. at 259, 126 S.Ct. 2479.

2

The remaining two Randall factors likewise weigh in favor of upholding the $500 limit. The majority's contrary conclusions do not pay adequate deference to the factual findings made by the district court after a seven-day bench trial and post-trial briefing, disregarding "one of the bread and butter principles of appellate review": "Trial courts find facts. We do not." Fisher v. Roe , 263 F.3d 906, 912 (9th Cir. 2001), overruled on other grounds by Payton v. Woodford , 346 F.3d 1204 (9th Cir. 2003).

a

First, the existing record does not "suggest[ ]" that Alaska's contribution limits "significantly restrict the amount of funding available for challengers to run competitive campaigns." Randall , 548 U.S. at 253, 126 S.Ct. 2479 (emphasis added). To the contrary, the district court found, and the panel majority previously agreed, based on an independent review of the evidence, see Thompson II , 909 F. 3d at 1035, that "candidates, whether challengers or incumbents , can run effective campaigns under the current limits and, to use [one of the expert's] words, ‘have done so.’ " Thompson I , 217 F. Supp. 3d at 1035 (emphasis added).

There is ample record support for the district court's finding that challengers can and have raised sufficient funds to run competitive campaigns under the $500 limit. For example, one expert testified that "challengers, on average, out-fundraised incumbents in Alaska's 2008 and 2010 state senate races." In the 2012 and 2014 election cycles, a few non-incumbent candidates raised over "$100,000 from individual contributions alone," setting aside contributions from political action committees ("PACs") and political parties—an amount that, according to the trial evidence, "allow[ed] a candidate to mount an effective campaign." Thompson II , 909 F.3d at 1038–39. Next, in the 2016 primary elections, "Alaska voters dispatched seven incumbents from the Alaska Legislature."¹ Thompson I , 217 F. Supp. 3d at 1036 & n.44. Moreover, a defense expert testified regarding the absence of "evidence that increasing the limits [from $500 to $1,000] [would] give[ ] challengers a better standing." Thompson did not present any compelling rebuttal evidence. Indeed, one of Thompson's witnesses, former state senator John Coghill, testified that the $500 limit required "more effort" and "broad[er]" outreach but had never prevented him from running an effective campaign. See Thompson I , 217 F. Supp. 3d at 1035. Although Thompson called Clark Bensen, the same expert whose testimony the Supreme Court cited in striking down Vermont's individual contribution limits in Randall , 548 U.S. at 253–54, 126 S.Ct. 2479, the district court reasonably declined to credit Bensen's testimony that candidates in competitive campaigns for Alaska offices would raise more money if the individual contribution limit was higher because Bensen admitted that his study "was based on exaggerated estimates and therefore flawed." Thompson I , 217 F. Supp. 3d at 1035. Additionally, Bensen testified that "his inflated estimates of lost revenue"—6 to 11% for the state house and 8 to 16% for the state senate—were "[p]robably almost twice as high as they should be" and, even so, nowhere "close to the percentages [of lost revenue] that troubled the Supreme Court in Randall ." See Randall , 548 U.S. at 253, 126 S.Ct. 2479 (crediting Bensen's finding that Vermont's challenged contribution limits "would have reduced the funds available in 1998 to Republican challengers in competitive races in amounts ranging from 18% to 53% of their total campaign income" (emphasis added)).

1 This evidence is more compelling and substantial than the "anecdotal evidence" in the Randall record reflecting one "competitive mayoral campaign" against an incumbent. Randall , 548 U.S. at 256, 126 S.Ct. 2479 ; cf. also McNeilly v. Land , 684 F.3d 611, 618 (6th Cir. 2012) (declining to find that Michigan's per-election-cycle limits—$500 for a candidate for state representative and $1,000 for a candidate for state senator—"significantly restrict[ed] the funding for challengers to run competitive campaigns" where the record did not contain "[t]he same quality or quantity of evidence" on that issue as in Randall ).

Bob Bell's testimony regarding his ultimately unsuccessful campaign to unseat an incumbent state senator in 2012 also does not "raise questions as to whether the $500 contribution limit establishes too low a ceiling to allow challengers to launch campaigns" and "run against incumbents competitively." Even though he did not win the election, Bell raised enough money to run a competitive campaign; he lost by a mere 59 votes out of 15,200—an election so close that Bell referred to the results as "a tie." Cf. id. at 255–56, 126 S.Ct. 2479 (citing study that defines competitive elections to include those that an "incumbent wins with less than 60% of vote"). Notably, Bell's 2012 senate campaign conducted the "most successful fundraiser in Alaska history."

In sum, because the record lacks evidence that the challenged limits "threaten[ ] to inhibit effective advocacy by those who seek election, particularly challengers," id. at 261, 126 S.Ct. 2479, this factor favors Alaska.

b

Second, Alaska has provided a "special justification" warranting the $500 limit because the existing record "indicat[es] that ... corruption (or its appearance)" in Alaska "is significantly more serious a matter than elsewhere." Id .

The district court found that Alaska is "highly, if not uniquely, vulnerable to corruption in politics and government." Thompson I , 217 F. Supp. 3d at 1029 ; see also Thompson III , 140 S. Ct. at 351–52 (Statement of Ginsburg, J.) (" ‘[S]pecial justification’ of this order may warrant Alaska's low individual contribution limit."). In support of this finding, the district court cited expert testimony regarding two factors that make the risk of corruption in Alaska particularly acute. Thompson I , 217 F. Supp. 3d at 1029. First, with "the second smallest legislature in the United States and the smallest senate," it takes only "ten votes [to] stop a legislative action such as an oil or gas tax increase from becoming law." Id . "Consequently, the incentive to buy a vote, and the chances of successfully doing so, are therefore higher in Alaska than in states with larger legislative bodies." Id. Second, Alaska relies on the oil and gas industry "for a majority of its revenues." Id. While 85 to 92% of Alaska's budget derives from the oil and gas industry, that industry is not responsible for more than 50% of any other state's budget. Id.

The district court's finding further rested on a "widely publicized" public corruption scandal in which 10% of Alaska's legislators exchanged political favors and votes for money from VECO, an oilfield services firm. Id. at 1030. Some of these legislators referred to themselves as the "Corrupt Bastards Club." Id. at 1030 n.18. News outlets played an FBI surveillance video showing one legislator, Representative Vic Kohring, accepting cash from VECO in exchange for his vote on pending oil tax legislation. Id. at 1030. After being criminally charged, Kohring went on to pen a newspaper column claiming that the only thing separating him from other Alaska lawmakers was that he got caught. Id. The publicity surrounding the VECO scandal bolsters the inference that corruption (or its appearance) is a more serious problem in Alaska than elsewhere.

Thompson disputes this finding. He argues that Alaska's small legislature and dependence on a single industry are the very factors that did not amount to a special justification in Randall . This analogy fails. Vermont's legislature is three times the size of Alaska's, and Thompson cites no evidence from the Randall record indicating that Vermont has historically relied almost exclusively on a single industry for its revenue or that it has experienced a public corruption scandal comparable to VECO. Cf. Br. of Resp'ts, Cross-Pet'rs Vermont Pub. Int. Rsch. Grp., Randall v. Sorrell , 548 U.S. 230, 126 S.Ct. 2479, 165 L.Ed.2d 482 (2006), 2006 WL 325190. Though the briefing in Randall noted growing concerns that Vermont's slate and bottle industries could "influence" the state legislature through "fundraising pressures," see id. , at *12–*13, the Supreme Court was unable to find "in the record" any evidence that corruption or its appearance uniquely threatened the integrity of Vermont's political system. Randall , 548 U.S. at 261, 126 S.Ct. 2479 (emphasis added). By contrast, the district court here upheld the challenged limits based, in part, on a well-supported finding that Alaska is "highly, if not uniquely vulnerable, to corruption in politics and government." Thompson I , 217 F. Supp. 3d at 1029. Thompson fails to cite any record evidence that undermines this finding.

3

Thompson contends that "the Supreme Court's per curiam opinion confirms that Alaska's limits cannot stand." Not so. Rather than dictate a particular outcome under Randall , the Supreme Court's order clarifies that Randall ’s plurality decision is controlling and narrows the panel's task on remand to consideration of the five Randall factors. See, e.g. , Thompson III , 140 S. Ct. at 351 (observing that "[t]he parties dispute whether there are pertinent special justifications" for Alaska's limits without opining who has the better of that dispute).² In sum, I would uphold Alaska's $500 individual-to-candidate contribution limit under the Randall factors "taken together." 548 U.S. at 261, 126 S.Ct. 2479.

2 Notably, none of the ten Circuit decisions cited in the Supreme Court's remand order—for the fact that they "correctly looked to Randall in reviewing campaign finance restrictions"—invalidated contribution limits under Randall . Thompson III , 140 S. Ct. at 350 n.*. See Nat'l Org. for Marriage v. McKee , 649 F.3d 34, 60–61 (1st Cir. 2011) (declining to invalidate, under Randall , Maine's campaign expenditure reporting threshold); Ognibene v. Parkes , 671 F.3d 174, 192 (2d Cir. 2011) (distinguishing Randall and upholding New York City's restrictions on contributions from individuals who have business dealings with the City (e.g., lobbyists) to $400 in city-wide elections, $320 for Borough offices, and $250 for City Council); Preston v. Leake , 660 F.3d 726, 739–40 (4th Cir. 2011) (upholding North Carolina's ban on contributions by registered lobbyists); Zimmerman v. City of Austin , 881 F.3d 378, 387–88 (5th Cir.) (declining to find any "danger signs" in connection with Austin's $350 limit on per-election contributions), cert. denied , ––– U.S. ––––, 139 S. Ct. 639, 202 L.Ed.2d 492 (2018) ; McNeilly , 684 F.3d at 617–20 (distinguishing Randall and affirming the district court's determination that a challenge to Michigan's caps on per-election contributions to candidates for state senate ($1,000) and state representative ($500) was not likely to succeed on the merits); Ill. Liberty PAC v. Madigan , 904 F.3d 463, 469–70 (7th Cir. 2018) (upholding, under Randall , $5,000 limit on individual-to-candidate contributions); Minn. Citizens Concerned for Life, Inc. v. Swanson , 640 F.3d 304, 319 n.9 (8th Cir. 2011) (declining to invalidate, under Randall , Minnesota's ban on direct corporate contributions to candidates), rev'd in part on other grounds , 692 F.3d 864 (8th Cir. 2012) (en banc); Indep. Inst. v. Williams , 812 F.3d 787, 790–91 (10th Cir. 2016) (merely citing Randall for standard of review without applying it because contribution limits were not at issue); Ala. Democratic Conf. v. Att'y Gen. of Ala. , 838 F.3d 1057, 1069–70 (11th Cir. 2016) (upholding PAC-to-PAC transfer ban under Randall ); Holmes v. Fed. Election Comm'n , 875 F.3d 1153, 1165 (D.C. Cir. 2017) (en banc) (upholding, under Randall , $2,600 per-election limit on individual contributions to candidates for federal office).

B

I would further uphold the $500 individual-to-group limit. Because the $500 individual-to-candidate limit is sufficiently tailored to advance Alaska's anticorruption interest, the $500 individual-to-group limit, a measure that prevents easy circumvention of the individual-to-candidate limit, withstands First Amendment scrutiny, too. The panel majority's vacated opinion adopted this very anticircumvention rationale, see Thompson II , 909 F.3d at 1039–40, and the Supreme Court's remand order does not take issue with that aspect of the vacated opinion. See Thompson III , 140 S. Ct. at 349–51.

1

In California Medical Association v. Federal Election Commission , 453 U.S. 182, 101 S.Ct. 2712, 69 L.Ed.2d 567 (1981) (" CalMed "), the Supreme Court upheld a $5,000 limit on individual contributions to PACs, concluding that these limits furthered the government's anticorruption interest by preventing individual contributors from circumventing the "limit on contributions to candidates ... by channeling funds" to candidates through such a committee. Id. at 197–98, 101 S.Ct. 2712 (plurality); see also id. at 202–03, 101 S.Ct. 2712 (Blackmun, J., concurring in part) (concluding that the limit on individual contributions to PACs was "narrowly drawn" as a "means of preventing evasion" of the $1,000 limit on individual-to-candidate contributions).

Since CalMed , the Court has continued to recognize circumvention of a constitutional individual-to-candidate contribution limit as a "valid theory of corruption" and preventing such circumvention ("anticircumvention") as an important state interest. Fed. Election Comm'n v. Colo. Republican Fed. Campaign Comm'n , 533 U.S. 431, 456, 121 S.Ct. 2351, 150 L.Ed.2d 461 (2001). In McCutcheon v. Federal Election Commission , 572 U.S. 185, 134 S.Ct. 1434, 188 L.Ed.2d 468 (2014), the Court held that the federal government had failed to carry its burden of demonstrating that its aggregate contribution limits, restricting how much a donor could contribute to all candidates or committees, "further[ed] its anticircumvention interest." Id. at 192–93, 211, 134 S.Ct. 1434 ; see also id. at 219, 134 S.Ct. 1434 (finding the "risk of circumvention" low). Nevertheless, the Supreme Court, as the panel majority previously recognized, did not "call into doubt anticircumvention as an important state interest." Thompson II , 909 F.3d at 1039 ; see also McCutcheon , 572 U.S. at 193, 134 S.Ct. 1434 (clarifying that the case did "not involve any challenge to the base limits," including the $5,000 limit on individual-to-PAC contributions, which the Court had "previously upheld as serving the permissible objective of combatting corruption"), 200–01 (citing CalMed without expressing any disapproval).

2

Under Alaska law, as few as two individuals may form a "group." Alaska Stat. § 15.13.400(9)(B). The ease with which a couple or business partners may form a group creates a significant circumvention risk. Cf. McCutcheon , 572 U.S. at 211, 219, 134 S.Ct. 1434 (finding the "risk of circumvention" low and the government's circumvention hypotheticals "implausible"). As Thompson observes, Alaska limits group-to-candidate contributions to $1,000. See Alaska Stat. § 15.13.070(c)(1). But contrary to his assertion, that limit does not address Alaska's circumvention concern. The $1,000 limit is still double the individual-to-candidate limit; thus, absent the individual-to-group limit, an individual could use a group to triple his or her contribution to a candidate—by contributing $500 directly to the candidate and $1,000 through the group. It is also worth noting that Alaska's individual-to-group limit does not include contributions made "to influence the outcome of a ballot proposition." Id. § 15.13.065(c); see also id. § 15.13.070(b)(1). Accordingly, the individual-to-group limit is tailored to target the "type of expression that implicates quid pro quo corruption concerns"—"spending money that can be directed to candidates ."

In sum, because the limit on individual-to-group contributions prevents contributors from giving "three times" the individual-to-candidate limit "by using a group as a simple pass-through device," it serves the state's legitimate anticircumvention interest and is constitutional.

II

Finally, I respectfully continue to disagree with the majority that Alaska's nonresident aggregate limit violates the First Amendment for reasons that are not affected by the Supreme Court's vacatur and remand. See Thompson II , 909 F.3d at 1044–49 (Thomas, C.J., concurring in part and dissenting in part).

A

To survive First Amendment scrutiny in this case, Alaska must establish that the nonresident aggregate contribution limit is justified by the risk of quid pro quo corruption or its appearance. And its burden is light.³ Alaska need only show that the risk of actual or perceived quid pro quo corruption by out-of-state actors is neither "illusory" nor "mere conjecture." Mont. Right to Life Ass'n v. Eddleman , 343 F.3d 1085, 1092 (9th Cir. 2003) (quoting Buckley v. Valeo , 424 U.S. 1, 27, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976) ; Shrink Mo. , 528 U.S. at 392, 120 S.Ct. 897 ); see also Lair v. Motl , 873 F.3d 1170, 1178 (9th Cir. 2017) (" Lair III ") (reaffirming burden of proof). After a seven-day bench trial, the district court concluded that Alaska had satisfied its burden. Its factual findings were not clearly erroneous, see Prete v. Bradbury , 438 F.3d 949, 960 (9th Cir. 2006) (describing standard), and its conclusions were amply supported by the record. Alaska demonstrated that nonresident contributions present a particular risk of quid pro quo corruption or its appearance.⁴ Alaska is uniquely vulnerable to exploitation by out-of-state actors for the reasons previously discussed: its "almost complete reliance" on the oil and gas industry for "a majority of its revenues" and the size of its legislature. Thompson I , 217 F. Supp. 3d at 1029. The economic benefits of natural resource extraction do not come without a cost. The interests of out-of-state oil companies are often at odds with the interests of some Alaska residents. "About 17 percent of Alaskans—or 120,000 people—live in rural areas, where 95 percent of households use fish and 86 percent use game for subsistence purposes[.]" Azmat Khan, Living off the Land in Rural Alaska , PBS, https://tinyurl.com/alaska-rural-econ (last visited July 22, 2021). Resource extraction has the potential to cause irremediable damage to Alaskan lands and culture: "any change that depletes wild resources, reduces access to wild areas and resources, or increases competition between user groups can create problems for subsistence[,]" which is "among the most highly valued parts of [Alaska] culture" and "essential ... to rural economies." Alaska Dep't of Fish & Game, Subsistence in Alaska: FAQs , https://www.adfg.alaska.gov/index.cfm?adfg=subsistence.faqs#QA13 (last visited July 22, 2021).

3 Because Thompson raised no challenge to the amount of the aggregate limit, the only question is whether "there is adequate evidence that the limitation furthers" Alaska's anti-corruption interest. Lair v. Bullock , 798 F.3d 736, 742 (9th Cir. 2015) ("Lair II ") (quoting Mont. Right to Life Ass'n v. Eddleman , 343 F.3d 1085, 1092 (9th Cir. 2003) ).

4 The Supreme Court has specifically rejected Thompson's argument that a ban is treated differently than a limit when it comes to connecting the regulation to the state's important interest. Fed. Election Comm'n v. Beaumont , 539 U.S. 146, 162, 123 S.Ct. 2200, 156 L.Ed.2d 179 (2003) ("It is not that the difference between a ban and a limit is to be ignored; it is just that the time to consider it is when applying scrutiny at the level selected[.]"). And there is no question that Alaska may limit campaign contributions to prevent quid pro quo corruption or its appearance. Shrink Mo. , 528 U.S. at 390, 120 S.Ct. 897. Thus, the issue here is essentially whether the state may draw a line between residents and non-residents.

Given the oil and gas industry's outsized impact on Alaska's economy, it is not difficult to see why, as the district court found, Alaska is dependent upon and therefore particularly vulnerable to corruption by out-of-state corporations, whose interests are likely to be indifferent to those of Alaska's residents. The district court was persuaded by trial testimony that "the unique combination of Alaska's small population, geographic isolation, and great natural resources make it extremely dependent on outside industry and interests." Thompson II , 909 F.3d at 1039. Alaska cannot afford to extract its natural resources without out-of-state corporations. Id . And because out-of-state corporations cannot extract without the cooperation of government, these corporations do all they can to influence state politics. Id.

As the Supreme Court has recognized, "the dangers of large, corrupt contributions and the suspicion that large contributions are corrupt are neither novel nor implausible." Shrink Mo. , 528 U.S. at 391, 120 S.Ct. 897. Thus, it is enough to demonstrate that out-of-state contributors are particularly interested in corrupting the political process in Alaska, as the State has easily done.

But the proof at trial was more than theoretical. The district court found that "natural resource extraction firms can and do exert pressure on their employees" to contribute to political campaigns in Alaska. Thompson I , 217 F. Supp. 3d at 1039. In other words, these out-of-state interests have found a way to circumvent the generally applicable contribution limits. And, as the district court determined, the publicity surrounding the VECO scandal supports Alaska's interest in limiting the appearance of quid pro quo corruption by out-of-state interests in order to preserve Alaskans’ belief in the integrity of their political system. Id . at 1031.

In sum, I would hold that Alaska's important anticorruption interest justifies a limit on nonresident speech. Nonresident contributions present a special risk of quid pro quo corruption that is neither "illusory" nor "mere conjecture." Lair III , 873 F.3d at 1178 (quoting Eddleman , 343 F.3d at 1092 ). Particularly in the aftermath of the VECO scandal, the nonresident aggregate contribution limit furthers Alaska's interest in preventing the appearance of corruption, thereby increasing "[c]onfidence in the integrity of [Alaska's] electoral processes," a value "essential to the functioning of our participatory democracy." Purcell v. Gonzalez , 549 U.S. 1, 4, 127 S.Ct. 5, 166 L.Ed.2d 1 (2006) (per curiam). The district court was entirely correct, and the record supports its conclusion.

B

The nonresident aggregate cap is also justified by a second important state interest: self-governance. I would hold that self-governance is a sufficiently important interest to justify the nonresident aggregate cap.

1

"[T]he right to govern is reserved to citizens." Foley v. Connelie , 435 U.S. 291, 297, 98 S.Ct. 1067, 55 L.Ed.2d 287 (1978). There is no question that Alaska may bar nonresidents from voting, no matter how tangible their interest in a state election, Holt Civic Club v. City of Tuscaloosa , 439 U.S. 60, 68–69, 99 S.Ct. 383, 58 L.Ed.2d 292 (1978), even though "[n]o right is more precious" than the right to vote, Wesberry v. Sanders , 376 U.S. 1, 17, 84 S.Ct. 526, 11 L.Ed.2d 481 (1964). Because of the need for responsiveness to local interests, states may also closely guard from nonresident interference those "functions that go to the heart of representative government," such as "state elective or important nonelective executive, legislative, and judicial positions[.]" Sugarman v. Dougall , 413 U.S. 634, 647, 93 S.Ct. 2842, 37 L.Ed.2d 853 (1973).

States should be able to prevent out-of-state interests from advancing candidates for whom the contributor cannot even vote. Campaign contributions are made primarily to directly influence the outcome of an election rather than to broadcast one's one political opinion. Beaumont , 539 U.S. at 161, 123 S.Ct. 2200 ("[C]ontributions lie closer to the edges than to the core of political expression."). Thus, they are "subject to relatively complaisant review." Id.

The nonresident aggregate limit furthers Alaska's important state interest in protecting state sovereignty in governance. It is "the choice, and right, of the people to be governed by their citizen peers." Foley , 435 U.S. at 296, 98 S.Ct. 1067. When out-of-state interests fund political campaigns, they place an obstacle between the people and their representatives. Alaska must be able to take measures to ensure that its legislators are responsive to the individuals that they represent, not to out-of-state interests.

Alaska's interest in protecting self-government is "important," as required under Eddleman ’s first prong. Lair II , 798 F.3d at 742 (quoting Eddleman , 343 F.3d at 1092 ). Indeed, on en banc review, we held that a state's interest in "securing the people's right to self-government" was "compelling" in the face of a First Amendment challenge to a law requiring municipal initiative proponents to be bonafide electors. Chula Vista Citizens for Jobs & Fair Competition v. Norris , 782 F.3d 520, 531 (9th Cir. 2015) (en banc). The Supreme Court reached a similar conclusion regarding residence requirements under an Equal Protection analysis. Dunn v. Blumstein , 405 U.S. 330, 343–44, 92 S.Ct. 995, 31 L.Ed.2d 274 (1972) (recognizing as "substantial" the government's interest in "preserv[ing] the basic conception of a political community").

2

Bluman v. Federal Election Commission , 800 F. Supp. 2d 281 (D.D.C. 2011), summarily aff'd , 565 U.S. 1104, 132 S. Ct. 1087, 181 L.Ed.2d 726 (2012) (mem.), decided by a three-judge panel of the D.C. District Court, is analogous. There, the court considered a federal law preventing foreign nationals from making not only contributions but also independent expenditures to influence federal elections. Id. at 282–83. Because spending money to influence an election is not only "speech" but also "participation in democratic self-government," foreign nationals may be subject to restrictions targeted at protecting sovereignty. Id. at 289.

In Bluman , the court recognized that "[p]olitical contributions and express-advocacy expenditures are an integral aspect of the process by which Americans elect officials to federal, state, and local government offices." Id. at 288. "[I]t is undisputed that the government may bar foreign citizens from voting and serving as elected officers"; "[i]t follows that the government may bar foreign citizens ... from participating in the campaign process that seeks to influence how voters will cast their ballots in the elections." Id.

Alaska presents an even stronger case than did the federal government in Bluman . There, the challenged law restricted individual expenditures as well as campaign contributions, and the court therefore applied strict scrutiny. Id. at 285–86 (citing McConnell v. Fed. Election Comm'n , 540 U.S. 93, 134–37, 124 S.Ct. 619, 157 L.Ed.2d 491 (2003) and Buckley , 424 U.S. at 20–23, 96 S.Ct. 612 ). Here, on the other hand, we need not identify a compelling government interest but only a "sufficiently important" one. Lair II , 798 F.3d at 742 (quoting Eddleman , 343 F.3d at 1092 ).

Nor can Bluman be distinguished on the ground that it involved a distinction between United States citizens and foreign nationals. "It has long been recognized that resident aliens enjoy the protections of the First Amendment." Price v. INS. , 962 F.2d 836, 841 (9th Cir. 1991) (internal citations omitted). The line drawn in Bluman separates citizens with the right to participate in government from foreign nationals subject to federal law but with no corollary right of participation. Alaska draws its line even more carefully by applying the aggregate contribution limit only to nonresidents.⁵

5 This, too, is why VanNatta v. Keisling , 151 F.3d 1215 (9th Cir. 1998), is immediately distinguishable, even if it remains good law and speaks to this precise issue, both of which propositions are questionable. VanNatta is distinguishable because it limited out-of-district contributions to candidates for state office. Id. at 1217. Further, as we noted in Eddleman , reliance on the Court's approach in VanNatta "fails to recognize the impact of the Supreme Court's ... decision in Shrink Missouri ." 343 F.3d at 1091 n.2. And the majority opinion in VanNatta is framed as a rejection of the state's evidence and legal argument rather than as setting forth a hard-and-fast rule regarding the constitutionality of all limits on out-of-district contributions. 151 F.3d at 1217–18.

3

I respectfully disagree that the Supreme Court has foreclosed this issue because it rejected other purported interests. Foundational to the judicial role is a recognition that "[w]ithout jurisdiction the court cannot proceed at all in any cause." Steel Co. v. Citizens for Better Env't , 523 U.S. 83, 94, 118 S.Ct. 1003, 140 L.Ed.2d 210 (1998) (quoting Ex parte McCardle , 74 U.S. (7 Wall.) 506, 514, 19 L.Ed. 264 (1868) ). Jurisdiction extends only to "Cases" and "Controversies." U.S. Const. art. III, § 2. It emphatically does not extend to issues that are not before a court. No court can reject a self-governance theory unless it is asked to do so. The Supreme Court has yet to take up this question; in resolving this controversy, it is not our role to apply a holding that does not exist.

4

"The Constitution limited but did not abolish the sovereign powers of the States, which retained ‘a residuary and inviolable sovereignty.’ " Murphy v. Nat'l Collegiate Athletic Ass'n , ––– U.S. ––––, 138 S. Ct. 1461, 1475, 200 L.Ed.2d 854 (2018) (quoting The Federalist No. 39, at 245 (James Madison) (Clinton Rossiter ed., 1961)). This basic principle arises from "a fundamental structural decision incorporated into the Constitution." Id.

Our federalist system is not binary; it does not simply pit the states—as a single entity—against federal power. Rather, it recognizes the sovereignty of each individual state. In the words of Justice Marshall, "[n]o political dreamer was ever wild enough to think of breaking down the lines which separate the States, and of compounding the American people into one common mass." McCulloch v. Maryland , 17 U.S. (4 Wheat.) 316, 403, 4 L.Ed. 579 (1819). Under our Constitution, "the people of each state compose a State, having its own government, and endowed with all the functions essential to separate and independent existence." Lane Cnty. v. Oregon , 74 U.S. (7 Wall.) 71, 76, 19 L.Ed. 101 (1868). "Not only, therefore, can there be no loss of separate and independent autonomy to the States, through their union under the Constitution, but ... the preservation of the States, and the maintenance of their governments, are as much within the design and care of the Constitution as the preservation of the Union and the maintenance of the National government." Texas v. White , 74 U.S. (7 Wall.) 700, 725, 19 L.Ed. 227 (1868).

In the current, highly partisan political climate, regional differences may be obscured by contentious national issues. Jessica Bulman-Pozen, Executive Federalism Comes to America , 102 VA. L. REV. 953, 962–63 (2016). However, "[e]ven at the level of national politics, ... there always remains a meaningful distinction between someone who is a citizen of the United States and of Georgia and someone who is a citizen of the United States and of Massachusetts." U.S. Term Limits, Inc. v. Thornton , 514 U.S. 779, 859, 115 S.Ct. 1842, 131 L.Ed.2d 881 (1995) (Thomas, J., dissenting).

Here, of course, we are not dealing with politics at a national level, but only with Alaska's ability to take measures to "represent and remain accountable to its own citizens." Printz v. United States , 521 U.S. 898, 920, 117 S.Ct. 2365, 138 L.Ed.2d 914 (1997) (internal citations omitted). State governments can and should be "more sensitive to the diverse needs" of their populations. Gregory v. Ashcroft , 501 U.S. 452, 458, 111 S.Ct. 2395, 115 L.Ed.2d 410 (1991). Alaska must have the right to prevent non-resident interests from taking hold of their elections. See Anthony Johnstone, Outside Influence , 13 ELECTION L. J. 117, 122–23(2014) ("No form of federalism, and therefore no form of government under the Constitution, works without limits on outside influence in the states."). Therefore, I disagree that Alaska's self-governance interest is not "sufficiently important" for purposes of limiting campaign contributions. Lair II , 798 F.3d at 742 (quoting Eddleman , 343 F.3d at 1092 ). III

For these reasons, I respectfully dissent, in part. I agree that Alaska's limit on political party contributions to individual candidates does not violate the First Amendment. However, I also would hold that Alaska's limitations on individual contributions to candidates and election-related groups as well as its nonresident aggregate contribution limit are constitutional. Thus, I would affirm the judgment of the district court in its entirety.