OPINION

Montanans select their judges through nonpartisan popular elections. In an effort to keep those elections nonpartisan, Montana has restricted judicial-campaign speech. One of those restrictions is before us—a rule that prohibits candidates from seeking, accepting, or using political endorsements in their campaigns. Mark French, a judicial candidate who wishes to seek and use such endorsements, claims that Montana’s rule violates his First Amendment rights," Montana argues that the rule is narrowly tailored to ensuring the impartiality and independence of Montana’s judiciary. The district court upheld the statute, and we agree. In light of the Supreme Court’s decision in Williams-Yulee v. Florida Bar, — U.S. -, 135 S.Ct. 1656 , 191 L.Ed.2d 570 (2015), and our decision in Wolfson v. Concannon, 811 F.3d 1176 (9th Cir. 2016) (en banc), we affirm the judgment.

Montana has declared that “[a]n independent, fair, and impartial judiciary is indispensable to [its] system, of justice.” Mont, Code of Judicial Conduct, Preamble (2009). Although that statement of principle must be universally acknowledged, American jurisdictions have chosen different means to secure it. See The Federalist No. 78, at 465 (C. Rossiter ed. 1961) (A. Hamilton) (arguing for the appointment of judges). Since 1935, Montana has decided to select its .judges through nonpartisan popular elections. See Mont. Code Ann. § 13-14-111 . Recognizing that mixing politics with judging could lead to injustice, Montana has prohibited all judges and candidates for judicial office from “engagpng] in political or campaign activity that is inconsistent with the independence, integrity, or impartiality of the judiciary.” Mont. Code of Judicial Conduct Canon 4. That broad prohibition applies to such activities as holding an office in 'or making speeches on behalf of a political organization, publicly endorsing political candidates, publicly identifying oneself as a political candidate, and otherwise using the names of political parties in judicial campaigns. Id. Rule 4.1; Mont. Code Ann. § 13-10-602 (2).

Only one restriction is at issue here. Rule 4.1(A)(7) of the Montana Code of Judicial Conduct provides: “[A] judge or judicial candidate shall, not ... seek, accept, or use endorsements from a political organization, or partisan or independent non-judicial office-holder or candidate ....” Mont. Code of Judicial Conduct Rule 4.1(A)(7). The Code defines a “political .organization” as “a political party or other group sponsored by or affiliated with a political party or candidate, the principal purpose of which is to further the election or appointment of. candidates, for political office.” Id. Terminology. If a judge or judicial candidate violates this endorsement provision, the Montana Judicial Standards Commission “shall recommend ... the censure, suspension, removal, or disability retirement of the judicial officer.”. Mont. Code Ann. § 3-1-1106 (3),

In 2014, Mark French ran as a candidate for justice of the peace in Sanders County. The Sanders County Republican Central Committee endorsed French’s candidacy, and two prominent Republican officeholders were willing to consider doing so if French had asked. Afraid of violating Rule 4.1(A)(7), French refrained from seeking or using these endorsements in his campaign. He ultimately lost the election, but intends to run again in' 2018. Although French would like to seek and use political endorsements during the next election cycle, he understands that he cannot do so as long as Rule 4.1(A)(7) remains in place.

French filed this action claiming that Rule 4.1(A)(7) violates his First Amendment rights and asking that the court enjoin its enforcement. The district court rejected French’s argument and' entered summary judgment against him. We review that decision de novo. See McIndoe v. Huntington Ingalls Inc., 817 F.3d 1170, 1173 (9th Cir. 2016).

II

The First Amendment, applicable to the states through the Fourteenth Amendment, prohibits the- government from “abridgirig the freedom of speech.” U.S. Const. amend. I; Stromberg v. California, 283 U.S. 359, 368 , 51 S.Ct. 532 , 75 L.Ed. 1117 (1931) (incorporating “the right of free speech” into the Due Process Clause of the Fourteenth Amendment). Content-based restrictions on judicial-campaign speech are subject to strict scrutiny under the First Amendment. See Republican Party of Minn. v. White (“White I”), 536 U.S. 765, 774 , 122 S.Ct. 2528 , 153 L.Ed.2d 694 (2002) (assuming strict scrutiny applies); Wolfson, 811 F.3d at 1180 (holding that strict scrutiny applies). To survive strict scrutiny, the government must show that “the restriction ‘furthers a compelling interest and is. narrowly tailored to achieve that interest.’ ” Citizens United v. Fed. Election Comm’n, 558 U.S. 310, 340 , 130 S.Ct. 876 , 175 L.Ed.2d 753 (2010) (citation omitted). “ ‘[I]t is the rare case’ in which a State demonstrates that a speech restriction is narrowly tailored to serve a compelling interest.... But those cases do arise.” Williams-Yulee, 135 S.Ct. at 1665-66 (citations omitted); see also Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 237 , 115 S.Ct. 2097 , 132 L.Ed.2d 158 (1995) (“[W]e wish to dispel the notion that strict scrutiny is ‘strict in theory, but fatal in fact.’ ” (citation omitted)).

Before determining whether Rule 4.1(A)(7) is narrowly tailored to achieve a compelling state interest, we must examine the development of First Amendment law in this murky area of judicial-campaign speech. The Supreme Court has addressed restrictions similar to Rule 4.1(A)(7) on two occasions—in White I and Williams-Yulee—providing mixed guidance on the proper analytical framework and producing some tension among the lower courts. We have addressed that tension in our en banc decision in Wolfson. Despite the confusion, we discern a clear shift in favor of state regulation—a shift that' renders many of French’s and his amicus curiae’s arguments no longer persuasive.

A

1

We begin our survey with White I. The Supreme Court there reviewed a Minnesota restriction on judicial elections that prohibited a candidate from “announcing] his or her views on disputed legal or political issues,” a prohibition that at the very least precluded “a judicial candidate from stating his views on any specific nonfaneiful legal question within the province of the court for which he [was] running.” White I, 536 U.S. at 768, 773 , 122 S.Ct. 2528 (citation omitted). The majority began by identifying a potential compelling interest Minnesota might have had in imposing the restriction: preserving both the actual and perceived impartiality of the state judiciary. Id. at 775-76 , 122 S.Ct. 2528 . The Court warned, however, that speaking of the need for an impartial judiciary in general terms would not do; instead, it was necessary to pinpoint the precise meaning of the term “impartial.” Id. at 775 , 122 S.Ct. 2528 . The majority offered three definitions. Id. at 775-84 , 122 S.Ct. 2528 .

, First, the term could mean a “lack of bias for or against either paHy to the proceeding.” Id. at 775 , 122 S.Ct. 2528 . But if that is what impartiality meant, the majority reasoned, the restriction was not narrowly tailored because it “[did] not restrict speech for or against particular parties, but rather speech for or against particular issues.” Id. at 776 , 122 S.Ct. 2528 . Second, impartiality could mean a “lack of preconception in favor of or against a particular legal view.” Id. at 777 , 122 S.Ct. 2528 . The Court held, however, that preserving such impartiality was not a compelling state interest because “[p]roof that a Justice’s mind at the time he joined the Court was a complete tabula rasa in the area of constitutional adjudication would be evidence of lack of qualification, not lack of bias.” Id. at 778 , 122 S.Ct. 2528 (citation omitted). Finally, “[a] third possible meaning of ‘impartiality’ ... might be described as open-mindedness.” Id. “This sort of impartiality seeks to guarantee each litigant, not an equal chance to win the legal points in the case, but at- least some chance of doing so.” Id. While recognizing that the state’s desire to ensure the open-mindedness of its judges might be compelling, the Court could not accept that Minnesota’s restriction was tailored to address this concern because it was “so woefully underinclusive.” Id. at 780 , 122 S.Ct. 2528 . Indeed, “statements in election campaigns are ... an infinitesimal portion of the public commitments to legal positions that judges (or judges-to-be) undertake,” for example, in legal opinions, public lectures, law review articles, and books. Id. at 779 , 122 S.Ct. 2528 . Because the restriction did not address such other public commitments, the Court concluded that the purpose behind' the restriction was “not openmindedness in the judiciary, but the undermining of judicial elections.” Id. at 782 , 122 S.Ct. 2528 .

In the aftermath of White I, few regulations of judicial-campaign speech withstood strict scrutiny. One of the most important decisions of that period, for our purposes, is the Eighth Circuit’s decision on remand from White I regarding the validity of a Minnesota restriction that was almost identical to Montana’s Rule 4.1(A)(7). See Republican Party of Minn. v. White (“White II”), 416 F.3d 738, 745 (8th Cir. 2005) (en banc). Minnesota prohibited judges or judicial candidates from “identify[ing] themselves as members of a political organization,” “attend[ing] political gatherings,” or “seek[ing], accepting] or us[ing] endorsements from a political organization.” Id. (quoting 52 Minn. Stat. Ann., Code of Judicial Conduct, Canon 5, subd. (A)(1)). The court referred to the restriction as the “partisan-activities” clause. Id.

Sitting en banc, a majority of the Eighth Circuit hewed closely to White 7’s framework, identifying the three meanings of judicial impartiality discussed in the Supreme Court’s decision and then analyzing whether any of these interests justified the restriction. Id. at 751-66 . As to the first two meanings of impartiality—lack of par-' ty bias and preconception toward a particular legal view—the court adopted the Supreme Court’s analysis, reasoning that the restriction involved in White I was not all that different from a restriction on assock ating with a particular political group. Id. at 753-56 . And as to the third type of impartiality—judicial open-mindedness— the court thought that, like the restriction in White I, the partisan-activities clause was “woefully underinclusive.” Id. at 756 . The clause prohibited “associative activities with a political party during a campaign,” but not at any time before that. Id. at 758 . More important still, the court continued, “it ma[de] little sense for the state to restrict [associational] activity only with political parties,”'but not with interest groups such as the National Rifle Association, the National Organization for Women, and the Christian Coalition, which could potentially harm judicial open-mindedness to the same extent. Id. at 759 . And although the court acknowledged that treating political parties differently might be “justified given political parties’ ‘powerful machinery,’ including a large membership, to enforce adherence to their views,” it countered that some Minnesota parties—such as the Constitution Party, the Natural Law Party, and the Green party— have “a more limited membership” and focus on “only a few'issues.” Id. at 760-61 n.12 (citation omitted). The court struck the partisan-activities clause under the First Amendment. Id. at 766 .

Other courts, including ours, followed the lead of White I and II and invalidated similar speech restrictions. In Sanders County, we ruled unconstitutional a Montana statute that made it a criminal offense for any political party to “endorse, contribute to, or make an expenditure to support or oppose a judicial candidate.” 698 F.3d at 744 (citing Mont. Code Ann. § 13-35-23 )< We accepted as true that “Montana has a compelling interest in maintaining a fair and independent judiciary,” but held that the statute was underinclusive because it forbade “judicial endorsements by political parties but not by other associations, individuals, corporations, special interest groups and the like.” Id. at 746-47 (quoting and relying on White II ⅛ discussion of underinclusivity). The panel majority also found no evidence that “preventing political parties from endorsing judicial candidates is a necessary prerequisite to maintaining a fair and independent judiciary,” especially given that many states “not only allow party endorsements but require party nominations.” Id. at 746 “If Montana were concerned that party endorsements might undermine elected judges’ independence,” the court concluded, “Montana could appoint its judges, with a bipartisan and expert panel making nominations—a less restrictive alternative currently practiced by several states.” Id. at 747 .

2

The strict First Amendment framework of White I underwent significant changes with the Supreme Court’s decision in Williams-Yulee. Before the Court was a challenge to Florida’s solicitation restriction, which prohibited judicial candidates from' personally soliciting funds but allowed them to establish committees to do so for them. Williams-Yulee, 135 S.Ct. at 1663 . The restriction was very similar to those invalidated by the Eighth Circuit in White II and the Sixth Circuit in Carey v. Wolnitzek, 614 F.3d 189, 201-07 (6th Cir. 2010). Employing reasoning -that contrasted sharply with White I, the Supreme Court upheld the regulation.

The Court began by “holding] ... what [it] assumed in White: A State may restrict the speech of a judicial candidate only if the restriction is narrowly tailored to serve a compelling interest.” Id. at 1665 (plurality' opinion). The Court then found that Florida had a compelling interest in “protecting the integrity” of its judiciary and “maintaining the public’s confidence in an impartial judiciary.” Id. at 1666 (citation omitted). In contrast to White I, however, the Court did not attempt to define precisely what judicial integrity or impartiality means. Instead, it emphasized that “[t]he concept of public confidence in judicial integrity does not easily reduce to precise definition, nor does it lend itself to proof by documentary record. But no one denies that it is genuine and compelling.” Id. at 1667. “Unlike the executive or the legislature, the judiciary ‘has no influence over either the sword or the purse... The judiciary’s authority therefore depends in large measure on the public’s willingness to respect and follow its decisions.” Id. at 1666 (quoting The Federalist No. 78, at 465).

After accepting Florida’s general interest in judicial impartiality, the Court considered the argument made in every case discussed above: the restriction was fatally underinclusive. Id. at 1668. The challenger argued that underinclusivity arose because Florida permitted indirect solicitations through committees and the writing of thank7you notes, which ensured that candidates knew the identity of donors. Id. In the pre-Williams-Yulee world, these arguments successfully convinced courts to invalidate similar solicitation clauses. See Carey, 614 F.3d at 205 (“Although the candidate himself may not solicit donations, his campaign committee may.... That leaves a rule preventing a candidate from sending a signed mass mailing to every voter in the district but permitting the candidate’s best friend to ask for a donation directly from an attorney who frequently practices before the court. Are not the risks of coercion and undue appearance far less with the first (prohibited) solicitation than the second (permitted) one?”); White II, 416 F.3d at 765-66 .

But the Court flatly rejected these arguments. Williams-Yulee, 135 S.Ct. at 1668 . “It is always somewhat counterintuitive to argue,” the Cburt reasoned, “that a law violates the First Amendment by abridging too little speech.” Id. And although underinclusivity might indicate that the law does not advance a compelling state interest, “the First Amendment imposes no freestanding ‘underinclusiveness limitation.’ ” Id. (citation omitted). “A State need not address all aspects of a problem in one fell swoop; policymakers may fqcus on their most pressing concerns.’’ Id. Florida’s prohibition on only, personal solicitations did precisely that. Despite the fact that a candidate’s campaign, committee could solicit funds on behalf of the candidate, “Florida ... ha[d] reasonably concluded that solicitation by the candidate personally creates a categorically different and more severe risk of undermining public confidence.” Id. at 1669 . Similarly, while permitting candidates to write thank you notes might “heighten[ ] the likelihood of actual bias by ensuring that judicial candidates know who supported their campaigns, and ensuring that the supporter knows that the candidate knows,” the compelling interest in an impartial judiciary “is implicated most directly by the candidate’s personal solicitation itself.’.’ Id.

Having dispensed with the argument that Florida prohibited too little speech, the Court turned to whether the solicitation provision restricted too much. Id. at 1670-71 . The challenger argued that her method of soliciting funds—mass mailing—presented such a small threat to the public confidence in the judiciary that Florida’s provision was unconstitutionally overinclusive. Id. at 1671 . Concluding ⅛ was not, the Court emphasized that the restriction must be only narrowly tailored, not “perfectly tailored,” because “the First Amendment does not confine a State to addressing evils in their most acute form.” Id. (citation omitted). The Court declined to “wade into [the] swamp” of unworkable line drawing, respecting Florida’s conclusion “that all personal solicitations by judicial candidates”—no matter what form they take—“create a public appearance that undermines confidence in the integrity of the judiciary.” Id. Accordingly, the Court ruled the solicitation provision a permissible restriction under the First Amendment. Id. at 1673 .

Williams-Yulee marked a palpable change in the approach to state regulations of. judicial-campaign- speech—a change perhaps best exemplified by our unanimous en banc decision in Wolfson. Wolfson involved two Arizona restrictions on judicial speech similar to that before us: a prohibition on public endorsements of political candidates, via monetary contributions or otherwise, and a restriction, on political campaigning generally. Wolfson, 811 F.3d at 1178 -79 & nn.2-3. In upholding these provisions, we relied almost exclusively on Williams-Yulee. Id. at 1180-86 . We first identified a broad compelling interest in preserving public confidence in the judiciary’s integrity without going through the interest analysis conducted in White I. Id. at 1182 . We then addressed and rejected the challenger’s contention that the restrictions were both under- and overinclusive.

The challenger’s underinclusivity argument was the same one accepted by the Eighth Circuit in White II and our court .in Sanders County: the restriction was not narrowly tailored because, among other things, it permitted judicial candidates to endorse persons and entities, other than political candidates. Id. at 1183 . But that argument, we held, was no longer persuasive in light of Williams-Yulee. Id. Although “Williams-Yulee may have been about a prohibition on direct candidate solicitations of campaign contributions, ... the Supreme Court’s reasoning was broad enough to encompass underinclusivity arguments aimed at other types of judicial candidate speech prohibitions such as [the endorsement and campaign prohibitions].” Id. There was no need to “question whether Arizona could have ... prohibited more types of endorsements or campaign participation” because Arizona was entitled to focus on only the most pressing concerns associated with “a judicial candidate actively engaging] in political campaigns.” Id. at 1184 .

As to overinclusivity, we acknowledged that Arizona’s restrictions reached actions that were unlikely to have any effect on judicial impartiality, such as a judicial candidate’s endorsement of the President of the United States. Id. But yet again, “Williams-Yulee forclose[d] [this] argument ].” Id. Following the lead of the Supreme Court, we declined to draw “unworkable and unnecessary line[s].” Id. at 1185 . It was simply not “our proper role to second-guess Arizona’s decision^ ]” to prohibit judicial candidates from engaging in political campaigns, even if that prohibition encompassed “political acts [that] present different levels of impropriety in different situations.” Id. We therefore held that Arizona’s restrictions were narrowly tailored to achieve Arizona’s interest in upholding public confidence in the judiciary. Id. at 1186 .

3

This summary evidences a change in the courts’ approach to state regulation of nonpartisan judicial elections. Although there are several important differences between the cases we have described, perhaps the most important one of all, at least for our purposes, is the way in which White I, White II, and Sanders County, on the one hand, and Williams-Yulee and Wolfson, on the other, approached the challengers’ un-derinclusivity arguments. And that difference might be more fundamental than it seems. According to White I, the underin-clusivity of Minnesota’s restriction revealed that its underlying purpose was not judicial impartiality, “but the undermining of judicial elections.” 536 U.S. at 782 , 122 S.Ct. 2528 ; see also Williams-Yulee, 135 S.Ct. at 1681 (Scalia, J., dissenting) (“[The Florida restriction’s] scope suggests that it has nothing to do with the appearances created by judges’ asking for money, and everything to do with hostility toward judicial campaigning.”). That is, the problem Minnesota faced stemmed from the very scheme of judicial elections, which threatened judicial impartiality in and of itself. White I, 536 U.S. at 782 , 122 S.Ct. 2528 (“[Ejlected judges—regardless of whether they have announced any views beforehand—always face the pressure of an electorate who might disagree with their rulings and therefore vote them off the bench.”). Minnesota’s attempts to address the problem without abandoning its election scheme served only to deprive voters of relevant information while failing to eliminate the underlying source of actual or perceived judicial bias. Id. at 788 , 122 S.Ct. 2528 (“[T]he greater power to dispense with elections altogether does not include the lesser power to conduct elections under conditions of state-imposed voter ignorance.” (alteration in original)). Consequently, after White I, a state was left with limited options in regulating its judicial elections because such regulations would be almost always underinclusive. That is why in Sanders County, we considered appointment of judges a “less restrictive alternative” able to accomplish the state’s goal of judicial impartiality. 698 F.3d at 747.

But the Williams-Yulee majority viewed things differently. It saw a middle ground where “policymakers [could] focus on their most pressing concerns” without completely eliminating the judicial bias (or perception thereof) associated with judicial elections generally. Williams-Yulee, 135 S.Ct. at 1668 . In other words, a state could both abridge some judicial-campaign speech and preserve its election system—“[t]he First Amendment does not put a State to [an] all-or-nothing choice.” Id. at 1670 . The Court thus sustained Florida’s prohibition on personal solicitations, despite the fact that other methods of solicitation might also undermine judicial impartiality, because a complete ban would put an end to judicial elections altogether. Id.; see also id. at 1681 (Scalia, J., dissenting) (“One cannot have judicial elections without judicial campaigns, and judicial campaigns without funds for campaigning, and funds for campaigning without asking for them.”). As our court in Wolfson held, moreover, this view of underinclusivity extends not just to restrictions on campaign contributions but also “encompass[es] un-derinclusivity arguments aimed at other types of judicial candidate speech prohibitions.” 811 F.3d at 1183 .

B

Against this background, we turn to Montana’s Rule 4.1(A)(7). As we have already mentioned, it prohibits judicial candidates like French from “seek[ing], accepting], or usfing] endorsements from a political organization, or partisan or independent non-judicial office-holder or candidate .... ” Mont. Code of Judicial Conduct Rule 4.1(A)(7). Because Rule 4.1(A)(7) is a content-based restriction on speech, we may uphold it only if it is narrowly-tailored to achieve some compelling state interest. Wolfson, 811 F.3d at 1180-81 .

We discern two such interests here. The first is an interest in both actual and perceived judicial impartiality. In Williams-Yulee, the Court noted “the regrettable but unavoidable appearance that judges who personally ask for money may diminish their integrity.” 135 S.Ct. at 1667 . Likewise, we note here the regrettable but unavoidable consequence that judges who personally ask for political endorsements may diminish the public’s faith in the impartiality of the judiciary, whether a judge’s actual impartiality is affected or not. Seeking and using political endorsements may create the appearance that a judge will favor certain politicians or political parties and thereby “undermine the public’s confidence that judges base rulings on law, and not on party affiliation.” Wolfson, 811 F.3d at 1183, 1186 . We need not define Montana’s interest in terms more specific than these, for “no one denies that [this interest] is genuine and compelling.” Williams-Yulee, 135 S.Ct. at 1667 .

Rule 4.1(A)(7) furthers a second interest that might be more compelling still: a related but distinct interest in a structurally independent judiciary. See Wolfson, 811 F.3d at 1186-88 (Berzon, J., concurring). If judicial candidates, including sitting judges running for reelection, regularly solicit and use endorsements from political parties, the public might view the judiciary as indebted to, dependent on, and in the end not different from the political branches. One way to preserve the distance between the judiciary and the political branches is to place.the judiciary on a different footing and do so in a way that is visible, to the public. The-federal system insulates the third, branch from partisan activities by separating judges from the direct-election process. Some states have followéd the federal model; others have adopted an appoihtment-and-retention-election model; and still others have decided on elections. These systems have their critics and their defenders. It is not'for us to choose among these systems because the U.S. Constitution does not prescribe any particular form for state judicial elections. What is sufficient for oür purposes is to observe that- these various models all treat the selection of judges differently from the processes for choosing our other public officials. That fact alone separates the judicial branch from the political branches. Montana has chosen to structure its third branch differently from the political branches, and we cannot fault its efforts to reinforce that choice in the manner in which it elects its judges. The Federalist No. 78 at 466 (‘“[TJhere is no liberty, if the po-wer of judging be not separated from the legislative and executive powers.’ ” (citation omitted)).

French and his amicus curiae nonetheless contend that Rule 4.1(A)(7) cannot survive strict scrutiny because it is underindusive, overinclusive, and otherwise insufficiently tailored to any interests Montana might have. Although these arguments might have been persuasive in the pre-Williams-Yulee era, they no longer carry the day. As we explain below, Rule 4.1(A)(7)’s scope is sufficiently narrowly tailored to pass muster under the First Amendment.

1

French’s primary claim is that Rule 4.1(A)(7) is fatally underindusive. To begin with, French correctly points out that the rule prohibits candidates from seeking and using endorsements from political organizations but does not forbid any interest groups, corporations, and other entities from making such endorsements. This same argument persuaded the court in White II to invalidate.a provision almost identical to Rule 4.1(A)(7), 416 F.3d at 758 , and was successful .in convincing our court in Sanders County to strike down a different Montana restriction, 698 F.3d. at 747. But White II and Sanders County were decided before Williams-Yulee clarified that “the First Amendment imposes no freestanding ‘underinclusiveness limitation’” and that “policymakers, may focus on their most-pressing concerns.” 135 S.Ct. at 1668 (citation omitted). Relying on these pronouncements, Wolfson held that an endorsement provision—one that prohibited judicial candidates from endorsing individuals running for a political office—was not invalid simply because it “allow[ed] judicial candidates to endorse public officials and non-candidates.” 811 F.3d at 1183 . We reached this conclusion despite- the- possibility that “Arizona could have ... prohibited more types of endorsements or campaign participation.” Id. at 1184 .

Wolfson ⅛ repudiation of French’s argument aside, political parties are simply not the same as interest groups and private individuals. Parties have comprehensive platfoms, take firm positions on a multitude of issues, and are capable of exerting more influence in an election than most (if not any) interest groups. The Supreme Court described the “salient” and “real-world differences between political parties and interest groups” in McConnell v. Federal Election Commission:

540 U.S. 98, 188 , 124 S.Ct. 619 , 157 L.Ed.2d 491 (2003), overruled on other grounds by Citizens United, 558 U.S. 310 , 130 S.Ct. 876 , 175 L.Ed.2d 753 .

Once we turn to Montana’s interest in judicial independence (as opposed to mere judicial impartiality) the differences between political parties and interest groups grow starker. An endorsement from a political party threatens the public perception of judicial independence to a greater degree than an endorsement from an interest group. In all cases, an endorsement suggests the possibility of a quid-pro-quo exchange in which a judge may rule favorably for the endorsing entity. But whereas a judge may only infrequently encounter litigation implicating an endorsing interest group, he or she is likely to often face legislation an endorsing political party has either supported or opposed. Dependence on an endorsing political party brings into question whether a judge will be able to independently interpret and review a given piece of legislation and thus goes to the core of the separation of powers. For all these reasons, Montana is well within its authority to focus its immediate concern on endorsements from political parties.

French next suggests that Rule 4.1(A)(7) is impermissibly underinclusive because Montana permits candidates to solicit and use political parties’ money but not their endorsements. We are. not persuaded.. Endorsements and campaign contributions are different kinds of support. An endorsement is a public and easily communicable show of solidarity. Although most campaign contributions are also public information, see Moiit. Code Ann. § 13-37-229, they are less forceful and less easily communicable. Unlike endorsements, information on campaign contributions typically requires extra work for voters to access. It would therefore not be surprising for judicial candidates to derive more value from endorsements from political parties and popular politicians (including politicians outside Montana) than from-even sizeable donations. Montana could reasonably conclude that endorsements are more suggestive of a quid-pro-quo exchange and pose a greater risk to the public perception of its judiciary than donations. Thus, we decline to disturb Montana’s determination that monetary donations “present markedly different appearance to the public,” Williams-Yulee, 135 -S.Ct. at 1669, than a candidate’s use of a political endorsement.

French’s remaining reasons for deetaing Rule 4.1(A)(7) unconstitutionally underin-clusive are the weakest. French complains that the -rule applies only during campaigns and only to endorsements from “non-judicial'office-holders.” But in order to create an impartial and independent judiciary, it makes perfect sense for Montana to prohibit the solicitation and use of endorsements during (as opposed to before) a judicial candidate’s campaign and limit those endorsements to political office holders and entities (as opposed to nonpartisan judges). It is almost self-evident that the dangers of actual and perceived bias and dependence are not nearly as great when the candidate is not yet running for office or when she uses endorsements from nonpartisan judges. According to French, Montana may ban the use of endorsements only if it bans the use of all endorsements from any individuals or entities at any time. But “[t]he First Amendment does not put a State to that all-or-nothing choice.” Williams-Yulee, 135 S.Ct. at 1670 .

French’s amicus appears to make an additional argument—one that was accepted in White II. It goes like this: A candidate’s discussion of her endorsements with the public is not all that different from a discussion of other important issues because a party label is just a “shorthand for the [numerous] views the candidate holds.” White II, 416 F.3d at 754 . And because White I made clear that a restriction on the candidates’ announcement of views is woefully underinclusive, a restriction on the use of political endorsements must be as well. We cannot accept this argument. The seeking and using of political endorsements is nothing like announcing one’s views on certain issues. An endorsement is a thing of value: it may attract voters’ attention, jumpstart a campaign, give assurance that the candidate has been vetted, or provide legitimacy to an unknown candidate and indicate that he or she is capable of mounting a successful campaign. Such things of value are usually not given out for free, and even when they are, the mere perception of quid pro quo in judicial campaigns might undermine the public’s trust in the impartiality and independence of its judiciary.

Along these same lines, French’s amicus suggests that Wolfson is distinguishable because the restriction there prohibited candidates from speaking about others but not themselves. While Wolfson did state that “Arizona’s prohibitions do not prevent judicial candidates from announcing their views on disputed legal and political subjects,” Rule 4.1(A)(7) here does not prohibit candidates from “announcing their views on disputed legal and political subjects.” 811 F.3d at 1185 . Candidates in Montana are still free to discuss political issues with their electorate. They can speak on abortion, criminal sentencing, healthcare, gun control, and dozens of other matters of controversy. What they cannot do is tell their electorate that a political party has given their candidacy a valuable stamp of approval. That restriction is not unconstitutionally underinclusive because it addresses a very specific concern present whenever a candidate for a nonpartisan office receives something of value from a partisan organization.

2

French and his amicus next argue that Rule 4.1(A)(7) is overinclusive because, unlike the solicitation restriction in Williams-Yulee, Montana does not allow even the candidates’ campaign committees to seek and use political .endorsements. Mont. Code of Judicial Conduct Rule 4.1(B) (providing that “[a] judge or judicial candidate shall take reasonable measures to ensure that other persons do not undertake, on behalf of the judge or judicial candidate, any activities prohibited in paragraph (A)”). But that is hardly a fair criticism of Rule 4.1(A)(7). Montana’s interests in an impartial and independent judiciary do not diminish simply because it is the candidate’s affiliates who go around telling voters about the political endorsements the candidate has received. The danger lies in the public losing trust in its judges from hearing political endorsements; it is irrelevant whether the candidate or the candidate’s committee delivers the message.

But even accepting that a committee engaging in prohibited conduct threatens judicial impartiality and independence to a lesser degree, the Supreme Court has told us that “the First Amendment does not confine a State to addressing evils in their most acute form.” Williams-Yulee, 135 S.Ct. at 1671 . Following the lead of Williams-Yulee and Wolfson, we decline to draw such arbitrary and unnecessary lines. Id.; Wolfson, 811 F.3d at 1185 . Montana has reasonably determined that both candidates and their committees pose a threat to its judiciary when they seek, accept, or use political endorsements in their campaigns. This “considered judgement ] deserve[s] our respect.” Williams-Yulee, 135 S.Ct. at 1671 .

3

Finally, French asserts that Rule 4.1(A)(7) is generally not narrowly tailored because Montana has presented no evidence showing that political endorsements cause harm. Indeed, as French rightly notes, several states not only allow but require political endorsements because their judicial elections are partisan. But neither of these two points defeats Rule 4.1(A)(7); the Court has not treated judicial elections as an either/or proposition, requiring any state that chooses to have judicial elections to conduct them like all other elections. As to the lack of evidence, the Supreme Court has flatly stated that “[t]he concept of public confidence in judicial integrity ... does [not] lend itself to proof by documentary record.” Id. at 1667 . Montana need not present empirical evidence of something as abstract as a decrease in actual or perceived judicial impartiality and independence for its rule to survive strict scrutiny. And as to the point regarding states with partisan judicial elections, neither Williams-Yulee nor Wolfson so much as thought about invalidating restrictions designed to preserve nonpartisanship in judicial elections simply because there are some states that have partisan elections and appear to be doing just fine. If that fact alone were sufficient to invalidate a restriction on judicial-eam-paign speech, then nonpartisan judicial elections could be themselves deemed unconstitutional. We decline to reach such a result.

Although French suggests that eliminating judicial elections altogether would be a less restrictive means to accomplishing Montana’s stated goals, Williams-Yulee and Wolfson foreclose that suggestion. Those cases confirm that the states have every right to devise and regulate a system of nonpartisan judicial elections. Williams-Yulee, 135 S.Ct. at 1671 ; Wolfson, 811 F.3d at 1185 . The Constitution does not demand that the states follow the federal model and appoint their judges, and if it permits the states to hold partisan judicial elections, we see no impediment to the states adopting nonpartisan judicial elections, as Montana has done.

[[Image here]]

We hold that Montana has compelling interests in an impartial and independent judiciary. Rule 4.1(A)(7) is narrowly tailored to those interests because it strikes an appropriate balance between a candidate’s speech and Montana’s interest in an independent and impartial judiciary. French’s arguments to the contrary are foreclosed by the Supreme Court’s decision in Williams-Yulee and our decision in Wolfson.

AFFIRMED.

. Montana is not alone in restricting the political speech of judges and judicial candidates. As of 2012, "[t]hirty-nine states have judicial elections, and nearly all have enacted laws to treat judicial elections differently from political elections.” Sanders Cty. Republican Cent. Comm. v. Bullock, 698 F.3d 741, 750 (9th Cir. 2012) (Schroeder, J., dissenting). . Montana doés not, however, prohibit political parties or organizations from endorsing - judicial candidates. See Sanders Cty., 698 F.3d at 749 . . Other circuits used analogous reasoning to strike various restrictions on judicial-campaign speech under White I. See Carey v. Wolnitzek, 614 F.3d 189, 201-07 (6th Cir. 2010) (invalidating Kentucky's regulations prohibiting judicial candidates from personally soliciting campaign funds and identifying themselves as members of political' parties); Siefert v. Alexander, 608 F.3d 974, 981-83, 990 (7th Cir. 2010) (striking Wisconsin’s statute prohibiting judicial candidates from being members of political parties). . Chief Justice Roberts’s opinion on the point .. commanded only a plurality of the Court, See id. at 1673 (Ginsburg, J,, concurring in part and concurring in the judgment). The dissenting justices, however, agreed with the plurality on this point. Id. at 1676 (Scalia, J., dissenting); see id. at 1682 (Kennedy, J., dissenting) (largely agreeing with Justice Scalia’s analysis); id. at 1685 (Alito, J., dissenting) (same). . Justice Scalia—the author of White I—wrote the principal dissent. He faulted the majority for failing to follow White I by permitting Florida "to invoke hazy concerns about judicial impartiality," which he thought were too malleable for a strict scrutiny analysis. Id. at 1677-78 (Scalia, J., dissenting). He then questioned whether "allowing personal solicitations would imperil public faith in judges," given the "coexistence of judicial elections and personal solicitations for most of our history.” Id. at 1678. And even if Florida's prohibition did improve the public reputation of judges, Justice Scalia reasoned that it would still fail strict scrutiny because of- its under- and overinclusivity. Id. at 1679-82. . We note that White I “neither assert[ed] nor implied] that the First Amendment requires campaigns for judicial office to sound the same as those for legislative office.” Id. at 783, 122 S.Ct. 2528 . But we think it is nonetheless clear that White I envisioned as permissible only those regulations that targeted threats to impartiality more salient and direct than those inherent in a system of judicial elections. For instance, White I thought it plausible that campaign promises in particular might pose a "special threat to openmind-edness” that could be potentially redressible through regulation. Id. at 780-81 , 122 S.Ct. 2528 . , See Methods of Judicial Selection, Nat’l Ctr. for St. Cts,, http://www.judicialselection.us/ judicial_selection/rnethods/selection_oL judges,cfm?state= (last visited Oct. 17, 2017).