Chula Vista Citizens for Jobs v. Donna Norris ( 2014 )


Menu:
  •                  FOR PUBLICATION
    
      UNITED STATES COURT OF APPEALS
           FOR THE NINTH CIRCUIT
    
    
    CHULA VISTA CITIZENS FOR JOBS            No. 12-55726
    AND FAIR COMPETITION; LORI
    KNEEBONE; LARRY BREITFELDER;               D.C. No.
    ASSOCIATED BUILDERS AND                 3:09-cv-00897-
    CONTRACTORS OF SAN DIEGO, INC.,           BEN-JMA
                 Plaintiffs-Appellants,
    
                     v.                        OPINION
    
    DONNA NORRIS; MAYOR CHERYL
    COX; PAMELA BENSOUSSAN; STEVE
    CASTANEDA; JOHN MCCANN, in his
    official capacity as Member of the
    Chula Vista City Council; RUDY
    RAMIREZ, JR., in his official
    Capacity as Member of the Chula
    Vista City Council,
                    Defendants-Appellees,
    
    STATE OF CALIFORNIA,
        Intervenor-Defendant–Appellee.
    
    
          Appeal from the United States District Court
             for the Southern District of California
           Roger T. Benitez, District Judge, Presiding
    
                  Argued and Submitted
           November 6, 2013—Pasadena, California
    2              CHULA VISTA CITIZENS V. NORRIS
    
                            Filed June 16, 2014
    
          Before: Diarmuid F. O’Scannlain, Susan P. Graber,
                  and Carlos T. Bea, Circuit Judges.
    
         Opinion by Judge O’Scannlain, in which Judge Graber
        joins, except as to Part IV, and in which Judge Bea joins,
           except as to Part III. Judge Graber filed an opinion
           dissenting as to Part IV. Judge Bea filed an opinion
                          concurring as to Part III.
    
    
                               SUMMARY*
    
    
                                Civil Rights
    
        The panel affirmed in part and reversed in part the district
    court’s summary judgment and remanded in an action
    brought by two associations and two individuals alleging that
    certain provisions of the California Elections Code pertaining
    to initiatives and referenda, as incorporated into the Chula
    Vista, California Charter, violated the First Amendment.
    
        Plaintiffs challenged the state and local requirements that:
    (1) official proponents of local ballot initiatives be electors,
    which excludes non-natural persons and thereby excludes
    associations; and (2) official initiative proponents identify
    themselves on the face of the initiative petitions.
    
    
    
    
      *
        This summary constitutes no part of the opinion of the court. It has
    been prepared by court staff for the convenience of the reader.
                 CHULA VISTA CITIZENS V. NORRIS                    3
    
        The panel first determined that as to the challenge to the
    elector requirement, abstention under Railroad Commission
    v. Pullman Co., 
    312 U.S. 496
     (1941), was not warranted
    because the challenge implicated the chilling of expression
    and the parties had not indicated that there were any pending
    actions in the California courts.
    
        Affirming the district court’s summary judgment to the
    defendants on the elector requirement, the panel held that
    associations do not have a First Amendment right to serve as
    official proponents of local ballot initiatives.
    
        Reversing the district court’s summary judgment to the
    defendants as to the petition-proponent disclosure
    requirement, the panel held that the requirement did not
    satisfy exacting scrutiny and therefore §§ 9202 and 9207 of
    the California Elections Code were invalid to the extent that
    the provisions require official initiative proponents to identify
    themselves on the face of initiative petitions.
    
        Concurring in part and dissenting in part, Judge Graber
    agreed with the majority opinion that the case was properly
    before the court, and concurred in Part III of the opinion,
    which held that the elector requirement passed constitutional
    muster. She wrote separately to dissent from Part IV of the
    opinion and stated that the petition-proponent disclosure
    requirement survived any level of review.
    
        Concurring, Judge Bea joined all of the majority opinion
    except for Section III, which analyzed the local and state
    requirements that official ballot initiative proponents be
    electors, thereby excluding associations. Judge Bea wrote
    that although he concurred in the result of Section III, he
    believed that the majority opinion employed an incorrect test
    4            CHULA VISTA CITIZENS V. NORRIS
    
    to determine whether the elector requirement burdened any
    First Amendment rights.
    
    
                             COUNSEL
    
    James Bopp, Jr., The Bopp Law Firm, PC, Terre Haute, IN,
    argued the cause for the Plaintiff-Appellant. Richard E.
    Coleson, The Bopp Law Firm, PC, Terre Haute, IN, filed the
    briefs for the plaintiff-appellant. With him on the briefs were
    James Bopp, Jr., The Bopp Law Firm, PC, Terre Haute, IN;
    Charles H. Bell, Jr., and Brian T. Hildreth, Bell, McAndrews
    & Hiltachk, LLP, Sacramento, CA; and Gary D. Leasure,
    Workman Leasure, LLP, San Diego, CA.
    
    Charles A. Bird, McKenna Long & Aldridge, LLP, San
    Diego, CA, argued the cause and filed a brief for the
    Defendant-Appellee.
    
    George Waters, Deputy Attorney General of California,
    Sacramento, CA, argued the cause and filed a brief for the
    Defendant-Intervenor–Appellee. With him on the brief were
    Kamala D. Harris, Attorney General of California; Douglas
    J. Woods, Senior Assistant Attorney General; and Peter A.
    Krause, Supervising Deputy Assistant Attorney General,
    Sacramento, CA.
                 CHULA VISTA CITIZENS V. NORRIS                 5
    
                             OPINION
    
    O’SCANNLAIN, Circuit Judge:
    
        We must decide whether associations have a First
    Amendment right to serve as official proponents of local
    ballot initiatives and the extent to which the same
    Amendment protects the anonymity of initiative proponents.
    
                                  I
    
                                  A
    
         This case arises from a political battle concerning labor
    unions. Chula Vista Citizens for Jobs and Fair Competition
    (“Chula Vista Citizens”), an unincorporated association, and
    Associated Builders and Contractors of San Diego, Inc., an
    incorporated association of construction-related businesses
    (“the Associations”), sought to place an initiative on the
    Chula Vista municipal ballot. As described by the title of the
    initiative, the proposed measure “mandat[ed] that the City or
    Redevelopment Agency not fund or contract for public works
    projects where there [was] a requirement to use only union
    employees.” The City of Chula Vista requires that initiative
    proponents be electors (“the elector requirement”), which
    excludes non-natural persons from serving as official
    proponents. Faced with this obstacle, Chula Vista Citizens
    asked two of its members, Lori Kneebone and Larry
    Breitfelder, to serve as proponents in place of the
    Associations. They agreed.
    
         Section 903 of the Chula Vista Charter incorporates the
    provisions of the California Elections Code that govern
    initiatives and referenda “so far as such provisions of the
    6              CHULA VISTA CITIZENS V. NORRIS
    
    Election Code are not in conflict with [the] Charter.” The
    code establishes several requirements that official proponents
    must meet to qualify an initiative. First, proponents must file
    a notice of intent to circulate an initiative petition for
    signatures, and such notice must be signed by at least one but
    not more than three proponents. Cal. Elec. Code § 9202(a)
    (the “notice-filing requirement”). Defendant Donna Norris,
    as the City Clerk, receives and processes these filings.
    Proponents must include the written text of the initiative and
    may include a 500-word statement of “reasons for the
    proposed petition.” Id. The City Attorney then provides a
    title and summary of the measure to the proponents. Id.
    § 9203.
    
        Because the City has a newspaper of general circulation,
    the proponents must publish the notice of intent, title, and
    summary in such newspaper and submit proof of publication
    to the City Clerk. Id. § 9205(a) (the “publication
    requirement”).1 Only at that point can the proponents begin
    circulating their petition for signatures. Id. § 9207.
    
        The initiative petition is typically divided into “sections”
    to facilitate gathering signatures. See id. § 9201. Each
    section of the petition must “bear a copy of the notice of
    intention and the title and summary prepared by the city
    attorney.” Id. § 9207. Because § 9202(a) requires
    proponents to sign the notice, the effect of § 9207 is that the
    identities of official proponents are disclosed to would-be
    signatories of the petition (the “petition-proponent disclosure
    requirement”). Proponents have 180 days to file the signed
    
    
        1
        Where no such newspaper exists for either the city or the county, the
    same information must be posted at three designated public places in the
    city. Cal. Elec. Code § 9205(b).
                  CHULA VISTA CITIZENS V. NORRIS                     7
    
    petitions with the City Clerk bearing the requisite number of
    signatures. Id. § 9208. The City Clerk informs the
    proponents whether they have gathered enough valid
    signatures to qualify the initiative for the ballot. Whether the
    initiative appears on the ballot or immediately becomes law
    depends on the number of signatures gathered and the actions
    taken by the City Council.
    
        Kneebone and Breitfelder made two attempts to qualify
    the initiative for the ballot. The first attempt (“First Petition”)
    began on August 28, 2008, with the filing of the notice of
    intent. Kneebone and Breitfelder later submitted 23,285
    signatures to Norris after having complied with all the
    requirements except one: They had not included their names
    on the notice that appeared on the circulated petitions.
    Instead, as Kneebone and Breitfelder later informed Norris,
    they printed the following statement at the end of each
    circulated petition: “Paid for by Chula Vista Citizens for Jobs
    and Fair Competition, major funding by Associated Builders
    & Contractors PAC and Associated General Contractors PAC
    to promote fair competition.” On November 12, 2008, Norris
    rejected the First Petition for failure to include the
    proponents’ signatures on the notice accompanying the
    circulated petitions.
    
         The Associations again asked Kneebone and Breitfelder
    to serve as proponents, which the pair again agreed to do.
    The second attempt (“Second Petition”) began with the notice
    filing on March 13, 2009.           It complied with all
    requirements—including the requirement that circulated
    petitions bear the proponents’ signatures—appeared on the
    June 8, 2010 municipal election ballot, and was approved by
    voters.
    8            CHULA VISTA CITIZENS V. NORRIS
    
                                  B
    
        On April 28, 2009, after Norris rejected their First
    Petition but before qualifying the Second Petition, the
    plaintiffs brought this 42 U.S.C. § 1983 suit in the Southern
    District of California seeking declaratory and injunctive
    relief. The complaint alleged that the elector and petition-
    proponent disclosure requirements, both facially and as
    applied, violate the First Amendment. On June 4, the
    plaintiffs moved for a preliminary injunction and for an
    expedited hearing. Because provisions of the state election
    code were at issue, the State of California intervened as a
    defendant.
    
        The district court held a hearing on the preliminary
    injunction motion on August 19. The next day, it ordered
    supplemental briefing as to whether the Elections Code did,
    in fact, require that official proponents be natural persons.
    On March 8, 2010, the district court denied the preliminary
    injunction motion as moot in light of the success of the
    Second Petition, and it stayed consideration of the § 1983 suit
    pending the Supreme Court’s decision in Doe v. Reed,
    
    561 U.S. 186
     (2010). When the district court lifted the stay,
    both sides filed motions for summary judgment. The district
    court granted summary judgment to Norris and her co-
    defendants on March 22, 2010. It entered its judgment on
    April 10, and plaintiffs timely appealed.
    
                                  II
    
        We must first determine whether the dispute over the
    elector requirement is properly before us. The parties
    disagree about whether the elector requirement is mandated
    by state law, municipal law, or the City’s interpretation of
                 CHULA VISTA CITIZENS V. NORRIS                   9
    
    either body of law. Relying on the Supreme Court’s decision
    in Railroad Commission v. Pullman Co., 
    312 U.S. 496
    (1941), Norris urges us to abstain from deciding the merits of
    this case if doing so would require us to resolve a contested
    issue of state law.
    
         “[W]hen a federal constitutional claim is premised on an
    unsettled question of state law, the federal court should stay
    its hand in order to provide the state courts an opportunity to
    settle the underlying state-law question and thus avoid the
    possibility of unnecessarily deciding a constitutional
    question.” Harris Cnty. Comm’rs Court v. Moore, 
    420 U.S. 77
    , 83 (1975). Pullman abstention counsels against deciding
    unnecessary federal constitutional questions, but it is also
    premised on “avoid[ing] federal-court error in deciding state-
    law questions antecedent to federal constitutional issues.”
    See Arizonans for Official English v. Arizona, 
    520 U.S. 43
    , 76
    (1997). Because abstention “does not implicate [federal
    courts’] subject matter jurisdiction,” we are “never required
    to apply Pullman.” Columbia Basin Apartment Ass’n v. City
    of Pasco, 
    268 F.3d 791
    , 802 (9th Cir. 2001). “Abstention is,
    of course, the exception and not the rule, and [the Supreme
    Court has] been particularly reluctant to abstain in cases
    involving facial challenges based on the First Amendment.”
    City of Houston v. Hill, 
    482 U.S. 451
    , 467 (1987) (internal
    citation omitted).
    
        We consider three factors when deciding whether
    Pullman abstention is appropriate: “(1) there are sensitive
    issues of social policy upon which the federal courts ought
    not to enter unless no alternative to its adjudication is open,
    (2) constitutional adjudication could be avoided by a state
    ruling, and (3) resolution of the state law issue is uncertain.”
    10              CHULA VISTA CITIZENS V. NORRIS
    
    Wolfson v. Brammer, 
    616 F.3d 1045
    , 1066 (9th Cir. 2010)
    (internal quotation marks omitted).
    
        The Supreme Court has held that abstention in the First
    Amendment context is disfavored because “the delay of
    state-court proceedings might itself effect the impermissible
    chilling of the very constitutional right [the plaintiff] seeks to
    protect.” Hill, 482 U.S. at 467–68 (internal quotation marks
    omitted). Our court has been particularly loath to abstain in
    First Amendment cases: “We have held that, in First
    Amendment cases, the first Pullman factor will almost never
    be present because the guarantee of free expression is always
    an area of particular federal concern.” Porter v. Jones,
    
    319 F.3d 483
    , 492 (9th Cir. 2003) (internal quotation marks
    omitted). This concern “applies to both facial and as-applied
    challenges.” Id. at 493. In fact, we have abstained only once
    in a First Amendment context, and that case had an “unusual
    procedural setting” because the “issue in question was already
    before the state supreme court.” Id. at 493–94. In every
    other procedural setting, we have rejected abstention.
    
         The challenge to the elector requirement implicates the
    chilling of expression. Hill, 482 U.S. at 467–68. Indeed, if
    the elector requirement is unconstitutional, the Associations
    are being completely deprived of a constitutional right.
    Moreover, the parties have not indicated that there are any
    pending actions in the California courts. Porter, 319 F.3d at
    492. In this context, the first Pullman factor is not met, and
    abstention is not warranted. Id. The merits are thus before
    us.2
    
    
         2
         Despite the successful qualification and passage of the initiative
    advocated by the plaintiffs, this case is not moot because it is “capable of
    repetition, yet evading review.” Fed. Election Comm’n v. Wis. Right to
                  CHULA VISTA CITIZENS V. NORRIS                      11
    
                                     III
    
        The First Amendment provides, “Congress shall make no
    law . . . abridging the freedom of speech, or of the press; or
    the right of the people peaceably to assemble, and to petition
    the Government for a redress of grievances.” U.S. Const.
    amend. I. By virtue of the Fourteenth Amendment, the First
    Amendment applies to actions by state governments.
    Everson v. Bd. of Educ., 
    330 U.S. 1
    , 8 (1947). The
    Associations contend that the elector requirement abridges
    the rights of speech, association, and petition. They further
    argue that strict scrutiny applies to the elector requirement
    and results in its invalidity. We begin with the threshold
    issue of whether the elector requirement implicates the First
    Amendment.
    
                                     A
    
        Although the Associations allege violations of speech,
    associational, and petition rights, our analysis will focus on
    the freedom of speech. The Associations mention their
    petition claim, but they provide no legal authority to support
    it. “A passing reference to an issue in a brief is not enough,
    and the failure to make arguments and cite authorities in
    support of an issue waives it.” Hamilton v. Southland
    Christian Sch., Inc., 
    680 F.3d 1316
    , 1319 (11th Cir. 2012);
    Acosta-Huerta v. Estelle, 
    7 F.3d 139
    , 144 (9th Cir. 1993).
    Thus, we do not address the Associations’ petition claim.
    
        The Associations also allege that the elector requirement
    is an unconstitutional condition on their right to associate for
    
    
    Life, Inc., 
    551 U.S. 449
    , 462 (2007); see also Davis v. Fed. Election
    Comm’n, 
    554 U.S. 724
    , 735 (2008).
    12           CHULA VISTA CITIZENS V. NORRIS
    
    purposes of political expression. See Speiser v. Randall,
    
    357 U.S. 513
    , 520–29 (1958). In the First Amendment
    context, the right to associate is not a free-standing right;
    rather, one has the right to associate for the purpose of
    engaging in activities protected by the First Amendment. Boy
    Scouts of Am. v. Dale, 
    530 U.S. 640
    , 647–48 (2000); Roberts
    v. U.S. Jaycees, 
    468 U.S. 609
    , 617–18 (1984). The
    Associations claim that serving as an official proponent is
    protected by the Free Speech Clause, and thus that they have
    a right to associate for the purpose of serving as official
    proponents. Thus, if serving as an official proponent is not an
    aspect of free speech, the condition imposed by the elector
    requirement does not violate the associational rights of the
    First Amendment. See Dale, 530 U.S. at 648 (“To determine
    whether a group is protected by the First Amendment’s
    expressive associational right, we must determine whether the
    group engages in ‘expressive association.’”); City of Dallas
    v. Stanglin, 
    490 U.S. 19
    , 25 (1989) (holding that, where the
    First Amendment does not protect a certain activity, there can
    be no First Amendment right of association to engage in that
    activity). Because the Associations’ claim regarding freedom
    of association depends on the success of their Free Speech
    claim, we focus on the latter claim.
    
        The question, then, is whether the elector requirement is
    a law “abridging the freedom of speech.” U.S. Const.
    amend. I.
    
                                  B
    
        To know whether the elector requirement abridges the
    freedom of speech, it is important to identify precisely what
    sort of infringement the requirement allegedly commits. The
                 CHULA VISTA CITIZENS V. NORRIS                  13
    
    Associations list several activities performed by official
    proponents that they contend are protected speech:
    
           [B]eing a proponent involves core political
           activity beyond ministerial acts of signing and
           filing things. A “proponent” begins with an
           idea about an issue, creates the text of an
           initiative to implement that idea, does the
           necessary publication of notices to qualify it,
           circulates petitions and/or arranges with
           others to do so, and advocates for the
           initiative.
    
         The Associations’ listing of these ostensibly expressive
    activities implies that associations are prohibited from
    engaging in them. But the Associations’ actions in this case
    belies that implication. As stated in their complaint, the
    Associations “decided to propose the Initiative.” “Chula
    Vista Citizens filed its required Clerk’s Version” of the
    initiative text, just as “Chula Vista Citizens published the
    Newspaper Version,” for which “[n]either Ms. Kneebone nor
    Mr. Breitfelder paid any money.” “Chula Vista Citizens
    hired The La Jolla Group to circulate the Petition in the City,”
    and, as the district court pointed out, the Associations were
    free to advocate for the initiative’s qualification and
    enactment. In short, the Associations were able to participate
    in all of the activities they mention.
    
        However, the Associations were dependent on Kneebone
    and Breitfelder as official proponents in order to engage in
    these activities. That is the gravamen of their alleged injury.
    The Associations believe the elector requirement violates the
    Free Speech Clause because, in their words, “speech-by-
    proxy is not a constitutionally permissible alternative because
    14              CHULA VISTA CITIZENS V. NORRIS
    
    it does not allow associations themselves to speak.” The
    Associations would rather have the legal authority to engage
    in these activities without relying on natural persons to serve
    as proxies, and that requires them to be official proponents.
    
        What the Associations seek, then, is the legal authority
    attaching to the status of an official proponent,3 and this
    amounts to a claim that serving as an official proponent is a
    form of “speech” protected by the First Amendment.
    
                                          C
    
        We must next determine the nature of the legal authority
    of official proponents. The Associations do not dispute that
    the initiative power is a legislative power. And rightly so. As
    the California Supreme Court has said, the initiative process
    “represents an exercise by the people of their reserved power
    to legislate.” Builders Ass’n of Santa Clara-Santa Cruz
    Cntys. v. Superior Court, 
    529 P.2d 582
    , 586 (Cal. 1974).4
    
    
         3
          The mere fact that the Associations have to rely on proxies to
    participate in these activities is insufficient, by itself, to violate the Free
    Speech Clause. If serving as an official proponent is not part of the
    freedom of speech, then it does not matter, for First Amendment purposes,
    that the Associations must rely on proxies who can serve as official
    proponents. See Citizens United v. Fed. Election Comm’n, 
    558 U.S. 310
    ,
    337–40 (2010).
         4
          In determining whether the authority of official proponents is
    legislative in nature, our inquiry is two-fold: (1) what powers, duties, and
    responsibilities are delegated to official proponents, and (2) whether those
    powers, duties, and responsibilities are legislative in character for First
    Amendment purposes? California law, and the state courts’ interpretation
    of California law, is dispositive as to the first question, but the second
    question—which determines whether the First Amendment is implicated
    by the elector requirement—is up to federal courts to decide.
                    CHULA VISTA CITIZENS V. NORRIS                         15
    
         Norris argues that the distinct role of proponents is to
    introduce legislation: “[T]he legal acts of a Proponent are acts
    of legislating, exercising the inherent, reserved power of
    citizens to legislate for the entity in which they reside.”
    Under this theory, because the initiative process is a
    lawmaking one, the activities that commence that process are
    analogous to the introduction of legislation. At least two
    California appellate courts support this description of the
    initiative process. San Francisco Forty-Niners v. Nishioka
    said the following:
    
             The initiative petition with its notice of
             intention is not a handbill or campaign flyer—
             it is an official election document subject to
             various restrictions by the Elections Code,
             including reasonable content requirements of
             truth.     It is the constitutionally and
    
    
    
          Our two-step analysis is no different than what we do in myriad areas
    of constitutional law. See, e.g., Memphis Light, Gas & Water Div. v.
    Craft, 
    436 U.S. 1
    , 9 (1978) (“The Fourteenth Amendment places
    procedural constraints on the actions of government that work a
    deprivation of interests enjoying the stature of ‘property’ within the
    meaning of the Due Process Clause. Although the underlying substantive
    interest is created by an independent source such as state law, federal
    constitutional law determines whether that interest rises to the level of a
    legitimate claim of entitlement protected by the Due Process Clause.”
    (internal quotation marks omitted)); see also Town of Castle Rock v.
    Gonzales, 
    545 U.S. 748
    , 756–57 (2005) (same). This method of analysis
    is all the more important when federal courts seek to analyze the structure
    of state government, implicating bedrock principles of federalism and state
    sovereignty. See Holt Civic Club v. City of Tuscaloosa, 
    439 U.S. 60
    , 71
    (1978) (“Government . . . is the science of experiment, and a State is
    afforded wide leeway when experimenting with the appropriate allocation
    of state legislative power.” (internal quotation marks and citation
    omitted)).
    16           CHULA VISTA CITIZENS V. NORRIS
    
           legislatively sanctioned method by which an
           election is obtained on a given initiative
           proposal.
    
    
    89 Cal. Rptr. 2d 388
    , 396 (Ct. App. 1999). Widders v.
    Furchtenicht stated that the legislative process begins once a
    petition is circulated for signatures: “An initiative is put
    before the people when they are asked to sign a petition to
    place it on the ballot . . . .” 
    84 Cal. Rptr. 3d 428
    , 438 (Ct.
    App. 2008) (internal quotation marks and citation omitted).
    If the activities involved in qualifying an initiative for the
    ballot start the legislative process, then official proponents
    exercise part of the legislative power.
    
        The Associations resist this characterization. They
    distinguish between placing an initiative on the ballot (which
    they concede is a legislative function) and asking electors to
    place an initiative on the ballot (which they contend is a non-
    legislative act). At oral argument, the Associations
    analogized initiative proponents to lobbyists: The official
    proponents come to the legislators (i.e., the electors) with a
    proposal and ask the legislators to introduce a bill (i.e., sign
    the petition to place the initiative on the ballot).
    
         The problem with the Associations’ proffered distinction
    is that the incidental role the Associations assign to official
    proponents is inconsistent with the responsibilities conferred
    on official proponents by the California Elections Code. As
    the California Supreme Court has said, “[O]fficial proponents
    of an initiative measure are recognized as having a distinct
    role—involving both authority and responsibilities that differ
    from other supporters of the measure.” Perry v. Brown,
    
    265 P.3d 1002
    , 1017–18 (Cal. 2011). Official proponents
    determine when the process will begin by filing the relevant
                 CHULA VISTA CITIZENS V. NORRIS                 17
    
    documents, Cal. Elec. Code § 9202(a), craft the text of the
    initiative that will be put before the people, id., ensure that
    the people know that the initiative process has commenced,
    id. § 9205(a)–(b), and exercise a measure of control over the
    arguments in favor of the initiative to which the people will
    be exposed, id. § 9287. Thus, the California Elections Code
    “place[s] an obligation upon the official proponents of an
    initiative measure to manage and supervise the process by
    which signatures for the initiative petition are obtained.”
    Perry, 265 P.3d at 1017. If public officials refuse to defend
    a successful initiative in court, official proponents may
    “intervene or [] participate as real parties in interest in a
    judicial proceeding to assert the state’s interest in the
    initiative’s validity and to appeal a judgment invalidating the
    measure.” Id. at 1025. But see Hollingsworth v. Perry,
    
    133 S. Ct. 2652
    , 2663–67 (2013) (holding that official
    proponents of California’s Proposition 8 lacked Article III
    standing in federal court). These rights and responsibilities
    are hardly consistent with the Associations’ minimalist
    characterization of official proponents.
    
         Perhaps most tellingly, unlike a lobbyist’s suggestion to
    a legislator, qualifying an initiative for the ballot is a
    necessary step for the people to exercise the initiative power.
    See Cal. Elec. Code § 9200 (authorizing municipal initiatives
    “pursuant to” the rules in the Elections Code); cf. Costa v.
    Superior Court, 
    128 P.3d 675
    , 685 (Cal. 2006) (discussing
    procedural challenges to ballot initiatives). In this critical
    respect, it is more like introducing legislation. Thus, by
    seeking the legal authority of official proponents, the
    Associations seek the legislative power of setting the
    initiative process in motion.
    18           CHULA VISTA CITIZENS V. NORRIS
    
                                  D
    
        We turn now to the question of whether serving as an
    official proponent, as we have described that status, is an
    aspect of the freedom of speech protected by the First
    Amendment.
    
        The Associations rely primarily on Meyer v. Grant,
    
    486 U.S. 414
     (1988). In Grant, Colorado forbade initiative
    proponents from employing paid petition circulators to gather
    signatures. Id. at 417. The Court held that “[t]he circulation
    of an initiative petition of necessity involves both the
    expression of a desire for political change and a discussion of
    the merits of the proposed change.” Id. at 421. Thus, it
    applied exacting scrutiny to the challenged ban. Id. at 420,
    428.
    
         As the district court astutely observed, Grant held that
    “advocation and circulation” of a petition is protected by the
    First Amendment, but no one disputed the legal status of the
    initiative proponents in Grant. Whether the activities of an
    official proponent are protected by the freedom of speech is
    a distinct question from whether serving as an official
    proponent (that is, having the legal authority attaching to
    official proponents) has the same protection. Thus, the issue
    presented by the Associations is unanswered by Grant.
    Indeed, it is one that neither the Supreme Court nor our
    circuit has decided.
    
        The Supreme Court has, however, addressed an analogous
    situation to the one presented in this case. In Nevada
    Commission on Ethics v. Carrigan, a state ethics law required
    public officers to recuse themselves from voting on matters
    in which they might reasonably be said to have a conflict of
                 CHULA VISTA CITIZENS V. NORRIS                   19
    
    interest. 
    131 S. Ct. 2343
    , 2346 (2011). Carrigan challenged
    the law, asserting that the First Amendment protected his
    right to vote in the city council. Id. at 2347.
    
        The Supreme Court held that “restrictions upon
    legislators’ voting are not restrictions upon legislators’
    protected speech.” Id. at 2350. Importantly, the Court cited
    the legislative nature of voting as the reason for its decision:
    “The Nevada Supreme Court thought a legislator’s vote to be
    protected speech because voting ‘is a core legislative
    function.’ We disagree, for the same reason.” Id. at 2347
    (internal citation omitted). The Court elaborated on this
    rationale: “[A] legislator’s vote is the commitment of his
    apportioned share of the legislature’s power to the passage or
    defeat of a particular proposal. The legislative power thus
    committed is not personal to the legislator but belongs to the
    people; the legislator has no personal right to it.” Id. at 2350.
    The Court went further and stated that “the act of voting [in
    a legislature] symbolizes nothing.” Id. Even if the legislative
    act of voting were expressive, the Court reasoned, the
    challenge would still fail because “[t]his Court has rejected
    the notion that the First Amendment confers a right to use
    governmental mechanics to convey a message.” Id. at 2351.
    
        Carrigan establishes that the legal authority attaching to
    a legislative office is not an aspect of the freedom of speech
    protected by the First Amendment. The Associations seek the
    legislative authority that comes with serving as official
    proponents. Following Carrigan, we conclude that serving
    20             CHULA VISTA CITIZENS V. NORRIS
    
    as an official proponent is not an aspect of speech within the
    meaning of the First Amendment.5
    
                                      E
    
        The Associations seem to think that because official
    proponents have authority to engage in expressive activities,
    such as the power to write the 500-word statement of reasons,
    the freedom of speech requires that they be permitted to be
    official proponents. But from the premise that certain
    activities are expressive, it does not follow that the legal
    authority to engage in such activities is part of the freedom of
    speech. This case presents that threshold issue: If serving as
    an official proponent is not part of the freedom of speech,
    then the expressive nature of official proponents’ activities is
    irrelevant.
    
        A contrary conclusion would produce absurd results. If
    the mere fact that an activity is expressive meant that there
    was a First Amendment right to engage in that activity,
    irrespective of the context in which the activity occurs, then
    the First Amendment would protect the right of any voter to
    participate in the debates of the state legislature. After all,
    such debates are highly expressive in nature. Yet, no one
    would maintain that the First Amendment prohibits limiting
    participation in such debates to members of the state
    legislature. Similarly, the exercise of an official proponents’
    authority, if expressive in nature, can be limited to those who
    qualify as official proponents. The First Amendment does
    
    
     5
       Carrigan’s holding was limited to the First Amendment. 131 S. Ct. at
    2350–51. We express no view here about whether the Associations’
    alleged right might be protected under other provisions of the Federal
    Constitution.
                  CHULA VISTA CITIZENS V. NORRIS                       21
    
    not require that associations be allowed to share in the
    legislative power simply because the exercise of such power
    might be expressive.
    
        The Supreme Court made this clear in Carrigan. In
    addition to upholding Nevada’s recusal law, the Court also
    upheld the recusal provision’s prohibition on advocacy. Id.
    at 2347. Because the recusal law was constitutional with
    respect to legislative voting on conflicted legislation, then it
    surely must also be the case, the Court reasoned, that the
    provision restricting who might advocate on that legislation
    was equally constitutional as a reasonable time, place, and
    manner restriction. Id. (citing Clark v. Cmty. for Creative
    Non-Violence, 
    468 U.S. 288
    , 293 (1984)). As the Supreme
    Court observed, “Legislative sessions would become massive
    town-hall meetings if those who had a right to speak were not
    limited to those who had a right to vote.” Id. So too here,
    any limit that the elector requirement might place on
    expression incidentally is a reasonable time, place, and
    manner restriction resulting from the initial, constitutional
    limitation on whom the people have designated to serve in
    this official role.
    
        As the Court emphasized in Carrigan, Doe v. Reed is
    consistent with Carrigan’s holding, id. at 2351, and it is
    consistent with the analysis here. Whereas Carrigan
    concerned whether the legal authority to exercise legislative
    power is protected by the freedom of speech, Doe concerned
    the extent to which the exercise of legislative power is
    protected.6 Doe did not analyze restrictions on who could
    
    
     6
       The concurrence claims that this manner of reconciling Carrigan and
    Doe departs from Supreme Court precedent, implying that its own
    approach is well-established in the U.S. Reports. See Concurrence at
    22              CHULA VISTA CITIZENS V. NORRIS
    
    sign initiative petitions; it discussed whether the signing of a
    petition was expressive.7 561 U.S. at 194–96. The
    Associations in this case seek the legal authority to exercise
    legislative power, which is why the analysis is governed by
    Carrigan.       Kneebone and Breitfelder, by contrast,
    undoubtedly have such authority, but they seek to exercise it
    in a certain way. Their challenge is governed by Doe.8 See
    infra Part IV.
    
        The challenge to the elector requirement asks whether the
    freedom of speech requires the people to delegate legislative
    power to associations, and Carrigan answers that it does not.9
    
    
    45–48.     Yet, other than the Supreme Court’s brief paragraph
    distinguishing Carrigan from Doe, see Carrigan, 131 S. Ct. at 2351, no
    federal court has described how Carrigan and Doe interact. Thus, any
    effort in this regard will break new ground, including that of the
    concurrence.
      7
         The concurrence is, therefore, quite wrong when it asserts that Doe
    controls the elector requirement analysis. The key question with regard
    to the elector requirement is whether California’s decision not to delegate
    legislative authority to associations violates the freedom of speech. Doe
    has nothing to say about that question.
      8
         This distinction between the legal authority to exercise legislative
    power and the exercise of such power explains why the dissent errs when
    it treats the challenges to the elector and petition-proponent disclosure
    requirements identically. See Dissent at 52. Only if we ignore Doe’s clear
    instruction, as the dissent would do, can we conclude that the legislative
    character of initiative petitions strips proponents of First Amendment
    protection.
      9
        Because we conclude that the elector requirement is constitutional,
    there is no need to resolve the parties’ dispute over whether the
    requirement is located in the California Elections Code, the City Charter,
    or some other source.
                    CHULA VISTA CITIZENS V. NORRIS                          23
    
                                        IV
    
       In their challenge to the petition-proponent disclosure
    requirement, Kneebone and Breitfelder contend that the
    compelled disclosure of their identities at the point of contact
    with signatories violates the freedom of speech.
    
                                        A
    
        The Supreme Court has never held that there is some
    “freewheeling right” to anonymity in the Constitution. Doe,
    561 U.S. at 218 n.4 (Stevens, J., concurring in part and
    concurring in judgment). Rather, the Court has said that the
    “decision to remain anonymous, like other decisions
    concerning omissions or additions to the content of a
    publication, is an aspect of the freedom of speech protected
    by the First Amendment.” McIntyre v. Ohio Elections
    Comm’n, 
    514 U.S. 334
    , 342 (1995). In the compelled
    disclosure context, the abridgment of the freedom of speech
    consists not in a violation of some amorphous “right to
    anonymity”; it consists in the “direct regulation of the content
    of speech,” id. at 345, or in the burden such disclosures place
    on speech by, for example, deterring the speaker from
    speaking, Buckley v. Valeo, 
    424 U.S. 1
    , 68 (1976); see also
    Citizens United, 558 U.S. at 480–83 (Thomas, J., concurring
    in part and dissenting in part). Our own precedent has
    
    
         The parties also dispute which standard of review applies to the
    elector requirement, but because we conclude that such requirement does
    not implicate the First Amendment, we do not proceed to resolve that
    question. See Ala. State Fed’n of Labor v. McAdory, 
    325 U.S. 450
    , 461
    (1945) (“It has long been [the Court’s] considered practice not to decide
    abstract, hypothetical or contingent questions, or to decide any
    constitutional question in advance of the necessity for its decision. . . .”
    (internal citations omitted)).
    24              CHULA VISTA CITIZENS V. NORRIS
    
    followed this basic framework. See ACLU v. Heller,
    
    378 F.3d 979
    , 987 (9th Cir. 2004) (describing the
    constitutional injury of compelled disclosure as the “direct
    regulation of the content of political speech”).
    
         We have never held that the content of a ballot initiative
    petition is part of an official proponent’s freedom of speech.
    The Supreme Court has recognized that the content of
    political handbills, McIntyre, 514 U.S. at 337–47, the speech
    of initiative petition circulators, Buckley v. Am. Constitutional
    Law Found., Inc., 
    525 U.S. 182
    , 197–200 (1999), and the
    signatures of initiative petition signatories, Doe, 561 U.S. at
    194–96, are protected speech, and thus the compelled
    disclosure of the speaker’s identity to the public constitutes
    a burden on such speech or a direct regulation thereof. But
    initiative petitions are official election documents, San
    Francisco Forty-Niners, 89 Cal. Rptr. 2d at 396, and the
    Court has not had occasion to consider whether the content of
    such documents constitutes protected speech.10
    
        However, because the parties to this litigation agree that
    the petition-proponent disclosure requirement is a regulation
    of political speech,11 we need not resolve that question. We
    
    
    
         10
            At least one of our sister circuits has implied that, in some
    circumstances, regulation of the content of initiative petitions would
    abridge the freedom of speech. Biddulph v. Mortham, 
    89 F.3d 1491
    , 1500
    (11th Cir. 1996) (per curiam) (“We obviously would be concerned about
    free speech and freedom-of-association rights were a state to enact
    initiative regulations that were content based or had a disparate impact on
    certain political viewpoints.”).
       11
          Norris limits her briefing to the elector requirement. California
    acknowledges that “an initiative petition is political speech” and states that
                    CHULA VISTA CITIZENS V. NORRIS                         25
    
    will assume—without deciding—that an official proponent’s
    decision to disclose his identity on the face of an initiative
    petition constitutes political speech, and under McIntyre, the
    compelled disclosure of such information is “a direct
    regulation of the content of speech” subject to First
    Amendment scrutiny. McIntyre, 514 U.S. at 345.
    
                                        B
    
       Of course, we must determine which standard of review
    governs our analysis of the petition-proponent disclosure
    requirement’s constitutionality.12
    
        The Supreme Court has “a series of precedents
    considering First Amendment challenges to disclosure
    requirements in the electoral context. These precedents have
    reviewed such challenges under what has been termed
    ‘exacting scrutiny.’” Doe, 561 U.S. at 196. “That standard
    
    
    “[t]here is no doubt that the challenged statutes, which govern the content
    of an initiative petition, trigger scrutiny under the First Amendment.”
     12
        Kneebone and Breitfelder bring both as-applied and facial challenges
    to the petition-proponent disclosure requirement. The nature of their
    argument, however, is a facial challenge: They claim that the requirement
    violates the freedom of speech no matter the identities or circumstances
    of the official proponents. To be sure, Kneebone and Breitfelder have
    asserted that they, in particular, have reasons for desiring to remain
    anonymous when serving as official proponents, but their arguments, if
    correct, preclude the idea that the requirement has a “plainly legitimate
    sweep” or that “circumstances exist under which [it] would be valid.”
    United States v. Stevens, 
    559 U.S. 460
    , 472–73 (2010) (internal quotation
    marks omitted). Because Kneebone and Breitfelder’s “claim and the relief
    that would follow . . . reach beyond the particular circumstances of these
    plaintiffs,” they must “satisfy our standards for a facial challenge to the
    extent of that reach.” Doe, 561 U.S. at 194.
    26           CHULA VISTA CITIZENS V. NORRIS
    
    requires a substantial relation between the disclosure
    requirement and a sufficiently important governmental
    interest.” Id. (internal quotation marks omitted). The
    “‘strength of the governmental interest must reflect the
    seriousness of the actual burden on First Amendment rights.’”
    Id. (quoting Davis, 554 U.S. at 744). Like the case before us,
    Doe considered the constitutionality of a law requiring the
    disclosure of identifying information—in that case, the
    identities of petition signatories. 561 U.S. at 190–95. The
    Court applied exacting scrutiny and upheld the law. Id. at
    197–202.
    
         Against this clearly articulated standard of review for
    compelled disclosure cases, Kneebone and Breitfelder argue
    that strict scrutiny should apply. They point to American
    Constitutional Law Foundation (“ACLF”) as an example of
    strict scrutiny employed in a disclosure context. The
    Supreme Court, however, has subsequently characterized the
    standard of review in ACLF as “exacting scrutiny.” See Doe,
    561 U.S. at 196. Moreover, it was precisely because the
    Court did not apply strict scrutiny that Justice Thomas wrote
    separately in ACLF. See 525 U.S. at 214–15 (Thomas, J.,
    concurring in the judgment). Thus, nothing in ACLF
    provides a basis for applying strict scrutiny to the petition-
    proponent disclosure requirement.
    
         California makes no effort to distinguish Doe. Rather, it
    simply asserts that public forum doctrine should govern our
    analysis. But the state points to no federal case that has
    adopted such an approach. California might instead have
    argued that the petition-proponent disclosure requirement
    relates to the “mechanics of the electoral process,” thus
    subjecting it to the potentially less-demanding “ordinary
    litigation test.” See McIntyre, 514 U.S. at 344–45. But Doe
                  CHULA VISTA CITIZENS V. NORRIS                   27
    
    forecloses that option. The Doe Court, faced with the
    compelled disclosure of signatories’ identities, rejected the
    argument that the legislative character of initiative petitions
    mandated a lesser form of scrutiny. 561 U.S. at 194–96. If
    exacting scrutiny applies to the compelled disclosure of
    signatories’ identities, there is no reason why official
    proponents’ identities should not receive the same protection.
    
        We therefore adhere to the Supreme Court’s “series of
    precedents” regarding compelled disclosure by subjecting the
    petition-proponent disclosure requirement to exacting
    scrutiny. Id. at 196; see also Wash. Initiatives Now (WIN) v.
    Rippie, 
    213 F.3d 1132
    , 1138–39 (9th Cir. 2000) (applying
    exacting scrutiny to a law compelling the disclosure of
    circulators’ identities, addresses, and compensation).
    
                                    C
    
        It remains for us to determine whether the petition-
    proponent disclosure requirement survives exacting scrutiny.
    In addressing this question, it is important to bear in mind
    that the statutory scheme, as incorporated by the City Charter,
    requires proponents to disclose their identities at three distinct
    moments in the initiative process: the filing of a signed notice
    with the City Clerk, Cal. Elec. Code § 9202(a), the
    publication of the notice in a newspaper of general
    circulation, id. § 9205, and the inclusion of the notice on each
    section of the circulated initiative petitions, id. §§ 9202(a),
    9207. Kneebone and Breitfelder only challenge the last
    requirement.
    
        The Supreme Court has described exacting scrutiny as a
    “strict test.” Buckley, 424 U.S. at 66. Although distinct from
    strict scrutiny, “exacting scrutiny is more than a rubber
    28            CHULA VISTA CITIZENS V. NORRIS
    
    stamp.” Minn. Citizens Concerned for Life, Inc. v. Swanson,
    
    692 F.3d 864
    , 876 (8th Cir. 2012). Indeed, “[t]he Supreme
    Court has not hesitated to hold laws unconstitutional under
    this standard.” Id. (collecting cases). As Buckley made clear,
    it is not enough for the state to have “some legitimate
    governmental interest”; the Court “also ha[s] insisted that
    there be a . . . ‘substantial relation’ between the governmental
    interest and the information required to be disclosed.” 424
    U.S. at 64. Moreover, it is the government’s burden to “show
    that its interests . . . are substantial, that those interests are
    furthered by the disclosure requirement, and that those
    interests outweigh the First Amendment burden the disclosure
    requirement imposes on political speech.” WIN, 213 F.3d at
    1138–39; see also Ctr. for Individual Freedom, Inc. v.
    Tennant, 
    706 F.3d 270
    , 282 (4th Cir. 2013); Minn. Citizens
    Concerned for Life, Inc., 692 F.3d at 877. Thus, the mere
    assertion of a connection between a vague interest and a
    disclosure requirement is insufficient.
    
         California asserts two interests in the petition-proponent
    disclosure requirement: (1) informing electors of an official
    proponent’s identity, and (2) “preserving the integrity of the
    electoral process.” Quoting Doe, the state claims that the
    latter interest “extends more generally to promoting
    transparency and accountability in the electoral process.”
    Doe, 561 U.S. at 198. The district court relied on both
    interests in sustaining the petition-proponent disclosure
    requirement.
    
                                    1
    
        California contends that the public has a right to know the
    identities of official proponents because an initiative is
    analogous to the introduction of legislation, and therefore “it
                 CHULA VISTA CITIZENS V. NORRIS                   29
    
    is no different from the requirement that every bill in the
    California Legislature be introduced by a member of the
    Legislature.” California believes that “[l]egislation is
    inherently a public act, regardless of the forum in which it
    takes place.” The district court agreed, relying on two of our
    cases that stressed the need for voters to know the identities
    of those participating in initiative campaigns. See Human
    Life of Wash., Inc. v. Brumsickle, 
    624 F.3d 990
     (9th Cir.
    2010); Cal. Pro-Life Council v. Getman, 
    328 F.3d 1088
     (9th
    Cir. 2003).
    
        Even assuming that this interest is sufficiently important
    to satisfy exacting scrutiny, California must demonstrate that
    the interest bears a substantial relation to the petition-
    proponent disclosure requirement. Kneebone and Breitfelder
    argue that because proponents must disclose their identities
    at two distinct moments before circulating a petition, any
    member of the public who wishes to learn the identities of
    official proponents can do so, and there is no need for
    disclosure on the face of the petition. California also cites the
    notice-filing and publication requirements, but it argues that
    these prior disclosures cut the other way: “[B]y the time
    proponents’ names are printed on initiative petitions, their
    identities are already known—the impact on proponents’
    privacy is negligible because their names have already been
    published in a newspaper of general circulation.”
    
        The precedents of the Supreme Court and this circuit have
    emphasized the importance of anonymity at the point of
    contact with voters. McIntyre v. Ohio Elections Commission
    established that “an author’s decision to remain anonymous,
    like other decisions concerning omissions or additions to the
    content of a publication, is an aspect of the freedom of speech
    protected by the First Amendment.” 514 U.S. at 342. In the
    30           CHULA VISTA CITIZENS V. NORRIS
    
    realm of political speech, anonymity is important because it
    “provides a way for a writer who may be personally
    unpopular to ensure that readers will not prejudge her
    message simply because they do not like its proponent.” Id.
    The Court therefore applied exacting scrutiny and struck
    down an Ohio statute requiring authors of any “form of
    general publication which is designed . . . to influence the
    voters in any election” to disclose their identities. Id. at 338
    n.3, 345–46.
    
        The Court extended McIntyre’s holding in ACLF. In that
    case, the Court applied exacting scrutiny to invalidate a
    Colorado requirement that petition circulators wear badges
    disclosing their identities at the point of contact with
    signatories, and it contrasted this invalid rule with the
    requirement that those same circulators submit affidavits to
    the state containing their names, addresses, and signatures:
    “Unlike a name badge worn at the time a circulator is
    soliciting signatures, the affidavit is separated from the
    moment the circulator speaks.” 525 U.S. at 198. The Court
    saw this separation in time as important because revealing
    one’s identity at the point of contact with signatories
    “operates when reaction to the circulator’s message is
    immediate and may be the most intense, emotional, and
    unreasoned.” Id. at 199 (internal quotation marks omitted).
    
        The Court observed that, when a circulator makes contact,
    “the circulator must endeavor to persuade electors to sign the
    petition,” id., a concern expressed in McIntyre’s statement
    that “an advocate may believe her ideas will be more
    persuasive if her readers are unaware of her identity,”
    514 U.S. at 342. For that reason, ACLF held that “the badge
    requirement compels personal name identification at the
    precise moment when the circulator’s interest in anonymity
                 CHULA VISTA CITIZENS V. NORRIS                  31
    
    is greatest.” 525 U.S. at 199. By contrast, the affidavit
    requirement was “responsive to the State’s concern” for
    providing the identifying information to the public, but it did
    so without interfering with the point of contact. Id. at 198.
    Thus, there was not a sufficient governmental interest to
    justify the badge requirement. Id. at 200.
    
        Our decision in WIN v. Rippie followed a similar chain of
    reasoning. WIN challenged a Washington law that compelled
    the disclosure of petition circulators’ identities, addresses,
    and compensation before and after an election. WIN,
    213 F.3d at 1134–35. These disclosures were “routinely filed
    during the circulation period,” which we said created a
    chilling effect on speech. Id. at 1138–39. Applying exacting
    scrutiny, we struck down the disclosure requirement. Id. at
    1140. Central to our holding was our judgment that the
    “interest in educating voters through campaign finance
    disclosure is more adequately served by a panoply of the
    State’s other requirements that have not been challenged.”
    Id. at 1139. Like ACLF, WIN illustrates that, where
    alternative means of furthering the state’s interest are
    available, it will be very difficult for a compelled disclosure
    law to survive exacting scrutiny.
    
        Heller remains our clearest articulation of the principles
    underlying McIntyre, ACLF, and WIN. In Heller, we
    invalidated a Nevada law that required “certain groups or
    entities publishing any material or information relating to an
    election, candidate or any question on a ballot to reveal on the
    publication the names and addresses of the publications’
    financial sponsors.” 378 F.3d at 981 (internal quotation
    marks omitted). Our holding rested on “[t]he constitutionally
    determinative distinction between on-publication identity
    disclosure requirements and after-the-fact reporting
    32           CHULA VISTA CITIZENS V. NORRIS
    
    requirements” that we said “has been noted and relied upon
    both by the Supreme Court and by this Circuit.” Id. at 991.
    We said ACLF stands for the following proposition: “[I]t is
    not just that a speaker’s identity is revealed, but how and
    when that identity is revealed, that matters in a First
    Amendment analysis of a state’s regulation of political
    speech.” Id. (emphasis added) (citing WIN, 213 F.3d at
    1138). For that reason, “requiring a publisher to reveal her
    identity on her election-related communication is
    considerably more intrusive than simply requiring her to
    report to a government agency for later publication how she
    spent her money.” Id. at 992. Because the Nevada law
    required the speaker to disclose her identity on the face of the
    election-related communication, we held the state’s asserted
    interests were inadequate to justify the burden on speech. Id.
    at 1002.
    
        In all of these precedents, the Supreme Court and this
    circuit have taken the view that “[t]he injury to speech is
    heightened” when speakers are compelled to disclose their
    identities “at the same time they deliver their political
    message.” ACLF, 525 U.S. at 199 (internal quotation marks
    omitted). Such is the case here, where the petition-proponent
    disclosure requirement forces official proponents to reveal
    their identities on the face of the petition. Forced disclosures
    of this kind are “significant encroachments on First
    Amendment rights.” Buckley, 424 U.S. at 64.
    
        These precedents also make clear that, where there are
    alternative methods of meeting the government’s asserted
    interests, the government’s task of justifying a compelled
    disclosure law becomes much more onerous. See ACLF,
    525 U.S. at 198–99; WIN, 213 F.3d at 1139. California
    contends that voters have an interest in knowing the identities
                   CHULA VISTA CITIZENS V. NORRIS                        33
    
    of official proponents, but such identities are already
    disclosed on two occasions before petition circulation can
    begin. Proponents must disclose their identities to the City
    Clerk when they file the notice of intent, and the Clerk must
    provide copies of the notice to “any person upon request.”
    Cal. Elec. Code §§ 9202(a), 9202.513. Additionally, there is
    the publication requirement. Id. § 9205(a)–(b). Voters who
    wish to know the identities of official proponents need only
    make a trip to the City Clerk’s office or search for the
    publication of the petition in their newspapers of general
    circulation.
    
        Like ACLF and McIntrye, the statutory scheme here
    “compels personal name identification at the precise moment
    when the [speaker’s] interest in anonymity is greatest.”
    ACLF, 525 U.S. at 199. Like Heller, the disclosure
    requirement in this case implicates the “constitutionally
    determinative distinction between on-publication identity
    disclosure requirements and [before-or-] after-the-fact
    reporting requirements.” 378 F.3d at 991. Like ACLF and
    WIN, there are alternative means of disclosure that are
    “responsive to the [public’s] concern” in knowing the
    identities of those involved in the initiative process. ACLF,
    525 U.S. at 198.          Under these circumstances, the
    informational interest does not bear a substantial relation to
    the petition-proponent disclosure requirement and fails
    exacting scrutiny.
    
    
    
    
     13
        Section 9202.5 was enacted by the California legislature in 2012 and
    took effect on January 1, 2013. See Cal. Elec. Code § 9202.5 (West
    2013).
    34           CHULA VISTA CITIZENS V. NORRIS
    
                                   2
    
        California also asserts an interest in maintaining the
    integrity of the electoral process.         Doe sustained a
    Washington disclosure law on the basis of a similar interest,
    see 561 U.S. at 197–98, and we will assume that the same
    interest is sufficiently important for purposes of this case.
    
        California provides no explanation for how its interest in
    the integrity of the electoral process relates to the petition-
    proponent disclosure requirement. It simply asserts the
    interest. The district court elaborated on the nature of this
    interest: “By requiring a proponent’s name to appear on the
    circulated copy of the ballot initiative, the local voters who
    consider the initiative may recognize whether the proponent
    qualifies as an elector.” The district court appeared to be
    saying that an anti-fraud interest underlay the petition-
    proponent disclosure requirement, an interest the Supreme
    Court found sufficiently important in Doe. Id.
    
        If the state is concerned about fraudulent proponents, as
    the district court suspected, it can protect against that
    possibility using the unchallenged disclosure requirements.
    See Cal. Elec. Code §§ 9202, 9202(a), 9205(a)–(b). At each
    of these stages, elections officials or the interested public can
    verify proponents’ qualifications. In Doe, Washington
    demonstrated that the existence of measures other than the
    disclosure requirement at issue did not alleviate the
    possibility of fraud and voter error. See, e.g., 561 U.S. at 198
    (pointing out that “the secretary’s verification and canvassing
    will not catch all invalid signatures”). It is California’s
    burden to show that the alternative methods of satisfying its
    anti-fraud goal are insufficient. WIN, 213 F.3d at 1138–39.
                    CHULA VISTA CITIZENS V. NORRIS                           35
    
    Not only has it failed to carry its burden; it has not even
    attempted to do so.
    
        California claims that, as was the case with Washington
    in Doe, its “interest in preserving electoral integrity is not
    limited to combating fraud.” 561 U.S. at 198. Rather, the
    interest “extends more generally to promoting transparency
    and accountability in the electoral process.” Id. California
    has not shown how the petition-proponent disclosure
    requirement serves that interest or why the alternative
    disclosure requirements are inadequate, relying instead on the
    bare pronouncement of its interest. That is insufficient to
    satisfy exacting scrutiny. See WIN, 213 F.3d at 1138–39; Ctr.
    for Individual Freedom, Inc., 706 F.3d at 282; Minn. Citizens
    Concerned for Life, Inc., 692 F.3d at 877.
    
                                         D
    
        A few responses to the dissent are in order. The dissent
    acknowledges that Supreme Court and Ninth Circuit
    precedent is deeply skeptical of compelled disclosure
    requirements like the one challenged in this case, but it claims
    that these cases “d[o] not apply here.” Dissent at 60. The
    dissent seems to argue that official proponents, when acting
    in their official capacity, have no right to speak anonymously
    during the initiative process due to the public nature of their
    office and the legislative character of an initiative petition.14
    
       14
           The dissent also implies that, because it relies exclusively on
    California’s asserted interest in the integrity of the electoral process to
    sustain the petition-proponent disclosure requirement, the doctrine of
    McIntyre, ACLF, Heller, and WIN is inapplicable. Dissent at 60. But Doe
    applied the doctrine of those cases in evaluating the constitutionality of
    Washington’s compelled disclosure law, even though Doe, like the
    dissent, relied exclusively on the interest in the integrity of the electoral
    36               CHULA VISTA CITIZENS V. NORRIS
    
    Id. Not a single precedent of this Court or of the Supreme
    Court has ever relied on such distinctions to sustain a
    compelled disclosure requirement, and the dissent does not
    cite any.15
    
        More fundamentally, it is incoherent for the dissent to
    deny that official proponents have no right to speak
    anonymously while simultaneously applying a form of
    scrutiny designed to safeguard that very right. See Dissent at
    55–56 (agreeing that exacting scrutiny applies). Compelled
    disclosure requirements are constitutionally suspect because
    the “decision to remain anonymous, like other decisions
    concerning omissions or additions to the content of a
    publication, is an aspect of the freedom of speech protected
    by the First Amendment.” McIntyre, 514 U.S. at 342. But if
    the dissent denies that the right to remain anonymous is an
    aspect of official proponents’ freedom of speech, then the
    petition-proponent disclosure requirement is not a “regulation
    of the content of speech,” id. at 345, and there is no reason to
    apply exacting scrutiny. See Doe, 561 U.S. at 219–21
    (Scalia, J., concurring in the judgment) (stating that the
    majority applied exacting scrutiny after finding that petition
    
    
    process. 561 U.S. at 195–202. The dissent cannot avoid the clear
    instructions of McIntyre, ACLF, Heller, and WIN merely by invoking the
    interest in electoral integrity.
      15
         To the extent the dissent relies on the fact that an initiative petition is
    a legislative document, Doe forecloses such argument. The signatories in
    Doe, no less than the official proponents in this case, were introducing
    legislation; their signatures were necessary for the petition to qualify for
    the ballot. In that case, Washington raised precisely the same argument
    the dissent makes here: because signatories are engaging in a “legally
    operative legislative act,” they were not entitled to the same level of First
    Amendment protection as they would be in other contexts. Doe, 561 U.S.
    at 195. The Supreme Court rejected that rationale. Id. at 195–96.
                   CHULA VISTA CITIZENS V. NORRIS                         37
    
    signatories have a right to anonymous political expression);
    Church of Am. Knights of the Ku Klux Klan v. Kerik,
    
    356 F.3d 197
    , 208–09 (2d Cir. 2004) (declining to apply
    exacting scrutiny “[b]ecause . . . plaintiffs’ right to
    anonymous speech is not implicated here”). In short, the
    dissent agrees we must apply a level of scrutiny that is only
    appropriate if official proponents have some right to speak
    anonymously, yet it denies that they have such a right. This
    confusion, we suggest, undermines the dissent’s analysis.
    
        The dissent’s confusion is compounded by its failure to
    resolve the fundamental problem with California’s argument:
    Even assuming that California has an important interest in
    forcing official proponents to disclose their identities, why
    must that disclosure occur on the face of initiative petitions?
    Kneebone and Breitfelder do not disagree that official
    proponents undertake duties and responsibilities that require
    disclosing proponents’ identities to the public. After all, they
    do not challenge the notice-filing and publication
    requirements. They simply wish to remain anonymous at the
    point of contact with voters. The dissent never explains why
    a particular form of disclosure—the petition-proponent
    disclosure requirement—is substantially related to the state’s
    interest when there are alternative, unchallenged means of
    disclosure.16
    
    
    
    
      16
         Significantly, California makes no effort to show why the petition-
    proponent disclosure requirement is needed given that the public can learn
    official proponents’ identities through the notice-filing and publication
    requirements. Even if the dissent were able to articulate such a
    justification, the burden is on California—not members of this Court—to
    do so. WIN, 213 F.3d at 1138–39; Ctr. for Individual Freedom, Inc.,
    706 F.3d at 282; Minn. Citizens Concerned for Life, Inc., 692 F.3d at 877.
    38           CHULA VISTA CITIZENS V. NORRIS
    
         Instead, the dissent merely asserts that the unchallenged
    disclosure provisions would “fail to satisfy the government’s
    interest in any meaningful or realistic sense.” Dissent at 62.
    The dissent may very well think that, but Supreme Court
    precedent is to the contrary. Any voter who wants to know
    the identities of official proponents before signing a petition
    can find out by visiting the City Clerk’s office or looking up
    the identities in the newspaper of general circulation, Cal.
    Elec. Code § 9202(a),§ 9202.5, § 9205, a far more accessible
    means of gaining information than Colorado voters had
    available to them after ACLF. Moreover, in ACLF, it was far
    less convenient for Colorado voters to seek out the affidavits
    of petition circulators than it was to have the circulators wear
    identification badges, but the Court nonetheless invalidated
    the badge requirement. 525 U.S. at 198–200. Perhaps the
    dissent has an explanation for why Colorado voters should
    have to work harder than California voters when evaluating
    initiative petitions, but we cannot fathom what it might be.
    
        The petition-proponent disclosure requirement does not
    satisfy exacting scrutiny.
    
                                   E
    
        The petition-proponent disclosure requirement is
    unconstitutional. Unlike the challenge to the elector
    requirement, none of the parties assert that the petition-
    proponent disclosure requirement exists apart from the state
    elections code. Thus, §§ 9202 and 9207 of the California
    Elections Code are invalid to the extent that they require
    official initiative proponents to identify themselves on the
    face of initiative petitions.
                 CHULA VISTA CITIZENS V. NORRIS                  39
    
                                   V
    
         We affirm the district court’s grant of summary judgment
    to the defendants as to the elector requirement, but we reverse
    its grant of summary judgment to the defendants as to the
    petition-proponent disclosure requirement. We therefore
    reverse the district court’s denial of summary judgment to the
    plaintiffs as to the petition-proponent disclosure requirement
    and remand so that it can enter an injunction consistent with
    this opinion. The parties shall bear their own costs.
    
      AFFIRMED in part, REVERSED in part, and
    REMANDED.
    
    
    
    BEA, Circuit Judge, concurring:
    
        I join all of the majority opinion except for Section III,
    which analyzes the local and state requirements that official
    ballot initiative proponents be electors, thereby excluding
    associations. Although I concur in the result of Section III,
    I see the issue differently. I believe that the majority opinion
    employs an incorrect test to determine whether acting as an
    official ballot initiative proponent is a legislative act with
    expressive content or an expressive act with legislative effect,
    and thus whether the elector requirement burdens any First
    Amendment rights. The test should not be whether a
    particular act has legislative effect or legislative character.
    The correct test, as stated in Doe and Carrigan, asks whether
    the individual—here the ballot initiative proponent—is
    exercising his own power, as does a ballot initiative signatory
    (Doe), or is exercising a governmental power that has been
    democratically apportioned to him, as does a legislator voting
    40            CHULA VISTA CITIZENS V. NORRIS
    
    in a legislature (Carrigan). When I apply this test I conclude
    that acting as a ballot initiative proponent is an expressive act,
    despite its legislative effect, and is protected by the First
    Amendment. Therefore, the elector requirement implicates
    the First Amendment and must be so analyzed. Nonetheless,
    I conclude the elector requirement satisfies at least the
    exacting scrutiny test applied to state electoral regulations
    under the First Amendment, and therefore I concur with the
    opinion’s conclusion upholding the elector requirement.
    
                                    I
    
                                    A
    
         The majority opinion incorrectly identifies what is the
    precise right sought to be vindicated by the appellants. In its
    Section III.B, the majority opinion suggests that the
    appellants seek “the legal authority attaching to the status of
    an official proponent.” Maj. Op. at 14. No, the precise right
    the appellants want is not merely an abstract “authority.” Put
    more simply and precisely, the appellants wish “to be
    initiative proponents.” Blue Br. at 6. Becoming an official
    ballot initiative proponent produces a particular quality of
    speech. When someone is qualified as an official proponent,
    he is able to speak from a particular vantage as an author of
    the proposal. The quality and impact of speech by official
    ballot initiative proponents is different from that of a mere
    member of the public. “Proponents” can be seen as
    sufficiently civic-minded to have taken the time and borne the
    cost to bring up a measure. They have also taken the risk of
    being identified with a particular political view. Because of
    their demonstrated interest and willingness to risk
    opprobrium, moreover, their speech is likely to carry greater
    weight once their position as ballot proponents is recognized.
                     CHULA VISTA CITIZENS V. NORRIS                             41
    
    By restricting who may serve in this role, Chula Vista and
    California restrict the range of speech non-electors and
    associations can exercise.
    
        The majority opinion also errs in holding that being an
    official ballot initiative proponent is a legislative act. The
    majority opinion holds that being an official ballot initiative
    proponent is legislative at core because California state cases
    and laws describe the ballot initiative proponent as akin to a
    legislator, and because the acts of a ballot initiative proponent
    are legislative in character and effect.
    
        With respect, this test is mistaken. The Supreme Court
    has explicitly held that expressive acts can still be protected
    by the First Amendment, even if they have a legislative effect
    under the state’s laws. Doe v. Reed, 
    561 U.S. 186
    , 194–96
    (2010).1 Indeed, if having legislative effect could deprive an
    expressive act of First Amendment protection, then
    
    
      1
        It should be noted that the respondent in Doe v. Reed made the same
    argument that the majority opinion does here, namely that acting as a
    ballot initiative signatory was a legislative act not protected by the First
    Amendment because the Supreme Court of Washington had declared it to
    be so. As the respondent’s brief before the Supreme Court in Doe stated,
    “The Washington Supreme Court has stated that the ‘exercise of the
    initiative power is an exercise of the reserved power of the people to
    legislate’ and that ‘[in] approving an initiative measure, the people
    exercise the same power of sovereignty as the legislature does when
    enacting a statute.’ Amalgamated Transit Union Local 587 v. State,
    
    11 P.3d 762
    , 779 (Wash. 2000).” Brief of Respondent at 24, Doe v. Reed,
    
    561 U.S. 186
     (2010) (No. 09–559), 
    2010 WL 1250504
    . The Supreme
    Court in Doe, however, did not even address this argument, and thus did
    not find it dispositive; nor should this panel. If acting as a ballot initiative
    signatory is still a First Amendment activity despite its characterization as
    a “legislative act” by state law, then this same logic should apply to acting
    as a ballot initiative proponent.
    42           CHULA VISTA CITIZENS V. NORRIS
    
    governments could limit political speech by granting
    legislative effect to particular speech acts. The majority
    opinion ignores the test the Supreme Court itself has already
    laid out in Doe and Carrigan to determine whether a
    particular political activity is speech protected by the First
    Amendment or is a legislative act with no First Amendment
    protection.
    
        The Doe Court stated the general rule that “[a]n
    individual express[ing] a view on a political matter” when
    participating in a political activity, whether or not it also
    happens to have a governmental effect. Id. at 194–95. The
    Court held that “sign[ing] a petition under Washington’s
    referendum procedure” was an example of an individual
    expressing a view on a political matter. Id. If the activity
    expresses a political view, then generally it “implicates a First
    Amendment right.” Id. at 195. Moreover, the First
    Amendment still protects that expressive activity even if it
    has “legal effect in the electoral process”:
    
            [S]igning a referendum petition may
            ultimately have the legal consequence of
            requiring the secretary of state to place the
            referendum on the ballot. But we do not see
            how adding such legal effect to an expressive
            activity somehow deprives that activity of its
            expressive component, taking it outside the
            scope of the First Amendment.
    
    Id.
    
        Carrigan created an exception to this general rule. When
    an individual performs a legislative act through his office as
                 CHULA VISTA CITIZENS V. NORRIS                 43
    
    legislator—such as a legislator voting in a legislature—that
    act is not protected by the First Amendment because:
    
           a legislator’s vote is the commitment of his
           apportioned share of the legislature’s power to
           the passage or defeat of a particular proposal.
           The legislative power thus committed is not
           personal to the legislator but belongs to the
           people; the legislator has no personal right to
           it.
    
    Nevada Comm’n on Ethics v. Carrigan, 
    131 S. Ct. 2343
    , 2345
    (2011). In other words, a legislative act is one that an
    individual performs not as an individual or as a “principal,”
    but as an elected representative, or an “agent” of the people,
    i.e. “a governmental act [an individual performs] as a
    representative of his constituents.” Id. at 2351 n.5.
    
        Carrigan stated that it was not in tension with Doe. Id. at
    2351. The Carrigan Court described Doe as holding that
    “core political speech,” such as signing a ballot initiative
    petition, “was not deprived of its protected status simply
    because, under state law, a petition that garnered a sufficient
    number of signatures would suspend the state law to which it
    pertained, pending a referendum.” Id. Such political speech,
    the Carrigan Court held, was distinguishable from the
    legislative, governmental act of a legislator voting in a
    legislature: “It is one thing to say that an inherently
    expressive act remains so despite its having governmental
    effect, but it is altogether another thing to say that a
    governmental act becomes expressive simply because the
    governmental actor wishes it to be so.” Id.
    44            CHULA VISTA CITIZENS V. NORRIS
    
        Likewise, even if an act is a legally effective part of the
    political process, it still may be an expressive act protected by
    the First Amendment. “[T]he State, having chosen to tap the
    energy and the legitimizing power of the democratic process,
    must accord the participants in that process the First
    Amendment rights that attach to their roles.” Doe, 561 U.S.
    at 195 (internal quotation marks, ellipsis, and brackets
    omitted). Only if the act done is authorized by a
    governmental, democratically apportioned representative
    power does it cease to be a personal act and thereby cease to
    be protected by the First Amendment.
    
         Thus, when a court must decide whether a political act is
    inherently expressive and therefore protected by the First
    Amendment or is a legislative, governmental act and
    therefore not at all protected by the First Amendment, the
    court must determine whether the act is more similar to the
    act in Doe (signing a ballot initiative) or the act in Carrigan
    (a legislator voting in a legislature). The relevant question is
    whether the politically expressive act is personal to the actor,
    or is merely exercised by him through democratic delegation
    of governmental power to him and therefore does not belong
    to him.
    
         Applied here, this test reveals that the act of being a ballot
    initiative proponent, just like the signing of a ballot initiative
    proposal in Doe, is at its core expressive, not legislative,
    despite its legislative effect. An official ballot initiative
    proponent is neither an elected nor democratically appointed
    position. The proponent has volunteered to exercise his own
    power vested in him by California state law, Cal. Elec. Code
    § 9202(a). He does not exercise a power that has been
    democratically apportioned to him as an “agent” of a
    governmental body. Therefore, the Chula Vista and
                 CHULA VISTA CITIZENS V. NORRIS                   45
    
    California elector requirements burden the First Amendment
    rights of those who desire to become official ballot initiative
    proponents. In other words, being a ballot initiative
    proponent is more like being a ballot initiative signatory
    (Doe) than it is like a legislator voting in a legislature
    (Carrigan).
    
        It may be argued that the electors of California, or those
    of Chula Vista, have “apportioned” to themselves the right to
    act as ballot initiative proponents, and thus that becoming the
    proponent of an initiative is the legislative act of an agent of
    the people. This argument would fail, however, because it is
    equally applicable to the ballot initiative signatories in Doe.
    The electors of Washington state “apportioned” to themselves
    the right to act as ballot initiative signatories just as the
    electors of California “apportioned” to themselves the right
    to act as ballot initiative proponents. The Supreme Court
    holding in Doe, however, shows that this level of
    “apportionment” does not convert the act from a personal,
    expressive one to a delegated, representative one. Therefore,
    when analyzed under the correct Doe-Carrigan framework,
    acting as an official ballot initiative proponent is a personal
    act of political expression that happens to have legislative
    effect, similar to acting as a ballot initiative signatory; it is
    not the governmental act of a democratically appointed agent
    of the people. The First Amendment therefore applies to any
    governmental restraints on such political expression.
    
                                   B
    
        Instead of applying the Doe-Carrigan framework to
    determine whether the activity at issue is protected by the
    First Amendment, the majority opinion applies its own test.
    It distinguishes between “the activities of an official
    46               CHULA VISTA CITIZENS V. NORRIS
    
    proponent” and “serving as an official proponent (that is,
    having the legal authority attaching to official proponents).”
    Maj. Op. at 18. As the majority opinion puts it, just because
    “certain activities are expressive, it does not follow that the
    legal authority to engage in such activities is part of the
    freedom of speech. This case presents that threshold issue.”2
    Maj. Op. at 20. Therefore, the majority opinion concludes,
    this case is controlled by Carrigan, and not by Doe:
    “Carrigan concerned whether the legal authority to exercise
    legislative power is protected by the freedom of speech,
    
    
      2
        The majority opinion suggests that “[a] contrary conclusion would
    produce absurd results.” As the majority opinion explains:
    
             If the mere fact that an activity is expressive meant that
             there was a First Amendment right to engage in that
             activity, irrespective of the context in which the activity
             occurs, then the First Amendment would protect the
             right of any voter to participate in the debates of the
             state legislature. After all, such debates are highly
             expressive in nature. Yet, no one would maintain that
             the First Amendment prohibits limiting participation in
             such debates to members of the state legislature.
    
    Maj. Op. at 20. This hypothetical reveals the error in the majority
    opinion’s reasoning. No one would argue that the calculus for
    determining whether the First Amendment protects a particular speech act
    is whether that act is expressive; this argument slays a mere straw man.
    Nor is legislative effect dispositive. As discussed above, the act in
    Carrigan was expressive in nature (and unprotected by the First
    Amendment) and the act in Doe was legislative in effect (and protected by
    the First Amendment). Neither fact was determinative. Instead, the
    correct test is whether the speech act is one that belongs to the individual
    and thus is constitutionally protected or one that stems from a
    governmental authority that has been democratically apportioned and thus
    can be limited by the entity which granted the authority. Acting as a
    legislator is a legislative activity. Acting as a ballot initiative signatory is
    a First Amendment activity.
                 CHULA VISTA CITIZENS V. NORRIS                   47
    
    [while] Doe concerned the extent to which the exercise of
    legislative power is protected.” Maj. Op. at 21.
    
        In other words, according to the majority opinion, if a
    given act is considered a speech act at core, as opposed to a
    legislative act, and thus is protected by the First Amendment,
    then governmental limits on how that speech is to be
    exercised would be analyzed under constitutional standards,
    but the First Amendment would provide no protection against
    governmental limits on who can exercise that speech. I see
    no constitutional basis for the distinction between possessing
    a right and exercising that right. Moreover, the majority
    opinion’s approach could allow governments to control the
    content of political speech by restricting the access to
    particular forms of political speech , and to avoid having
    those restrictions analyzed under the First Amendment.
    
        Indeed, Citizens United refutes the majority opinion’s
    “threshold question” argument. The majority opinion here
    states that, because the state did not grant associations the
    right to be official ballot initiative proponents, the panel need
    not analyze the law under the First Amendment. Citizens
    United, on the other hand, examined the constitutionality of
    a statute and found it unconstitutional precisely because
    associations were excluded from participating in a particular
    activity. Citizen United v. FEC, 
    558 U.S. 310
     (2010). In
    Citizens United, 2 U.S.C. § 441b prevented associations from
    making expenditures in connection to an election to public
    office. The Supreme Court examined the “threshold
    question” that the majority opinion here discusses and held
    that “[p]remised on mistrust of governmental power, the First
    Amendment stands against attempts to disfavor certain
    subjects or viewpoints. Prohibited, too, are restrictions
    distinguishing among different speakers, allowing speech by
    48           CHULA VISTA CITIZENS V. NORRIS
    
    some but not others.” Id. (internal quotation marks and
    citations omitted). “We find no basis,” the Court went on,
    “for the proposition that, in the context of political speech,
    the Government may impose restrictions on certain
    disfavored speakers.” Id. at 341. Thus, even the “threshold
    question” of who gets to participate in a particular speech act
    must be analyzed under the First Amendment.
    
        Finally, under Citizens United, associations have the right
    to assert these First Amendment protections. Id. at 365
    (“[T]he Government may not suppress political speech on the
    basis of the speaker’s corporate identity.”).
    
                                  II
    
        Because the majority opinion concludes that the elector
    requirement does not burden the First Amendment at all, it
    does not address what scrutiny should be applied to determine
    its constitutionality. Because I conclude that the elector
    requirement does burden the First Amendment, however, I
    now address the proper scrutiny to be applied.
    
        Under Citizens United, it would seem at first blush that,
    because the “elector requirement” burdens political speech on
    the basis of the identity of the speaker, it should be analyzed
    under strict scrutiny. See Citizens United, 558 U.S. at 340
    (“Laws that burden political speech are subject to strict
    scrutiny, which requires the Government to prove that the
    restriction furthers a compelling interest and is narrowly
    tailored to achieve that interest.” (internal quotation marks
    omitted)).
    
       The language of the Supreme Court in Doe, however,
    suggests an indeterminate but deferential level of scrutiny for
                 CHULA VISTA CITIZENS V. NORRIS                   49
    
    restrictions in the ballot initiative process such as the elector
    requirement. The six-person majority in Doe stated:
    
            [T]he electoral context is [not] irrelevant to
            the nature of our First Amendment review.
            We allow States significant flexibility in
            implementing their own voting systems. . . .
            The State’s interest in preserving the integrity
            of the electoral process is undoubtedly
            important. States allowing ballot initiatives
            have considerable leeway to protect the
            integrity and reliability of the initiative
            process, as they have with respect to election
            processes generally.
    
    Doe, 561 U.S. at 197 (internal quotation marks omitted); see
    id. at 212–13 (Sotomayor, J., concurring) (“States enjoy
    considerable leeway to . . . specify the requirements for
    obtaining ballot access . . . . As the Court properly recognizes,
    each of these structural decisions inevitably affects—at least
    to some degree—the individual’s right to speak about
    political issues and to associate with others for political ends.
    For instance, requiring petition signers to be registered voters
    . . . no doubt limits the ability or willingness of some
    individuals to undertake the expressive act of signing a
    petition. Regulations of this nature, however, stand a step
    removed from the communicative aspect of petitioning, and
    the ability of States to impose them can scarcely be
    doubted.”) (citations and internal quotation marks omitted)
    (emphasis added). Doe therefore suggests that regulations of
    the electoral process, such as the “elector” requirement, while
    burdening the First Amendment, are subject to a deferential
    level of scrutiny, less demanding than strict scrutiny. We
    must therefore conclude, on the basis of the deferential
    50           CHULA VISTA CITIZENS V. NORRIS
    
    language in Doe, that not all political expression has the same
    protection, and that burdens that state and local governments
    impose on political expression while those governments are
    “implementing their own voting systems,” Doe, 561 U.S. at
    2819, must be analyzed under a level of scrutiny more
    deferential than strict scrutiny.
    
        Considering the deferential language of Doe, and because
    the elector requirement burdens “core political speech,”
    McIntyre v. Ohio Elections Comm’n, 
    514 U.S. 334
    , 347
    (1995), the level of scrutiny used here should be exacting
    scrutiny, which requires a “substantial relation” between the
    elector requirement and a “sufficiently important”
    governmental interest. Doe, 561 U.S. at 195.
    
        The interest that California asserts as the basis for the
    “elector requirement” here is “that only the people
    themselves shall exercise the right of self-government
    reserved in the initiative.” California and Chula Vista assert
    an interest that only those with “skin in the game,” i.e.
    electors, who will be affected by the measure, should initiate
    the referendum process. The California state and local
    governments want only civic-minded locals, who presumably
    would have knowledge of local affairs and would themselves
    be affected by the referendum, to participate in the initiative
    process. Otherwise, carpetbaggers, who themselves would
    not bear the full cost of initiating a proposal and who would
    not be burdened by the effects of the referendum, could
    hijack the initiative process and dictate a state’s referendum
    agenda. Considering the language of Doe, such an interest is
    “sufficiently important,” and thus satisfies that prong of the
    exacting scrutiny test.
                   CHULA VISTA CITIZENS V. NORRIS                       51
    
        The second part of the exacting scrutiny test requires that
    the elector requirement be “substantially related” to the
    governmental interest. The elector requirement could be
    more narrowly tailored to achieve its objective. In particular,
    the elector requirement is overinclusive, in that it prevents
    from being an official ballot initiative proponent an
    association made up entirely of electors, even though such an
    association would not detract from the governmental interest
    of insuring that only those with “skin in the game” initiate the
    referendum process. The elector requirement, therefore,
    because it is overinclusive, does not comply with the narrow
    tailoring requirement of strict scrutiny. See Citizens United,
    558 U.S. at 340. Under Doe’s more deferential level of
    exacting scrutiny applied here, however, the elector
    requirement is substantially related to the governmental
    interest.3 It ensures that only those with “skin in the game”
    will initiate the referendum process. Moreover, although it
    prevents electors from initiating ballot proposals in
    association with each other, it does not prevent them from
    doing so as individuals. Therefore, I conclude that the elector
    requirement satisfies exacting scrutiny; thus, I would reach
    the same outcome as the majority opinion.
    
                                      III
    
       Like the majority opinion, I think that the elector
    requirement does not violate the Constitution. I, however,
    would follow closely the test articulated in Doe and Carrigan.
    This would ensure that core political expression is analyzed
    under the correct constitutional framework and would prevent
    
    
    
     3
       Doe was decided June 24, 2010, five months after Citizens United was
    handed down.
    52             CHULA VISTA CITIZENS V. NORRIS
    
    core political expression from being denied the protection of
    the First Amendment.
    
    
    
    GRABER, Circuit Judge, concurring in part and dissenting in
    part:
    
         Two groups of Plaintiffs1 mount challenges to two
    restrictions that the people of California have placed on their
    initiative process:        (1) the requirement that official
    proponents be electors, that is, individual voters; and (2) the
    requirement that each petition section list the name of at least
    one official proponent. I agree with the majority opinion that
    the case is properly before us, and I concur in Part III, which
    holds that the elector requirement passes constitutional
    muster. I write separately to dissent from Part IV. The
    majority opinion properly recognizes in Part III that the role
    of an official proponent of an initiative petition in California
    is like that of a legislator. But the majority fails to apply this
    analogy equally to Part IV. Following the analogy of official
    proponent as legislator to its logical end, the disclosure
    requirement survives any level of review.
    
       The overarching question begins and ends with the role of
    the official proponent within the California lawmaking
    process. Although the California Constitution does not
    describe the full contours of the official proponent’s role, the
    California legislature has fleshed it out in a series of statutes.
    
    
      1
        Plaintiffs comprising the Associations are Chula Vista Citizens for
    Jobs and Fair Competition and the Associated Builders and Contractors,
    Inc. (“Associations”). Plaintiffs comprising the Individual Plaintiffs are
    Lori Kneebone and Larry Breitfelder (“Individual Plaintiffs”).
                    CHULA VISTA CITIZENS V. NORRIS                         53
    
         Under the California Elections Code, an official
    proponent enjoys a special relationship to the initiative that
    continues long after the advocacy process is complete. See,
    e.g., Cal. Elec. Code §§ 9202, 9205, 9207. In particular,
    official proponents: (1) bear the obligation “to manage and
    supervise the process by which signatures for the initiative
    petition are obtained”; (2) “control the arguments in favor of
    an initiative measure,” including by serving as gatekeeper for
    all ballot arguments, providing arguments afforded priority
    status on the ballot, controlling all rebuttal ballot arguments,
    and retaining the ability to withdraw ballot arguments at any
    time; and (3) are allowed to intervene, both before and after
    the initiative is passed, in litigation affecting the initiated
    statute, and to appeal state court rulings adverse to the
    initiative’s validity. Perry v. Brown, 
    265 P.3d 1002
    , 1017–18
    (Cal. 2011). But see Hollingsworth v. Perry, 
    133 S. Ct. 2652
    ,
    2662 (2013) (holding that the authority of the official
    proponent to intervene in court proceedings pertaining to an
    initiative is insufficient, without more, to create Article III
    standing). In addition to having special duties beyond those
    of ordinary supporters of an initiative, “the official
    proponents of an initiative measure are recognized as having
    a distinct role—involving both authority and responsibilities
    that differ from other supporters of the measure,” Perry,
    265 P.3d at 1017–18, and the California Supreme Court has
    equated the role of a proponent to that of an elected legislator
    to whom the people have delegated lawmaking power.2 It is
    
     2
       In fact, California law gives an official proponent more authority than
    a legislator who, despite having sponsored and championed a piece of
    legislation through the California legislature, would not have a right to
    intervene in court on behalf of the legislation after it had been codified.
    See Perry, 265 P.3d at 1021 (noting that legislators would not be afforded
    the ability to intervene on behalf of a law that they had sponsored, before
    holding that official proponents could so intervene). The United States
    54              CHULA VISTA CITIZENS V. NORRIS
    
    the distinct character of this role that informs the First
    Amendment analysis for both challenges.
    
         The Individual Plaintiffs challenge a requirement of the
    California Elections Code that the text of the petitions
    disclose the name of at least one official proponent. The
    California Elections Code, as incorporated by the Chula Vista
    City Charter, requires an official proponent of a ballot
    initiative to provide a name and signature at three distinct
    points during the initiative process. First, at least one of the
    official proponents must provide a name and signature to the
    City Clerk on the Notice of Intent to Circulate Petition when
    the document is first filed. Cal. Elec. Code § 9202. Second,
    the Notice of Intent to Circulate Petition, containing the
    signature and name of at least one official proponent, must be
    published in a newspaper of general circulation within the
    city and county. Cal. Elec. Code §§ 9202, 9205. Finally, the
    California Elections Code mandates that each section of the
    petition bear a copy of the Notice of Intent to Circulate
    Petition, which would necessarily include the name and
    signature of any official proponent who signed the form
    initially. Id. § 9207.
    
        The Individual Plaintiffs mount a facial challenge to only
    the final requirement, that each section of the petition bear a
    copy of the Notice of Intent to Circulate Petition. They
    contend that this content-based restriction, affecting the text
    
    
    Supreme Court has held that an official proponent under California law is
    not equivalent to an elected, public official for Article III purposes under
    the Federal Constitution. Hollingsworth, 133 S. Ct. at 2662. The
    California Supreme Court, however, retains supreme authority to define
    the role of an official proponent under state law. See, e.g., Pembaur v.
    City of Cincinnati, 
    475 U.S. 469
    , 481–82 (1986); R.R. Comm’n of Cal. v.
    L.A. Ry. Corp., 
    280 U.S. 145
    , 152 (1929).
                 CHULA VISTA CITIZENS V. NORRIS                  55
    
    of the petition, impermissibly chills core political speech by
    forcing speakers to disclose their identities at the point of
    contact with potential signatories. Because this disclosure is
    required at the point of contact with voters, the Individual
    Plaintiffs urge us to review the disclosure regime with strict
    scrutiny.
    
        Anonymous political advocacy, such as the debate
    between “Publius” in the Federalist Papers and his detractors
    “Federal Farmer” and “Cato,” has played a fundamental role
    in the development of our constitutional framework.
    “Anonymity is a shield from the tyranny of the majority. It
    thus exemplifies the purpose behind the Bill of Rights, and of
    the First Amendment in particular: to protect unpopular
    individuals from retaliation—and their ideas from
    suppression—at the hand of an intolerant society.” McIntyre
    v. Ohio Elections Comm’n, 
    514 U.S. 334
    , 356 (1995) (citation
    omitted). It is perhaps because of this essential value that our
    free speech tradition has hesitated to allow a government to
    exclude speech from the marketplace of ideas merely because
    that speech does not disclose its source. See id. at 342–43.
    But compelled disclosure is not a direct prohibition on
    speech. Doe v. Reed, 
    130 S. Ct. 2811
    , 2818 (2010); see also
    Citizens United v. Fed. Election Comm’n, 
    558 U.S. 310
    , 366
    (2010) (holding that “disclosure requirements may burden
    the ability to speak, but they . . . do not prevent anyone from
    speaking” (citation and internal quotation marks omitted)).
    Striking a balance between those two points, the Supreme
    Court has applied “‘exacting scrutiny’” to disclosure regimes
    in the electoral context, which “requires a ‘substantial
    relation’ between the disclosure requirement and a
    ‘sufficiently important’ governmental interest.” Citizens
    United, 558 U.S. at 366–67 (quoting Buckley v. Valeo,
    
    424 U.S. 1
    , 64, 66 (1976) (per curiam)). Following Doe and
    56             CHULA VISTA CITIZENS V. NORRIS
    
    Citizens United, I would apply exacting scrutiny to the
    disclosure regime.
    
         The government maintains that the disclosure requirement
    is a reasonable regulation of the initiative process that serves
    two sufficiently important state interests: (1) to preserve the
    integrity of the initiative process; and (2) to inform
    signatories “as to who is formally proposing the legislation.”3
    Because I find the state’s interest in preserving the integrity
    of the electoral process sufficiently important, indeed
    compelling, and substantially related to a narrowly tailored
    disclosure regime, I would find the regime constitutional
    under any level of scrutiny.
    
         The Supreme Court has consistently recognized that the
    government’s interest in preserving the integrity of the
    electoral process is sufficiently important to survive exacting
    scrutiny. Doe, 130 S. Ct. at 2819. “States allowing ballot
    initiatives have considerable leeway to protect the integrity
    and reliability of the initiative process, as they have with
    respect to election processes generally.” Id. (internal
    quotation marks omitted). The government’s interest in
    preserving the integrity of elections is especially strong in the
    context of fraud, but the interest “is not limited to combating
    fraud” and “also extends more generally to promoting
    transparency and accountability in the electoral process,
    which the State argues is ‘essential to the proper functioning
    of a democracy.’” Id.
    
    
     3
       I do not reach the question whether the people’s informational interest
    is sufficiently important, because I would hold that the government’s
    interest in preserving the integrity of the electoral process alone is
    sufficiently important to sustain the minimal burden on official
    proponents.
                 CHULA VISTA CITIZENS V. NORRIS                  57
    
         In the federal context, “[t]he public nature of federal
    lawmaking is constitutionally required.” Id. at 2834 (Scalia,
    J., dissenting) (quoting U.S. Const art. I, § 5, cl. 3: “‘Each
    House shall keep a Journal of its Proceedings, and from time
    to time publish the same, excepting such Parts as may in their
    Judgment require Secrecy[.]’”). The lawmaking process is
    kept transparent for good reason: Knowing the identities of
    lawmakers and their actions plays an important role in
    allowing the public to evaluate officials and hold them
    accountable. “In a republic where the people are sovereign,
    the ability of the citizenry to make informed choices among
    candidates for office is essential, for the identities of those
    who are elected will inevitably shape the course that we
    follow as a nation.” Buckley, 424 U.S. at 14–15.
    
         Similarly, the local government has an “essential” interest
    in preserving an electoral process in which members of the
    California public who are considering whether to sign an
    initiative petition know for whom they are expressing support
    as the official proponent when they sign a petition—and to
    whom they will delegate certain lawmaking duties if the
    petition is successful. The government’s interest in
    supporting the integrity of the electoral process by providing
    the public with the identity of an official proponent is not
    directed solely at preventing fraud. The electoral process
    would be degraded if potential signers have no way of
    knowing whether their signatures are delegating lawmaking
    duties to a desirable proponent for the initiative, who will
    present arguments on behalf of the initiative and defend the
    initiative in a manner with which the signers agree. As noted,
    in California official proponents play a central role, both
    during the lawmaking process and after their initiative is
    enacted. Perry, 265 P.3d at 1017–18. An ineffective official
    proponent: (1) could fail to manage and supervise the
    58            CHULA VISTA CITIZENS V. NORRIS
    
    initiative process; (2) could fail to file the petition with the
    state; (3) could make poor choices regarding arguments and
    statements for the ballot; (4) would receive priority status for
    even the weakest arguments on the ballot; (5) could fail to
    mount, or could withdraw from the ballot, the better
    arguments; and (6) could fail to defend the initiative in court
    proceedings. Id.
    
        On issues of public importance, potential signers could
    face multiple initiatives on the same topic. In order to make
    an informed decision about which of the initiatives to support,
    potential signers would need to know the differences in
    content among the various initiatives. But the voters would
    also need to know the identities of the official proponents for
    each initiative so that the voters could evaluate how those
    official proponents would present the important public issue
    at hand. Because the official proponent serves an important
    role in the lawmaking process and is delegated duties in the
    lawmaking process far beyond that of an advocate, the
    government has an essential interest in preserving an electoral
    process that allows voters to know to whom they are
    delegating lawmaking power when signing a particular
    petition.
    
         This “essential” interest clearly outweighs the minor
    actual burden, if any, on the official proponents who must
    disclose their identities. The role that these individuals wish
    to fill is itself a public legislative role that is akin to the role
    of an elected legislator. The voluntary undertaking of a
    California proponent’s role entails other duties (beyond the
    initial filing) that require disclosure of the official
    proponent’s identity, for example, monitoring the integrity of
    the petition-circulation process, crafting arguments for the
    ballot, and intervening in court proceedings. Other circuits
                 CHULA VISTA CITIZENS V. NORRIS                59
    
    have recognized that candidates for public office have no
    First Amendment interest in anonymity by virtue of their
    voluntary undertaking of a public role. See, e.g., Majors v.
    Abell, 
    317 F.3d 719
    , 722 (7th Cir. 2003) (“[The plaintiff’s]
    standing might be questioned on the ground that a candidate
    has no [free speech] interest in anonymity that the statute
    might protect; for there are no anonymous candidates.”).
    Similarly, the role sought by these individuals is one that
    necessarily requires public disclosure of identity. Even
    assuming that an official proponent has a First Amendment
    anonymity interest, the potential burden is negligible.
    
        In a different context, the Supreme Court has expressed
    skepticism that an informational interest can sustain regimes
    that compel disclosure of the identity of an advocate at the
    point of contact with voters, or signatories in the initiative
    context. In Buckley v. American Constitutional Law
    Foundation, Inc. (“ACLF”), 
    525 U.S. 182
    , 200 (1999), the
    Court struck down a Colorado statute requiring every petition
    circulator to wear a badge bearing the circulator’s name,
    because the public’s informational interest in identifying the
    advocate was insufficiently important to justify chilling
    political speech and potentially to subject circulators to
    harassment at the apex of face-to-face political advocacy.
    Similarly, in McIntyre, the Supreme Court struck down an
    Ohio ban on handbills that failed to identify the name of the
    advocate, because the public’s informational interest in
    knowing the identity of the advocate “means nothing more
    than the provision of additional information that may either
    buttress or undermine the argument in a document.” 514 U.S.
    at 348. Beyond expressing concerns over fear of harassment
    that could chill political speech, the Court noted that
    disclosure of the advocates’ identity at the point of contact
    with voters could weaken the effectiveness of the speech:
    60           CHULA VISTA CITIZENS V. NORRIS
    
    “[A]n advocate may believe her ideas will be more persuasive
    if her readers are unaware of her identity.” Id. at 342; see
    also ACLU of Nev. v. Heller, 
    378 F.3d 979
    , 994 (9th Cir.
    2004) (“[F]ar from enhancing the reader’s evaluation of a
    message, identifying the publisher can interfere with that
    evaluation by requiring the introduction of potentially
    extraneous information at the very time the reader encounters
    the substance of the message.”).
    
        We have adopted the Court’s skepticism of an
    informational interest in the identity of the advocate at the
    point of contact with voters. See Heller, 378 F.3d at 995
    (holding that an informational interest in allowing the public
    to evaluate the advocacy document is not sufficiently
    important to sustain compelled disclosure of an advocate’s
    identity on the document itself); (WIN) Wash. Initiatives Now
    v. Rippie, 
    213 F.3d 1132
    , 1140 (9th Cir. 2000) (holding that
    an informational interest is not sufficiently important to
    sustain compelled disclosure of advocates’ identities during
    the circulation period).
    
        But this doctrine does not apply here for two reasons.
    First, the statute is directed toward the government’s interest
    in preserving the integrity of elections, an interest that the
    Supreme Court has recognized as sufficient to support
    mandated disclosure of identity. Doe, 130 S. Ct. at 2820.
    Second, the proponent of a California initiative is asking
    voters to allow her to serve an official public role and to
    allow her to act on the voters’ behalf in the legislative
    process, not just recounting an idea as an advocate, and a
    petition is an official legislative form, not a pamphlet or
    advocacy document. See Timmons v. Twin Cities Area New
    Party, 
    520 U.S. 351
    , 363 (1997) (holding that the First
    Amendment does not provide “a right to use the ballot itself
                 CHULA VISTA CITIZENS V. NORRIS                  61
    
    to send a particularized message”). Here, the disclosure
    regime seeks to disclose the identity not of an advocate, but
    of an individual who serves as an official proponent and
    representative of the signers in the lawmaking process—a
    role akin to a candidate for office and recognized explicitly as
    distinct from that of an advocate under California law. See
    Perry, 265 P.3d at 1017–18 (holding that “the official
    proponents of an initiative measure are recognized as having
    a distinct role—involving both authority and responsibilities
    that differ from other supporters of the measure”). The
    government’s interest in alerting potential signatories to the
    official proponent’s identity, as a representative of the
    initiative, is markedly different than the interests at stake in
    McIntyre, Citizens United, and ACLF—that is, knowing the
    identity of a mere advocate in order to evaluate an argument.
    
        The Individual Plaintiffs respond that the state’s interest
    is satisfied, or lessened, by two other required points of
    disclosure—at the points of application and publication in a
    newspaper of general circulation. Cal. Elec. Code §§ 9202,
    9205. That argument fails to recognize that the signing of a
    petition in this context is not simply agreeing to the content
    of an initiative; it also is an expression of support for that
    particular official proponent. If the name of the official
    proponent is not printed on the petition, every elector who is
    considering whether to sign would be required to match the
    petition with public records or newspaper publications in
    order to glean whom the elector is designating as the official
    proponent. The two earlier points of disclosure identified by
    the Individual Plaintiffs would provide no identifying
    information whatsoever to an elector who is approached on
    the street with a petition; the elector would lack sufficient
    information to allow for an informed decision whether to
    62           CHULA VISTA CITIZENS V. NORRIS
    
    sign. Such a disclosure scheme clearly would fail to satisfy
    the government’s interest in any meaningful or realistic sense.
    
        By analogy, if ballots listed only the platforms of
    candidates for an office, but not the candidates’ names,
    undoubtedly we would not find it sufficient that voters were
    able to access the names and platforms of candidates from
    public records or from local media in order to guide voting
    choices. We do not permit federal candidates for public
    office to remain anonymous at the point of contact with
    voters, nor do we force voters to support federal candidates
    without knowing the candidates’ identities. So too here, we
    should uphold the decision by the citizens of Chula Vista to
    prohibit anonymous candidates for an official legislative role.
    
        Finally, the Supreme Court has allowed those resisting
    disclosure to mount a successful First Amendment challenge
    where “they can show a reasonable probability that the
    compelled disclosure of personal information will subject
    them to threats, harassment, or reprisals from either
    Government officials or private parties.” Doe, 130 S. Ct. at
    2820 (internal quotation marks and brackets omitted). The
    Individual Plaintiffs, however, have provided no evidence
    that shows a likelihood of harassment, and they have
    effectively conceded that they experienced no harassment in
    response to their service as official proponents to Chula Vista
    Measure G. Moreover, given the public role that an official
    proponent serves in the lawmaking process, some public
    pressure must be expected in order to hold that official
    proponent accountable to a good faith performance of his
    duties. Accordingly, Plaintiffs have failed to show that the
    compelled disclosure would subject them to threats,
    harassment, or unreasonable reprisals from government
    officials or the public.
                CHULA VISTA CITIZENS V. NORRIS                63
    
        In sum, I would hold that the disclosure regime survives
    exacting scrutiny because it is substantially related and
    narrowly tailored to the government’s interest in preserving
    the integrity and transparency of the electoral process by
    providing voters with the identity of the official proponent.
    Accordingly, I respectfully dissent from Part IV. I would
    affirm the judgment in full.