delivered the opinion of the Court.

The issue on appeal is whether a limitation of $250 on contributions to committees formed to support or oppose ballot measures violates the First Amendment.

*292I

The voters of Berkeley, Cal., adopted the Election Reform Act of 1974, Ord. No. 4700-N. S., by initiative. The campaign ordinance so enacted placed limits on expenditures and contributions in campaigns involving both candidates and ballot measures.1 Section 602 of the ordinance provides:

“No person shall make, and no campaign treasurer shall solicit or accept, any contribution which will cause the total amount contributed by such person with respect to a single election in support of or in opposition to a measure to exceed two hundred and fifty dollars ($250).”2

Appellant Citizens Against Rent Control is an unincorporated association formed to oppose a ballot measure at issue in the April 19, 1977, election. The ballot measure would have imposed rent control on many of Berkeley’s rental units. To make its views on the ballot measure known, Citizens Against Rent Control raised more than $108,000 from ap*293proximately 1,300 contributors. It accepted nine contributions over the $250 limit. Those nine contributions totaled $20,850, or $18,600 more than if none of the contributions exceeded $250. Pursuant to § 604 of the ordinance,3 appellee Berkeley Fair Campaign Practices Commission, 20 days before the election, ordered appellant Citizens Against Rent Control to pay $18,600 into the city treasury.

Two weeks before the election, Citizens Against Rent Control sought and obtained a temporary restraining order prohibiting enforcement of §§ 602 and 604. The ballot measure relating to rent control was defeated. The Superior Court subsequently granted Citizens Against Rent Control’s motion for summary judgment, declaring that §602 was invalid on its face because it violated the First Amendment of the United States Constitution and Art. I, § 2, of the California Constitution. A panel of the California Court of Appeal unanimously affirmed that conclusion.

The California Supreme Court, dividing 4-3, reversed. 27 Cal. 3d 819, 614 P. 2d 742 (1980). Citing Buckley v. Valeo, 424 U. S. 1 (1976), the majority announced that it would strictly scrutinize § 602. It concluded that the section furthered compelling governmental interests because it ensured that special interest groups could not “corrupt” the initiative process by spending large amounts to support or oppose a ballot measure. Such corruption, the court found, could produce apathetic voters; these governmental interests were held to outweigh the First Amendment interests infringed upon. Finally, it concluded that § 602 accomplished its goal *294by the least restrictive means available. The California Supreme Court did not consider the disclosure requirements of the ordinance a sufficient prophylaxis to dispel perceptions of corruption.4

We noted probable jurisdiction, 450 U. S. 908 (1981), and we reverse.

II

The appellees concede that the challenged ordinance has an impact on First Amendment rights; the parties disagree only as to the extent of the impact. Long ago this Court admonished that with respect to the First Amendment:

“[T]he power to regulate must be so exercised as not, in attaining a permissible end, unduly to infringe the protected freedom.” Cantwell v. Connecticut, 310 U. S. 296, 304 (1940).

This was but another way of saying that regulation of First Amendment rights is always subject to exacting judicial review.

We begin by recalling that the practice of persons sharing common views banding together to achieve a common end is deeply embedded in the American political process. The 18th-century Committees of Correspondence and the pamphleteers were early examples of this phenomena and the Federalist Papers were perhaps the most significant and lasting example. The tradition of volunteer committees for collective action has manifested itself in myriad community and public activities; in the political process it can focus on a candidate or on a ballot measure. Its value is that by collective effort individuals can make their views known, when, individually, their voices would be faint or lost.

*295The Court has long viewed the First Amendment as protecting a marketplace for the clash of different views and conflicting ideas. That concept has been stated and restated almost since the Constitution was drafted. The voters of the city of Berkeley adopted the challenged ordinance which places restrictions on that marketplace. It is irrelevant that the voters rather than a legislative body enacted § 602, because the voters may no more violate the Constitution by enacting a ballot measure than a legislative body may do so by enacting legislation.

Ill

A

The Court has acknowledged the importance of freedom of association in guaranteeing the right of people to make their voices heard on public issues:

“Effective advocacy of both public and private points of view, particularly controversial ones, is undeniably enhanced by group association, as this Court has more than once recognized by remarking upon the close nexus between the freedoms of speech and assembly.” NAACP v. Alabama, 357 U. S. 449, 460 (1958).

More recently the Court stated: “The First Amendment protects political association as well as political expression.” Buckley v. Valeo, supra, at 15.

Buckley also made clear that contributors cannot be protected from the possibility that others will make larger contributions:

“[T]he concept that government may restrict the speech of some elements of our society in order to enhance the relative voice of others is wholly foreign to the First Amendment, which was designed ‘to secure “the widest possible dissemination of information from diverse and antagonistic sources,”’ and ‘“to assure unfettered inter*296change of ideas for the bringing about of political and social changes desired by the people.”’ New York Times Co. v. Sullivan, [376 U. S.], at 266, 269, quoting Associated Press v. United States, 326 U. S. 1, 20 (1945), and Roth v. United States, 354 U. S., at 484. The First Amendment’s protection against governmental abridgment of free expression cannot properly be made to depend on a person’s financial ability to engage in public discussion. Cf. Eastern R. Conf. v. Noerr Motors, 365 U. S. 127, 139 (1961).” 424 U. S., at 48-49.

The Court went on to note that the freedom of association “is diluted if it does not include the right to pool money through contributions, for funds are often essential if ‘advocacy’ is to be truly or optimally ‘effective.’ ” Id., at 65-66.6 Under the Berkeley ordinance an affluent person can, acting alone, spend without limit to advocate individual views on a ballot measure. It is only when contributions are made in concert with one or more others in the exercise of the right of association that they are restricted by § 602.

There are, of course, some activities, legal if engaged in by one, yet illegal if performed in concert with others, but political expression is not one of them. To place a Spartan limit— or indeed any limit — on individuals wishing to band together to advance their views on a ballot measure, while placing none on individuals acting alone, is clearly a restraint on the right of association. Section 602 does not seek to mute the voice of one individual, and it cannot be allowed to hobble the collective expressions of a group.

Buckley identified a single narrow exception to the rule that limits on political activity were contrary to the First *297Amendment. The exception relates to the perception of undue influence of large contributors to a candidate:

“To the extent that large contributions are given to secure a political quid pro quo from current and potential office holders, the integrity of our system of representative democracy is undermined. . . .

“. . . Congress could legitimately conclude that the avoidance of the appearance of improper influence ‘is also critical... if confidence in the system of representative Government is not to be eroded to a disasterous extent.’ [CSC v. Letter Carriers,] 413 U. S., at 565.” 424 U. S., at 26-27.

Buckley thus sustained limits on contributions to candidates and their committees.

Federal Courts of Appeals have recognized that Buckley does not support limitations on contributions to committees formed to favor or oppose ballot measures. In C & C Plywood Corp. v. Hanson, 583 F. 2d 421 (1978), the Ninth Circuit struck down a Montana statute prohibiting corporate contributions supporting or opposing ballot measures. In so doing the court noted:

“The state interest in preventing corruption of officials, which provided the basis for the Supreme Court’s finding in Buckley that restrictions could permissibly be placed on contributions, is not at issue here.” Id., at 425.

Similarly, the Fifth Circuit interpreted Buckley to hold that

“[t]he sole governmental interest that the Supreme Court recognized as a justification for restricting contributions was the prevention of quid pro quo corruption between a contributor and a candidate.” Let’s Help Florida v. McCrary, 621 F. 2d 195, 199 (1980).

In First National Bank of Boston v. Bellotti, 435 U. S. 765 (1978), we held that a state could not prohibit corporations any more than it could preclude individuals from making con*298tributions or expenditures advocating views on ballot measures. The Bellotti Court relied on Buckley to strike down state legislative limits on advocacy relating to ballot measures:

“Referenda are held on issues, not candidates for public office. The risk of corruption perceived in cases involving candidate elections [citations omitted] simply is not present in a popular vote on a public issue. To be sure, corporate advertising may influence the outcome of the vote; this would be its purpose. But the fact that advocacy may persuade the electorate is hardly a reason to suppress it: The Constitution ‘protects expression which is eloquent no less than that which is unconvincing.’ Kingsley Int’l Pictures Corp. v. Regents, 360 U. S., at 689.” 435 U. S., at 790 (footnote omitted).

Notwithstanding Buckley and Bellotti, the city of Berkeley argues that § 602 is necessary as a prophylactic measure to make known the identity of supporters and opponents of ballot measures. It is true that when individuals or corporations speak through committees, they often adopt seductive names that may tend to conceal the true identity of the source. Here, there is no risk that the Berkeley voters will be in doubt as to the identity of those whose money supports or opposes a given ballot measure since contributors must make their identities known under § 112 of the ordinance, which requires publication of lists of contributors in advance of the voting. See n. 4, supra.

Contributions by individuals to support concerted action by a committee advocating a position on a ballot measure is beyond question a very significant form of political expression. As we have noted, regulation of First Amendment rights is always subject to exacting judicial scrutiny. Supra, at 294. The public interest allegedly advanced by § 602 — identifying the sources of support for and opposition to ballot measures — is insubstantial because voters may identify those *299sources under the provisions of § 112. In addition, the record in this case does not support the California Supreme Court’s conclusion that §602 is needed to preserve voters’ confidence in the ballot measure process. Cf. Bellotti, supra, at 789-790. It is clear, therefore, that § 602 does not advance a legitimate governmental interest significant enough to justify its infringement of First Amendment rights.6

B

Apart from the impermissible restraint on freedom of association, but virtually inseparable from it in this context, § 602 imposes a significant restraint on the freedom of expression of groups and those individuals who wish to express their views through committees. As we have noted, an individual may make expenditures without limit under § 602 on a ballot measure but may not contribute beyond the $250 limit when joining with others to advocate common views. The contribution limit thus automatically affects expenditures, and limits on expenditures operate as a direct restraint on freedom of expression of a group or committee desiring to engage in political dialogue concerning a ballot measure.

Whatever may be the state interest or degree of that interest in regulating and limiting contributions to or expenditures of a candidate or a candidate’s committees there is no significant state or public interest in curtailing debate and discussion of a ballot measure. Placing limits on contributions which in turn limit expenditures plainly impairs freedom of expression. The integrity of the political system will be adequately protected if contributors are identified in a *300public filing revealing the amounts contributed; if it is thought wise, legislation can outlaw anonymous contributions.

IV

A limit on contributions in this setting need not be analyzed exclusively in terms of the right of association or the right of expression. The two rights overlap and blend; to limit the right of association places an impermissible restraint on the right of expression. The restraint imposed by the Berkeley ordinance on rights of association and in turn on individual and collective rights of expression plainly contravenes both the right of association and the speech guarantees of the First Amendment. Accordingly, the judgment of the California Supreme Court is reversed, and the case is remanded for proceedings not inconsistent with this opinion.

Reversed and remanded.

concurring in the judgment.

The contribution limitations at issue here encroach directly on political expression and association. Thus, Berkeley’s ordinance cannot survive constitutional challenge unless it withstands “exacting scrutiny.” First National Bank of Boston v. Bellotti, 435 U. S. 765, 786 (1978). To meet this rigorous standard of review, Berkeley must demonstrate that its ordinance advances a sufficiently important governmental interest and employs means “ ‘closely drawn to avoid unnecessary abridgment’” of First Amendment freedoms. Ibid, (quoting Buckley v. Valeo, 424 U. S. 1, 25 (1976)).

We would hold that Berkeley has neither demonstrated a genuine threat to its important governmental interests nor employed means closely drawn to avoid unnecessary abridgment of protected activity. In Buckley, this Court upheld limitations on contributions to candidates as necessary to prevent contributors from corrupting the representatives to whom the people have delegated political decisions. But curtailment of speech and association in a ballot measure campaign, where the people themselves render the ultimate political decision, cannot be justified on this basis.

Nor has Berkeley proved a genuine threat to its interest in maintaining voter confidence in government. We would not deny the legitimacy of that interest. Indeed, in Bellotti, this Court explicitly recognized that “[preserving the integrity of the electoral process, preventing corruption, . . . ‘sustaining] the active, alert responsibility of the individual citizen in a democracy for the wise conduct of government,’” and “[preservation of the individual citizen’s confidence in government” are “interests of the highest importance” in ballot measure elections. 435 U. S., at 788-789, citing and quoting *303United States v. Automobile Workers, 352 U. S. 567, 570, 575 (1957). We did not find those interests threatened in Bellotti, however, in part because the State failed to show “by record or legislative findings that corporate advocacy threatened imminently to undermine democratic processes” or “the confidence of the citizenry in government.” 435 U. S., at 789-790. The city’s evidentiary support in this case is equally sparse.

Finally, Berkeley does not justify its contribution limit as necessary to encourage disclosure. We cannot accept the Court’s conclusion that that interest is “insubstantial,” given the Court’s concession that “when individuals or corporations speak through committees, they often adopt seductive names that may tend to conceal the true identity of the source.” Ante, at 298. Yet Berkeley need not impose a $250 ceiling on contributions to encourage disclosure so long as it vigorously enforces its already stringent disclosure laws. Ante, at 294, n. 4.

We need say no more in order to reverse. Accordingly, we concur in the judgment.

dissenting.

In Buckley v. Valeo, 424 U. S. 1 (1976), the Court upheld restrictions on contributions but struck down limits on expenditures in campaigns for federal office that Congress, the body most expert in the matter, thought equally essential to protect the integrity of the election process. Two years later, a bare majority of the Court, substituting its judgment for that of the Massachusetts Legislature, invalidated that State’s prohibition on corporate spending in referendum elections. First National Bank of Boston v. Bellotti, 435 U. S. 765 (1978). Disagreeing with the Court’s assumption that those regulations inhibited the free interplay of political advocacy, I would have upheld the expenditure limitations at issue in Buckley and the restrictions contested in Bellotti.

This case poses a less encompassing regulation on campaign activity, one tailored to the odd measurements of *304Buckley and Bellotti. Precisely because it reflects these decisions, the ordinance regulates contributions but not expenditures and does not prohibit corporate spending.1 It is for that very reason perhaps that the effectiveness of the ordinance in preserving the integrity of the referendum process is debatable. Even so, the result here illustrates that the Buckley framework is most problematical and strengthens my belief that there is a proper role for carefully drafted limitations on expenditures.

Even under Buckley, however, the Berkeley ordinance represents such a negligible intrusion on expression and association that the measure should be upheld. The ordinance certainly does not go beyond what I understand the First Amendment to permit. For both these reasons, I dissent.

I — I

The Berkeley ordinance does not control the quantity or content of speech. Unlike the statute in Bellotti, it does not completely prohibit contributions and expenditures. Any person or company may contribute up to $250. If greater spending is desired, it must be made as an expenditure, and expenditures are not limited or otherwise controlled. Individuals also remain completely unfettered in their ability to join interested groups or otherwise directly participate in the campaign.

*305The Court reaches the conclusion that the ordinance is unconstitutional only by giving Buckley the most extreme reading and by essentially giving the Berkeley ordinance no reading at all. It holds that the contributions involved here are “beyond question a very significant form of political expression.” Ante, at 298. Yet in Buckley the Court found that contribution limitations “entai[l] only a marginal restriction upon the contributor’s ability to engage in free communication.” 424 U. S., at 20-21. As with contributions to candidates, ballot measure contributions “involv[e] speech by someone other than the contributor” and a limitation on such donations “does not in any way infringe the contributor’s freedom to discuss candidates and issues.” Id., at 21. Indeed what today has become “a very significant form of political expression” was held just last Term to involve only “some limited element of protected speech.” California Medical Assn. v. FEC, 453 U. S. 182, 196, n. 16 (1981) (Marshall, J., joined by Brennan, White, and Stevens, JJ.). “‘Speech by proxy,’” we said, “is not the sort of advocacy that this Court in Buckley found entitled to full First Amendment protection.” Id., at 196.

The Court also finds that the freedom of association is im-permissibly compromised by not allowing persons to contribute unlimited funds to committees organized to support or oppose a ballot measure. However, in Buckley, the Court observed that contribution ceilings “leav[e] persons free to engage in independent political expression, to associate actively through volunteering their services, and to assist to a limited but nonetheless substantial extent in supporting candidates and committees with financial resources.” 424 U. S., at 28. Associational rights, it was thought, were seriously impinged only by expenditure ceilings — there by virtue of precluding associations from effectively amplifying the voice of their adherents, “the original basis for the recognition of First Amendment protection of the freedom of association.” Id., at 22. See NAACP v. Alabama, 357 U. S. 449, *306460 (1958). The Court’s concern that this ordinance will “hobble the collective expressions of a group” ante, at 296, is belied by the fact that appellants, having already met their campaign budget, ended all fundraising almost a month before the election.

It is bad enough that the Court overstates the extent to which First Amendment interests are implicated. But the Court goes on to assert that the ordinance furthers no legitimate public interest and cannot survive “any degree of scrutiny.” Apparently the Court assumes this to be so because the ordinance is not directed at quid pro quos between large contributors and candidates for office, “the single narrow exception” for regulation that it viewed Buckley as endorsing. The Buckley Court, however, found it “unnecessary to look beyond the Act’s primary purpose,” the prevention of corruption, to uphold the contribution limits, and thus did not consider other possible interests for upholding the restriction. Indeed, at least since United States v. Automobile Workers, 352 U. S. 567, 575 (1957), the Court has recognized that “sustaining the active alert responsibility of the individual citizen in a democracy for the wise conduct of government” is a valid state interest. The Bellotti Court took care to note that this objective, along with “[preserving the integrity of the electoral process [and] the individual citizen’s confidence in government” “are interests of the highest importance.” 435 U. S., at 788-789.

In Bellotti, the Court found inadequate evidence in the record to support these interests, but it suggested that some regulation of corporate spending might be justified if “corporate advocacy threatened imminently to undermine democratic processes, thereby denigrating rather than serving First Amendment interests.” Id., at 789. The Court suggested that such a situation would arise if it could be shown that “the relative voice of corporations ha[d] been overwhelming [and] . . . significant in influencing referenda.” *307Id., at 789-790. It is quite possible that such a test is fairly-met in this case. Large contributions, mainly from corporate sources, have skyrocketed as the role of individuals has declined.2 Staggering disparities have developed between spending for and against various ballot measures.3 While it *308is not possible to prove that heavy spending “bought” a victory on any particular ballot proposition, there is increasing evidence that large contributors are at least able to block the adoption of measures through the initiative process.4 Recognition that enormous contributions from a few institutional sources can overshadow the efforts of individuals may have discouraged participation in ballot measure campaigns5 and undermined public confidence in the referendum process.

By restricting the size of contributions, the Berkeley ordinance requires major contributors to communicate directly with the voters. If the ordinance has an ultimate impact on speech, it will be to assure that a diversity of views will be presented to the voters. As such, it will “facilitate and enlarge public discussion and participation in the electoral process, goals vital to a self-governing people.” Buckley, 424 U. S., at 92-93. Of course, entities remain free to make major direct expenditures. But because political communica*309tions must state the source of funds, voters will be able to identify the source of such messages and recognize that the communication reflects, for example, the opinion of a single powerful corporate interest rather than the views of a large number of individuals. As the existence of disclosure laws in many States suggests,6 information concerning who supports or opposes a ballot measure significantly affects voter evaluation of the proposal.7 The Court asserts, without elaboration, that existing disclosure requirements suffice to inform voters of the identity of contributors. Yet, the inadequacy of disclosure laws was a major reason for the adoption of the Berkeley ordinance. Section 101(d) of the ordinance constitutes a finding by the people of Berkeley that “the influence of large campaign contributors is increased because existing laws for disclosure of campaign receipts and expenditures have proved to be inadequate.”

Admittedly, Berkeley cannot present conclusive evidence of a causal relationship between major undisclosed expenditures and the demise of the referendum as a tool of direct democracy. But the information available suffices to demonstrate that the voters had valid reasons for adopting contribution ceilings. It was on a similar foundation that the Court upheld contribution limits in Buckley and California Medical Assn. v. FEC, 453 U. S. 182 (1981). In my view, the ordinance survives scrutiny under the Buckley and Bellotti cases.

II

There are other grounds for sustaining the ordinance. I continue to believe that because the limitations are content-*310neutral, and because many regulatory actions will indirectly affect speech in the same manner as regulations in the sphere of campaign finance, “the argument that money is speech and that limiting the flow of money to the speaker violates the First Amendment proves entirely too much.” Buckley, supra, at 262 (White, J., concurring in part and dissenting in part). Every form of regulation — from taxes to compulsory bargaining — has some effect on the ability of individuals and corporations to engage in expressive activity. We must therefore focus on the extent to which expressive and associational activity is restricted by the Berkeley ordinance. That First Amendment interests are implicated should begin, not end, our inquiry. When the infringement is as slight and ephemeral as it is here, the requisite state interest to justify the regulation need not be so high.

The interests which justify the Berkeley ordinance can properly be understood only in the context of the historic role of the initiative in California. “California’s entire history demonstrates the repeated use of referendums to give citizens a voice on questions of public policy.” James v. Valtierra, 402 U. S. 137, 141 (1971). From its earliest days, it was designed to circumvent the undue influence of large corporate interests on government decisionmaking.8 It served, as President Wilson put it, as a “gun behind the door” to keep political bosses and legislators honest. In more recent years, concerned that the heavy financial participation by corporations in referendum contests has undermined this tool of direct democracy, the voters of California enacted by *311initiative in 1974 the Political Reform Act, which limited expenditures in statewide ballot measure campaigns,9 and Berkeley voters adopted the ordinance at issue in this case. The role of the initiative in California cannot be separated from its purpose of preventing the dominance of special interests. That is the very history and purpose of the initiative in California, and similarly it is the purpose of ancillary regulations designed to protect it. Both serve to maximize the exchange of political discourse. As in Bellotti, “[t]he Court’s fundamental error is its failure to realize that the state regulatory interests . . . are themselves derived from the First Amendment.” 435 U. S., at 803-804 (White, J., dissenting).

Perhaps, as I have said, neither the city of Berkeley nor the State of California can “prove” that elections have been or can be unfairly won by special interest groups spending large sums of money, but there is a widespread conviction in legislative halls, as well as among citizens, that the danger is real. I regret that the Court continues to disregard that hazard.

concurring.

I agree that the judgment of the Supreme Court of California must be reversed in this case. Unlike the factual situation in First National Bank of Boston v. Bellotti, 435 U. S. 765 (1978), the Berkeley ordinance was not aimed only at corporations, but sought to impose an across-the-board limitation on the size of contributions to committees formed to support or oppose ballot measure referenda. While one of the appellants here, Mason-McDuffie, is a California corporation, there is no indication that the Berkeley ordinance was aimed at corporations as opposed to individuals. Therefore, my dissenting opinion -in First National Bank of Boston v. Bellotti, supra, which relied on the corporate shield which the State had granted to corporations as a form of quid pro quo for the limitation does not come into play. Buckley v. Valeo, 424 U. S. 1 (1976), holds that in this situation there is no state interest which could justify a limitation on the exercise of rights guaranteed under the First and Fourteenth Amendments to the United States Constitution.

concurring in the judgment.

The Court today holds that a local ordinance restricting the amount of money that an individual can contribute to a committee organized to support or oppose a ballot measure violates the right to freedom of speech and association guaranteed by the First Amendment. In reaching this conclusion, however, the Court fails to indicate whether or not it attaches any constitutional significance to the fact that the Berkeley ordinance seeks to limit contributions as opposed to direct expenditures. As Justice White correctly notes in dissent, beginning with our decision in Buckley v. Valeo, 424 U. S. 1 (1976), this Court has always drawn a distinction between restrictions on contributions, and direct limitations on the amount an individual can expend for his own speech. As we noted last term in California Medical Assn. v. FEC, 453 U. S. 182, 196 (1980) (Marshall, J., joined by Brennan, White, and Stevens, JJ.), the “‘speech by proxy’” that is achieved through contributions to a political campaign committee “is not the sort of political advocacy that this Court in Buckley found entitled to full First Amendment protection.”

Because the Court’s opinion is silent on the standard of review it is applying to this contributions limitation, I must assume that the Court is following our consistent position that this type of governmental action is subjected to less rigorous scrutiny than a direct restriction on expenditures. The city of Berkeley seeks to justify its ordinance on the ground that it is necessary to maintain voter confidence in government. If I found that the record before the California Supreme Court disclosed sufficient evidence to justify the conclusion that large contributions to ballot measure committees undermined the “confidence of the citizenry in government,” First National Bank of Boston v. Bellotti, 435 U. S. 765, 790 (1978), I would join Justice White in dissent on the ground that the State had demonstrated a sufficient governmental interest to sustain the indirect infringement on First Amend*302ment interests resulting from the operation of the Berkeley ordinance. Like Justices Blackmun and O’Connor, however, I find no such evidentiary support in this record. I therefore concur in the judgment.