with whom Justice Ginsbueg, Justice Breyer, and Justice Sotomayor join, concurring in part and dissenting in part.
The real issue in this case concerns how, not if, the appellant may finance its electioneering. Citizens United is a wealthy nonprofit corporation that runs a political action committee (PAC) with millions of dollars in assets. Under the Bipartisan Campaign Reform Act of 2002 (BCRA), it could have used those assets to televise and promote Hillary: The Movie wherever and whenever it wanted to. It also could have spent unrestricted sums to broadcast Hillary at any time other than the 30 days before the last primary election. Neither Citizens United’s nor any other corporation’s speech has been “banned,” ante, at 319. All that the parties dispute is whether Citizens United had a right to use the funds in its general treasury to pay for broadcasts during the 30-day period. The notion that the First Amendment *394dictates an affirmative answer to that question is, in my judgment, profoundly misguided. Even more misguided is the notion that the Court must rewrite the law relating to campaign expenditures by for-profit corporations and unions to decide this case.
The basic premise underlying the Court’s ruling is its iteration, and constant reiteration, of the proposition that the First Amendment bars regulatory distinctions based on a speaker’s identity, including its “identity” as a corporation. While that glittering generality has rhetorical appeal, it is not a correct statement of the law. Nor does it tell us when a corporation may engage in electioneering that some of its shareholders oppose. It does not even resolve the specific question whether Citizens United may be required to finance some of its messages with the money in its PAC. The conceit that corporations must be treated identically to natural persons in the political sphere is not only inaccurate but also inadequate to justify the Court’s disposition of this case.
In the context of election to public office, the distinction between corporate and human speakers is significant. Although they make enormous contributions to our society, corporations are not actually members of it. They cannot vote or run for office. Because they may be managed and controlled by nonresidents, their interests may conflict in fundamental respects with the interests of eligible voters. The financial resources, legal structure, and instrumental orientation of corporations raise legitimate concerns about their role in the electoral process. Our lawmakers have a compelling constitutional basis, if not also a democratic duty, to take measures designed to guard against the potentially deleterious effects of corporate spending in local and national races.
The majority’s approach to corporate electioneering marks a dramatic break from our past. Congress has placed special limitations on campaign spending by corporations ever since the passage of the Tillman Act in 1907, ch. 420, 34 Stat. 864. We have unanimously concluded that this “reflects a *395permissible assessment of the dangers posed by those entities to the electoral process,” FEC v. National Right to Work Comm., 459 U. S. 197, 209 (1982) (NRWC), and have accepted the “legislative judgment that the special characteristics of the corporate structure require particularly careful regulation,” id., at 209-210. The Court today rejects a century of history when it treats the distinction between corporate and individual campaign spending as an invidious novelty born of Austin v. Michigan Chamber of Commerce, 494 U. S. 652 (1990). Relying largely on individual dissenting opinions, the majority blazes through our precedents, overruling or disavowing a body of case law including FEC v. Wisconsin Right to Life, Inc., 551 U. S. 449 (2007) (WRTL), McConnell v. FEC, 540 U. S. 93 (2003), FEC v. Beaumont, 539 U. S. 146 (2003), FEC v. Massachusetts Citizens for Life, Inc., 479 U. S. 238 (1986) (MCFL), NRWC, 459 U. S. 197, and California Medical Assn. v. FEC, 453 U. S. 182 (1981).
In his landmark concurrence in Ashwander v. TVA, 297 U. S. 288, 346 (1936), Justice Brandéis stressed the importance of adhering to rules the Court has “developed ... for its own governance” when deciding constitutional questions. Because departures from those rules always enhance the risk of error, I shall review the background of this case in some detail before explaining why the Court’s analysis rests on a faulty understanding of Austin and McConnell and of our campaign finance jurisprudence more generally.1 I regret the length of what follows, but the importance and novelty of the Court’s opinion require a full response. Although *396I concur in the Court’s decision to sustain BCRA’s disclosure provisions and join Part IV of its opinion, I emphatically dissent from its principal holding.
I
The Court’s ruling threatens to undermine the integrity of elected institutions across the Nation. The path it has taken to reach its outcome will, I fear, do damage to this institution. Before turning to the question whether to overrule Austin and part of McConnell, it is important to explain why the Court should not be deciding that question.
Scope of the Case
The first reason is that the question was not properly brought before us. In declaring § 203 of BCRA facially unconstitutional on the ground that corporations’ electoral expenditures may not be regulated any more stringently than those of individuals, the majority decides this case on a basis relinquished below, not included in the questions presented to us by the litigants, and argued here only in response to the Court’s invitation. This procedure is unusual and inadvisable for a court.2 Our colleagues’ suggestion that “we are asked to reconsider Austin and, in effect, McConnell,” ante, at 319, would be more accurate if rephrased to state that “we have asked ourselves” to reconsider those cases.
In the District Court, Citizens United initially raised a facial challenge to the constitutionality of §203. App. 23a-*39724a. In its motion for summary judgment, however, Citizens United expressly abandoned its facial challenge, 1:07-cv-2240-RCL-RWR, Docket Entry No. 52, pp. 1-2 (May 16, 2008), and the parties stipulated to the dismissal of that claim, id., Nos. 53 (May 22,2008), 54 (May 23, 2008), App. 6a. The District Court therefore resolved the case on alternative grounds,3 and in its jurisdictional statement to this Court, Citizens United properly advised us that it was raising only “an as-applied challenge to the constitutionality of... BCRA §203.” Juris. Statement 5. The jurisdictional statement never so much as cited Austin, the key case the majority today overrules. And not one of the questions presented suggested that Citizens United was surreptitiously raising the facial challenge to § 203 that it previously agreed to dismiss. In fact, not one of those questions raised an issue based on Citizens United’s corporate status. Juris. Statement (i). Moreover, even in its merits briefing, when Citizens United injected its request to overrule Austin, it never sought a declaration that § 203 was facially unconstitutional as to all corporations and unions; instead it argued only that the statute could not be applied to it because it was “fimded overwhelmingly by individuals. ” Brief for Appellant 29; see also id., at 10, 12, 16, 28 (affirming “as applied” character of challenge to §203); Tr. of Oral Arg. 4-9 (Mar. 24, 2009) (coun*398sel for Citizens United conceding that § 203 could be applied to General Motors); id., at 55 (counsel for Citizens United stating that “we accept the Court’s decision in [WRTL]”).
“ ‘It is only in exceptional cases coming here from the federal courts that questions not pressed or passed upon below are reviewed,’ ” Youakim v. Miller, 425 U. S. 231, 234 (1976) (per curiam) (quoting Duignan v. United States, 274 U. S. 195, 200 (1927)), and it is “only in the most exceptional cases” that we will consider issues outside the questions presented, Stone v. Powell, 428 U. S. 465, 481, n. 15 (1976). The appellant in this case did not so much as assert an exceptional circumstance, and one searches the majority opinion in vain for the mention of any. That is unsurprising, for none exists.
Setting the case for reargument was a constructive step, but it did not cure this fundamental problem. Essentially, five Justices were unhappy with the limited nature of the case before us, so they changed the case to give themselves an opportunity to change the law.
As-Applied and Facial Challenges
This Court has repeatedly emphasized in recent years that “[fjacial challenges are disfavored.” Washington State Grange v. Washington State Republican Party, 552 U. S. 442, 450 (2008); see also Ayotte v. Planned Parenthood of Northern New Eng., 546 U. S. 320, 329 (2006) (“[T]he ‘normal rule’ is that ‘partial, rather .than facial, invalidation is the required course,’ such that a ‘statute may ... be declared invalid to the extent that it reaches too far, but otherwise left intact’ ” (quoting Brockett v. Spokane Arcades, Inc., 472 U. S. 491, 504 (1985); alteration in original)). By declaring § 203 facially unconstitutional, our colleagues have turned an as-applied challenge into a facial challenge, in defiance of this principle.
This is not merely a technical defect in the Court’s decision. The unnecessary resort to a facial inquiry “run[s] con*399trary to the fundamental principle of judicial restraint that courts should neither anticipate a question of constitutional law in advance of the necessity of deciding it nor formulate a rule of constitutional law broader than is required by the precise facts to which it is to be applied.” Washington State Grange, 552 U. S., at 450 (internal quotation marks omitted). Scanting that principle “threaten[s] to short circuit the democratic process by preventing laws embodying the will of the people from being implemented in a manner consistent with the Constitution.” Id., at 451. These concerns are heightened when judges overrule settled doctrine upon which the legislature has relied. The Court operates with a sledge hammer rather than a scalpel when it strikes down one of Congress’ most significant efforts to regulate the role that corporations and unions play in electoral polities. It compounds the offense by implicitly striking down a great many state laws as well.
The problem goes still deeper, for the Court does all of this on the basis of pure speculation. Had Citizens United maintained a facial challenge, and thus argued that there are virtually no circumstances in which BCRA §203 can be applied constitutionally, the parties could have developed, through the normal process of litigation, a record about the actual effects of § 203, its actual burdens and its actual benefits, on all manner of corporations and unions.4 “Claims of facial invalidity often rest on speculation,” and consequently “raise the risk of premature interpretation of statutes on the *400basis of factually barebones records.” Id., at 450 (internal quotation marks omitted). In this case, the record is not simply incomplete or unsatisfactory; it is nonexistent. Congress crafted BCRA in response to a virtual mountain of research on the corruption that previous legislation had failed to avert. The Court now negates Congress’ efforts without a shred of evidence on how §203 or its state-law counterparts have been affecting any entity other than Citizens United.5
Faced with this gaping empirical hole, the majority throws up its hands. Were we to confine our inquiry to Citizens United’s as-applied challenge, it protests, we would commence an “extended” process of “draw[ing], and then redrawing], constitutional lines based on the particular media or technology used to disseminate political speech from a particular speaker.” Ante, at 326. While tacitly acknowledging that some applications of § 203 might be found constitutional, the majority thus posits a future in which novel First Amendment standards must be devised on an ad hoc basis, and then leaps from this unfounded prediction to the unfounded conclusion that such complexity counsels the abandonment of all normal restraint. Yet it is a pervasive *401feature of regulatory systems that unanticipated events, such as new technologies, may raise some unanticipated difficulties at the margins. The fluid nature of electioneering communications does not make this case special. The fact that a Court can hypothesize situations in which a statute might, at some point down the line, pose some unforeseen as-applied problems, does not come close to meeting the standard for a facial challenge.6
The majority proposes several other justifications for the sweep of its ruling. It suggests that a facial ruling is necessary because, if the Court were to continue on its normal course of resolving as-applied challenges as they present themselves, that process would itself run afoul of the First Amendment. See, e. g., ante, at 326 (as-applied review process “would raise questions as to the courts’ own lawful authority”); ibid. (“Courts, too, are bound by the First Amendment”). This suggestion is perplexing. Our colleagues elsewhere trumpet “our duty ‘to say what the law is,’ ” even when our predecessors on the bench and our counterparts in Congress have interpreted the law differently. Ante, at 365 (quoting Marbury v. Madison, 1 Cranch 137, 177 (1803)). We do not typically say what the law is not as a hedge against future judicial error. The possibility that later courts will misapply a constitutional provision does not give
*402us a basis for pretermitting litigation relating to that provision.7
The majority suggests that a facial ruling is necessary because anything less would chill too much protected speech. See ante, at 326-327, 329, 333-336. In addition to begging the question what types of corporate spending are constitutionally protected and to what extent, this claim rests on the assertion that some significant number of corporations have been cowed into quiescence by FEC “ ‘censorship].’ ” Ante, at 335. That assertion is unsubstantiated, and it is hard to square with practical experience. It is particularly hard to square with the legal landscape following WRTL, which held that a corporate communication could be regulated under §203 only if it was “susceptible of no reasonable interpretation other than as an appeal to vote for or against a specific candidate.” 551 U. S., at 470 (opinion of Roberts, C. J.) (emphasis added). The whole point of this test was to make §203 as simple and speech-protective as possible. The Court does not explain how, in the span of a single election cycle, it has determined The Chief Justice’s project to be a failure. In this respect, too, the majority’s critique of line-drawing collapses into a critique of the as-applied review method generally.8
*403The majority suggests that, even though it expressly dismissed its facial challenge, Citizens United nevertheless preserved it — not as a freestanding “claim,” but as a potential argument in support of “a claim that the FEC has violated its First Amendment right to free speech.” Ante, at 330; see also ante, at 376 (Roberts, C. J., concurring) (describing Citizens United’s claim as: “[T]he Act violates the First Amendment”). By this novel logic, virtually any submission could be reconceptualized as “a claim that the Government has violated my rights,” and it would then be available to the Court to entertain any conceivable issue that might be relevant to that claim’s disposition. Not only the as-applied/facial distinction, but the basic relationship between litigants and courts, would be upended if the latter had free rein to construe the former’s claims at such high levels of generality. There would be no need for plaintiffs to argue their case; they could just cite the constitutional provisions they think relevant, and leave the rest to us.9
Finally, the majority suggests that though the scope of Citizens United’s claim may be narrow, a facial ruling is necessary as a matter of remedy. Relying on a law review article, it asserts that Citizens United’s dismissal of the facial challenge does not prevent us “ ‘from making broader pronouncements of invalidity in properly “as-applied” cases.’” Ante, at 331 (quoting Fallon, As-Applied and Facial Challenges and *404Third-Party Standing, 113 Harv. L. Rev. 1321, 1339 (2000) (hereinafter Fallon)); accord, ante, at 376 (opinion of Roberts, C. J.) (“Regardless whether we label Citizens United’s claim a ‘facial’ or ‘as-applied’ challenge, the consequences of the Court’s decision are the same”). The majority is on firmer conceptual ground here. Yet even if one accepts this part of Professor Fallon’s thesis, one must proceed to ask which as-applied challenges, if successful, will “properly” invite or entail invalidation of the underlying statute.10 The paradigmatic case is a judicial determination that the legislature acted with an impermissible purpose in enacting a provision, as this carries the necessary implication that all fiiture as-applied challenges to the provision must prevail. See Fallon 1339-1340.
Citizens United’s as-applied challenge was not of this sort. Until this Court ordered reargument, its contention was that BCRA § 208 could not lawfully be applied to a feature-length video-on-demand film (such as Hillary) or to a nonprofit corporation exempt from taxation under 26 U. S. C. § 501(c)(4)11 and funded overwhelmingly by individuals (such as itself). See Brief for Appellant 16-41. Success on either of these claims would not necessarily carry any implications for the validity of § 203 as applied to other types of broadcasts, other *405types of corporations, or unions. It certainly would not invalidate the statute as applied to a large for-profit corporation. See Tr. of Oral Arg. 8, 4 (Mar. 24, 2009) (counsel for Citizens United emphasizing that appellant is “a small, nonprofit organization, which is very much like [an MCFL corporation],” and affirming that its argument “definitely would not be the same” if Hillary were distributed by General Motors).12 There is no legitimate basis for resurrecting a facial challenge that dropped out of this case 20 months ago.
Narrower Grounds
It is all the more distressing that our colleagues have manufactured a facial challenge, because the parties have advanced numerous ways to resolve the case that would facilitate electioneering by nonprofit advocacy corporations such as Citizens United, without toppling statutes and precedents. Which is to say, the majority has transgressed yet another “cardinal” principle of the judicial process: “[I]f it is not necessary to decide more, it is necessary not to decide more,” PDK Labs. Inc. v. Drug Enforcement Admin., 362 F. 3d 786, *406799 (CADC 2004) (Roberts, J., concurring in part and concurring in judgment).
Consider just three of the narrower grounds of decision that the majority has bypassed. First, the Court eould have ruled, on statutory grounds, that a feature-length film distributed through video-on-demand does not qualify as an “electioneering communication” under §203 of BCRA, 2 U. S. C. § 441b. BCRA defines that term to encompass certain communications transmitted by “broadcast, cable, or satellite.” § 434(f)(3)(A). When Congress was developing BCRA, the video-on-demand medium was still in its infancy, and legislators were focused on a very different sort of programming: short advertisements run on television or radio. See McConnell, 540 U. S., at 207. The sponsors of BCRA acknowledge that the FEC’s implementing regulations do not clearly apply to video-on-demand transmissions. See Brief for Senator John McCain et al. as Amici Curiae 17-18. In light of this ambiguity, the distinctive characteristics of video-on-demand, and “[t]he elementary rule . . . that every reasonable construction must be resorted to, in order to save a statute from unconstitutionality,” Hooper v. California, 155 U. S. 648, 657 (1895), the Court could have reasonably ruled that §203 does not apply to Hillary.13
Second, the Court could have expanded the MCFL exemption to cover § 501(c)(4) nonprofits that accept only a de minimis amount of money from for-profit corporations. Citizens United professes to be such a group: Its brief says it “is funded predominantly by donations from individuals who support [its] ideological message.” Brief for Appellant 5. Numerous Courts of Appeals have held that de minimis business support does not, in itself, remove an otherwise *407qualifying organization from the ambit of MCFL.14 This Court could have simply followed their lead.15
Finally, let us not forget Citizens United’s as-applied constitutional challenge. Precisely because Citizens United looks so much like the MCFL organizations we have exempted from regulation, while a feature-length video-on-demand film looks so unlike the types of electoral advocacy Congress has found deserving of regulation, this challenge is a substantial one. As the appellant’s own arguments show, the Court could have easily limited the breadth of its constitutional holding had it declined to adopt the novel notion that speakers and speech acts must always be treated identically — and always spared expenditures restrictions — in the political realm. Yet the Court nonetheless turns its back on the as-applied review process that has been a staple of campaign finance litigation since Buckley v. Valeo, 424 U. S. 1 *408(1976) (per curiam), and that was affirmed and expanded just two Terms ago in WRTL, 551 U. S. 449.
This brief tour of alternative grounds on which the ease could have been decided is not meant to show that any of these grounds is ideal, though each is perfectly “valid,” ante, at 329 (majority opinion).16 It is meant to show that there were principled, narrower paths that a Court that was serious about judicial restraint could have taken. There was also the straightforward path: applying Austin and McConnell, just as the District Court did in holding that the funding of Citizens United's film can be regulated under them. The only thing preventing the majority from affirming the District Court, or adopting a narrower ground that would retain Austin, is its disdain for Austin.
II
The final principle of judicial process that the majority violates is the most transparent: stare decisis. I am not an absolutist when it comes to stare decisis, in the campaign finance area or in any other. No one is. But if this principle is to do any meaningful work in supporting the rule of law, it must at least demand a significant justification, beyond the preferences of five Justices, for overturning settled doctrine. “[A] decision to overrule should rest on some special reason *409over and above the belief that a prior case was wrongly decided.” Planned Parenthood of Southeastern Pa. v. Casey, 505 U. S. 833, 864 (1992). No such justification exists in this case, and to the contrary there are powerful prudential reasons to keep faith with our precedents.17
The Court’s central argument for why stare decisis ought to be trumped is that it does not like Austin. The opinion “was not well reasoned,” our colleagues assert, and it conflicts with First Amendment principles. Ante, at 363. This, of course, is the Court’s merits argument, the many defects in which we will soon consider. I am perfectly willing to concede that if one of our precedents were dead wrong in its reasoning or irreconcilable with the rest of our doctrine, there would be a compelling basis for revisiting it. But neither is true of Austin, as I explain at length in Parts III and IV, infra, at 414-478, and restating a merits argument with additional vigor does not give it extra weight in the stare decisis calculus.
Perhaps in recognition of this point, the Court supplements its merits case with a smattering of assertions. The Court proclaims that “Austin is undermined by experience since its announcement.” Ante, at 364. This is a curious claim to make in a case that lacks a developed record. The majority has no empirical evidence with which to substantiate the claim; we just have its ipse dixit that the real world has not been kind to Austin. Nor does the majority bother to specify in what sense Austin has been “undermined. ” Instead it treats the reader to a string of non sequiturs: “Our Nation’s speech dynamic is changing,” ante, at 364; “[speakers have become adept at presenting citizens with sound bites, talking points, and scripted messages,” ibid.; “[corporations ... do not have monolithic views,” ibid. How any *410of these ruminations weakens the force of stare decisis escapes my comprehension.18
The majority also contends that the Government’s hesitation to rely on Austin’s antidistortion rationale “diminished]” “the principle of adhering to that precedent.” Ante, at 363; see also ante, at 382 (opinion of Roberts, C. J.) (Government’s litigating position is “most importan[t]” factor undermining Austin). Why it diminishes the value of stare decisis is left unexplained. We have never thought fit to overrule a precedent because a litigant has taken any particular tack. Nor should we. Our decisions can often be defended on multiple grounds, and a litigant may have strategic or case-specific reasons for emphasizing only a subset of them. Members of the public, moreover, often rely on our bottom-line holdings far more than our precise legal arguments; surely this is true for the legislatures that have been regulating corporate electioneering since Austin. The task of evaluating the continued viability of precedents falls to this Court, not to the parties.19
*411Although the majority opinion spends several pages making these surprising arguments, it says almost nothing about the standard considerations we have used to determine stare decisis value, such as the antiquity of the precedent, the workability of its legal rule, and the reliance interests at stake. It is also conspicuously silent about McConnell, even though the McConnell Court’s decision to uphold BCRA § 203 relied not only on the antidistortion logic of Austin but also on the statute’s historical pedigree, see, e. g., 540 U. S., at 115-132, 223-224, and the need to preserve the integrity of federal campaigns, see id., at 126-129, 205-208, and n. 88.
We have recognized that “[s]tare decisis has special force when legislators or citizens ‘have acted in reliance on a previous decision, for in this instance overruling the decision would dislodge settled rights and expectations or require an extensive legislative response.’ ” Hubbard v. United States, 514 U. S. 695, 714 (1995) (plurality opinion) (quoting Hilton v. South Carolina Public Railways Comm’n, 502 U. S. 197, 202 (1991)). Stare decisis protects not only personal rights involving property or contract but also the ability of the elected branches to shape their laws in an effective and coherent fashion. Today’s decision takes away a power that we have long permitted these branches to exercise. State legislatures have relied on their authority to regulate corporate electioneering, confirmed in Austin, for more than a century.20 The Federal Congress has relied on this authority for a comparable stretch of time, and it specifically relied on Austin throughout the years it spent developing and de*412bating BCR A. The total record it compiled was 100,000 pages long.21 Pulling out the rug beneath Congress after affirming the constitutionality of §203 six years ago show's great disrespect for a coequal branch.
By removing one of its central components, today’s ruling makes a hash out of BCRA’s “delicate and interconnected regulatory scheme.” McConnell, 540 U. S., at 172. Consider just one example of the distortions that will follow: Political parties are barred under BCRA from soliciting or spending “soft money,” funds that are not subject to the statute’s disclosure requirements or its source and amount limitations. 2 U. S. C. § 441i; McConnell, 540 U. S., at 122-126. Going forward, corporations and unions will be free to spend as much general treasury money as they wish on ads that support or attack specific candidates, whereas national parties will not be able to spend a dime of soft money on ads of any kind. The Court’s ruling thus dramatically enhances the role of corporations and unions — and the narrow interests they represent — vis-a-vis the role of political parties— and the broad coalitions they represent — in determining who will hold public office.22
Beyond the reliance interests at stake, the other stare decisis factors also cut against the Court. Considerations of antiquity are significant for similar reasons. McConnell is only six years old, but Austin has been on the books for two decades, and many of the statutes called into question by today’s opinion have been on the books for a half century or more. The Court points to no intervening change in circumstances that warrants revisiting Austin. Certainly nothing *413relevant has changed since we decided WRTL two Terms ago. And the Court gives no reason to think that Austin and McConnell are unworkable.
In fact, no one has argued to us that Austin’s rule has proved impracticable, and not a single for-profit corporation, union, or State has asked us to overrule it. Quite to the contrary, leading groups representing the business community,23 organized labor,24 and the nonprofit sector,25 together with more than half of the States26 urge that we preserve Austin. As for McConnell, the portions of BCR A it upheld may be prolix, but all three branches of Government have worked to make §203 as user-friendly as possible. For instance, Congress established a special mechanism for expedited review of constitutional challenges, see note following 2 U. S. C. §437h; the FEC has established a standardized process, with clearly defined safe harbors, for corporations to claim that a particular electioneering communication is permissible under WRTL, see 11 CFR §114.15 (2009);27 and, as noted above, The Chief Justice crafted his controlling opinion in WRTL with the express goal of maximizing clarity and administrability, 551 U. S., at 469-470, 473-474. The case for stare decisis may be bolstered, we have said, when *414subsequent rulings “have reduced the impact” of a precedent “while reaffirming the decision’s core ruling.” Dickerson v. United States, 530 U. S. 428, 443 (2000).28
In the end, the Court’s rejection of Austin and McConnell comes down to nothing more than its disagreement with their results. Virtually every one of its arguments was made and rejected in those cases, and the majority opinion is essentially an amalgamation of resuscitated dissents. The only relevant thing that has changed since Austin and McConnell is the composition of this Court. Today’s ruling thus strikes at the vitals of stare decisis, “the means by which we ensure that the law will not merely change erratically, but will develop in a principled and intelligible fashion” that “permits society to presume that bedrock principles are founded in the law rather than in the proclivities of individuals.” Vasquez v. Hillery, 474 U. S. 254, 265 (1986).
III
The novelty of the Court’s procedural dereliction and its approach to stare decisis is matched by the novelty of its ruling on the merits. The ruling rests on several premises. First, the Court claims that Austin and McConnell have “banned” corporate speech. Second, it claims that the First Amendment precludes regulatory distinctions based on speaker identity, including the speaker’s identity as a corpo*415ration. Third, it claims that Austin and McConnell were radical outliers in our First Amendment tradition and our campaign finance jurisprudence. Each of these claims is wrong.
The So-Called “Ban”
Pervading the Court’s analysis is the ominous image of a “categorical ba[n]” on corporate speech. Ante, at 361. Indeed, the majority invokes the specter of a “ban” on nearly every page of its opinion. Ante, at 319, 321, 324, 327, 328, 329, 330, 333, 337, 339, 340, 343, 344, 345, 346, 347, 349, 351, 354, 355, 358, 360, 361, 362, 364, 369. This characterization is highly misleading, and needs to be corrected.
In fact it already has been. Our cases have repeatedly pointed out that, “[c]ontrary to the [majority’s] critical assumptions,” the statutes upheld in Austin and McConnell do “not impose an absolute ban on all forms of corporate political spending.” Austin, 494 U. S., at 660; see also McConnell, 540 U. S., at 203-204; Beaumont, 539 U. S., at 162-163. For starters, both statutes provide exemptions for PACs, separate segregated funds established by a corporation for political purposes. See 2 U. S. C. § 441b(b)(2)(C); Mich. Comp. Laws Ann. §169.255 (West 2005). “The ability to form and administer separate segregated funds,” we observed in McConnell, “has provided corporations and unions with a constitutionally sufficient opportunity to engage in express advocacy. That has been this Court’s unanimous view.” 540 U. S., at 203.
Under BCRA, any corporation’s “stockholders and their families and its executive or administrative personnel and their families” can pool their resources to finance electioneering communications. 2 U. S. C. § 441b(b)(4)(A)(i). A significant and growing number of corporations avail themselves of this option;29 during the most recent election cycle, *416corporate and union PACs raised nearly a billion dollars.30 Administering a PAC entails some administrative burden, but so does complying with the disclaimer, disclosure, and reporting requirements that the Court today upholds, see ante, at 366-367, and no one has suggested that the burden is severe for a sophisticated for-profit corporation. To the extent the majority is worried about this issue, it is important to keep in mind that we have no record to show how substantial the burden really is, just the majority’s own unsupported factfinding, see ante, at 337-339. Like all other natural persons, every shareholder of every corporation remains entirely free under Austin and McConnell to do however much electioneering she pleases outside of the corporate form. The owners of a “mom & pop” store can simply place ads in their own names, rather than the store’s. If ideologically aligned individuals wish to make unlimited expenditures through the corporate form, they may utilize an MCFL organization that has policies in place to avoid becoming a conduit for business or union interests. See MCFL, 479 U. S., at 263-264.
The laws upheld in Austin and McConnell leave open many additional avenues for corporations’ political speech. Consider the statutory provision we are ostensibly evaluating in this case, BCRA § 203. It has no application to genuine issue advertising — a category of corporate speech Congress found to be far more substantial than election-related advertising, see McConnell, 540 U. S., at 207 — or to Internet, *417telephone, and print advocacy.31 Like numerous statutes, it exempts media companies' news stories, commentaries, and editorials from its electioneering restrictions, in recognition of the unique role played by the institutional press in sustaining public debate.32 See 2 U. S. C. § 434(f)(3)(B)(i); McConnell, 540 U. S., at 208-209; see also Austin, 494 U. S., at 666-668. It also allows corporations to spend unlimited sums on political communications with their executives and shareholders, § 441b(b)(2)(A); 11 CFR § 114.3(a)(1), to fund additional PAC activity through trade associations, 2 U. S. C. § 441b(b)(4)(D), to distribute voting guides and voting records, 11 CFR §§ 114.4(c)(4)-(5), to underwrite voter registration and voter turnout activities, § 114.3(c)(4); § 114.4(c)(2), to host fundraising events for candidates within certain limits, *418§ 114.4(c); § 114.2(f)(2), and to publicly endorse candidates through a press release and press conference, § 114.4(c)(6).
At the time Citizens United brought this lawsuit, the only types of speech that could be regulated under §203 were: (1) broadcast, cable, or satellite communications;33 (2) capable of reaching at least 50,000 persons in the relevant electorate;34 (3) made within 30 days of a primary or 60 days of a general federal election;35 (4) by a labor union or a nonMCFL, nonmedia corporation;36 (5) paid for with general treasury funds;37 and (6) “susceptible of no reasonable interpretation other than as an appeal to vote for or against a specific candidate.”38 The category of communications meeting all of these criteria is not trivial, but the notion that corporate political speech has been “suppressfed] . . . altogether,” ante, at 319, that corporations have been “exelu[ded] . . . from the general public dialogue,” ante, at 341, or that a work of fiction such as Mr. Smith Goes to Washington might be covered, ante, at 371-372, is nonsense.39 Even the plaintiffs in McConnell, who had every incentive to depict BORA as negatively as possible, declined to argue that § 203’s prohibition on certain uses of general treasury funds amounts to a complete ban. See 540 U. S., at 204.
*419In many ways, then, §203 functions as a source restriction or a time, place, and manner restriction. It applies in a viewpoint-neutral fashion to a narrow subset of advocacy messages about clearly identified candidates for federal office, made during discrete time periods through discrete channels. In the case at hand, all Citizens United needed to do to broadcast Hillary right before the primary was to abjure business contributions or use the funds in its PAC, which by its own account is “one of the most active conservative PACs in America,” Citizens United Political Victory Fund, http://www.cupvf.org/.40
So let us be clear: Neither Austin nor McConnell held or implied that corporations may be silenced; the FEC is not a “censor”; and in the years since these cases were decided, corporations have continued to play a major role in the national dialogue. Laws such as §203 target a class of communications that is especially likely to corrupt the political process, that is at least one degree removed from the views of individual citizens, and that may not even reflect the views of those who pay for it. Such laws burden political speech, and that is always a serious matter, demanding careful scrutiny. But the majority’s incessant talk of a “ban” aims at a straw man.
Identity-Based Distinctions
The second pillar of the Court’s opinion is its assertion that “the Government cannot restrict political speech based on the speaker’s . . . identity.” Ante, at 346; accord, ante, at 319, 340-341, 342-343, 346-347, 347, 348, 349, 350, 364, 365. *420The case on which it relies for this proposition is First Nat. Bank of Boston v. Bellotti, 435 U. S. 765 (1978). As I shall explain, infra, at 442-446, the holding in that case was far narrower than the Court implies. Like its paeans to unfettered discourse, the Court’s denunciation of identity-based distinctions may have rhetorical appeal but it obscures reality.
“Our jurisprudence over the past 216 years has rejected an absolutist interpretation” of the First Amendment. WRTL, 551 U. S., at 482 (opinion of Roberts, C. J.). The First Amendment provides that “Congress shall make no law ... abridging the freedom of speech, or of the press.” Apart perhaps from measures designed to protect the press, that text might seem to permit no distinctions of any kind. Yet in a variety of contexts, we have held that speech can be regulated differentially on account of the speaker’s identity, when identity is understood in categorical or institutional terms. The Government routinely places special restrictions on the speech rights of students,41 prisoners,42 members of the Armed Forces,43 foreigners,44 and its own employees.45 *421When such restrictions are justified by a legitimate governmental interest, they do not necessarily raise constitutional problems.46 In contrast to the blanket rule that the majority espouses, our cases recognize that the Government’s interests may be more or less compelling with respect to different classes of speakers,47 cf. Minneapolis Star & Tribune Co. v. Minnesota Comm’r of Revenue, 460 U. S. 575, 585 (1983) (“[Djifferential treatment” is constitutionally suspect “unless justified by some special characteristic” of the regulated class of speakers (emphasis added)), and that the constitutional rights of certain categories of speakers, in certain contexts, “‘are not automatically coextensive with the rights’ ” that are normally accorded to members of our soci*422ety, Morse v. Frederick, 551 U. S. 393, 396-397, 404 (2007) (quoting Bethel School Dist. No. 403 v. Fraser, 478 U. S. 675, 682 (1986)).
The free speech guarantee thus does not render every other public interest an illegitimate basis for qualifying a speaker’s autonomy; society could scarcely function if it did. It is fair to say that our First Amendment doctrine has “frowned on” certain identity-based distinctions, Los Angeles Police Dept. v. United Reporting Publishing Corp., 528 U. S. 32, 47, n. 4 (1999) (Stevens, J., dissenting), particularly those that may reflect invidious discrimination or preferential treatment of a politically powerful group. But it is simply incorrect to suggest that we have prohibited all legislative distinctions based on identity or content. Not even close.
The election context is distinctive in many ways, and the Court, of course, is right thát the First Amendment closely guards political speech. But in this context, too, the authority of legislatures to enact viewpoint-neutral regulations based on content and identity is well settled. We have, for example, allowed state-run broadcasters to exclude independent candidates from televised debates. Arkansas Ed. Television Comm’n v. Forbes, 523 U. S. 666 (1998).48 We have upheld statutes that prohibit the distribution or display of campaign materials near a polling place. Burson v. Freeman, 504 U. S. 191 (1992).49 Although we have not reviewed *423them directly, we have never cast doubt on laws that place special restrictions on campaign spending by foreign nationals. See, e. g., 2 U. S. C. § 441e(a)(1). And we have consistently approved laws that bar Government employees, but not others, from contributing to or participating in political activities. See n. 45, supra. These statutes burden the political expression of one class of speakers, namely, civil servants. Yet we have sustained them on the basis of longstanding practice and Congress’ reasoned judgment that certain regulations which leave “untouched full participation ... in political decisions at the ballot box,” Civil Service Comm’n v. Letter Carriers, 413 U. S. 548, 556 (1973) (internal quotation marks omitted), help ensure that public officials are “sufficiently free from improper influences,” id., at 564, and that “confidence in the system of representative Government is not... eroded to a disastrous extent,” id., at 565.
The same logic applies to this case with additional force because it is the identity of corporations, rather than individuals, that the Legislature has taken into account. As we have unanimously observed, legislatures are entitled to decide “that the special characteristics of the corporate structure require particularly careful regulation” in an electoral context. NRWC, 459 U. S., at 209-210.50 Not only has the distinctive potential of corporations to corrupt the electoral process long been recognized, but within the area of campaign finance, corporate spending is also “furthest from the core of political expression, since corporations’ First Amendment speech and association interests are derived largely *424from those of their members and of the public in receiving information,” Beaumont, 539 U. S., at 161, n. 8 (citation omitted). Campaign finance distinctions based on corporate identity tend to be less worrisome, in other words, because the “speakers” are not natural persons, much less members of our political community, and the governmental interests are of the highest order. Furthermore, when corporations, as a class, are distinguished from noncorporations, as a class, there is a lesser risk that regulatory distinctions will reflect invidious discrimination or political favoritism.
If taken seriously, our colleagues’ assumption that the identity of a speaker has no relevance to the Government’s ability to regulate political speech would lead to some remarkable conclusions. Such an assumption would have accorded the propaganda broadcasts to our troops by “Tokyo Rose” during World War II the same protection as speech by Allied commanders. More pertinently, it would appear to afford the same protection to multinational corporations controlled by foreigners as to individual Americans: To do otherwise, after all, could “ ‘enhance the relative voice’ ” of some (i. e., humans) over others (i. e., nonhumans). Ante, at 349-350 (quoting Buckley, 424 U. S., at 49).51 Under the *425majority’s view, I suppose it may be a First Amendment problem that corporations are not permitted to vote, given that voting is, among other things, a form of speech.52
In short, the Court dramatically overstates its critique of identity-based distinctions, without ever explaining why corporate identity demands the same treatment as individual identity. Only the most wooden approach to the First Amendment could justify the unprecedented line it seeks to draw.
Our First Amendment Tradition
A third fulcrum of the Court’s opinion is the idea that Austin and McConnell are radical outliers, “aberration[s],” in our First Amendment tradition. Ante, at 355; see also ante, at 361, 372 (professing fidelity to “our law and our tradition”). The Court has it exactly backwards. It is today’s holding that is the radical departure from what had been settled First Amendment law. To see why, it is useful to take a long view.
1. Original Understandings
Let us start from the beginning. The Court invokes “ancient First Amendment principles,” ante, at 319 (internal quotation marks omitted), and original understandings, ante, at 353-354, to defend today’s ruling, yet it makes only a perfunctory attempt to ground its analysis in the principles or *426understandings of those who drafted and ratified the Amendment. Perhaps this is because there is not a scintilla of evidence to support the notion that anyone believed it would preclude regulatory distinctions based on the corporate form. To the extent that the Framers’ views are discernible and relevant to the disposition of this case, they would appear to cut strongly against the majority’s position.
This is not only because the Framers and their contemporaries conceived of speech more narrowly than we now think of it, see Bork, Neutral Principles and Some First Amendment Problems, 47 Ind. L. J. 1, 22 (1971), but also because they held very different views about the nature of the First Amendment right and the role of corporations in society. Those few corporations that existed at the founding were authorized by grant of a special legislative charter.53 Corporate sponsors would petition the legislature, and the legislature, if amenable, would issue a charter that specified the corporation’s powers and purposes and “authoritatively fixed *427the scope and content of corporate organization,” including “the internal structure of the corporation.” J. Hurst, The Legitimacy of the Business Corporation in the Law of the United States 1780-1970, pp. 15-16 (1970) (reprinted 2004). Corporations were created, supervised, and conceptualized as quasi-public entities, “designed to serve a social function for the state.” Handlin & Handlin, Origins of the American Business Corporation, 5 J. Econ. Hist. 1, 22 (1945). It was “assumed that [they] were legally privileged organizations that had to be closely scrutinized by the legislature because their purposes had to be made consistent with public welfare.” R. Seavoy, Origins of the American Business Corporation, 1784-1855, p. 5 (1982).
The individualized charter mode of incorporation reflected the “cloud of disfavor under which corporations labored” in the early years of this Nation. 1 W. Fletcher, Cyclopedia of the Law of Corporations §2, p. 8 (rev. ed. 2006); see also Louis K. Liggett Co. v. Lee, 288 U. S. 517, 548-549 (1933) (Brandeis, J., dissenting) (discussing fears of the “evils” of business corporations); L. Friedman, A History of American Law 194 (2d ed. 1985) (“The word 'soulless’ constantly recurs in debates over corporations.... Corporations, it was feared, could concentrate the worst urges of whole groups of men”). Thomas Jefferson famously fretted that corporations would subvert the Republic.54 General incorporation statutes, and widespread acceptance of business corporations as socially useful actors, did not emerge until the 1800’s. See Hansmann & Kraakman, The End of History for Corporate Law, 89 Geo. L. J. 439, 440 (2001) (hereinafter Hansmann & Kraakman) (“[A]ll general business corporation statutes appear to date from well after 1800”).
*428The Framers thus took it as a given that corporations could be comprehensively regulated in the service of the public welfare. Unlike our colleagues, they had little trouble distinguishing corporations from human beings, and when they constitutionalized the right to free speech in the First Amendment, it was the free speech of individual Americans that they had in mind.55 While individuals might join together to exercise their speech rights, business corporations, at least, were plainly not seen as facilitating such assoeiational or expressive ends. Even “the notion that business corporations could invoke the First Amendment would probably have been quite a novelty,” given that “at the time, the legitimacy of every corporate activity was thought to rest entirely in a concession of the sovereign.” Shelledy, Autonomy, Debate, and Corporate Speech, 18 Hastings Const. L. Q. 541, 578 (1991); cf. Trustees of Dartmouth College v. Woodward, 4 Wheat. 518, 636 (1819) (Marshall, *429C. J.) (“A corporation is an artificial being, invisible, intangible, and existing only in contemplation of law. Being the mere creature of law, it possesses only those properties which the charter of its creation confers upon it”); Eule, Promoting Speaker Diversity: Austin and Metro Broadcasting, 1990 S. Ct. Rev. 105, 129 (“The framers of the First Amendment could scarcely have anticipated its application to the corporation form. That, of course, ought not to be dispositive. What is compelling, however, is an understanding of who was supposed to be the beneficiary of the free speech guaranty — the individual”). In light of these background practices and understandings, it seems to me implausible that the Framers believed “the freedom of speech” would extend equally to all corporate speakers, much less that it would preclude legislatures from taking limited measures to guard against corporate capture of elections.
The Court observes that the Framers drew on diverse intellectual sources, communicated through newspapers, and aimed to provide greater freedom of speech than had existed in England. Ante, at 353. From these (accurate) observations, the Court concludes that “[t]he First Amendment was certainly not understood to condone the suppression of political speech in society’s most salient media.” Ibid. This conclusion is far from certain, given that many historians believe the Framers were focused on prior restraints on publication and did not -understand the First Amendment to “prevent the subsequent punishment of such [publications] as may be deemed contrary to the public welfare.” Near v. Minnesota ex rel. Olson, 283 U. S. 697, 714 (1931) (internal quotation marks omitted). Yet, even if the majority’s conclusion were correct, it would tell us only that the First Amendment was understood to protect political speech in certain media. It would tell us little about whether the Amendment was understood to protect general treasury electioneering expenditures by corporations, and to what extent.
*430As a matter of original expectations, then, it seems absurd to think that the First Amendment prohibits legislatures from taking into account the corporate identity of a sponsor of electoral advocacy. As a matter of original meaning, it likewise seems baseless — unless one evaluates the First Amendment’s “principles,” ante, at 319, 363, or its “purpose,” ante, at 376 (opinion of Roberts, C. J.), at such a high level of generality that the historical understandings of the Amendment cease to be a meaningful constraint on the judicial task. This case sheds a revelatory light on the assumption of some that an impartial judge’s application of an originalist methodology is likely to yield more determinate answers, or to play a more decisive role in the decisional process, than his or her views about sound policy.
Justice Scalia criticizes the foregoing discussion for failing to adduce statements from the founding era showing that corporations were understood to be excluded from the First Amendment’s free speech guarantee. Ante, at 386, 393. Of course, Justice Scalia adduces no statements to suggest the contrary proposition, or even to suggest that the contrary proposition better reflects the kind of right that the drafters and ratiflers of the Free Speech Clause thought they were enshrining. Although Justice Scalia makes a perfectly sensible argument that an individual’s right to speak entails a right to speak with others for a common cause, cf. MCFL, 479 U. S. 238, he does not explain why those two rights must be precisely identical, or why that principle applies to electioneering by corporations that serve no “common cause.” Ante, at 392. Nothing in his account dislodges my basic point that members of the founding generation held a cautious view of corporate power and a narrow view of corporate rights (not that they “despised” corporations, ante, at 386), and that they conceptualized speech in individualistic terms. If no prominent Framer bothered to articulate that corporate speech would have lesser status than individual speech, that may well be because the contrary proposition— *431if not also the very notion of “corporate speech” — was inconceivable.56
Justice Scalia also emphasizes the unqualified nature of the First Amendment text. Ante, at 386, 392-393. Yet he would seemingly read out the Free Press Clause: How else could he claim that my purported views on newspapers must track my views on corporations generally? Ante, at 390.57 Like virtually all modern lawyers, Justice Scalia presumably believes that the First Amendment restricts the Executive, even though its language refers to Congress alone. In any event, the text only leads us back to the questions who or what is guaranteed “the freedom of speech,” and, just as critically, what that freedom consists of and under what circumstances it may be limited. Justice Scalia appears to believe that because corporations are created and utilized by individuals, it follows (as night the day) that their electioneering must be equally protected by the First Amendment *432and equally immunized from expenditure limits. See ante, at 391-392. That conclusion certainly does not follow as a logical matter, and Justice Scalia fails to explain why the original public meaning leads it to follow as a matter of interpretation.
The truth is we cannot be certain how a law such as BCR A § 203 meshes with the original meaning of the First Amendment.58 I have given several reasons why I believe the Constitution would have been understood then, and ought to be understood now, to permit reasonable restrictions on corporate electioneering, and I will give many more reasons in the pages to come. The Court enlists the Framers in its defense without seriously grappling with their understandings of corporations or the free speech right, or with the republican principles that underlay those understandings.
In fairness, our campaign finance jurisprudence has never attended very closely to the views of the Framers, see Randall v. Sorrell, 548 U. S. 230, 280 (2006) (Stevens, J., dissenting), whose political universe differed profoundly from that of today. We have long since held that corporations are covered by the First Amendment, and many legal scholars have long since rejected the concession theory of the corporation. But “historical context is usually relevant,” ibid, (internal quotation marks omitted), and in light of the Court’s effort to cast itself as guardian of ancient values, it pays to remember that nothing in our constitutional history dictates today’s outcome. To the contrary, this history helps illuminate just how extraordinarily dissonant the decision is.
2. Legislative and Judicial Interpretation
A century of more recent history puts to rest any notion that today’s ruling is faithful to our First Amendment tradi*433tion. At the federal level, the express distinction between corporate and individual political spending on elections stretches back to 1907, when Congress passed the Tillman Act, ch. 420, 34 Stat. 864, banning all corporate contributions to candidates. The Senate Report on the legislation observed that “[t]he evils of the use of [corporate] money in connection with political elections are so generally recognized that the committee deems it unnecessary to make any argument in favor of the general purpose of this measure. It is in the interest of good government and calculated to promote purity in the selection of public officials.” S. Rep. No. 3056, 59th Cong., 1st Sess., 2 (1906). President Roosevelt, in his 1905 annual message to Congress, declared:
“ ‘All contributions by corporations to any political committee or for any political purpose should be forbidden by law; directors should not be permitted to use stockholders’ money for such purposes; and, moreover, a prohibition of this kind would be, as far as it went, an effective method of stopping the evils aimed at in corrupt practices acts.’ ” United States v. Automobile Workers, 352 U. S. 567, 572 (1957) (quoting 40 Cong. Rec. 96).
The Court has surveyed the history leading up to the Tillman Act several times, see WRTL, 551 U. S., at 508-510 (Souter, J., dissenting); McConnell, 540 U. S., at 115; Automobile Workers, 352 U. S., at 570-575, and I will refrain from doing so again. It is enough to say that the Act was primarily driven by two pressing concerns: first, the enormous power corporations had come to wield in federal elections, with the accompanying threat of both actual corruption and a public perception of corruption; and second, a respect for the interest of shareholders and members in preventing the use of their money to support candidates they opposed. See ibid.; United States v. CIO, 335 U. S. 106, 113 (1948); Winkler, “Other People’s Money”: Corporations, Agency Costs, and Campaign Finance Law, 92 Geo. L. J. 871 (2004).
*434Over the years, the limitations on corporate political spending have been modified in a number of ways, as Congress responded to changes in the American economy and political practices that threatened to displace the commonweal. Justice Souter recently traced these developments at length.59 WRTL, 551 U. S., at 507-519 (dissenting opinion); see also McConnell, 540 U. S., at 115-133; McConnell, 251 F. Supp. 2d, at 188-205. The Taft-Hartley Act of 1947 is of special significance for this case. In that Act passed more than 60 years ago, Congress extended the prohibition on corporate support of candidates to cover not only direct contributions, but independent expenditures as well. Labor Management Relations Act, 1947, § 304, 61 Stat. 159. The bar on contributions “was being so narrowly construed” that corporations were easily able to defeat the purposes of the Act by supporting candidates through other means. WRTL, 551 U. S., at 511 (Souter, J., dissenting) (citing S. Rep. No. 1, 80th Cong., 1st Sess., 38-39 (1947)).
Our colleagues emphasize that in two cases from the middle of the 20th century, several Justices wrote separately to criticize the expenditure restriction as applied to unions, even though the Court declined to pass on its constitutionality. Ante, at 343-344. Two features of these cases are of far greater relevance. First, those Justices were writing separately; which is to say, their position failed to command a majority. Prior to today, this was a fact we found signifi*435cant in evaluating precedents. Second, each case in this line expressed support for the principle that corporate and union political speech financed with PAC funds, collected voluntarily from the organization’s stockholders or members, receives greater protection than speech financed with general treasury funds.60
This principle was carried forward when Congress enacted comprehensive campaign finance reform in the Federal Election Campaign Act of 1971 (FECA), 86 Stat. 3, which retained the restriction on using general treasury funds for contributions and expenditures, 2 U. S. C. § 441b(a). FECA *436codified the option for corporations and unions to create PACs to finance contributions and expenditures forbidden to the corporation or union itself. § 441b(b).
By the time Congress passed FECA in 1971, the bar on corporate contributions and expenditures had become such an accepted part of federal campaign finance regulation that when a large number of plaintiffs, including several nonprofit corporations, challenged virtually every aspect of FECA in Buckley, 424 U. S. 1, no one even bothered to argue that the bar as such was unconstitutional. Buckley famously (or infamously) distinguished direct contributions from independent expenditures, id., at 58-59, but its silence on corporations only reinforced the understanding that corporate expenditures could be treated differently from individual expenditures. “Since our decision in Buckley, Congress’ power to prohibit corporations and unions from using funds in their treasuries to finance advertisements expressly advocating the election or defeat of candidates in federal elections has been firmly embedded in our law.” McConnell, 540 U. S., at 203.
Thus, it was unremarkable, in a 1982 case holding that Congress could bar nonprofit corporations from soliciting nonmembers for PAC funds, that then-justice Rehnquist wrote for a unanimous Court that Congress’ “careful legislative adjustment of the federal electoral laws, in a cautious advance, step by step, to account for the particular legal and economic attributes of corporations . . . warrants considerable deference,” and “reflects a permissible assessment of the dangers posed by those entities to the electoral process.” NRWC, 459 U. S., at 209 (internal quotation marks and citation omitted). “The governmental interest in preventing both actual corruption and the appearance of corruption of elected representatives has long been recognized,” the unanimous Court observed, “and there is no reason why it may not ... be accomplished by treating . . . corporations . . . differently from individuals.” Id., at 210-211.
*437The corporate/individual distinction was not questioned by the Court’s disposition, in 1986, of a challenge to the expenditure restriction as applied to a distinctive type of nonprofit corporation. In MCFL, 479 U. S. 238, we stated again “that ‘the special characteristics of the corporate structure require particularly careful regulation,’ ” id., at 256 (quoting NRWC, 459 U. S., at 209-210), and again we acknowledged that the Government has a legitimate interest in “regulat[ing] the substantial aggregations of wealth amassed by the special advantages which go with the corporate form,” 479 U. S., at 257 (internal quotation marks omitted). Those aggregations can distort the “free trade in ideas” crucial to candidate elections, ibid, (internal quotation marks omitted), at the expense of members or shareholders who may disagree with the object of the expenditures, id., at 260. What the Court held by a 5-to-4 vote was that a limited class of corporations must be allowed to use their general treasury funds for independent expenditures, because Congress’ interests in protecting shareholders and “restriet[ing] ‘the influence of political war chests funneled through the corporate form,’ ” id., at 257 (quoting FEC v. National Conservative Political Action Comm., 470 U. S. 480, 501 (1985) (NCPAC), did not apply to corporations that were structurally insulated from those concerns.61
It is worth remembering for present purposes that the four MCFL dissenters, led by Chief Justice Rehnquist, thought the Court was carrying the First Amendment too *438far. They would have recognized congressional authority to bar general treasury electioneering expenditures even by this class of nonprofits; they acknowledged that “the threat from corporate political activity will vary depending on the particular characteristics of a given corporation,” but believed these “distinctions among corporations” were “distinctions in degree,” not “in kind,” and thus “more properly drawn by the Legislature than by the Judiciary.” 479 U. S., at 268 (opinion of Rehnquist, C. J.) (internal quotation marks omitted). Not a single Justice suggested that regulation of corporate political speech could be no more stringent than of speech by an individual.
Four years later, in Austin, 494 U. S. 652, we considered whether corporations falling outside the MCFL exception could be barred from using general treasury funds to make independent expenditures in support of, or in opposition to, candidates. We held they could be. Once again recognizing the importance of “the integrity of the marketplace of political ideas” in candidate elections, MCFL, 479 U. S., at 257, we noted that corporations have “special advantages — such as limited liability, perpetual life, and favorable treatment of the accumulation and distribution of assets,” 494 U. S., at 658-659 — that allow them to spend prodigious general treasury sums on campaign messages that have “little or no correlation” with the beliefs held by actual persons, id., at 660. In light of the corrupting effects such spending might have on the political process, ibid., we permitted the State of Michigan to limit corporate expenditures on candidate elections to corporations’ PACs, which rely on voluntary contributions and thus “reflect actual public support for the political ideas espoused by corporations,” ibid. Notwithstanding our colleagues’ insinuations that Austin deprived the public of general “ideas,” “facts,” and “ ‘knowledge,’ ” ante, at 354, 355, the decision addressed only candidate-focused expenditures and gave the State no license to regulate corporate spending on other matters.
*439In the 20 years since Austin, we have reaffirmed its holding and rationale a number of times, see, e. g., Beaumont, 539 U. S., at 153-156, most importantly in McConnell, 540 U. S. 93, where we upheld the provision challenged here, § 203 of BCRA.62 Congress crafted § 203 in response to a problem created by Buckley. The Buckley Court had construed FECA’s definition of prohibited “expenditures” narrowly to avoid any problems of constitutional vagueness, holding it applicable only to “communications that expressly advocate the election or defeat of a clearly identified candidate,” 424 U. S., at 80, i. e., statements containing so-called “magic words” like “ ‘vote for,’ ‘elect,’ ‘support,’ ‘cast your ballot for,’ ‘Smith for Congress,’ ‘vote against,’ ‘defeat,’ [or] ‘reject,’” id., at 43-44, and n. 52. After Buckley, corporations and unions figured out how to circumvent the limits on express advocacy by using sham “issue ads” that “eschewed the use of magic words” but nonetheless “advocatefd] the election or defeat of clearly identified federal candidates.” McConnell, 540 U. S., at 126. “Corporations and unions spent hundreds *440of millions of dollars of their general funds to pay for these ads.” Id., at 127. Congress passed § 203 to address this circumvention, prohibiting corporations and unions from using general treasury funds for electioneering communications that “refe[r] to a clearly identified candidate,” whether or not those communications use the magic words. 2 U. S. C. § 434(f)(3)(A)(i)(I).
When we asked in McConnell “whether a compelling governmental interest justifie[d]” §203, we found the question “easily answered”: “We have repeatedly sustained legislation aimed at 'the corrosive and distorting effects of immense aggregations of wealth that are accumulated with the help of the corporate form and that have little or no correlation to the public’s support for the corporation’s political ideas.’” 540 U. S., at 205 (quoting Austin, 494 U. S., at 660). These precedents “represent respect for the legislative judgment that the special characteristics of the corporate structure require particularly careful regulation.” 540 U. S., at 205 (internal quotation marks omitted). “Moreover, recent cases have recognized that certain restrictions on corporate electoral involvement permissibly hedge against' “circumvention of [valid] contribution limits.””’ Ibid, (quoting Beaumont, 539 U. S., at 155, in turn quoting FEC v. Colorado Republican Federal Campaign Comm., 533 U. S. 431, 456, and n. 18 (2001) (Colorado II); alteration in original). BCR A, we found, is faithful to the compelling governmental interests in “ ‘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 the government,’ ” and maintaining “ 'the individual citizen’s confidence in government.’” 540 U. S., at 206-207, n. 88 (quoting Bellotti, 435 U. S., at 788-789; some internal quotation marks and brackets omitted). What made the answer even easier than it might have been otherwise was the option to form PACs, which give corporations, at the least, *441“a constitutionally sufficient opportunity to engage in” independent expenditures. 540 U. S., at 203.
3. Buckley and Bellotti
Against this extensive background of congressional regulation of corporate campaign spending, and our repeated affirmation of this regulation as constitutionally sound, the majority dismisses Austin as “a significant departure from ancient First Amendment principles,” ante, at 319 (internal quotation marks omitted). How does the majority attempt to justify this claim? Selected passages from two cases, Buckley, 424 U. S. 1, and Bellotti, 435 U. S. 765, do all of the work. In the Court’s view, Buckley and Bellotti decisively rejected the possibility of distinguishing corporations from natural persons in the 1970’s; it just so happens that in every single case in which the Court has reviewed campaign finance legislation in the decades since, the majority failed to grasp this truth. The Federal Congress and dozens of state legislatures, we now know, have been similarly deluded.
The majority emphasizes Buckley’s statement that “ ‘[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.’” Ante, at 349-350 (quoting 424 U. S., at 48-49); ante, at 379 (opinion of Roberts, C. J.). But this elegant phrase cannot bear the weight that our colleagues have placed on it. For one thing, the Constitution does, in fact, permit numerous “restrictions on the speech of some in order to prevent a few from drowning out the many”: for example, restrictions on ballot access and on legislators’ floor time. Nixon v. Shrink Missouri Government PAC, 528 U. S. 377, 402 (2000) (Breyer, J., concurring). For another, the Buckley Court used this line in evaluating “the ancillary governmental interest in equalizing the relative ability of individuals and groups to influence the outcome of elections.” 424 U. S., at 48. It is not apparent why this is relevant to the case *442before us. The majority suggests that Austin rests on the foreign concept of speech equalization, ante, at 350; ante, at 379-381 (opinion of ROBERTS, C. J.), but we made it clear in Austin (as in several cases before and since) that a restriction on the way corporations spend their money is no mere exercise in disfavoring the voice of some elements of our society in preference to others. Indeed, we expressly ruled that the compelling interest supporting Michigan’s statute was not one of “ ‘equalizing] the relative influence of speakers on elections,’ ” Austin, 494 U. S., at 660 (quoting id., at 705 (Kennedy, J., dissenting)), but rather the need to confront the distinctive corrupting potential of corporate electoral advocacy financed by general treasury dollars, id., at 659-660.
For that matter, it should go without saying that when we made this statement in Buckley, we could not have been casting doubt on the restriction on corporate expenditures in candidate elections, which had not been challenged as “foreign to the First Amendment,” ante, at 350 (quoting Buckley, 424 U. S., at 49), or for any other reason. Buckley’s independent expenditure analysis was focused on a very different statutory provision, 18 U. S. C. § 608(e)(1) (1970 ed., Supp. V). It is implausible to think, as the majority suggests, ante, at 346, that Buckley covertly invalidated FECA’s separate corporate and union campaign expenditure restriction, §610 (now codified at 2 U. S. C. §441b), even though that restriction had been on the books for decades before Buckley and would remain on the books, undisturbed, for decades after.
The case on which the majority places even greater weight than Buckley, however, is Bellotti, 435 U. S. 765, claiming it “could not have been clearer” that Bellotti’s holding forbade distinctions between corporate and individual expenditures like the one at issue here, ante, at 346. The Court’s reliance is odd. The only thing about Bellotti that could not be clearer is that it declined to adopt the majority’s position. Bellotti ruled, in an explicit limitation on the scope of its holding, that “our consideration of a corporation’s right to *443speak on issues of general public interest implies no comparable right in the quite different context of participation in a political campaign for election to public office.” 435 U. S., at 788, n. 26; see also id., at 787-788 (acknowledging that the interests in preserving public confidence in Government and protecting dissenting shareholders may be “weighty ... in the context of partisan candidate elections”). Bellotti, in other words, did not touch the question presented in Austin and McConnell, and the opinion squarely disavowed the proposition for which the majority cites it.
The majority attempts to explain away the distinction Bellotti drew — between general corporate speech and campaign speech intended to promote or prevent the election of specific candidates for office — as inconsistent with the rest of the opinion and with Buckley. Ante, at 347, 357-360. Yet the basis for this distinction is perfectly coherent: The anticorruption interests that animate regulations of corporate participation in candidate elections, the “importance” of which “has never been doubted,” 435 U. S., at 788, n. 26, do not apply equally to regulations of corporate participation in referenda. A referendum cannot owe a political debt to a corporation, seek to curry favor with a corporation, or fear the corporation’s retaliation. Cf. Austin, 494 U. S., at 678 (Stevens, J., concurring); Citizens Against Rent Control/Coalition for Fair Housing v. Berkeley, 454 U. S. 290, 299 (1981). The majority likewise overlooks the fact that, over the past 30 years, our cases have repeatedly recognized the candidate/issue distinction. See, e. g., Austin, 494 U. S., at 659; NCPAC, 470 U. S., at 495-496; FCC v. League of Women Voters of Cal., 468 U. S. 364, 371, n. 9 (1984); NRWC, 459 U. S., at 210, n. 7. The Court’s critique of Bellotti’s footnote 26 puts it in the strange position of trying to elevate Bellotti to canonical status, while simultaneously disparaging a critical piece of its analysis as unsupported and irreconcilable with Buckley. Bellotti, apparently, is both the font of all wisdom and internally incoherent.
*444The Bellotti Court confronted a dramatically different factual situation from the one that confronts us in this case: a state statute that barred business corporations’ expenditures on some referenda but not others. Specifically, the statute barred a business corporation “from making contributions or expenditures ‘for the purpose of . . . influencing or affecting the vote on any question submitted to the voters, other than one materially affecting any of the property, business or assets of the corporation,’ ” 435 U. S., at 768 (quoting Mass. Gen. Laws Ann., ch. 55, § 8 (West Supp. 1977); alteration in original), and it went so far as to provide that referenda related to income taxation would not “ ‘be deemed materially to affect the property, business or assets of the corporation,’ ” 435 U. S., at 768. As might be guessed, the legislature had enacted this statute in order to limit corporate speech on a proposed state constitutional amendment to authorize a graduated income tax. The statute was a transparent attempt to prevent corporations from spending money to defeat this amendment, which was favored by a majority of legislators but had been repeatedly rejected by the voters. See id., at 769-770, and n. 3. We said that “where, as here, the legislature’s suppression of speech suggests an attempt to give one side of a debatable public question an advantage in expressing its views to the people, the First Amendment is plainly offended.” Id., at 785-786 (footnote omitted).
Bellotti thus involved a viewpoint-discriminatory statute, created to effect a particular policy outcome. Even Justice Rehnquist, in dissent, had to acknowledge that “a very persuasive argument could be made that the [Massachusetts Legislature], desiring to impose a personal income tax but more than once defeated in that desire by the combination of the Commonwealth’s referendum provision and corporate expenditures in opposition to such a tax, simply decided to muzzle corporations on this sort of issue so that it could succeed in its desire.” Id., at 827, n. 6. To make matters *445worse, the law at issue did not make any allowance for corporations to spend money through PACs. Id., at 768, n. 2 (opinion of the Court). This really was a complete ban on a specific, preidentified subject. See MCFL, 479 U. S., at 259, n. 12 (stating that 2 U. S. C. §441b’s expenditure restriction “is of course distinguishable from the complete foreclosure of any opportunity for political speech that we invalidated in the state referendum context in . . . Bellotti” (emphasis added)).
The majority grasps a quotational straw from Bellotti, that speech does not fall entirely outside the protection of the First Amendment merely because it comes from a corporation. Ante, at 346-347. Of course not, but no one suggests the contrary, and neither Austin nor McConnell held otherwise. They held that even though the expenditures at issue were subject to First Amendment scrutiny, the restrictions on those expenditures were justified by a compelling state interest. See McConnell, 540 U. S., at 205; Austin, 494 U. S., at 658, 660. We acknowledged in Bellotti that numerous “interests of the highest importance” can justify campaign finance regulation. 435 U. S., at 788-789. But we found no evidence that these interests were served by the Massachusetts law. Id., at 789. We left open the possibility that our decision might have been different if there had been “record or legislative findings that corporate advocacy threatened imminently to undermine democratic processes, thereby denigrating rather than serving First Amendment interests.” Ibid.
Austin and McConnell, then, sit perfectly well with Bellotti. Indeed, all six Members of the Austin majority had been on the Court at the time of Bellotti, and none so much as hinted in Austin that they saw any tension between the decisions. The difference between the cases is not that Austin and McConnell rejected First Amendment protection for corporations whereas Bellotti accepted it. The difference is that the statute at issue in Bellotti smacked of viewpoint *446discrimination, targeted one class of corporations, and provided no PAC option; and the State has a greater interest in regulating independent corporate expenditures on candidate elections than on referenda, because in a functioning democracy the public must have faith that its representatives owe their positions to the people, not to the corporations with the deepest pockets.
* * *
In sum, over the course of the past century Congress has demonstrated a recurrent need to regulate corporate participation in candidate elections to “ ‘[p]reserv[e] the integrity of the electoral process, preven[t] corruption, . . . sustai[n] the active, alert responsibility of the individual citizen,’” protect the expressive interests of shareholders, and “ ‘[p]reserv[e] . . . the individual citizen’s confidence in government.’” McConnell, 540 U. S., at 206-207, n. 88 (quoting Bellotti, 435 U. S., at 788-789; first alteration in original). These understandings provided the combined impetus behind the Tillman Act in 1907, see Automobile Workers, 352 U. S., at 570-575, the Taft-Hartley Act in 1947, see WRTL, 551 U. S., at 511 (Souter, J., dissenting), FECA in 1971, see NRWC, 459 U. S., at 209-210, and BCRA in 2002, see McConnell, 540 U. S., at 126-132. Continuously for over 100 years, this line of “[c]ampaign finance reform has been a series of reactions to documented threats to electoral integrity obvious to any voter, posed by large sums of money from corporate or union treasuries.” WRTL, 551 U. S., at 522 (Souter, J., dissenting). Time and again, we have recognized these realities in approving measures that Congress and the States have taken. None of the cases the majority cites is to the contrary. The only thing new about Austin was the dissent, with its stunning failure to appreciate the legitimacy of interests recognized in the name of democratic integrity since the days of the Progressives.
*447IV
Having explained why this is not an appropriate case in which to revisit Austin and McConnell and why these decisions sit perfectly well with “First Amendment principles,” ante, at 319, 363, I come at last to the interests that are at stake. The majority recognizes that Austin and McConnell may be defended on anticorruption, antidistortion, and shareholder protection, rationales. Ante, at 348-362. It badly errs both in explaining the nature of these rationales, which overlap and complement each other, and in applying them to the ease at hand.
The Anticorruption Interest
Undergirding the majority’s approach to the merits is the claim that the only “sufficiently important governmental interest in preventing corruption or the appearance of corruption” is one that is “limited to quid pro quo corruption.” Ante, at 359. This is the same “crabbed view of corruption” that was espoused by Justice Kennedy in McConnell and squarely rejected by the Court in that case. 540 U. S., at 152. While it is true that we have not always spoken about corruption in a clear or consistent voice, the approach taken by the majority cannot be right, in my judgment. It disregards our constitutional history and the fundamental demands of a democratic society.
On numerous occasions we have recognized Congress’ legitimate interest in preventing the money that is spent on elections from exerting an “‘undue influence on an officeholder’s judgment’ ” and from creating “ ‘the appearance of such influence,’” beyond the sphere of quid pro quo relationships. Id., at 150; see also, e.g., id., at 143-144, 152-154; Colorado II, 533 U. S., at 441; Shrink Missouri, 528 U. S., at 389. Corruption can take many forms. Bribery may be the paradigm case. But the difference between selling a vote and selling access is a matter of degree, not kind. And sell*448ing access is not qualitatively different from giving special preference to those who spent money on one's behalf. Corruption operates along a spectrum, and the majority’s apparent belief that quid pro quo arrangements can be neatly demarcated from other improper influences does not accord with the theory or reality of politics. It certainly does not accord with the record Congress developed in passing BCRA, a record that stands as a remarkable testament to the energy and ingenuity with which corporations, unions, lobbyists, and politicians may go about scratching each other’s backs — and which amply supported Congress’ determination to target a limited set of especially destructive practices.
The District Court that adjudicated the initial challenge to BCRA pored over this record. In a careful analysis, Judge Kollar-Kotelly made numerous findings about the corrupting consequences of corporate and union independent expenditures in the years preceding BCRA’s passage. See McConnell, 251 F. Supp. 2d, at 555-560, 622-625; see also id., at 804-805, 813, n. 143 (Leon, J.) (indicating agreement). As summarized in her own words:
“The factual findings of the Court illustrate that corporations and labor unions routinely notify Members of Congress as soon as they air electioneering communications relevant to the Members’ elections. The record also indicates that Members express appreciation to organizations for the airing of these election-related advertisements. Indeed, Members of Congress are particularly grateful when negative issue advertisements are run by these organizations, leaving the candidates free to run positive advertisements and be seen as 'above the fray.’ Political consultants testify that campaigns are quite aware of who is running advertisements on the candidate’s behalf, when they are being run, and where they are being run. Likewise, a prominent lob*449byist testifies that these organizations use issue advocacy as a means to influence various Members of Congress.
“The Findings also demonstrate that Members of Congress seek to have corporations and unions run these advertisements on their behalf. The Findings show that Members suggest that corporations or individuals make donations to interest groups with the understanding that the money contributed to these groups will assist the Member in a campaign. After the election, these organizations often seek credit for their support----Finally, a large majority of Americans (80%) are of the view that corporations and other organizations that engage in electioneering communications, which benefit specific elected officials, receive special consideration from those officials when matters arise that affect these corporations and organizations.” Id., at 623-624 (citations and footnote omitted).
Many of the relationships of dependency found by Judge Kollar-Kotelly seemed to have a quid pro quo basis, but other arrangements were more subtle. Her analysis shows the great difficulty in delimiting the precise scope of the quid pro quo category, as well as the adverse consequences that all such arrangements may have. There are threats of corruption that are far more destructive to a democratic society than the odd bribe. Yet the majority’s understanding of corruption would leave lawmakers impotent to address all but the most discrete abuses.
Our “undue influence” cases have allowed the American people to cast a wider net through legislative experiments designed to ensure, to some minimal extent, “that officeholders will decide issues ... on the merits or the desires of their constituencies,” and not “according to the wishes of those who have made large financial contributions” — or expenditures — “valued by the officeholder.” McConnell, 540 *450U. S., at 153.63 When private interests are seen to exert outsized control over officeholders solely on account of the money spent on (or withheld from) their campaigns, the result can depart so thoroughly “from what is pure or correct” in the conduct of Government, Webster’s Third New International Dictionary 512 (1966) (defining “corruption”), that it amounts to a “subversion ... of the . . . electoral process,” Automobile Workers, 352 U. S., at 575. At stake in the legislative efforts to address this threat is therefore not only the legitimacy and quality of Government but also the public’s faith therein, not only “the capacity of this democracy to represent its constituents [but also] the confidence of its citizens in their capacity to govern themselves,” WRTL, 551 U. S., at 507 (Souter, J., dissenting). “Take away Congress’ authority to regulate the appearance of undue influence and ‘the cynical assumption that large donors call the tune could jeopardize the willingness of voters to take part in democratic governance.’ ” McConnell, 540 U. S., at 144 (quoting Shrink Missouri, 528 U. S., at 390).64
*451The cluster of interrelated interests threatened by such undue influence and its appearance has been well captured under the rubric of “democratic integrity.” WRTL, 551 U. S., at 522 (Souter, J., dissenting). This value has underlined a century of state and federal efforts to regulate the role of corporations in the electoral process.65
Unlike the majority’s myopic focus on quid pro quo scenarios and the free-floating “First Amendment principles” on which it rests so much weight, ante, at 319, 363, this broader understanding of corruption has deep roots in the Nation’s history. “During debates on the earliest [campaign finance] reform acts, the terms ‘corruption’ and ‘undue influence’ were used nearly interchangeably.” Pasquale, Reclaiming Egalitarianism in the Political Theory of Campaign Finance Reform, 2008 U. Ill. L. Rev. 599, 601. Long before Buckley, we appreciated that “[t]o say that Congress is without power to pass appropriate legislation to safeguard ... an election from the improper use of money to influence the result is to deny to the nation in a vital particular the power of self protection.” Burroughs v. United States, 290 U. S. 534, 545 (1934). And whereas we have no evidence to support the notion that the Framers would have wanted corporations to have the same rights as natural persons in the electoral context, we have ample evidence to suggest that they would *452have been appalled by the evidence of corruption that Congress unearthed in developing BCRA and that the Court today discounts to irrelevance. It is fair to say that “[t]he Framers were obsessed with corruption,” Teachout 348, which they understood to encompass the dependency of public officeholders on private interests, see id., at 373-374; see also Randall, 548 U. S., at 280 (Stevens, J., dissenting). They discussed corruption “more often in the Constitutional Convention than factions, violence, or instability.” Teachout 352. When they brought our constitutional order into being, the Framers had their minds trained on a threat to republican self-government that this Court has lost sight of.
Quid Pro Quo Corruption
There is no need to take my side in the debate over the scope of the anticorruption interest to see that the Court’s merits holding is wrong. Even under the majority’s “crabbed view of corruption,” McConnell, 540 U. S., at 152, the Government should not lose this case.
“The importance of the governmental interest in preventing [corruption through the creation of political debts] has never been doubted.” Bellotti, 435 U. S., at 788, n. 26. Even in the eases that have construed the anticorruption interest most narrowly, we have never suggested that such quid pro quo debts must take the form of outright vote buying or bribes, which have long been distinct crimes. Rather, they encompass the myriad ways in which outside parties may induce an officeholder to confer a legislative benefit in direct response to, or anticipation of, some outlay of money the parties have made or will make on behalf of the officeholder. See McConnell, 540 U. S., at 143 (“We have not limited [the anticorruption] interest to the elimination of cash-for-votes exchanges. In Buckley, we expressly rejected the argument that antibribery laws provided a less restrictive alternative to FECA’s contribution limits, noting that such laws ‘deal[t] with only the most blatant and specific attempts *453of those with money to influence governmental action’” (quoting 424 U. S., at 28; alteration in original)).. It has likewise never been doubted that “[o]f almost equal concern as the danger of actual quid pro quo arrangements is the impact of the appearance of corruption.” Id., at 27. Congress may “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 disastrous extent.” Ibid, (internal quotation marks omitted; alteration in original). A democracy cannot function effectively when its constituent members believe laws are being bought and sold.
In theory, our colleagues accept this much. As applied to BCRA §203, however, they conclude “[t]he anticorruption interest is not sufficient to displace the speech here in question.” Ante, at 357.
Although the Court suggests that Buckley compels its conclusion, ante, at 356-360, Buckley cannot sustain this reading. It is true that, in evaluating FECA's ceiling on independent expenditures by all persons, the Buckley Court found the governmental interest in preventing corruption “inadequate.” 424 U. S., at 45. But Buckley did not evaluate corporate expenditures specifically, nor did it rule out the possibility that a future Court might find otherwise. The opinion reasoned that an expenditure limitation covering only express advocacy (i e., magic words) would likely be ineffectual, ibid., a problem that Congress tackled in BCRA, and it concluded that “the independent advocacy restricted by [FECA § 608(e)(1)] does not presently appear to pose dangers of real or apparent corruption comparable to those identified with large campaign contributions,” id., at 46 (emphasis added). Buckley expressly contemplated that an anticorruption rationale might justify restrictions on independent expenditures at a later date, “because it may be that, in some circumstances, ‘large independent expenditures pose the same dangers of actual or apparent quid pro quo *454arrangements as do large contributions.’ ” WRTL, 551 U. S., at 478 (opinion of Roberts, C. J.) (quoting Buckley, 424 U. S., at 45). Certainly Buckley did not foreclose this possibility with respect to electioneering communications made with corporate general treasury funds, an issue the Court had no occasion to consider.
The Austin Court did not rest its holding on quid pro quo corruption, as it found the broader corruption implicated by the antidistortion and shareholder protection rationales a sufficient basis for Michigan’s restriction on corporate electioneering. 494 U. S., at 658-660. Concurring in that opinion, I took the position that “the danger of either the fact, or the appearance, of quid pro quo relationships [also] provides an adequate justification for state regulation” of these independent expenditures. Id., at 678. I did not see this position as inconsistent with Buckley’s analysis of individual expenditures. Corporations, as a class, tend to be more attuned to the complexities of the legislative process and more directly affected by tax and appropriations measures that receive little public scrutiny; they also have vastly more money with which to try to buy access and votes. See Supp. Brief for Appellee 17 (stating that the Fortune 100 companies earned revenues of $13.1 trillion during the last election cycle). Business corporations must engage the political process in instrumental terms if they are to maximize shareholder value. The unparalleled resources, professional lobbyists, and single-minded focus they bring to this effort, I believed, make quid pro quo corruption and its appearance inherently more likely when they (or their conduits or trade groups) spend unrestricted sums on elections.
It is with regret rather than satisfaction that I can now say that time has borne out my concerns. The legislative and judicial proceedings relating to BCRA generated a substantial body of evidence suggesting that, as corporations grew more and more adept at crafting “issue ads” to help *455or harm a particular candidate, these nominally independent expenditures began to corrupt the political process in a very direct sense. The sponsors of these ads were routinely granted special access after the campaign was over; “candidates and officials knew who their friends were,” McConnell, 540 U. S., at 129. Many corporate independent expenditures, it seemed, had become essentially interchangeable with direct contributions in their capacity to generate quid pro quo arrangements. In an age in which money and television ads are the coin of the campaign realm, it is hardly surprising that corporations deployed these ads to curry favor with, and to gain influence over, public officials.
The majority appears to think it decisive that the BCR A record does not contain “direct examples of votes being exchanged for... expenditures.” Ante, at 360 (internal quotation marks omitted). It would have been quite remarkable if Congress had created a record detailing such behavior by its own Members. Proving that a specific vote was exchanged for a specific expenditure has always been next to impossible: Elected officials have diverse motivations, and no one will acknowledge that he sold a vote. Yet, even if “[i]ngratiation and access . . . are not corruption” themselves, ibid., they are necessary prerequisites to it; they can create both the opportunity for, and the appearance of, quid pro quo arrangements. The influx of unlimited corporate money into the electoral realm also creates new opportunities for the mirror image of quid pro quo deals: threats, both explicit and implicit. Starting today, corporations with large war chests to deploy on electioneering may find democratically elected bodies becoming much more attuned to their interests. The majority both misreads the facts and draws the wrong conclusions when it suggests that the BCRA record provides “only scant evidence that independent expenditures . .. ingratiate,” and that, “in any event,” none of it matters. Ibid.
*456In her analysis of the record, Judge Kollar-Kotelly documented the pervasiveness of this ingratiation and explained its significance under the majority’s own touchstone for defining the scope of the antieorruption rationale, Buckley. See McConnell, 251 F. Supp. 2d, at 555-560, 622-625. Witnesses explained how political parties and candidates used corporate independent expenditures to circumvent FECA’s “hard-money” limitations. See, e. g., id., at 478-479. One former Senator candidly admitted to the District Court that “‘[candidates whose campaigns benefit from [phony “issue ads”] greatly appreciate the help of these groups. In fact, Members will also be favorably disposed to those who finance these groups when they later seek access to discuss pending legislation.’” Id., at 556 (quoting declaration of Sen. Dale Bumpers). One prominent lobbyist went so far as to state, in uncontroverted testimony, that “‘unregulated expenditures — whether soft money donations to the parties or issue ad campaigns — can sometimes generate far more influence than direct campaign contributions.’ ” Ibid, (quoting declaration of Wright Andrews; emphasis added). In sum, Judge Kollar-Kotelly found, “[t]he record powerfully demonstrates that electioneering communications paid for with the general treasury funds of labor unions and corporations endears those entities to elected officials in a way that could be perceived by the public as corrupting.” Id., at 622-623. She concluded that the Government’s interest in preventing the appearance of corruption, as that concept was defined in Buckley, was itself sufficient to uphold BCRA §203. 251 F. Supp. 2d, at 622-625. Judge Leon agreed. See id., at 804-805 (dissenting only with respect to the Wellstone Amendment’s coverage of MCFL corporations).
When the McConnell Court affirmed the judgment of the District Court regarding § 203, we did not rest our holding on a narrow notion of quid pro quo corruption. Instead we relied on the governmental interest in combating the unique forms of corruption threatened by corporations, as recog*457nized in Austin’s antidistortion and shareholder protection rationales, 540 U. S., at 205 (citing Austin, 494 U. S., at 660), as well as the interest in preventing circumvention of contribution limits, 540 U. S., at 128-129, 205, 206, n. 88. Had we felt constrained by the view of today’s Court that quid pro quo corruption and its appearance are the only interests that count in this field, ante, at 348-362, we of course would have looked closely at that issue. And as the analysis by Judge Kollar-Kotelly reflects, it is a very real possibility that we would have found one or both of those interests satisfied and §203 appropriately tailored to them.
The majority’s rejection of the Buckley anticorruption rationale on the ground that independent corporate expenditures “do not give rise to [quid pro quo] corruption or the appearance of corruption,” ante, at 357, is thus unfair as well as unreasonable. Congress and outside experts have generated significant evidence corroborating this rationale, and the only reason we do not have any of the relevant materials before us is that the Government had no reason to develop a record at trial for a facial challenge the plaintiff had abandoned. The Court cannot both sua sponte choose to relitigate McConnell on appeal and then complain that the Government has failed to substantiate its case. If our colleagues were really serious about the interest in preventing quid pro quo corruption, they would remand to the District Court with instructions to commence evidentiary proceedings.66
*458The insight that even technically independent expenditures can be corrupting in much the same way as direct contributions is bolstered by our decision last year in Caperton v. A. T. Massey Coal Co., 556 U. S. 868 (2009). In that case, Don Blankenship, the chief executive officer of a corporation with a lawsuit pending before the West Virginia high court, spent large sums on behalf of a particular candidate, Brent Benjamin, running for a seat on that court. “In addition to contributing the $1,000 statutory maximum to Benjamin’s campaign committee, Blankenship donated almost $2.5 million to ‘And For The Sake Of The Kids,’ ” a § 527 corporation that ran ads targeting Benjamin’s opponent. Id., at 873. “This was not all. Blankenship spent, in addition, just over $500,000 on independent expenditures . . . ‘ “to support. . . Brent Benjamin.””’ Ibid, (second alteration in original). Applying its common sense, this Court accepted petitioners’ argument that Blankenship’s “pivotal role in getting Justice Benjamin elected created a constitutionally intolerable probability of actual bias” when Benjamin later declined to recuse himself from the appeal by Blankenship’s corporation. Id., at 882. “Though n[o] . . . bribe or criminal influence” was involved, we recognized that “Justice Benjamin would nevertheless feel a debt of gratitude to Blankenship for his extraordinary efforts to get him elected.” Ibid. “The difficulties of inquiring into actual bias,” we further noted, “simply underscore the need for objective rules,” id., at 883 — rules which will perforce turn on the appearance of bias rather than its actual existence.
In Caperton, then, we accepted the premise that, at least in some circumstances, independent expenditures on candidate elections will raise an intolerable specter of quid pro quo corruption. Indeed, this premise struck the Court as so intuitive that it repeatedly referred to Blankenship’s spending on behalf of Benjamin — spending that consisted of *45999.97% independent expenditures ($3 million) and 0.03% direct contributions ($1,000) — as a “contribution.” See, e.g., id., at 872 (“The basis for the [recusal] motion was that the justice had received campaign contributions in an extraordinary amount from” Blankenship); id., at 873 (referencing “Blankenship’s $3 million in contributions”); id., at 884 (“Blankenship contributed some $3 million to unseat the incumbent and replace him with Benjamin”); id., at 885 (“Blankenship’s campaign contributions . . . had a significant and disproportionate influence on the electoral outcome”). The reason the Court so thoroughly conflated expenditures and contributions, one assumes, is that it realized that some expenditures may be functionally equivalent to contributions in the way they influence the outcome of a race, the way they are interpreted by the candidates and the public, and the way they taint the decisions that the officeholder thereafter takes.
Caperton is illuminating in several additional respects. It underscores the old insight that, on account of the extreme difficulty of proving corruption, “prophylactic measures, reaching some [campaign spending] not corrupt in purpose or effect, [may be] nonetheless required to guard against corruption.” Buckley, 424 U. S., at 30; see also Shrink Missouri, 528 U. S., at 392, n. 5. It underscores that “certain restrictions on corporate electoral involvement” may likewise be needed to “hedge against circumvention of valid contribution limits.” McConnell, 540 U. S., at 205 (internal quotation marks and brackets omitted); see also Colorado II, 533 U. S., at 456 (“[A]ll Members of the Court agree that circumvention is a valid theory of corruption”). It -underscores that for-profit corporations associated with electioneering communications will often prefer to use nonprofit conduits with “misleading names,” such as And For The Sake Of The Kids, “to conceal their identity” as the sponsor of those communications, thereby frustrating the utility of dis*460closure laws. McConnell, 540 U. S., at 128; see also id., at 196-197.
And it underscores that the consequences of today’s holding will not be limited to the legislative or executive context. The majority of the States select their judges through popular elections. At a time when concerns about the conduct of judicial elections have reached a fever pitch, see, e. g., O’Connor, Justice for Sale, Wall St. Journal, Nov. 15, 2007, p. A25; Brief for Justice at Stake et al. as Amici Curiae 2, the Court today unleashes the floodgates of corporate and union general treasury spending in these races. Perhaps “Caperton motions” will catch some of the worst abuses. This will be small comfort to those States that, after today, may no longer have the ability to place modest limits on corporate electioneering even if they believe such limits to be critical to maintaining the integrity of their judicial systems.
Deference and Incumbent Self-Protection
Rather than show any deference to a coordinate branch of Government, the majority thus rejects the anticorruption rationale without serious analysis.67 Today’s opinion provides no clear rationale for’being so dismissive of Congress, but the prior individual opinions on which it relies have offered one: the incentives of the legislators who passed BCRA. Section 203, our colleagues have suggested, may be little more than “an incumbency protection plan,” McConnell, 540 U. S., at 306 (Kennedy, J., concurring in judgment in part and dissenting in part); see also id., at 249-250, 260-263 (Scalia, J., concurring in part, concurring in judgment in part, and dissenting in part), a disreputable attempt at legislative self-dealing rather than an earnest effort to facilitate First Amendment values and safeguard the legitimacy *461of our political system. This possibility, the Court apparently believes, licenses it to run roughshod over Congress’ handiwork.
In my view, we should instead start by acknowledging that “Congress surely has both wisdom and experience in these matters that is far superior to ours.” Colorado Republican Federal Campaign Comm. v. FEC, 518 U. S. 604, 650 (1996) (Stevens, J., dissenting). Many of our campaign finance precedents explicitly and forcefully affirm the propriety of such presumptive deference. See, e. g., McConnell, 540 U. S., at 158; Beaumont, 539 U. S., at 155-156; NRWC, 459 U. S., at 209-210. Moreover, “[j]udicial deference is particularly warranted where, as here, we deal with a congressional judgment that has remained essentially unchanged throughout a century of careful legislative adjustment.” Beaumont, 539 U. S., at 162, n. 9 (internal quotation marks omitted); cf. Shrink Missouri, 528 U. S., at 391 (“The quantum of empirical evidence needed to satisfy heightened judicial scrutiny of legislative judgments will vary up or down with the novelty and plausibility of the justification raised”). In America, incumbent legislators pass the laws that govern campaign finance, just like all other laws. To apply a level of scrutiny that effectively bars them from regulating electioneering whenever there is the faintest whiff of self-interest, is to deprive them of the ability to regulate electioneering.
This is not to say that deference would be appropriate if there were a solid basis for believing that a legislative action was motivated by the desire to protect incumbents or that it will degrade the competitiveness of the electoral process.68 *462See League of United Latin American Citizens v. Perry, 548 U. S. 399, 447 (2006) (Stevens, J., concurring in part and dissenting in part); Vieth v. Jubelirer, 541 U. S. 267, 317 (2004) (Stevens, J., dissenting). Along with our duty to balance competing constitutional concerns, we have a vital role to play in ensuring that elections remain at least minimally open, fair, and competitive. But it is the height of recklessness to dismiss Congress’ years of bipartisan deliberation and its reasoned judgment on this basis, without first confirming that the statute in question was intended to be, or will function as, a restraint on electoral competition. “Absent record evidence of invidious discrimination against challengers as a class, a court should generally be hesitant to invalidate legislation which on its face imposes evenhanded restrictions.” Buckley, 424 U. S., at 31.
We have no record evidence from which to conclude that BCRA § 203, or any of the dozens of state laws that the Court today calls into question, reflects or fosters such invidious discrimination. Our colleagues have opined that “ ‘any restriction upon a type of campaign speech that is equally available to challengers and incumbents tends to favor incumbents.’” McConnell, 540 U. S., at 249 (opinion of SCALIA, J.). This kind of airy speculation could easily be turned on its head. The electioneering prohibited by § 203 might well tend to favor incumbents, because incumbents have pre-existing relationships with corporations and unions, and groups that wish to procure legislative benefits may tend to support the candidate who, as a sitting officeholder, is already in a position to dispense benefits and is statistically likely to retain office. If a corporation’s goal is to induce officeholders to do its bidding, the corporation would do well to cultivate stable, long-term relationships of dependency.
So we do not have a solid theoretical basis for condemning §203 as a front for incumbent self-protection, and it seems equally if not more plausible that restrictions on corporate electioneering will be self-denying. Nor do we have a good *463empirical case for skepticism, as the Court’s failure to cite any empirical research attests. Nor does the legislative history give reason for concern. Congress devoted years of careful study to the issues underlying BCRA; “[f]ew legislative proposals in recent years have received as much sustained public commentary or news coverage”; “[political scientists and academic experts . . . with no self-interest in incumbent protectio[n] were central figures in pressing the case for BCRA”; and the legislation commanded bipartisan support from the outset. Pildes, The Supreme Court 2003 Term Foreword: The Constitutionalization of Democratic Politics, 118 Harv. L. Rev. 28, 137 (2004). Finally, it is important to remember just how incumbent-friendly congressional races were prior to BCRA’s passage. As the Solicitor General aptly remarked at the time, “the evidence supports overwhelmingly that incumbents were able to get re-elected under the old system just fine.” Tr. of Oral Arg. in McConnell v. FEC, O. T. 2003, No. 02-1674, p. 61. “It would be hard to develop a scheme that could be better for incumbents.” Id., at 63.
In this case, then, “there is no convincing evidence that th[e] important interests favoring expenditure limits are fronts for incumbency protection.” Randall, 548 U. S., at 279 (Stevens, J., dissenting). “In the meantime, a legislative judgment that ‘enough is enough’ should command the greatest possible deference from judges interpreting a constitutional provision that, at best, has an indirect relationship to activity that affects the quantity ... of repetitive speech in the marketplace of ideas.” Id., at 279-280. The majority cavalierly ignores Congress’ factual findings and its constitutional judgment: It acknowledges the validity of the interest in preventing corruption, but it effectively discounts the value of that interest to zero. This is quite different from conscientious policing for impermissibly anticompetitive motive or effect in a sensitive First Amendment context. *464It is the denial of Congress’ authority to regulate corporate spending on elections.
Austin and Corporate Expenditures
Just as the majority gives short shrift to the general societal interests at stake in campaign finance regulation, it also overlooks the distinctive considerations raised by the regulation of corporate expenditures. The majority fails to appreciate that Austin’s antidistortion rationale is itself an anticorruption rationale, see 494 U. S., at 660 (describing “a different type of corruption”), tied to the special concerns raised by corporations. Understood properly, “antidistortion” is simply a variant on the classic governmental interest in protecting against improper influences on officeholders that debilitate the democratic process. It is manifestly not just an “ ‘equalizing’ ” ideal in disguise. Ante, at 350 (quoting Buckley, 424 U. S., at 48).69
*4651. Antidistortion
The fact that corporations are different from human beings might seem to need no elaboration, except that the majority opinion almost completely elides it. Austin set forth some of the basic differences. Unlike natural persons, corporations have “limited liability” for their owners and managers, “perpetual life,” separation of ownership and control, “and favorable treatment of the accumulation and distribution of assets ... that enhance their ability to attract capital and to deploy their resources in ways that maximize the return on their shareholders’ investments.” 494 U. S., at 658-659. Unlike voters in U. S. elections, corporations may be foreign controlled.70 Unlike other interest groups, business corporations have been “effectively delegated responsibility for ensuring society’s economic welfare”;71 they inescapably structure the life of every citizen. “ ‘[T]he resources in the treasury of a business corporation,’ ” furthermore, “ ‘are not an indication of popular support for the corporation’s political ideas.’” Id., at 659 (quoting MCFL, 479 U. S., at 258). “ ‘They reflect instead the economically motivated decisions of investors and customers. The availability of these resources may make a corporation a formidable political presence, even though the power of the corporation may be no reflection of the power of its ideas.’ ” 494 U. S., at 659 (quoting MCFL, 479 U. S., at 258).72
*466It might also be added that corporations have no consciences, no beliefs, no feelings, no thoughts, no desires. Corporations help structure and facilitate the activities of human beings, to be sure, and their “personhood” often serves as a useful legal fiction. But they are not themselves members of “We the People” by whom and for whom our Constitution was established.
These basic points help explain why corporate electioneering is not only more likely to impair compelling governmental interests, but also why restrictions on that electioneering are less likely to encroach upon First Amendment freedoms. One fundamental concern of the First Amendment is to “protec[t] the individual’s interest in self-expression.” Consolidated Edison Co. of N. Y. v. Public Serv. Comm’n of N. Y., 447 U. S. 530, 534, n. 2 (1980); see also Bellotti, 435 U. S., at 777, n. 12. Freedom of speech helps “make men free to develop their faculties,” Whitney v. California, 274 U. S. 357, 375 (1927) (Brandéis, J., concurring), it respects their “dignity and choice,” Cohen v. California, 403 U. S. 15, 24 (1971), and it facilitates the value of “individual self-realization,” Redish, The Value of Free Speech, 130 U. Pa. L. Rev. 591, 594 (1982). Corporate speech, however, is derivative speech, speech by proxy. A regulation such as BCRA § 203 may affect the way in which individuals disseminate certain messages through the corporate form, but it does not prevent anyone from speaking in his or her own voice. “Within the realm of [campaign spending] generally,” corporate *467spending is “furthest from the core of political expression.” Beaumont, 539 U. S., at 161, n. 8.
It is an interesting question “who” is even speaking when a business corporation places an advertisement that endorses or attacks a particular candidate. Presumably it is not the customers or employees, who typically have no say in such matters. It cannot realistically be said to be the shareholders, who tend to be far removed from the day-to-day decisions of the firm and whose political preferences may be opaque to management. Perhaps the officers or directors of the corporation have the best claim to be the ones speaking, except their fiduciary duties generally prohibit them from using corporate funds for personal ends. Some individuals associated with the corporation must make the decision to place the ad, but the idea that these individuals are thereby fostering their self-expression or cultivating their critical faculties is fanciful. It is entirely possible that the corporation’s electoral message will conflict with their personal convictions. Take away the ability to use general treasury funds for some of those ads, and no one’s autonomy, dignity, or political equality has been impinged upon in the least.
Corporate expenditures are distinguishable from individual expenditures in this respect. I have taken the view that a legislature may place reasonable restrictions on individuals’ electioneering expenditures in the service of the governmental interests explained above, and in recognition of the fact that such restrictions are not direct restraints on speech but rather on its financing. See, e. g., Randall, 548 U. S., at 273 (dissenting opinion). But those restrictions concededly present a tougher case, because the primary conduct of actual, flesh-and-blood persons is involved. Some of those individuals might feel that they need to spend large sums of money on behalf of a particular candidate to vindicate the intensity of their electoral preferences. This is obviously not the situation with business corporations, as their routine practice of giving “substantial sums to both major national *468parties” makes pellucidly clear. McConnell, 540 U. S., at 148. “[Co]rporate participation” in elections, any business executive will tell you, “is more transactional than ideological.” Supp. Brief for Committee for Economic Development as Amicus Curiae 10.
In this transactional spirit, some corporations have affirmatively urged Congress to place limits on their electioneering communications. These corporations fear that officeholders will shake them down for supportive ads, that they will have to spend increasing sums on elections in an ever-escalating arms race with their competitors, and that public trust in business will be eroded. See id., at 10-19. A system that effectively forces corporations to use their shareholders’ money both to maintain access to, and to avoid retribution from, elected officials may ultimately prove more harmful than beneficial to many corporations. It can impose a kind of implicit tax.73
In short, regulations such as § 203 and the statute upheld in Austin impose only a limited burden on First Amendment freedoms not only because they target a narrow subset of expenditures and leave untouched the broader “public dialogue,” ante, at 341, but also because they leave untouched *469the speech of natural persons. Recognizing the weakness of a speaker-based critique of Austin, the Court places primary emphasis not on the corporation’s right to electioneer, but rather on the listener’s interest in hearing what every possible speaker may have to say. The Court’s central argument is that laws such as § 203 have “ ‘deprived [the electorate] of information, knowledge and opinion vital to its function,’” ante, at 354 (quoting CIO, 335 U. S., at 144 (Rutledge, J., concurring in result)), and this, in turn, “interferes with the ‘open marketplace’ of ideas protected by the First Amendment,” ante, at 354 (quoting New York State Bd. of Elections v. Lopez Torres, 552 U. S. 196, 208 (2008)).
There are many flaws in this argument. If the overriding concern depends on the interests of the audience, surely the public’s perception of the value of corporate speech should be given important weight. That perception today is the same as it was a century ago when Theodore Roosevelt delivered the speeches to Congress that, in time, led to the limited prohibition on corporate campaign expenditures that is overruled today. See WRTL, 551 U. S., at 509-510 (Souter, J., dissenting) (summarizing President Roosevelt’s remarks). The distinctive threat to democratic integrity posed by corporate domination of politics was recognized at “the inception of the republic” and “has been a persistent theme in American political life” ever since. Regan 302. It is only certain Members of this Court, not the listeners themselves, who have agitated for more corporate electioneering.
Austin recognized that there are substantial reasons why a legislature might conclude that unregulated general treasury expenditures will give corporations “unfai[r] influence” in the electoral process, 494 U. S., at 660, and distort public debate in ways that undermine rather than advance the interests of listeners. The legal structure of corporations allows them to amass and deploy financial resources on a scale few natural persons can match. The structure of a business corporation, furthermore, draws a line between the *470corporation’s economic interests and the political preferences of the individuals associated with the corporation; the corporation must engage the electoral process with the aim “to enhance the profitability of the company, no matter how persuasive the arguments for a broader or conflicting set of priorities,” Brief for American Independent Business Alliance as Amicus Curiae 11; see also ALI, Principles of Corporate Governance: Analysis and Recommendations § 2.01(a), p. 55 (1992) (“[A] corporation . . . should have as its objective the conduct of business activities with a view to enhancing corporate profit and shareholder gain”). In a state election such as the one at issue in Austin, the interests of nonresident corporations may be fundamentally adverse to the interests of local voters. Consequently, when corporations grab up the prime broadcasting slots on the eve of an election, they can flood the market with advocacy that bears “little or no correlation” to the ideas of natural persons or to any broader notion of the public good, 494 U. S., at 660. The opinions of real people may be marginalized. “The expenditure restrictions of [2 U. S. C.] § 441b are thus meant to ensure that competition among actors in the political arena is truly competition among ideas.” MCFL, 479 U. S., at 259.
In addition to this immediate drowning out of noncorporate voices, there may be deleterious effects that follow soon thereafter. Corporate “domination” of electioneering, Austin, 494 U. S., at 659, can generate the impression that corporations dominate our democracy. When citizens turn on their televisions and radios before an election and hear only corporate electioneering, they may lose faith in their capacity, as citizens, to influence public policy. A Government captured by corporate interests, they may come to believe, will be neither responsive to their needs nor willing to give their views a fair hearing. The predictable result is cynicism and disenchantment: an increased perception that large spenders “ ‘call the tune’ ” and a reduced “ ‘willingness of voters to take part in democratic governance.’” McConnell, *471540 U. S., at 144 (quoting Shrink Missouri, 528 U. S., at 390). To the extent that corporations are allowed to exert undue influence in electoral races, the speech of the eventual winners of those races may also be chilled. Politicians who fear that a certain corporation can make or break their reelection chances may be cowed into silence about that corporation. On a variety of levels, unregulated corporate electioneering might diminish the ability of citizens to “hold officials accountable to the people,” ante, at 339, and disserve the goal of a public debate that is “uninhibited, robust, and wide-open,” New York Times Co. v. Sullivan, 376 U. S. 254, 270 (1964). At the least, I stress again, a legislature is entitled to credit these concerns and to take tailored measures in response.
The majority’s unwillingness to distinguish between corporations and humans similarly blinds it to the possibility that corporations’ “war chests” and their special “advantages” in the legal realm, Austin, 494 U. S., at 659 (internal quotation marks omitted), may translate into special advantages in the market for legislation. When large numbers of citizens have a common stake in a measure that is under consideration, it may be very difficult for them to coordinate resources on behalf of their position. The corporate form, by contrast, “provides a simple way to channel rents to only those who have paid their dues, as it were. If you do not own stock, you do not benefit from the larger dividends or appreciation in the stock price caused by the passage of private interest legislation.” Sitkoff, Corporate Political Speech, Political Extortion, and the Competition for Corporate Charters, 69 U. Chi. L. Rev. 1103, 1113 (2002). Corporations, that is, are uniquely equipped to seek laws that favor their owners, not simply because they have a lot of money but because of their legal and organizational structure. Remove all restrictions on their electioneering, and the door may be opened to a type of rent seeking that is “far more destructive” than what noncorporations are capable of. *472Ibid. It is for reasons such as these that our campaign finance jurisprudence has long appreciated that “the ‘differing structures and purposes’ of different entities ‘may require different forms of regulation in order to protect the integrity of the electoral process.’ ” NRWC, 459 U. S., at 210 (quoting California Medical Assn., 453 U. S., at 201).
The Court’s facile depiction of corporate electioneering assumes away all of these complexities. Our colleagues ridicule the idea of regulating expenditures based on “nothing more” than a fear that corporations have a special “ability to persuade,” ante, at 382 (opinion of Roberts, C. J.), as if corporations were our society’s ablest debaters and viewpoint-neutral laws such as §203 were created to suppress their best arguments. In their haste to knock down yet another straw man, our colleagues simply ignore the fundamental concerns of the Austin Court and the legislatures that have passed laws like § 203: to safeguard the integrity, competitiveness, and democratic responsiveness of the electoral process. All of the majority’s theoretical arguments turn on a proposition with undeniable surface appeal but little grounding in evidence or experience, “that there is no such thing as too much speech,” Austin, 494 U. S., at 695 (Scalia, J., dissenting).74 If individuals in our society had infinite free time to listen to and contemplate every last bit of speech uttered by anyone, anywhere; and if broadcast advertisements had no special ability to influence elections apart from the merits of their arguments (to the extent they make any); and if legislators always operated with nothing less than perfect virtue; then I suppose the majority’s premise would be sound. In the real world, we have seen, corporate domination of the airwaves prior to an election may decrease the average listener’s exposure to relevant viewpoints, and it may diminish citizens’ willingness and capacity to participate in the democratic process.
*473None of this is to suggest that corporations can or should be denied an opportunity to participate in election campaigns or in any other public forum (much less that a work of art such as Mr. Smith Goes to Washington may be banned), or to deny that some corporate speech may contribute significantly to public debate. What it shows, however, is that Austin’s “concern about corporate domination of the political process,” id., at 659, reflects more than a concern to protect governmental interests outside of the First Amendment. It also reflects a concern to facilitate First Amendment values by preserving some breathing room around the electoral “marketplace” of ideas, ante, at 335, 350, 354, 367, 369, the marketplace in which the actual people of this Nation determine how they will govern themselves. The majority seems oblivious to the simple truth that laws such as §203 do not merely pit the anticorruption interest against the First Amendment, but also pit competing First Amendment values against each other. There are, to be sure, serious concerns with any effort to balance the First Amendment rights of speakers against the First Amendment rights of listeners. But when the speakers in question are not real people and when the appeal to “First Amendment principles” depends almost entirely on the listeners’ perspective, ante, at 319, 363, it becomes necessary to consider how listeners will actually be affected.
In critiquing Austin’s antidistortion rationale and campaign finance regulation more generally, our colleagues place tremendous weight on the example of media corporations. See ante, at 351-354, 361-362; ante, at 372-373, 382 (opinion of Roberts, C. J.); ante, at 390 (opinion of Scalia, J.). Yet it is not at all clear that Austin would permit § 203 to be applied to them. The press plays a unique role not only in the text, history, and structure of the First Amendment but also in facilitating public discourse; as the Austin Court explained, “media corporations differ significantly from other corporations in that their resources are devoted to the collec*474tion of information and its dissemination to the public,” 494 U. S., at 667. Our colleagues have raised some interesting and difficult questions about Congress’ authority to regulate electioneering by the press, and about how to define what constitutes the press. But that is not the case before us. Section 203 does not apply to media corporations, and even if it did, Citizens United is not a media corporation. There would be absolutely no reason to consider the issue of media corporations if the majority did not, first, transform Citizens United's as-applied challenge into a facial challenge and, second, invent the theory that legislatures must eschew all “identity”-based distinctions and treat a local nonprofit news outlet exactly the same as General Motors.75 This calls to mind George Berkeley’s description of philosophers: “[W]e have first raised a dust, and then complain we cannot see.” Principles of Human Knowledge/Three Dialogues 38, ¶ 3 (R. Woolhouse ed. 1988).
It would be perfectly understandable if our colleagues feared that a campaign finance regulation such as § 203 may be counterproductive or self-interested, and therefore attended carefully to the choices the Legislature has made. But the majority does not bother to consider such practical matters, or even to consult a record; it simply stipulates that “enlightened self-government” can arise only in the absence of regulation. Ante, at 339. In light of the distinctive features of corporations identified in Austin, there is no valid basis for this assumption. The marketplace of ideas is not actually a place where items — or laws — are meant to be bought and sold, and when we move from the realm of eco*475nomics to the realm of corporate electioneering, there may be no “reason to think the market ordering is intrinsically good at all,” Strauss 1386.
The Court’s blinkered and aphoristic approach to the First Amendment may well promote corporate power at the cost of the individual and collective self-expression the Amendment was meant to serve. It will undoubtedly cripple the ability of ordinary citizens, Congress, and the States to adopt even limited measures to protect against corporate domination of the electoral process. Americans may be forgiven if they do not feel the Court has advanced the cause of self-government today.
2. Shareholder Protection
There is yet another way in which laws such as § 203 can serve First Amendment values. Interwoven with Austin’s concern to protect the integrity of the electoral process is a concern to protect the rights of shareholders from a kind of coerced speech: electioneering expenditures that do not “reflee[t] [their] support.” 494 U. S., at 660-661. When corporations use general treasury funds to praise or attack a particular candidate for office, it is the shareholders, as the residual claimants, who are effectively footing the bill. Those shareholders who disagree with the corporation’s electoral message may find their financial investments being used to undermine their political convictions.
The PAC mechanism, by contrast, helps ensure that those who pay for an electioneering communication actually support its content and that managers do not use general treasuries to advance personal agendas. Ibid. It “ ‘allows corporate political participation without the temptation to use corporate funds for political influence, quite possibly at odds with the sentiments of some shareholders or members.’” McConnell, 540 U. S., at 204 (quoting Beaumont, 539 U. S., at 163). A rule that privileges the use of PACs thus does more than facilitate the political speech of like-minded share*476holders; it also curbs the rent seeking behavior of executives and respects the views of dissenters. Austin’s acceptance of restrictions on general treasury spending “simply allows people who have invested in the business corporation for purely economic reasons” — the vast majority of investors, one assumes — “to avoid being taken advantage of, without sacrificing their economic objectives.” Winkler, Beyond Bellotti, 32 Loyola (LA) L. Rev. 133, 201 (1998).
The concern to protect dissenting shareholders and union members has a long history in campaign finance reform. It provided a central motivation for the Tillman Act in 1907 and subsequent legislation, see Pipefitters v. United States, 407 U. S. 385, 414-415 (1972); Winkler, 92 Geo. L. J., at 887-900, and it has been endorsed in a long line of our cases, see, e. g., McConnell, 540 U. S., at 204-205; Beaumont, 539 U. S., at 152-154; MCFL, 479 U. S., at 258; NRWC, 459 U. S., at 207-208; Pipefitters, 407 U. S., at 414-416; see also n. 60, supra. Indeed, we have unanimously recognized the governmental interest in “proteet[ing] the individuals who have paid money into a corporation or union for purposes other than the support of candidates from having that money used to support political candidates to whom they may be opposed.” NRWC, 459 U. S., at 207-208.
The Court dismisses this interest on the ground that abuses of shareholder money can be corrected “through the procedures of corporate democracy,” ante, at 362 (internal quotation marks omitted), and, it seems, through Internet-based disclosures, ante, at 370-371.76 I fail to understand *477how this addresses the concerns of dissenting union members, who will also be affected by today’s ruling, and I fail to understand why the Court is so confident in these mechanisms. By “corporate democracy,” presumably the Court means the rights of shareholders to vote and to bring derivative suits for breach of fiduciary duty. In practice, however, many corporate lawyers will tell you that “these rights are so limited as to be almost nonexistent,” given the internal authority wielded by boards and managers and the expansive protections afforded by the business judgment rule. Blair & Stout 320; see also id., at 298-315; Winkler, 32 Loyola (LA) L. Rev., at 165-166, 199-200. Modern technology may help make it easier to track corporate activity, including electoral advocacy, but it is utopian to believe that it solves the problem. Most American households that own stock do so through intermediaries such as mutual funds and pension plans, see Evans, A Requiem for the Retail Investor? 95 Va. L. Rev. 1105 (2009), which makes it more difficult both to monitor and to alter particular holdings. Studies show that a majority of individual investors make no trades at all during a given year. Id., at 1117. Moreover, if the corporation in question operates a PAC, an investor who sees the company’s ads may not know whether they are being funded through the PAC or through the general treasury.
If and when shareholders learn that a corporation has been spending general treasury money on objectionable electioneering, they can divest. Even assuming that they reliably learn as much, however, this solution is only partial. The injury to the shareholders’ expressive rights has already occurred; they might have preferred to keep that corporation’s stock in their portfolio for any number of economic reasons; and they may incur a capital gains tax or other penalty from selling their shares, changing their pension plan, or the like. The shareholder protection rationale has been criticized as underinclusive, in that corporations also spend money on lobbying and charitable contributions in ways that any particu*478lar shareholder might disapprove. But those expenditures do not implicate the selection of public officials, an area in which “the interests of unwilling... corporate shareholders [in not being] forced to subsidize that speech” “are at their zenith.” Austin, 494 U. S., at 677 (Brennan, J., concurring). And in any event, the question is whether shareholder protection provides a basis for regulating expenditures in the weeks before an election, not whether additional types of corporate communications might similarly be conditioned on voluntariness.
Recognizing the limits of the shareholder- protection rationale, the Austin Court did not hold it out as an adequate and independent ground for sustaining the statute in question. Rather, the Court applied it to reinforce the anti-distortion rationale, in two main ways. First, the problem of dissenting shareholders shows that even if electioneering expenditures can advance the political views of some members of a corporation, they will often compromise the views of others. See, e. g., id., at 663 (discussing risk that corporation’s “members may be ... reluctant to withdraw as members even if they disagree with [its] political expression”). Second, it provides an additional reason, beyond the distinctive legal attributes of the corporate form, for doubting that these “expenditures reflect actual public support for the political ideas espoused,” id., at 660. The shareholder protection rationale, in other words, bolsters the conclusion that restrictions on corporate electioneering can serve both speakers’ and listeners’ interests, as well as the anticorruption interest. And it supplies yet another reason why corporate expenditures merit less protection than individual expenditures.
V
Today’s decision is backwards in many senses. It elevates the majority’s agenda over the litigants’ submissions, facial attacks over as-applied claims, broad constitutional theories *479over narrow statutory grounds, individual dissenting opinions over precedential holdings, assertion over tradition, absolutism over empiricism, rhetoric over reality. Our colleagues have arrived at the conclusion that Austin must be overruled and that §203 is facially unconstitutional only after mischaracterizing both the reach and rationale of those authorities, and after bypassing or ignoring rules of judicial restraint used to cabin the Court’s lawmaking power. Their conclusion that the societal interest in avoiding corruption and the appearance of corruption does not provide an adequate justification for regulating corporate expenditures on candidate elections relies on an incorrect description of that interest, along with a failure to acknowledge the relevance of established facts and the considered judgments of state and federal legislatures over many decades.
In a democratic society, the longstanding consensus on the need to limit corporate campaign spending should outweigh the wooden application of judge-made rules. The majority’s rejection of this principle “elevate[s] corporations to a level of deference which has not been seen at least since the days when substantive due process was regularly used to invalidate regulatory legislation thought to unfairly impinge upon established economic interests.” Bellotti, 435 U. S., at 817, n. 13 (White, J., dissenting). At bottom, the Court’s opinion is thus a rejection of the common sense of the American people, who have recognized a need to prevent corporations from undermining self-government since the founding, and who have fought against the distinctive corrupting potential of corporate electioneering since the days of Theodore Roosevelt. It is a strange time to repudiate that common sense. While American democracy is imperfect, few outside the majority of this Court would have thought its flaws included a dearth of corporate money in politics.
I would affirm the judgment of the District Court.
Specifically, Part I, infra, at 396-408, addresses the procedural history of the case and the narrower grounds of decision the majority has bypassed. Part II, infra, at 408-414, addresses stare decisis. Part III, infra, at 414-446, addresses the Court’s assumptions that BCRA “bans” corporate speech, that identity-based distinctions may not be drawn in the political realm, and that Austin and McConnell were outliers in our First Amendment tradition. Part IV, infra, at 447-478, addresses the Court’s treatment of the anticorruption, antidistortion, and shareholder protection rationales for regulating corporate electioneering.
See Yee v. Escondido, 503 U. S. 519, 535 (1992) (“[U]nder this Court’s Rule 14.1(a), only the questions set forth in the petition, or fairly included therein, will be considered by the Court” (internal quotation marks and alteration omitted)); Wood v. Allen, ante, at 304 (“[T]he fact that petitioner discussed [an] issue in the text of his petition for certiorari does not bring it before us. Rule 14.1(a) requires that a subsidiary question be fairly included in the question presented for our review” (internal quotation marks and brackets omitted)); Cooper Industries, Inc. v. Aviall Services, Inc., 543 U. S. 157, 168-169 (2004) (“We ordinarily do not decide in the first instance issues not decided below” (internal quotation marks omitted)).
The majority states that, in denying Citizens United’s motion for a preliminary injunction, the District Court “addressed” the facial validity of BCRA § 203. Ante, at 330. That is true, in the narrow sense that the court observed the issue was foreclosed by McConnell v. FEC, 540 U. S. 93 (2003). See 530 F. Supp. 2d 274, 278 (DC 2008) (per curiam). Yet as explained above, Citizens United subsequently dismissed its facial challenge, so that by the time the District Court granted the Federal Election Commission’s (FEC) motion for summary judgment, App. 261a-262a, any question about statutory validity had dropped out of the case. That latter ruling by the District Court was the “final decision” from which Citizens United appealed to this Court under BCRA § 403(a)(3). As regards the lower court decision that has come before us, the claim that § 203 is facially unconstitutional was neither pressed nor passed upon in any form.
Shortly before Citizens United mooted the issue by abandoning its facial challenge, the Government advised the District Court that it “require[d] time to develop a factual record regarding [the] facial challenge.” 1:07-cv-2240-RCL-RWR, Docket Entry No. 47, p. 4 (Mar. 26, 2008). By reinstating a claim that Citizens United abandoned, the Court gives it a perverse litigating advantage over its adversary, which was deprived of the opportunity to gather and present information necessary to its rebuttal.
In fact, we do not even have a good evidentiary record of how §208 has been affecting Citizens United, which never submitted to the District Court the details of Hillary’s funding or its own finances. We likewise have no evidence of how §203 and comparable state laws were expected to affect corporations and unions in the future.
It is true, as the majority points out, that the McConnell Court evaluated the facial validity of §203 in light of an extensive record. See ante, at 331-332. But that record is not before us in this case. And in any event, the majority’s argument for striking down § 203 depends on its contention that the statute has proved too “chilling” in practice — and in particular on the contention that the controlling opinion in WRTL, 551 U. S. 449 (2007), failed to bring sufficient clarity and “breathing space” to this area of law. See ante, at 329, 333-336. We have no record with which to assess that claim. The Court complains at length about the burdens of complying with § 203, but we have no meaningful evidence to show how regulated corporations and unions have experienced its restrictions.
Our cases recognize a “type of facial challenge in the First Amendment context under which a law may he overturned as impermissibly overbroad because a substantial number of its applications are unconstitutional.” Washington State Grange v. Washington State Republican Party, 552 U. S. 442, 449, n. 6 (2008) (internal quotation marks omitted). Citizens United has not made an overbreadth argument, and “[w]e generally do not apply the strong medicine of overbreadth analysis where the parties fail to describe the instances of arguable overbreadth of the contested law,” ibid, (internal quotation marks omitted). If our colleagues nonetheless concluded that §203’s fatal flaw is that it affects too much protected speech, they should have invalidated it for overbreadth and given guidance as to which applications are permissible, so that Congress could go about repairing the error.
Also perplexing is the majority’s attempt to pass blame to the Government for its litigating position. By “holding] out the possibility of ruling for Citizens United on a narrow ground yet refrain[ing] from adopting that position,” the majority says, the Government has caused “added uncertainty [that] demonstrates the necessity to address the question of statutory validity.” Ante, at 333. Our colleagues have apparently never heard of an alternative argument. Like every litigant, the Government would prefer to win its case outright; failing that, it would prefer to lose on a narrow ground. The fact that there are numerous different ways this case could be decided, and that the Government acknowledges as much, does not demonstrate anything about the propriety of a facial ruling.
The majority’s “chilling” argument is particularly inapposite with respect to 2 U. S. C. § 441b’s longstanding restriction on the use of corporate general treasury funds for express advocacy. If there was ever any significant uncertainty about what counts as the functional equivalent of ex*403press advocacy, there has been little doubt about what counts as express advocacy since the “magic words” test of Buckley v. Valeo, 424 U. S. 1, 44, n. 52 (1976) (per curiam). Yet even though Citizens United’s briefs never once mention §441b’s restriction on express advocacy; even though this restriction does not generate chilling concerns; and even though no one has suggested that Hillary counts as express advocacy; the majority nonetheless reaches out to opine that this statutory provision is “invalid” as well. Ante, at 365.
The majority adds that the distinction between facial and as-applied challenges does not have “some automatic effect” that mechanically controls the judicial task. Ante, at 331. I agree, but it does not follow that in any given ease we should ignore the distinction, much less invert it.
Professor Fallon proposes an intricate answer to this question that the majority ignores. Fallon 1327-1359. It bears mention that our colleagues have previously cited Professor Fallon’s article for the exact opposite point from the one they wish to make today. In Gonzales v. Carhart, 550 U. S. 124 (2007), the Court explained that “[fit is neither our obligation nor within our traditional institutional role to resolve questions of constitutionality with respect to each potential situation that might develop,” and “[f]or this reason, ‘[a]s-applied challenges are the basic building blocks of constitutional adjudication.’” Id., at 168 (opinion for the Court by Kennedy, J.) (quoting Fallon 1328 (second alteration in original)).
Internal Revenue Code § 501(c)(4) applies, inter alia, to nonprofit organizations “operated exclusively for the promotion of social welfare,. .. the net earnings of which are devoted exclusively to charitable, educational, or recreational purposes.”
The Chief Justice is therefore much too quick when he suggests that, “[e]ven if considered in as-applied terms, a holding in this case that the Act may not be applied to Citizens United — because corporations as well as individuals enjoy the pertinent First Amendment rights — would mean that any other corporation raising the same challenge would also win.” Ante, at 376 (concurring opinion). That conclusion would only follow if the Court were to ignore Citizens United’s plausible as-applied arguments and instead take the implausible position that all corporations and all types of expenditures enjoy the same First Amendment protections, which always trump the interests in regulation. At times, the majority appears to endorse this extreme view. At other times, however, it appears to suggest that nonprofit corporations have a better claim to First Amendment protection than for-profit corporations, see ante, at 337, 355, “advocacy” organizations have a better claim than other nonprofits, ante, at 337, domestic corporations have a better claim than foreign corporations, ante, at 362, small corporations have a better claim than large corporations, ante, at 354-356, and printed matter has a better claim than broadcast communications, ante, at 349. The majority never uses a multinational business corporation in its hypotheticals.
The Court entirely ignores this statutory argument. It concludes that §203 applies to Hillary on the basis of the film’s content, ante, at 324-326, without considering the possibility that §203 does not apply to video-on-demand transmissions generally.
See Colorado Right to Life Comm., Inc. v. Coffman, 498 F. 3d 1137, 1148 (CA10 2007) (adopting this rule and noting that “every other circuit to have addressed this issue” has done likewise); Brief for Independent Sector as Amicus Curiae 10-11 (collecting cases). The Court rejects this solution in part because the Government “merely suggests] it” and “does not say that it agrees with the interpretation.” Ante, at 328, 329. Our colleagues would thus punish a defendant for showing insufficient excitement about a ground it has advanced, at the same time that they decide the ease on a ground the plaintiff expressly abandoned. The Court also protests that a de minimis standard would “requir[e] intricate case-by-case determinations.” Ante, at 329. But de minimis tests need not be intricate at all. A test that granted MCFL status to § 501(c)(4) organizations if they received less than a fixed dollar amount of business donations in the previous year, or if such donations represent less than a fixed percentage of their total assets, would be perfectly easy to understand and administer.
Another bypassed ground, not briefed by the parties, would have been to revive the Snowe-Jeffords Amendment in BORA § 203(c), allowing certain nonprofit corporations to pay for electioneering communications with general treasury funds, to the extent they can trace the payments to individual contributions. See Brief for National Rifle Association as Amicus Curiae 5-15 (arguing forcefully that Congress intended this result).
The Chief Justice finds our discussion of these narrower solutions “quite perplexing” because we suggest that the Court should “latch on to one of them in order to avoid reaching the broader constitutional question,” without doing the same ourselves. Ante, at 375. There is nothing perplexing about the matter, because we are not similarly situated to our colleagues in the majority. We do not share their view of the First Amendment. Our reading of the Constitution would not lead us to strike down any statutes or overturn any precedents in this case, and we therefore have no occasion to practice constitutional avoidance or to vindicate Citizens United’s as-applied challenge. Each of the arguments made above is surely at least as strong as the statutory argument the Court accepted in last year’s Voting Rights Act case, Northwest Austin Municipal Util. Dist. No. One v. Holder, 557 U. S. 193 (2009).
I will have more to say shortly about the merits — about why Austin and McConnell are not doctrinal outliers, as the Court contends, and why their logic is not only defensible but also compelling. For present purposes, I limit the discussion to siore-decisis-speeific considerations.
The Chief Justice suggests that Austin has been undermined by subsequent dissenting opinions. Ante, at 380. Under this view, it appears that the more times the Court stands by a precedent in the face of requests to overrule it, the weaker that precedent becomes. The Chief Justice further suggests that Austin “is uniquely destabilizing because it threatens to subvert our Court’s decisions even outside” its particular facts, as when we applied its reasoning in McConnell-. Ante, at 380. Once again, the theory seems to be that the more we utilize a precedent, the more we call it into question. For those who believe Austin was correctly decided — as the Federal Government and the States have long believed, as the majority of Justices to have served on the Court since Austin have believed, and as we continue to believe — there is nothing “destabilizing” about the prospect of its continued application. It is gutting campaign finance laws across the country, as the Court does today, that will be destabilizing.
Additionally, the majority cites some recent scholarship challenging the historical account of campaign finance law given in United States v. Automobile Workers, 352 U. S. 567 (1957). Ante, at 363-364. Austin did not so much as allude to this historical account, much less rely on it. Even *411if the scholarship cited by the majority is correct that certain campaign finance reforms were less deliberate or less benignly motivated than Automobile Workers suggested, the point remains that this body of law has played a significant and broadly accepted role in American political life for decades upon decades.
See Brief for State of Montana et al. as Amici Curiae 5-13; see also Supp, Brief for Senator John McCain et al. as Amici Curiae 1a-8a (listing 24 States that presently limit or prohibit independent electioneering expenditures from corporate general treasuries).
Magleby, The Importance of the Record in McConnell v. FEC, 3 Election L. J. 285 (2004).
To be sure, the majority may respond that Congress can correct the imbalance by removing BCRA’s soft-money limits. Cf. Tr. of Oral Arg. 24 (Sept. 9, 2009) (query of Kennedy, J.). But this is no response to any legislature that takes campaign finance regulation seriously. It merely illustrates the breadth of the majority’s deregulatory vision.
See Brief for Committee for Economic Development as Amicus Curiae; Brief for American Independent Business Alliance as Amicus Curiae. But see Supp. Brief for Chamber of Commerce of the United States of America as Amicus Curiae.
See Brief for American Federation of Labor and Congress of Industrial Organizations as Amicus Curiae 3, 9.
See Brief for Independent Sector as Amicus Curiae 16-20.
See Brief for State of Montana et al. as Amici Curiae.
The FEC established this process following the Court’s June 2007 decision in that case, 551 U. S. 449. In the brief interval between the establishment of this process and the 2008 election, corporations and unions used it to make $108.5 million in electioneering communications. Supp. Brief for Appellee 22-23; FEC, Electioneering Communication Summary, online at http://fee.gov/finance/disclosure/ECSummary.shtml (all Internet materials as visited Jan. 18, 2010, and available in Clerk of Court’s case file).
Concededly, Austin and McConnell were constitutional decisions, and we have often said that “claims of stare decisis are at their weakest in that field, where our mistakes cannot be corrected by Congress.” Vieth v. Jubelirer, 541 U. S. 267, 305 (2004) (plurality opinion). As a general matter, this principle is a sound one. But the principle only takes on real force when an earlier ruling has obstructed the normal democratic process; it is the fear of making “mistakes [that] cannot be corrected by Congress,” ibid., that motivates us to review constitutional precedents with a more critical eye. Austin and McConnell did not obstruct state or congressional legislative power in any way. Although it is unclear how high a bar today’s decision will pose to future attempts to regulate corporate electioneering, it will clearly restrain much legislative action.
See FEC, Number of Federal PAC’s Increases, http://fec.gov/press/ press2008/20080812paccount.shtml.
See Supp. Brief for Appellee 16 (citing FEC statistics placing this figure at $840 million). The majority finds the PAC option inadequate in part because “[a] PAC is a separate association from the corporation.” Ante, at 337. The formal “separateness” of PACs from their host corporations — which administer and control the PACs but which cannot funnel general treasury funds into them or force members to support them — is, of course, the whole point of the PAC mechanism.
Roaming far afield from the case at hand, the majority worries that the Government will use § 203 to ban books, pamphlets, and blogs. Ante, at 333, 337, 349, 364. Yet by its plain terms, §203 does not apply to printed material. See 2 U. S. C. §434(f)(3)(A)(i); see also 11 CFR § 100.29(c)(1) (“[Electioneering communication does not include communications appearing in print media”). And in light of the ordinary understanding of the terms “broadcast, cable, [and] satellite,” 2 U. S. C. §434(f)(3)(A)(i), coupled with Congress’ dear aim of targeting “a virtual torrent of televised election-related ads,” McConnell, 540 U. S., at 207, we highly doubt that §203 could be interpreted to apply to a Web site or book that happens to be transmitted at some stage over airwaves or cable lines, or that the FEC would ever try to do so. See 11 CFR § 100.26 (exempting most Internet communications from regulation as advertising); § 100.155 (exempting uncompensated Internet activity from regulation as an expenditure); Supp. Brief for Center for Independent Media et al. as Amici Curiae 14 (explaining that “the FEC has consistently construed [BCRA’s] media exemption to apply to a variety of non-traditional media”). If it should, the Government acknowledges “there would be quite [a] good as-applied challenge.” Tr. of Oral Arg. 65 (Sept. 9, 2009).
As the Government points out, with a media corporation there is also a lesser risk that investors will not understand, learn about, or support the advocacy messages that the corporation disseminates. Supp. Reply Brief for Appellee 10. Everyone knows and expects that media outlets may seek to influence elections in this way.
2 U. S. C. §434(f)(3)(A)(i).
§ 434(f)(3)(C).
§ 434(f)(3)(A)(i)(II).
§ 441b(b); McConnell, 540 U. S, at 211.
§441b(b)(2)(C).
WRTL, 551 U. S. 449, 470 (2007) (opinion of Roberts, C. J.).
It is likewise nonsense to suggest that the FEC’s ‘“business is to censor.’” Ante, at 335 (quoting Freedman v. Maryland, 380 U. S. 51, 57 (1965)). The FEC’s business is to administer and enforce the campaign finance laws. The regulatory body at issue in Freedman was a state board of censors that had virtually unfettered discretion to bar distribution of motion picture films it deemed not to be “ “moral and proper. ’ ” See id., at 52-53, and n. 2. No movie could be shown in the State of Maryland that was not first approved and licensed by the board of censors. Id., at 52, n. 1. It is an understatement to say that Freedman is not on point, and the majority’s characterization of the FEC is deeply disconcerting.
Citizens United has administered this PAC for over a decade. See Defendant FEC’s Memorandum in Opposition to Plaintiff’s Second Motion for Preliminary Injunction in No. 07-2240 (ARR, RCL, RWR) (DC), p. 20. Citizens United also operates multiple “527” organizations that engage in partisan political activity. See Defendant FEC’s Statement of Material Facts as to Which There Is No Genuine Dispute in No. 07-2240 (DC), n 22-24.
See, e. g., Bethel School Dist. No. 403 v. Fraser, 478 U. S. 675, 682 (1986) ("[T]he constitutional rights of students in public school are not automatically coextensive with the rights of adults in other settings”).
See, e. g., Jones v. North Carolina Prisoners’ Labor Union, Inc., 433 U. S. 119, 129 (1977) (“In a prison context, an inmate does.not retain those First Amendment rights that are inconsistent with his status as a prisoner or with the legitimate penological objectives of the corrections system” (internal quotation marks omitted)).
See, e. g., Parker v. Levy, 417 U. S. 733, 758 (1974) (“While the members of the military are not excluded from the protection granted by the First Amendment, the different character of the military community and of the military mission requires a different application of those protections”).
See, e. g., 2 U. S. C. § 441e(a)(1) (foreign nationals may not directly or indirectly make contributions or independent expenditures in connection with a U. S. election).
See, e. g., Civil Service Comm’n v. Letter Carriers, 413 U. S. 548, 550 (1973) (upholding statute prohibiting Executive Branch employees from taking “an active part in political management or in political campaigns” *421(internal quotation marks omitted)); Public Workers v. Mitchell, 330 U. S. 75 (1947) (same); United States v. Wurzbach, 280 U. S. 396, 398 (1930) (upholding statute prohibiting federal employees from making contributions to Members of Congress for “any political purpose whatever” (internal quotation marks omitted)); Ex parte Curtis, 106 U. S. 371 (1882) (upholding statute prohibiting certain federal employees from giving money to other employees for political purposes).
The majority states that the cases just cited are “inapposite” because they “stand only for the proposition that there are certain governmental functions that cannot operate without some restrictions on particular kinds of speech.” Ante, at 341. The majority’s creative suggestion that these cases stand only for that one proposition is quite implausible. In any event, the proposition lies at the heart of this case, as Congress and half the state legislatures have concluded, over many decades, that their core functions of administering elections and passing legislation cannot operate effectively without some narrow restrictions on corporate electioneering paid for by general treasury funds.
Outside of the law, of course, it is a commonplace that the identity and incentives of the speaker might be relevant to an assessment of his speech. See Aristotle, Poetics § 11.2(vi), pp. 43-44 (M. Heath transl. 1996) (“In evaluating any utterance or action, one must take into account not just the moral qualities of what is actually done or said, but also the identity of the agent or speaker, the addressee, the occasion, the means, and the motive”). The insight that the identity of speakers is a proper subject of regulatory concern, it bears noting, motivates the disclaimer and disclosure provisions that the Court today upholds.
I dissented in Forbes because the broadcaster’s decision to exclude the respondent from its debate was done “on the basis of entirely subjective, ad hoc judgments,” 523 U. S., at 690, that suggested anticompetitive viewpoint discrimination, id., at 693-694, and lacked a compelling justification. Needless to say, my concerns do not apply to the instant case.
The law at issue in Burson was far from unusual. “[A]ll 50 States,” the Court observed, “limit access to the areas in or around polling places.” 504 U. S., at 206 (plurality opinion); see also Note, 91 Ky. L. J. 715, 729, n. 89, 747-769 (2003) (collecting statutes). I dissented in Burson because the evidence adduced to justify Tennessee’s law was “exceptionally thin,” 504 U. S., at 219, and “the reason for [the] restriction [had] disappeared]” *423over time, id., at 223. “In short,” I concluded, “Tennessee ha[d] failed to point to any legitimate interest that would justify its selective regulation of campaign-related expression.” Id., at 225. These criticisms are inapplicable to the ease before us.
They are likewise entitled to regulate media corporations differently from other corporations “to ensure that the law ‘does not hinder or prevent the institutional press from reporting on, and publishing editorials about, newsworthy events.’ ” McConnell, 540 U. S., at 208 (quoting Austin v. Michigan Chamber of Commerce, 494 U. S. 652, 668 (1990)).
The Court all but confesses that a categorical approach to speaker identity is untenable when it acknowledges that Congress might be allowed to take measures aimed at “preventing foreign individuals or associations from influencing our Nation’s political process.” Ante, at 362. Such measures have been a part of U. S. campaign finance law for many years. The notion that Congress might lack the authority to distinguish foreigners from citizens in the regulation of electioneering would certainly have surprised the Framers, whose “obsession with foreign influence derived from a fear that foreign powers and individuals had no basic investment in the well-being of the country.” Teachout, The Anti-Corruption Principle, 94 Cornell L. Rev. 341, 393, n. 245 (2009) (hereinafter Teachout); see also U. S. Const., Art. I, §9, cl. 8 (“[N]o Person holding any Office of Profit or Trust. . . shall, -without the Consent of the Congress, accept of any present, Emolument, Office, or Title, of any kind whatever, from any King, Prince, or foreign State”). Professor Teachout observes that a cor*425poration might be analogized to a foreign power in this respect, “inasmuch as its legal loyalties necessarily exclude patriotism.” Teachout 393, n. 245.
See A. Bickel, The Supreme Court and the Idea of Progress 59-60 (1978); A. Meiklejohn, Political Freedom: The Constitutional Powers of the People 39-40 (1965); Tokaji, First Amendment Equal Protection: On Discretion, Inequality, and Participation, 101 Mich. L. Rev. 2409, 2508-2509 (2003). Of course, voting is not speech in a pure or formal sense, but then again neither is a campaign expenditure; both are nevertheless communicative acts aimed at influencing electoral outcomes. Cf. Strauss, Corruption, Equality, and Campaign Finance Reform, 94 Colum. L. Rev. 1369, 1383-1384 (1994) (hereinafter Strauss).
Scholars have found that only a handful of business corporations were issued charters during the colonial period, and only a few hundred during all of the 18th century. See E. Dodd, American Business Corporations Until 1860, p. 197 (1954); L. Friedman, A History of American Law 188-189 (2d ed. 1985); Baldwin, American Business Corporations Before 1789, 8 Am. Hist. Rev. 449, 450-459 (1903). Justice Scalia quibbles with these figures; whereas we say that “a few hundred” charters were issued to business corporations during the 18th century, he says that the number is “approximately 335.” Ante, at 387 (concurring opinion). Justice Scalia also raises the more serious point that it is improper to assess these figures by today’s standards, ibid., though I believe he fails to substantiate his claim that “the corporation was a familiar figure in American economic life” by the century’s end, ibid, (internal quotation marks omitted). His formulation of that claim is also misleading, because the relevant reference point is not 1800 but the date of the First Amendment’s ratification, in 1791. And at that time, the number of business charters must have been significantly smaller than 335, because the pace of chartering only began to pick up steam in the last decade of the 18th century. More than half of the century’s total business charters were issued between 1796 and 1800. Friedman, History of American Law, at 189.
See Letter from Thomas Jefferson to Tom Logan (Nov. 12, 1816), in 12 The Works of Thomas Jefferson 42, 44 (P. Ford ed. 1905) (“I hope we shall . . . crush in [its] birth the aristocracy of our monied corporations which dare already to challenge our government to a trial of strength and bid defiance to the laws of our country”).
In normal usage then, as now, the term “speech” referred to oral communications by individuals. See, e. g., 2 S. Johnson, Dictionary of the English Language 1853-1854 (4th ed. 1773) (reprinted 1978) (listing as primary definition of “speech”: “The power of articulate utterance; the power of expressing thoughts by vocal words”); 2 N. Webster, American Dictionary of the English Language (1828) (reprinted 1970) (listing as primary definition of “speech”: “The faculty of uttering articulate sounds or words, as in human beings; the faculty of expressing thoughts by words or articulate sounds. Speech was given to man by his Creator for the noblest purposes”). Indeed, it has been “claimed that the notion of institutional speech ... did not exist in post-revolutionary America.” Fagundes, State Actors as First Amendment Speakers, 100 Nw. U. L. Rev. 1637, 1654 (2006); see also Bezanson, Institutional Speech, 80 Iowa L. Rev. 735, 775 (1995) (“In the intellectual heritage of the eighteenth century, the idea that free speech was individual and personal was deeply rooted and clearly manifest in the writings of Locke, Milton, and others on whom the framers of the Constitution and the Bill of Rights drew”). Given that corporations were conceived of as artificial entities and do not have the technical capacity to “speak,” the burden of establishing that the Framers and ratifiers understood “the freedom of speech” to encompass corporate speech is, I believe, far heavier than the majority acknowledges.
Postratifieation practice bolsters the conclusion that the First Amendment, “as originally understood,” ante, at 353, did not give corporations political speech rights on a par with the rights of individuals. Well into the modern era of general incorporation statutes, “[t]he common law was generally interpreted as prohibiting corporate political participation,” First Nat. Bank of Boston v. Bellotti, 435 U. S. 765, 819 (1978) (White, J., dissenting), and this Court did not recognize any First Amendment protections for corporations until the middle part of the 20th century, see ante, at 342 (listing cases).
In fact, the Free Press Clause might be toned against Justice Scalia, for two reasons. First, we learn from it that the drafters of the First Amendment did draw distinctions — explicit distinctions — between types of “speakers,” or speech outlets or forms. Second, the Court’s strongest historical evidence all relates to the Framers’ views on the press, see ante, at 353-354; ante, at 388-390 (Scalia, J., concurring), yet while the Court tries to sweep this evidence into the Free Speech Clause, the Free Press Clause provides a more natural textual home. The text and history highlighted by our colleagues suggests why one type of corporation, those that are part of the press, might be able to claim special First Amendment status, and therefore why some kinds of “identity”-based distinctions might be permissible after all. Once one accepts that much, the intellectual edifice of the majority opinion crumbles.
Cf. L. Levy, Legacy of Suppression: Freedom o'f Speech and Press in Early American History 4 (1960) (“The meaning of no other clause of the Bill of Rights at the time of its framing and ratification has been so obscure to us” as the Free Speech and Press Clause).
As the majority notes, there is some academic debate about the precise origins of these developments. Ante, at 363-364; see also n. 19, supra. There is always some academic debate about such developments; the motives of legislatures are never entirely clear or unitary. Yet the basic shape and trajectory of 20th-century campaign finance reform are clear, and one need not take a naive or triumphalist view of this history to find it highly relevant. The Court’s skepticism does nothing to mitigate the absurdity of its claim that Austin and McConnell were outliers. Nor does it alter the fact that five Justices today destroy a longstanding American practice.
See Pipefitters v. United States, 407 U.S. 385, 409, 414-415 (1972) (reading the statutory bar on corporate and union campaign spending not to apply to “the voluntary donations of employees,” when maintained in a separate account, because “[t]he dominant [legislative] concern in requiring that contributions be voluntary was, after all, to protect the dissenting stockholder or union member”); Automobile Workers, 352 U. S., at 592 (advising the District Court to consider on remand whether the broadcast in question was “paid for out of the general dues of the union membership or [whether] the funds [could] be fairly said to have been obtained on a voluntary basis”); United States v. CIO, 335 U. S. 106, 123 (1948) (observing that “funds voluntarily contributed [by union members or corporate stockholders] for election purposes” might not be covered by the expenditure bar). Both the Pipefitters and the Automobile Workers Courts approvingly referenced Congress’ goal of reducing “the effect of aggregated wealth on federal elections,” understood as wealth drawn from a corporate or union general treasury without the stockholders’ or members’ “free and knowing choice.” Pipefitters, 407 U. S., at 416; see Automobile Workers, 352 U. S., at 582.
The two dissenters in Pipefitters would not have read the statutory provision in question, a successor to § 304 of the Taft-Hartley Act, to allow such robust use of corporate and union funds to finance otherwise prohibited electioneering. “This opening of the door to extensive corporate and union influence on the elective and legislative processes,” Justice Powell wrote, “must be viewed with genuine concern. This seems to me to be a regressive step as contrasted with the numerous legislative and judicial actions in recent years designed to assure that elections are indeed free and representative.” 407 U. S., at 450 (opinion of Powell, J., joined by Burger, C. J.).
Specifically, these corporations had to meet three conditions. First, they had to be formed “for the express purpose of promoting political ideas,” so that their resources reflected political support rather than commercial success. MCFL, 479 U. S., at 264. Next, they had to have no shareholders, so that “persons connected with the organization will have no economic disincentive for disassociating with it if they disagree with its political activity.” Ibid. Finally, they could not be “established by a business corporation or a labor union,” nor “accept contributions from such entities,” lest they “serv[e] as conduits for the type of direct spending that creates a threat to the political marketplace.” Ibid.
According to The Chief Justice, we are “erroneou[s]” in claiming that McConnell and Beaumont “ 'reaffirmed’ ” Austin. Ante, at 376-377. In both cases, the Court explicitly relied on Austin and quoted from it at length. See 540 U. S., at 204-205; 539 U. S., at 153-155, 158, 160, 163; see also ante, at 332 (opinion of the Court) (“The holding and validity of Austin were essential to the reasoning of the McConnell majority opinion”); Brief for Appellants National Rifle Association et al., O. T. 2003, No. 02-1675, p. 21 (“Beaumont reaffirmed ... the Austin rationale for restricting expenditures”). The McConnell Court did so in the teeth of vigorous protests by Justices in today’s majority that Austin should be overruled. See ante, at 332 (citing relevant passages); see also Beaumont, 539 U. S., at 163-164 (Kennedy, J., concurring in judgment). Both Courts also heard criticisms of Austin from parties or amici. See Brief for Appellants Chamber of Commerce of the United States et al., O. T. 2003, No. 02-1756, p. 35, n. 22; Reply Brief for Appellants/Cross-Appellees Senator Mitch McConnell et al., O. T. 2003, No. 02-1674, pp. 13-14; Brief for Pacific Legal Foundation as Amicus Curiae in FEC v. Beaumont, O. T. 2002, No. 02-403, passim. If this does not qualify as reaffirmation of a precedent, then I do not know what would.
Cf. Nixon v. Shrink Missouri Government PAC, 528 U. S. 377, 389 (2000) (recognizing “the broader threat from politicians too compliant with the wishes of large contributors”)- Though discrete in scope, these experiments must impose some meaningful limits if they are to have a chance at functioning effectively and preserving the public’s trust. “Even if it occurs only occasionally, the potential for such undue influence is manifest. And unlike straight cash-for-votes transactions, such corruption is neither easily detected nor practical to criminalize.” McConnell, 540 U.S., at 153. There should be nothing controversial about the proposition that the influence being targeted is “undue.” In a democracy, officeholders should not make public decisions with the aim of placating a financial benefactor, except to the extent that the benefactor is seen as representative of a larger constituency or its arguments are seen as especially persuasive.
The majority declares by fiat that the appearance of undue influence by high-spending corporations “will not cause the electorate to lose faith in our democracy.” Ante, at 360. The electorate itself has consistently indicated otherwise, both in opinion polls, see McConnell v. FEC, 251 F. Supp. 2d 176, 557-558, 623-624 (DC 2003) (opinion of Kollar-Kotelly, J.), and in the laws its representatives have passed, and our colleagues have no basis for elevating their own optimism into a tenet of constitutional law.
Quite distinct from the interest in preventing improper influences on the electoral process, I have long believed that “a number of [other] purposes, both legitimate and substantial, may justify the imposition of reasonable limitations on the expenditures permitted during the course of any single campaign.” Davis v. FEC, 554 U. S. 724, 751 (2008) (opinion concurring in part and dissenting in part). In my judgment, such limitations may be justified to the extent they are tailored to “improving the quality of the exposition of ideas” that voters receive, ibid., “freeing] candidates and their staffs from the interminable burden of fundraising,” ibid. (internal quotation marks omitted), and “protecting] equal access to the political arena,” Randall v. Sorrell, 548 U. S. 230, 278 (2006) (Stevens, J., dissenting) (internal quotation marks omitted). I continue to adhere to these beliefs, but they have not been briefed by the parties or amici in this case, and their soundness is immaterial to its proper disposition.
In fact, the notion that the “electioneering communications” covered by §203 can breed quid pro quo corruption or the appearance of such corruption has only become more plausible since we decided McConnell. Recall that The Chief Justice’s controlling opinion in WRTL subsequently limited BCRA’s definition of “electioneering communications” to those that are “susceptible of no reasonable interpretation other than as an appeal to vote for or against a specific candidate.” 551 U. S., at 470. The upshot was that after WRTL, a corporate or union expenditure could be regulated under §203 only if everyone would understand it as an endorsement of or attack on a particular candidate for office. It does not *458take much imagination to perceive why this type of advocacy might be especially apt to look like or amount to a deal or a threat.
“We must give weight” and “due deference” to Congress’ efforts to dispel corruption, the Court states at one point. Ante, at 361. It is unclear to me what these maxims mean, but as applied by the Court they clearly do not entail “deference” in any normal sense of that term.
Justice Breyer has suggested that we strike the balance as follows: “We should defer to [the legislature’s] political judgment that unlimited spending threatens the integrity of the electoral process. But we should not defer in respect to whether its solution . .. insulates legislators from effective electoral challenge.” Shrink Missouri, 528 U. S., at 403-404 (concurring opinion).
The Chief Justice denies this, ante, at 380-382, citing scholarship that has interpreted Austin to endorse an equality rationale, along with an article by Justice Thurgood Marshall’s former law clerk that states that Marshall, the author of Austin, accepted “equality of opportunity” and “equalizing access to the political process” as bases for campaign finance regulation, Garrett, New Voices in Politics: Justice Marshall’s Jurisprudence on Law and Polities, 52 How. L. J. 655, 667-668 (2009) (internal quotation marks omitted). It is fair to say that Austin can bear an egalitarian reading, and I have no reason to doubt this characterization of Justice Marshall’s beliefs. But the fact that Austin can be read a certain way hardly proves The Chief Justice’s charge that there is nothing more to it. Many of our precedents can bear multiple readings, and many of our doctrines have some “equalizing” implications but do not rest on an equalizing theory: for example, our takings jurisprudence and numerous rules of criminal procedure. More importantly, the Austin Court expressly declined to rely on a speech-equalization rationale, see 494 U. S., at 660, and we have never understood Austin to stand for such a rationale. Whatever his personal views, Justice Marshall simply did not write the opinion that The Chief Justice suggests he did; indeed, he “would have viewed it as irresponsible to write an opinion that boldly staked out a rationale based on equality that no one other than perhaps Justice White would have even considered joining,” Garrett, 52 How. L. J., at 674.
In state elections, even domestic corporations may be “foreign” controlled in the sense that they are incorporated in another jurisdiction and primarily owned and operated by out-of-state residents.
Regan, Corporate Speech and Civic Virtue, in Debating Democracy’s Discontent 289, 302 (A. Allen & M. Regan eds. 1998) (hereinafter Regan).
Nothing in this analysis turns on whether the corporation is conceptualized as a grantee of a state concession, see, e. g., Trustees of Dartmouth College v. Woodward, 4 Wheat. 518, 636 (1819) (Marshall, C. J.), a nexus of explicit and implicit contracts, see, e. g., F. Easterbrook & D. Fischel, The Economic Structure of Corporate Law 12 (1991), a mediated hierarchy of stakeholders, see, e. g., Blair & Stout, A Team Production Theory of Corpo*466rate Law, 85 Va. L. Rev. 247 (1999) (hereinafter Blair & Stout), or any other recognized model. Austin referred to the structure and the advantages of corporations as “state-conferred” in several places, 494 U. S., at 660, 665, 667, but its antidistortion argument relied only on the basic descriptive features of corporations, as sketched above. It is not necessary to agree on a precise theory of the corporation to agree that corporations differ from natural persons in fundamental ways, and that a legislature might therefore need to regulate them differently if it is human welfare that is the object of its concern. Cf. Hansmann & Kraakman 441, n. 5.
Not all corporations support BCRA § 203, of course, and not all corporations are large business entities or their tax-exempt adjuncts. Some nonprofit corporations are created for an ideological purpose. Some closely held corporations are strongly identified with a particular owner or founder. The fact that § 203, like the statute at issue in Austin, regulates some of these corporations’ expenditures does not disturb the analysis above. See 494 U. S., at 661-665. Small-business owners may speak in their own names, rather than the business’, if they wish to evade §203 altogether. Nonprofit corporations that want to make unrestricted electioneering expenditures may do so if they refuse donations from businesses and unions and permit members to disassociate without economic penalty. See MCFL, 479 U. S. 238, 264 (1986). Making it plain that their decision is not motivated by a concern about BCRA’s coverage of nonprofits that have ideological missions but lack MCFL status, our colleagues refuse to apply the Snowe-Jeffords Amendment or the lower courts’ de minimis exception to MCFL. See ante, at 327-329.
Of course, no presiding person in a courtroom, legislature, classroom, polling place, or family dinner would take this hyperbole literally.
Under the majority’s view, the legislature is thus damned if it does and damned if it doesn’t. If the legislature gives media corporations an exemption from electioneering regulations that apply to other corporations, it violates the newly minted First Amendment rule against identity-based distinctions. If the legislature does not give media corporations an exemption, it violates the First Amendment rights of the press. The only way out of this invented bind: no regulations whatsoever.
I note that, among the many other regulatory possibilities it has left open, ranging from new versions of §203 supported by additional evidence of quid pro quo corruption or its appearance to any number of tax incentive or public financing schemes, today’s decision does not require that a legislature rely solely on these mechanisms to protect shareholders. Legislatures remain free in their incorporation and tax laws to condition the types of activity in which corporations may engage, including electioneering activity, on specific disclosure requirements or on prior express approval by shareholders or members.