with whom Justice Alito joins, concurring.
The Government urges us in this case to uphold a direct prohibition on political speech. It asks us to embrace a theory of the First Amendment that would allow censorship not only of television and radio broadcasts, but of pamphlets, *373posters, the Internet, and virtually any other medium that corporations and unions might find useful in expressing their views on matters of public concern. Its theory, if accepted, would empower the Government to prohibit newspapers from running editorials or opinion pieces supporting or opposing candidates for office, so long as the newspapers were owned by corporations — as the major ones are. First Amendment rights could be confined to individuals, subverting the vibrant public discourse that is at the foundation of our democracy.
The Court properly rejects that theory, and I join its opinion in full. The First Amendment protects more than just the individual on a soapbox and the lonely pamphleteer. I write separately to address the important principles of judicial restraint and stare decisis implicated in this case.
I
Judging the constitutionality of an Act of Congress is “the gravest and most delicate duty that this Court is called on to perform.” Blodgett v. Holden, 275 U. S. 142, 147-148 (1927) (Holmes, J., concurring). Because the stakes are so high, our standard practice is to refrain from addressing constitutional questions except when necessary to rule on particular claims before us. See Ashwander v. TVA, 297 U. S. 288, 346-348 (1936) (Brandeis, J., concurring). This policy underlies both our willingness to construe ambiguous statutes to avoid constitutional problems and our practice “ ‘never to formulate a rule of constitutional law broader than is required by the precise facts to which it is to be applied.’” United States v. Raines, 362 U. S. 17, 21 (1960) (quoting Liverpool, New York & Philadelphia S. S. Co. v. Commissioners of Emigration, 113 U. S. 33, 39 (1885)).
The majority and dissent are united in expressing allegiance to these principles. Ante, at 329; post, at 405-406 (Stevens, J., concurring in part and dissenting in part). *374But I cannot agree with my dissenting colleagues on how these principles apply in this case.
The majority’s step-by-step analysis accords with our standard practice of avoiding broad constitutional questions except when necessary to decide the case before us. The majority begins by addressing — and quite properly rejecting — Citizens United’s statutory claim that 2 U. S. C. § 441b does not actually cover its production and distribution of Hillary: The Movie (hereinafter Hillary). If there were a valid basis for deciding this statutory claim in Citizens United’s favor (and thereby avoiding constitutional adjudication), it would be proper to do so. Indeed, that is precisely the approach the Court took just last Term in Northwest Austin Municipal Util. Dist. No. One v. Holder, 557 U. S. 193 (2009), when eight Members of the Court agreed to decide the case on statutory grounds instead of reaching the appellant’s broader argument that the Voting Rights Act is unconstitutional.
It is only because the majority rejects Citizens United’s statutory claim that it proceeds to consider the group’s various constitutional arguments, beginning with its narrowest claim (that Hillary is not the functional equivalent of express advocacy) and proceeding to its broadest claim (that Austin v. Michigan Chamber of Commerce, 494 U. S. 652 (1990), should be overruled). This is the same order of operations followed by the controlling opinion in Federal Election Comm’n v. Wisconsin Right to Life, Inc., 551 U. S. 449 (2007) (WRTL). There the appellant was able to prevail on its narrowest constitutional argument because its broadcast ads did not qualify as the functional equivalent of express advocacy; there was thus no need to go on to address the broader claim that McConnell v. Federal Election Comm’n, 540 U. S. 93 (2003), should be overruled. WRTL, 551 U. S., at 482; id., at 482-483 (Alito, J., concurring). This case is different — not, as the dissent suggests, because the approach taken in WRTL has been deemed a “failure,” post, at 402, *375but because, in the absence of any valid narrower ground of decision, there is no way to avoid Citizens United’s broader constitutional argument.
The dissent advocates an approach to addressing Citizens United’s claims that I find quite perplexing. It presumably agrees with the majority that Citizens United’s narrower statutory and constitutional arguments lack merit — otherwise its conclusion that the group should lose this case would make no sense. Despite agreeing that these narrower arguments fail, however, the dissent argues that the majority should nonetheless latch on to one of them in order to avoid reaching the broader constitutional question of whether Austin remains good law. It even suggests that the Court’s failure to adopt one of these concededly meritless arguments is a sign that the majority is not “serious about judicial restraint.” Post, at 408.
This approach is based on a false premise: that our practice of avoiding .unnecessary (and unnecessarily broad) constitutional holdings somehow trumps our obligation faithfully to interpret the law. It should go without saying, however, that we cannot embrace a narrow ground of decision simply because it is narrow; it must also be right. Thus while it is true that “[i]f it is not necessary to decide more, it is necessary not to decide more,” post, at 405 (internal quotation marks omitted), sometimes it is necessary to decide more. There is a difference between judicial restraint and judicial abdication. When constitutional questions are “indispensably necessary” to resolving the case at hand, “the court must meet and decide them.” Ex parte Randolph, 20 F. Cas. 242, 254 (No. 11,558) (CC Va. 1883) (Marshall, C. J.).
Because it is necessary to reach Citizens United’s broader argument that Austin should be overruled, the debate over whether to consider this claim on an as-applied or facial basis strikes me as largely beside the point. Citizens United has standing — it is being injured by the Government’s enforcement of the Act. Citizens United has a constitutional *376claim — the Act violates the First Amendment, because it prohibits political speech. The Government has a defense— the Act may be enforced, consistent with the First Amendment, against corporations. Whether the claim or the defense prevails is the question before us.
Given the nature of that claim and defense, it makes no difference of any substance whether this case is resolved by invalidating the statute on its face or only as applied to Citizens United. Even if considered in as-applied terms, a holding in this ease 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. Likewise, a conclusion that the Act may be applied to Citizens United — because it is constitutional to prohibit corporate political speech — would similarly govern future cases. Regardless whether we label Citizens United’s claim a “facial” or “as-applied” challenge, the consequences of the Court’s decision are the same.1
II
The text and purpose of the First Amendment point in the same direction: Congress may not prohibit political speech, even if the speaker is a corporation or union. What makes this case difficult is the need to confront our prior decision in Austin.
This is the first case in which we have been asked to overrule Austin, and thus it is also the first in which we have had reason to consider how much weight to give stare decisis in assessing its continued validity. The dissent erroneously *377declares that the Court “reaffirmed” Austin’s holding in subsequent cases — namely, Federal Election Comm’n v. Beaumont, 539 U. S. 146 (2003); McConnell; and WRTL. Post, at 439-441. Not so. Not a single party in any of those cases asked us to overrule Austin, and as the dissent points out, post, at 396-398, the Court generally does not consider constitutional arguments that have not properly been raised. Austin’s validity was therefore not directly at issue in the cases the dissent cites. The Court's unwillingness to overturn Austin in those eases cannot be understood as a reaffirmation of that decision.
A
Fidelity to precedent — the policy of stare decisis — is vital to the proper exercise of the judicial function. “Stare decisis is the preferred course because it promotes the evenhanded, predictable, and consistent development of legal principles, fosters reliance on judicial decisions, and contributes to the actual and perceived integrity of the judicial process.” Payne v. Tennessee, 501 U. S. 808, 827 (1991). For these reasons, we have long recognized that departures from precedent are inappropriate in the absence of a “special justification.” Arizona v. Rumsey, 467 U. S. 203, 212 (1984).
At the same time, stare decisis is neither an “inexorable command,” Lawrence v. Texas, 539 U. S. 558, 577 (2003), nor “a mechanical formula of adherence to the latest decision,” Helvering v. Hallock, 309 U. S. 106, 119 (1940), especially in constitutional cases, see United States v. Scott, 437 U. S. 82, 101 (1978). If it were, segregation would be legal, minimum wage laws would be unconstitutional, and the Government could wiretap ordinary criminal suspects without first obtaining warrants. See Plessy v. Ferguson, 163 U. S. 537 (1896), overruled by Brown v. Board of Education, 347 U. S. 483 (1954); Adkins v. Children’s Hospital of D. C., 261 U. S. 525 (1923), overruled by West Coast Hotel Co. v. Parrish, 300 U. S. 379 (1937); Olmstead v. United States, 277 U. S. 438 (1928), overruled by Katz v. United States, *378389 U. S. 347 (1967). As the dissent properly notes, none of us has viewed stare decisis in such absolute terms. Post, at 408; see also, e. g., Randall v. Sorrell, 548 U. S. 230, 274-281 (2006) (Stevens, J., dissenting) (urging the Court to overrule its invalidation of limits on independent expenditures on political speech in Buckley v. Valeo, 424 U. S. 1 (1976) (per curiam)).
Stare decisis is instead a “principle of policy.” Helvering, supra, at 119. When considering whether to reexamine a prior erroneous holding, we must balance the importance of having constitutional questions decided against the importance of having them decided right. As Justice Jackson explained, this requires a “sober appraisal of the disadvantages of the innovation as well as those of the questioned case, a weighing of practical effects of one against the other.” Jackson, Decisional Law and Stare Decisis, 30 A. B. A. J. 334 (1944).
In conducting this balancing, we must keep in mind that stare decisis is not an end in itself. It is instead “the means by which we ensure that the law will not merely change erratically, but will develop in a principled and intelligible fashion.” Vasquez v. Hillery, 474 U. S. 254, 265 (1986). Its greatest purpose is to serve a constitutional ideal — the rule of law. It follows that in the unusual circumstance when fidelity to any particular precedent does more to damage this constitutional ideal than to advance it, we must be more willing to depart from that precedent.
Thus, for example, if the precedent under consideration itself departed from the Court’s jurisprudence, returning to the “‘intrinsically sounder’ doctrine established in prior cases” may “better serv[e] the values of stare decisis than would following [the] more recently decided case inconsistent with the decisions that came before it.” Adarand Constructors, Inc. v. Peña, 515 U. S. 200, 231 (1995); see also Helvering, supra, at 119; Randall, supra, at 274 (Stevens, J., dissenting). Abrogating the errant precedent, rather than *379reaffirming or extending it, might better preserve the law’s coherence and curtail the precedent’s disruptive effects.
Likewise, if adherence to a precedent actually impedes the stable and orderly adjudication of future cases, its stare decisis effect is also diminished. This can happen in a number of circumstances, such as when the precedent’s validity is so hotly contested that it cannot reliably function as a basis for decision in future cases, when its rationale threatens to upend our settled jurisprudence in related areas of law, and when the precedent’s underlying reasoning has become so discredited that the Court cannot keep the precedent alive without jury-rigging new and different justifications to shore up the original mistake. See, e. g., Pearson v. Callahan, 555 U. S. 223, 235 (2009); Montejo v. Louisiana, 556 U. S. 778, 792 (2009) (stare decisis does not control when adherence to the prior decision requires “fundamentally revising its theoretical basis”).
B
These considerations weigh against retaining our decision in Austin. First, as the majority explains, that decision was an “aberration” insofar as it departed from the robust protections we had granted political speech in our earlier cases. Ante, at 355; see also Buckley, supra; First Nat. Bank of Boston v. Bellotti, 435 U. S. 765 (1978). Austin undermined the careful line that Buckley drew to distinguish limits on contributions to candidates from limits on independent expenditures on speech. Buckley rejected the asserted government interest in regulating independent expenditures, concluding that “restrict[ing] the speech of some elements of our society in order to enhance the relative voice of others is wholly foreign to the First Amendment.” 424 U. S., at 48-49; see also Bellotti, supra, at 790-791; Citizens Against Rent Control/Coalition for Fair Housing v. Berkeley, 454 U. S. 290, 295 (1981). Austin, however, allowed the Government to prohibit these same expenditures out of concern for “the corrosive and distorting effects of immense aggrega*380tions of wealth” in the marketplace of ideas. 494 U. S., at 660. Austin's reasoning was — and remains — inconsistent with Buckley's explicit repudiation of any government interest in “equalizing the relative ability of individuals and groups to influence the outcome of elections.” 424 U. S., at 48-49.
Austin was also inconsistent with Bellotti’s clear rejection of the idea that “speech that otherwise would be within the protection of the First Amendment loses that protection simply because its source is a corporation.” 435 U. S., at 784. The dissent correctly points out that Bellotii involved a referendum rather than a candidate election, and that Bellota itself noted this factual distinction, id., at 788, n. 26; post, at 442-443. But this distinction does not explain why corporations may be subject to prohibitions on speech in candidate elections when individuals may not.
Second, the validity of Austin’s rationale — itself adopted over two “spirited dissents,” Payne, 501 U. S., at 829 — has proved to be the consistent subject of dispute among Members of this Court ever since. See, e. g., WRTL, 551 U. S., at 483 (Scalia, J., joined by Kennedy and Thomas, JJ., concurring in part and concurring in judgment); McConnell, 540 U. S., at 247, 264, 286 (opinions of Scalia, Thomas, and Kennedy, JJ.); Beaumont, 539 U. S., at 163, 164 (opinions of Kennedy and Thomas, JJ.). The simple fact that one of our decisions remains controversial is, of course, insufficient to justify overruling it. But it does undermine the precedent’s ability to contribute to the stable and orderly development of the law. In such circumstances, it is entirely appropriate for the Court — which in this case is squarely asked to reconsider Austin’s validity for the first time — to address the matter with a greater willingness to consider new approaches capable of restoring our doctrine to sounder footing.
Third, the Austin decision is uniquely destabilizing because it threatens to subvert our Court’s decisions even outside the particular context of corporate express advocacy. *381The First Amendment theory underlying Austin’s holding is extraordinarily broad. Austin’s logic would authorize government prohibition of political speech by a category of speakers in the name of equality — a point that most scholars acknowledge (and many celebrate), but that the dissent denies. Compare, e. g., Garrett, New Voices in Polities: Justice Marshall’s Jurisprudence on Law and Politics, 52 How. L. J. 655, 669 (2009) (Austin “has been understood by most commentators to be an opinion driven by equality considerations, albeit disguised in the language of ‘political corruption’ ”), with post, at 464 (Austin’s rationale “is manifestly not just an ‘equalizing’ ideal in disguise”).2
It should not be surprising, then, that Members of the Court have relied on Austin’s expansive logic to justify greater incursions on the First Amendment, even outside the original context of corporate advocacy on behalf of candidates running for office. See, e. g., Davis v. Federal Election Comm’n, 554 U. S. 724, 756 (2008) (Stevens, J., concurring in part and dissenting in part) (relying on Austin and other cases to justify restrictions on campaign spending by individual candidates, explaining that “there is no reason that their logic — specifically, their concerns about the corrosive and distorting effects of wealth on our political process — is not equally applicable in the context of individual wealth”); McConnell, supra, at 203-209 (extending Austin beyond its original context to cover not only the “functional equivalent” of express advocacy by corporations, but also *382electioneering speech conducted by labor unions). The dissent in this case succumbs to the same temptation, suggesting that Austin justifies prohibiting corporate speech because such speech might unduly influence “the market for legislation.” Post, at 471. The dissent reads Austin to permit restrictions on corporate speech based on nothing more than the fact that the corporate form may help individuals coordinate and present their views more effectively. Post, at 471-472. A speaker’s ability to persuade, however, provides no basis for government regulation of free and open public debate on what the laws should be.
If taken seriously, Austin’s logic would apply most directly to newspapers and other media corporations. They have a more profound impact on public discourse than most other speakers. These corporate entities are, for the time being, not subject to §441b’s otherwise generally applicable prohibitions on corporate political speech. But this is simply a matter of legislative grace. The fact that the law currently grants a favored position to media corporations is no reason to overlook the danger inherent in accepting a theory that would allow government restrictions on their political speech. See generally McConnell, supra, at 283-286 (Thomas, J., concurring in part, concurring in judgment in part, and dissenting in part).
These readings of Austin do no more than carry that decision’s reasoning to its logical endpoint. In doing so, they highlight the threat Austin poses to First Amendment rights generally, even outside its specific factual context of corporate express advocacy. Because Austin is so difficult to confine to its facts — and because its logic threatens to undermine our First Amendment jurisprudence and the nature of public discourse more broadly — the costs of giving it stare decisis effect are unusually high.
Finally and most importantly, the Government’s own effort to defend Austin — or, more accurately, to defend something that is not quite Austin — underscores its weakness as *383a precedent of the Court. The Government concedes that Austin “is not the most lucid opinion,” yet asks us to reaffirm its holding. Tr. of Oral Arg. 62 (Sept. 9, 2009). But while invoking stare decisis to support this position, the Government never once even mentions the compelling interest that Austin relied upon in the first place: the need to diminish “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.” 494 U. S., at 660.
Instead of endorsing Austin on its own terms, the Government urges us to reaffirm Austin’s specific holding on the basis of two new and potentially expansive interests — the need to prevent actual or apparent quid pro quo corruption, and the need to protect corporate shareholders. See Supp. Brief for Appellee 8-10, 12-13. Those interests may or may not support the result in Austin, but they were plainly not part of the reasoning on which Austin relied.
To its credit, the Government forthrightly concedes that Austin did not embrace either of the new rationales it now urges upon us. See, e. g., Supp. Brief for Appellee 11 (“The Court did not decide in Austin .. . whether the compelling interest in preventing actual or apparent corruption provides a constitutionally sufficient justification for prohibiting the use of corporate treasury funds for independent electioneering”); Tr. of Oral Arg. 45 (Sept. 9, 2009) (“Austin did not articulate what we believe to be the strongest compelling interest”); id., at 61 (“[The Court:] I take it we have never accepted your shareholder protection interest. This is a new argument. [The Government:] I think that that’s fair”); id., at 64 (“[The Court:] In other words, you are asking us to uphold Austin on the basis of two arguments, two principles, two compelling interests we have never accepted, in [the context of limits on political expenditures]. [The Government:] [I]n this particular context, fair enough”).
*384To be clear: The Court in Austin nowhere relied upon the only arguments the Government now raises to support that decision. In fact, the only opinion in Austin endorsing the Government’s argument based on the threat of quid pro quo corruption was Justice Stevens’s concurrence. 494 U. S., at 678. The Court itself did not do so, despite the fact that the concurrence highlighted the argument. Moreover, the Court’s only discussion of shareholder protection in Austin appeared in a section of the opinion that sought merely to distinguish Austin’s facts from those of Federal Election Comm’n v. Massachusetts Citizens for Life, Inc., 479 U. S. 238 (1986). Austin, supra, at 663. Nowhere did Austin suggest that the goal of protecting shareholders is itself a compelling interest authorizing restrictions on First Amendment rights.
To the extent that the Government’s case for reaffirming Austin depends on radically reconceptualizing its reasoning, that argument is at odds with itself. Stare decisis is a doctrine of preservation, not transformation. It counsels deference to past mistakes, but provides no justification for making new ones. There is therefore no basis for the Court to give precedential sway to reasoning that it has never accepted, simply because that reasoning happens to support a conclusion reached on different grounds that have since been abandoned or discredited.
Doing so would undermine the rule-of-law values that justify stare decisis in the first place. It would effectively license the Court to invent and adopt new principles of constitutional law solely for the purpose of rationalizing its past errors, without a proper analysis of whether those principles have merit on their own. This approach would allow the Court’s past missteps to spawn future mistakes, undercutting the very rule-of-law values that stare decisis is designed to protect.
None of this is to say that the Government is barred from making new arguments to support the outcome in Austin. *385On the contrary, it is free to do so. And of course the Court is free to accept them. But the Government’s new arguments must stand or fall on their own; they are not entitled to receive the special deference we accord to precedent. They are, as grounds to support Austin, literally imprecedented. Moreover, to the extent the Government relies on new arguments — and declines to defend Austin on its own terms — we may reasonably infer that it lacks confidence in that decision’s original justification.
Because continued adherence to Austin threatens to subvert the “principled and intelligible” development of our First Amendment jurisprudence, Vasquez, 474 U. S., at 265, I support the Court’s determination to overrule that decision.
* * *
We have had two rounds of briefing in this case, two oral arguments, and 54 amicus briefs to help us carry out our obligation to decide the necessary constitutional questions according to law. We have also had the benefit of a comprehensive dissent that has helped ensure that the Court has considered all the relevant issues. This careful consideration convinces me that Congress violates the First Amendment when it decrees that some speakers may not engage in political speech at election time, when it matters most.
The dissent suggests that I am “much too quick” to reach this conclusion because I “ignore” Citizens United’s narrower arguments. Post, at 405, n. 12. But in fact I do not ignore those arguments; on the contrary, I (and my colleagues in the majority) appropriately consider and reject them on their merits, before addressing Citizens United’s broader claims. Supra, at 373-375; ante, at 322-329.
See also, e. g., R. Hasen, The Supreme Court and Election Law: Judging Equality from Baker v. Carr to Bush v. Gore 114 (2003) (“Austin represents the first and only case [before McConnell] in which a majority of the Court accepted, in deed if not in word, the equality rationale as a permissible state interest”); Strauss, Corruption, Equality, and Campaign Finance Reform, 94 Colum. L. Rev. 1369, and n. 1 (1994) (noting that Austin’s rationale was based on equalizing political speech); Ashdown, Controlling Campaign Spending and the “New Corruption”: Waiting for the Court, 44 Vand. L. Rev. 767, 781 (1991); Eule, Promoting Speaker Diversity: Austin and Metro Broadcasting, 1990 S. Ct. Rev. 105, 108-111.