with whom Justice White and Justice Blackmun join, dissenting.
In 14th-century London the bakers’ guild regulated the economics of the craft and the quality of its product. In the year 1316, it was adjudged that one Richard de Lughte-burghe “should have the punishment of the hurdle” because he sold certain loaves of bread in London; the bread had been baked in Suthwerke, rather than London, and the loaves were not of “the proper weight.”1 Thus Richard had vio*583lated a guild restriction designed to protect the economic interests of the local bakers2 as well as a restriction designed to protect the public from the purchase of inferior products.
For centuries the common law of restraint of trade has been concerned with restrictions on entry into particular professions and occupations. As the case of the Suthwerke baker illustrates, the restrictions imposed by medieval English guilds served two important but quite different purposes. The guilds limited the number of persons who might engage in a particular craft in order to be sure that there was enough work available to enable guild members to earn an adequate livelihood.3 They also protected the public by ensuring that apprentices, journeymen, and master craftsmen would have the skills that were required for their work. In numerous occupations today, licensing requirements4 may serve *584either or both of the broad purposes of the medieval guild restrictions.
The risk that private regulation of market entry, prices, or output may be designed to confer monopoly profits on members of an industry at the expense of the consuming public has been the central concern of both the development of the common law of restraint of trade and our antitrust jurisprudence. At the same time, the risk that the free market may not adequately protect the public from purveyors of inferior goods and services has provided a legitimate justification for the public regulation of entry into a wide variety of occupations. Private regulation is generally proscribed by the antitrust laws; public regulation is generally consistent with antitrust policy. A potential conflict arises, however, whenever government delegates licensing power to private parties whose economic interests may be served by limiting the number of competitors who may engage in a particular trade. In fact private parties have used licensing to advance their own interests in restraining competition at the expense of the public interest. See generally Gellhorn, The Abuse of Occupational Licensing, 44 U. Chi. L. Rev. 6 (1976).
The potential conflict with the antitrust laws may be avoided in either of two ways. The State may itself formulate the governing standards and administer the procedures *585that determine whether or not particular applicants are qualified. When the State itself governs entry into a profession, the evils associated with giving power over a market to those who stand to benefit from inhibiting entry into that market are absent. For that reason, state action of that kind, even if it is specifically designed to control output and to regulate prices, does not violate the antitrust laws. Parker v. Brown, 317 U. S. 341 (1943). Alternatively, the State may delegate to private parties the authority to formulate the standards and to determine the qualifications of particular applicants. When that authority is delegated to those with a stake in the competitive conditions within the market, there is a risk that public power will be exercised for private benefit. To minimize that risk, state policies displacing competition must be “clearly and affirmatively expressed” and must be appropriately supervised. See Community Communications Co. v. Boulder, 455 U. S. 40 (1982); California Retail Liquor Dealers Assn. v. Midcal Aluminum, Inc., 445 U. S. 97, 103-106 (1980).
In this case respondent has been unable to obtain a license to practice law in Arizona. He alleges that this is not because of any doubts about his competence as a lawyer, but because petitioners have engaged in an anticompetitive conspiracy in which they have used the Arizona bar examination to artificially limit the number of persons permitted to practice law in that State. Petitioners claim that the alleged conspiracy is not actionable under § 1 of the Sherman Act, 15 U. S. C. § 1, because it represents the decision of the State. But petitioners do not identify any state body that has decided that it is in the public interest to limit entry of even fully qualified persons into the Arizona Bar. Indeed, the conspiracy that is alleged is not the product of any regulatory scheme at all; there is no evidence that any criterion except competence has been adopted by Arizona as the basis for granting licenses to practice law. The conspiracy respondent has alleged is private; market participants are allegedly *586attempting to protect their competitive position through a misuse of their powers. Yet the Court holds that this conspiracy is cloaked in the State’s immunity from the antitrust laws. In my judgment, the competitive ideal of the Sherman Act may not be so easily escaped.
I
Petitioners are members of the Arizona Supreme Court’s Committee on Examinations and Admissions. The Arizona Supreme Court established the Committee to recommend applicants for admission to the Arizona Bar; it consists of seven members of the State Bar selected from a list of nominees supplied by the Arizona State Bar Association’s Board of Governors.5 Petitioners administered the 1974 bar examination which respondent took and failed. In his complaint, respondent alleged that after the scores of each candidate were known, petitioners selected a particular score which would equal the passing grade. The complaint alleges that the petitioners would adjust the grading formula in order to limit the number of persons who could enter the market and compete with members of the Arizona Bar. In this manner, respondent was “artificially prevented from entering into competition as an attorney in the State of Arizona.”6
The Arizona Supreme Court has instructed petitioners to recommend for admission to the Bar “[a]ll applicants who receive a passing grade in the general examination and who are found to be otherwise qualified . . . .”7 There is no indication that any criterion other than competence is appropriate under the Supreme Court’s Rules for regulating admission to the Bar.8 Indeed with respect to respondent’s application *587for admission, the Arizona Supreme Court wrote: “The practice of law is not a privilege but a right, conditioned solely upon the requirement that a person have the necessary mental, physical and moral qualifications.” Application of Ronwin, 113 Ariz. 357, 358, 555 P. 2d 315, 316 (1976), cert. denied, 430 U. S. 907 (1977). In short, one looks in vain in Arizona law, petitioners’ briefs, or the pronouncements of the Arizona Supreme Court for an articulation of any policy beside that of admitting only competent attorneys to practice in Arizona.
Thus, respondent does not challenge any state policy. He contests neither the decision to license those who wish to practice law, nor the decision to require a certain level of competence, as measured in a bar examination, as a precondition to licensing. Instead, he challenges an alleged decision to exclude even competent attorneys from practice in Arizona in order to protect the interests of the Arizona Bar.
As we have often reiterated in cases that involve the sufficiency of a pleading, a federal court may not dismiss a complaint for failure to state a claim unless it appears beyond doubt, even when the complaint is liberally construed, that the plaintiff can prove no set of facts which would entitle him to relief.9 The allegations of the complaint must be taken as true for purposes of a decision on the pleadings.10
A judge reading a complaint of this kind is understandably somewhat skeptical. It seems highly improbable that members of the profession entrusted by the State Supreme Court *588with a public obligation to administer an examination system that will measure applicants’ competence would betray that trust, and secretly subvert that system to serve their private ends. Nevertheless, the probability that respondent will not prevail at trial is no justification for dismissing the complaint. “Indeed it may appear on the face of the pleadings that a recovery is very remote and unlikely but that is not the test.” Scheuer v. Rhodes, 416 U. S. 232, 236 (1974). The Court does not purport to justify dismissal of this complaint by reference to the low probability that respondent will prevail at trial. Instead, it substantially broadens the doctrine of antitrust immunity, using an elephant gun to kill a flea.
II
If respondent were challenging a restraint of trade imposed by the sovereign itself, this case would be governed by Parker v. Brown, 317 U. S. 341 (1943), which held that the Sherman Act does not apply to the sovereign acts of States. See id., at 350-352. As the Court points out, the Arizona Supreme Court exercises sovereign power with respect to admission to the Arizona Bar; hence if the challenged conduct were that of the court, it would be immune under Parker. Ante, at 567-569.11 The majority’s conclusion that the challenged action was that of the Arizona Supreme Court is, however, plainly wrong. Respondent alleged that the decision to place an artificial limit on the number of lawyers was made by petitioners — not by the State Supreme Court. There is no contention that petitioners made that decision at the direction or behest of the Supreme Court. That court is not a petitioner, nor was it named as a defendant in respondent’s complaint. Nor, unlike the Court, have petitioners suggested that the Arizona Supreme Court played any part in establishing the grading standards for the bar examination *589or made any independent decision to admit or reject any individual applicant for admission to the Bar.12 Because respondent is not challenging the conduct of the Arizona Supreme Court, Parker is simply inapplicable.
Any possible claim that the challenged conduct is that of the State Supreme Court is squarely foreclosed by Goldfarb v. Virginia State Bar, 421 U. S. 773 (1975). There an antitrust action was brought challenging minimum-fee schedules published by a county bar association and enforced by the State Bar pursuant to its mandate from the Virginia Supreme Court to regulate the practice of law in that State. After acknowledging that the State Bar was a state agency which had enforced the schedules pursuant to the authority granted it by the State Supreme Court, we stated a simple test for antitrust immunity:
“The threshold inquiry in determining if an anticompet-itive activity is state action of the type the Sherman Act was not meant to proscribe is whether the activity is required by the State acting as sovereign. Here we need not inquire further into the state-action question because it cannot fairly be said that the State of Virginia through its Supreme Court Rules required the anticompet-itive activities of either respondent. Respondents have pointed to no Virginia statute requiring their activities; state law simply does not refer to fees, leaving regulation of the profession to the Virginia Supreme Court; although the Supreme Court’s ethical codes mention advisory fee schedules they do not direct either respondent *590to supply them, or require the type of price floor which arose from respondents’ activities.” Id., at 790 (emphasis supplied) (citations omitted).
In Bates v. State Bar of Arizona, 433 U. S. 350 (1977), the Court applied the Goldfarb test to a disciplinary rule restricting advertising by Arizona attorneys that the Supreme Court itself “has imposed and enforces,” 433 U. S., at 353:
“In the instant case ... the challenged restraint is the affirmative command of the Arizona Supreme Court under its Rules 27(a) and 29(a) and its Disciplinary Rule 2-101(B). That court is the ultimate body wielding the State’s power over the practice of law, see Ariz. Const., Art. 3; In re Bailey, 30 Ariz. 407, 248 P. 29 (1926), and, thus, the restraint is ‘compelled by direction of the State acting as a sovereign.’ 421 U. S., at 791 (footnote omitted).” Id., at 359-360.
The test stated in Goldfarb and Bates is that the sovereign must require the restraint. Indeed, that test is derived from Parker itself: “We find nothing in the language of the Sherman Act or in its history which suggests that its purpose was to restrain a state or its officers or agents from activities directed by its legislature [or supreme court].” 317 U. S., at 350-351 (emphasis supplied). Here, the sovereign is the State Supreme Court, not petitioners, and the court did not require petitioners to grade the bar examination as they did.13 The fact that petitioners are part of a state agency under the direction of the sovereign is insufficient to cloak them in the sovereign’s immunity; that much was also decided in Goldfarb:
*591“The fact that the State Bar is a state agency for some limited purposes does not create an antitrust shield that allows it to foster anticompetitive practices for the benefit of its members. The State Bar, by providing that deviation from County Bar minimum fees may lead to disciplinary action, has voluntarily joined in what is essentially a private anticompetitive activity, and in that posture cannot claim it is beyond the reach of the Sherman Act.” 421 U. S., at 791-792 (footnotes and citation omitted).
“Goldfarb therefore made it clear that, for purposes of the Parker doctrine, not every act of a state agency is that of the State as sovereign.” Lafayette v. Louisiana Power & Light Co., 435 U. S. 389, 410 (1978) (plurality opinion). Rather, “anticompetitive actions of a state instrumentality not compelled by the State acting as sovereign are not immune from the antitrust laws.” Id., at 411, n. 41. See also id., at 425 (opinion of Burger, C. J.); Cantor v. Detroit Edison Co., 428 U. S. 579, 604 (1976) (opinion of Burger, C. J.). An antitrust attack falls under Parker only when it challenges a decision of the sovereign and not the decision of the state bar which indisputably is not the sovereign. See California Retail Liquor Dealers Assn. v. Midcal Aluminum, Inc., 445 U. S. 97, 104-105 (1980).14 Here no decision of the sovereign, the Arizona Supreme Court, is attacked;15 only a *592conspiracy of petitioners which was neither compelled nor directed by the sovereign is at stake. Since there is no claim that the court directed petitioners to artificially reduce the number of lawyers in Arizona, petitioners cannot utilize the sovereign’s antitrust immunity.16
The majority’s confused analysis is illustrated by its difficulty in identifying the sovereign conduct which it thinks is at issue here. To support its conclusion that the challenged action is that of the Arizona Supreme Court, the majority suggests that what respondent challenges is the court’s decision to deny respondent’s application for admission to the Bar. Ante, at 577-578, n. 30. I find nothing in the record to indicate that the court ever made such a decision. Respondent’s complaint alleges only that petitioners “announced the results” of the bar examination. App. 9. In their answer, petitioners admitted this and added nothing else of significance. Id., at 17. The Rules of the Supreme Court do not call for the court to deny the application of a person who has failed the bar examination; rather they state only that any “applicant aggrieved by any decision of the Committee . . . may within 20 days after such occurrence file a verified pe*593tition with this Court for a review.” Ariz. Sup. Ct. Rule 28(c) XII. Yet the Court disavows reliance on the Supreme Court’s denial of Ronwin’s petition, ante, at 577-578, n. 30,17 and with good reason, see n. 15, supra.18 Thus, if the Supreme Court did not itself deny Ronwin’s application, if its denial of Ronwin’s petition for review is irrelevant, and if the only criterion it ever required petitioners to employ was competence, it is difficult to see why petitioners should have immunity from the requirements of federal law if, as alleged, they took the initiative in employing a criterion other than competence. “It is not enough that. . . anticompetitive conduct is ‘prompted’ by state action; rather, anticompetitive activities must be compelled by direction of the State acting as a sovereign.” Goldfarb, 421 U. S., at 791.
I — 1 HH
It is, of course, true that the Arizona Supreme Court delegated to petitioners the task of administering the bar exam, and retained the authority to review or revise any action taken by petitioners. However, neither of these fac*594tors is sufficient to accord petitioners immunity under the Sherman Act.
In Bates, the Court held that the State Bar’s restrictions on attorney advertising qualified for antitrust immunity, 433 U. S., at 359-362, because “the state policy requiring the anticompetitive restraint as part of a comprehensive regulatory system, was one clearly articulated and affirmatively expressed as state policy, and that the State’s policy was actively supervised by the State Supreme Court as the policymaker.” Lafayette, 435 U. S., at 410 (plurality opinion) (footnote omitted). This Court has since “adopted the principle, expressed in the plurality opinion in Lafayette, that anticompetitive restraints engaged in by state municipalities or subdivisions must be ‘clearly articulated and affirmatively expressed as state policy’ in order to gain an antitrust exemption.” Community Communications Co. v. Boulder, 455 U. S., at 51, n. 14 (quoting Midcal, 445 U. S., at 105).19
Here there is nothing approaching a clearly articulated and affirmatively expressed state policy favoring an artificial limit on the number of lawyers licensed to practice in Arizona. Indeed, the majority does not attempt to argue that petitioners satisfy this test. The only articulated policy to be found in Arizona law is that competent lawyers should be admitted to practice; indeed this is the only policy petitioners articulate in this Court. An agreement of the type alleged in respondent’s complaint is entirely unrelated to any “clearly articulated and affirmatively expressed” policy of Arizona. While the Arizona Supreme Court may have permitted petitioners to grade and score respondent’s bar examination as they did, Parker itself indicates that “a state does not give immunity to those who violate the Sherman Act by authorizing them to violate it, or by declaring that their action is lawful.. . .” 317 U. S., at 351. The Arizona Supreme Court *595may permit the challenged restraint, but it has hardly required it as a consequence of some affirmatively expressed and clearly articulated policy. What we said of a state home-rule provision that permitted but did not require municipalities to adopt a challenged restraint on competition applies fully here:
“[Plainly the requirement of ‘clear articulation and affirmative expression’ is not satisfied when the State’s position is one of mere neutrality respecting the municipal actions challenged as anticompetitive. A State that allows its municipalities to do as they please can hardly be said to have ‘contemplated’ the specific anticompet-itive actions for which municipal liability is sought. . . . Acceptance of such a proposition — that the general grant of power to enact ordinances necessarily implies state authorization to enact specific anticompetitive ordinances — would wholly eviscerate the concepts of ‘clear articulation and affirmative expression’ that our precedents require.” Boulder, 455 U. S., at 55-56 (emphasis in original).
Unless the Arizona Supreme Court affirmatively directed petitioners to restrain competition by limiting the number of otherwise qualified lawyers admitted to practice in Arizona, it simply cannot be said that its position is anything more than one of neutrality; mere authorization for anti-competitive conduct is wholly insufficient to satisfy the test for antitrust immunity. See Midcal, 445 U. S., at 105-106; Lafayette, 435 U. S., at 414-415 (plurality opinion).20 No *596affirmative decision of the Arizona Supreme Court to restrain competition by limiting the number of qualified persons admitted to the Bar is disclosed on the present record. The alleged conspiracy to introduce a factor other than competence into the bar examination process is not the product of a clearly articulated and affirmatively expressed state policy and hence does not qualify for antitrust immunity.21
IV
The conclusion that enough has been alleged in the complaint to survive a motion to dismiss does not warrant the further conclusion that the respondent is likely to prevail at *597trial, or even that his case is likely to survive a motion for summary judgment. For it is perfectly clear that the admissions policy that is described in the Arizona Supreme Court’s Rules does not offend the Sherman Act. Any examination procedure will place a significant barrier to entry into the profession; moreover, a significant measure of discretion must be employed in the administration of testing procedures. Yet ensuring that only the competent are licensed to serve the public is entirely consistent with the Sherman Act. See Goldfarb, 421 U. S., at 792-793.22
The Court is concerned about the danger that because thousands of aspirants fail to pass bar examinations every year, “affirmance of the Court of Appeals in this case could well invite numerous suits” questioning bar examiners’ motives; the Court fears that the burdens of discovery and trial and “the threat of treble damages” will deter “ ‘able citizens’ from performing this essential public service.” Ante, at 580-581, n. 34. The Court is, I submit, unduly alarmed.23 A *598denial of antitrust immunity in this case would not necessarily pose any realistic threat of liability, or even of prolonged litigation. Respondent must first produce sufficient evidence that petitioners have indeed abused their public trust to survive summary judgment, a task that no doubt will prove formidable.24 Moreover, petitioners’ motives will not necessarily be relevant to respondent’s case. If the proof demonstrates that petitioners have adopted a reasonable means for regulating admission to the Arizona Bar on the basis of competence, respondent will be unable to show the requisite adverse effect on competition even if the subjective motivation of one or more bar examiners was tainted by sinister self-interest. Indeed, even if respondent can show that he was “arbitrarily” denied admission to the Bar for reasons unrelated to his qualifications, unless he can also show that this occurred as part of an anticompetitive scheme, his antitrust claim will fail.
In any event, there is true irony in the Court’s reliance on these concerns. In essence, the Court is suggesting that a special protective shield should be provided to lawyers because they — unlike bakers, engineers, or the members of any other craft — may not have sufficient confidence in the ability of our legal system to identify and reject unmeritorious claims to be willing to assume the ordinary risks of litigation associated with the performance of civic responsibilities. I do not share the Court’s fear that the administration of bar *599examinations by court-appointed lawyers cannot survive the scrutiny associated with rather ordinary litigation that persons in most other walks of life are expected to endure.
The Court also no doubt believes that lawyers — or at least those leaders of the bar who are asked to serve as bar examiners — will always be faithful to their fiduciary responsibilities. Though I would agree that the presumption is indeed a strong one, nothing in the sweeping language of the Sherman Act justifies carving out rules for lawyers inapplicable to any other profession. In Goldfarb we specifically rejected such parochialism. Indeed, the argument that it is unwise or unnecessary to require the petitioners to comply with the Sherman Act “is simply an attack upon the wisdom of the longstanding congressional commitment to the policy of free markets and open competition embodied in the antitrust laws.” Boulder, 455 U. S., at 56. We should not ignore that commitment today.
Denial of antitrust immunity in this case would hardly leave the State helpless to cope with felt exigencies; should it wish to do so, the Arizona Supreme Court remains free to give petitioners an affirmative direction to engage in the precise conduct that respondent has alleged. The antitrust laws hardly create any inescapable burdens for the State; they simply require that decisions to displace the free market be made overtly by public officials subject to public accountability, rather than secretly in the course of a conspiracy involving representatives of a private guild accountable to the public indirectly if at all. See id., at 56-57; Lafayette, 435 U. S., at 416-417 (plurality opinion). “The national policy in favor of competition cannot be thwarted by casting such a gauzy cloak of state involvement over what is essentially a private price-fixing arrangement.” Midcal, 445 U. S., at 106.
The practical concerns identified by the Court pale when compared with the principle that should govern the decision *600of this case. The rule of law that applies to this case is applicable to countless areas of the economy in which arbitrary restraints on entry may impose the very costs on the consuming public which the antitrust laws were designed to avoid.25 Experience in the administration of the Sherman Act has demonstrated that there is a real risk that private associations that purport merely to regulate professional standards may in fact use their powers to restrain competition which threatens their members.26 It is little short of irresponsible to tear a gaping hole in the fabric of antitrust law simply because we may be confident that respondent will be unable to prove what he alleges.
*601Frivolous cases should be treated as exactly that, and not as occasions for fundamental shifts in legal doctrine.27 Our legal system has developed procedures for speedily disposing of unfounded claims; if they are inadequate to protect petitioners from vexatious litigation, then there is something wrong with those procedures, not with the law of antitrust immunity. That body of law simply does not permit the Sherman Act to be displaced when neither the state legislature nor the state supreme court has expressed any desire to preclude application of the antitrust laws to the conduct of those who stand to benefit from restraints of trade. A healthy respect for state regulatory policy does not require immunizing those who abuse their public trust; such a thin veneer of state involvement is insufficient justification for casting aside the competitive ideal of the Sherman Act. The commitment to free markets and open competition that has evolved over the centuries and is embodied in the Sherman Act should be sturdy enough to withstand petitioners’ flimsy claim. That claim might have merited the support of the 14th-century guilds; today it should be accorded the “punishment of the hurdle.”
I respectfully dissent.
H. Riley, Memorials of London and London Life in the XIIIth, XIVth, and XVth Centuries 119-120 (1868). The punishment is described in a footnote as “[b]eing drawn on a hurdle through the principal streets of the City.” Id., at 119, n. 5.
“The principal reason for the existence of the gild was to preserve to its own members the monopoly of trade. No one not in the gild merchant of the town could buy or sell there except under conditions imposed by the gild. Foreigners coming from other countries or traders from other English towns were prohibited from buying or selling in any way that might interfere with the interest of the gildsmen. They must buy and sell at such times and in such places and only such articles as were provided by the gild regulations.” E. Cheyney, An Introduction to the Industrial and Social History of England 52-53 (1920).
“The craft gilds existed usually under the authority of the town government, though frequently they obtained authorization or even a charter from the crown. They were formed primarily to regulate and preserve the monopoly of their own occupations in their own town, just as the gild merchant existed to regulate the trade of the town in general. No one could carry on any trade without being subject to the organization which controlled that trade.” Id., at 55.
Professor Handler has pointed out:
“Entry into various fields of endeavor is guarded by numerous licensing restrictions. Licenses are demanded of physicians and surgeons, dentists, optometrists, pharmacists and druggists, nurses, midwives, chiropodists, veterinarians, certified public accountants, lawyers, architects, engineers and surveyors, shorthand reporters, master plumbers, undertakers and embalmers, real estate brokers, junk dealers, pawnbrokers, ticket agents, liquor dealers, private detectives, auctioneers, milk dealers, peddlers, *584master pilots and steamship engineers, weighmasters, forest guides, motion picture operators, itinerant retailers on boats, employment agencies, commission merchants of farm produce, and manufacturers of frozen desserts, concentrated feeds, and commercial fertilizers. No factory, cannery, place of public assembly, laundry, cold storage warehouse, shooting gallery, bowling alley and billiard parlor, or place of storage of explosives can be operated nor can industrial house work be carried on without registration or license. Licenses are also required for the sale of minnows, use of fishing nets, and the operation of educational institutions, correspondence schools, filling stations and motor vehicles. Motion pictures cannot be exhibited unless licensed, and canal boats must be registered.” M. Handler, Cases and Other Materials on Trade Regulation 3-4 (1937) (footnotes omitted).
Ariz. Sup. Ct. Rule 28(a).
See App. 10-11.
Ariz. Sup. Ct. Rule 28(c) VIII.
Petitioners certainly do not suggest the existence of any other criterion under Arizona law. To the contrary, at oral argument they expressly acknowledged that there is no state policy adopting any criterion but competence for admission to the Bar. Tr. of Oral Arg. 22-24.
See McClain v. Real Estate Bd. of New Orleans, 444 U. S. 232, 246-247 (1980); Lake Country Estates v. Tahoe Regional Planning Agency, 440 U. S. 391, 397, n. 11 (1979); Hospital Building Co. v. Trustees of Rex Hospital, 425 U. S. 738, 746 (1976); Scheuer v. Rhodes, 416 U. S. 232, 236 (1974); Conley v. Gibson, 355 U. S. 41, 45-46 (1957).
See Hughes v. Rowe, 449 U. S. 5, 10 (1980) (per curiam); Cruz v. Beto, 405 U. S. 319, 322 (1972) (per curiam); California Motor Transport Co. v. Trucking Unlimited, 404 U. S. 508, 515-516 (1972); Jenkins v. McKeithen, 395 U. S. 411, 421 (1969) (plurality opinion); Walker Process Equipment, Inc. v. Food Machinery & Chemical Corp., 382 U. S. 172, 174-175 (1965).
See Bates v. State Bar of Arizona, 433 U. S. 350, 359-360 (1977).
It should be noted that petitioners do not advance the imaginative argument on which this Court’s decision rests — that the examination procedure is merely advisory and that the Arizona Supreme Court itself “made the final decision on each applicant.” Ante, at 578 (footnote omitted). Presumably petitioners are more familiar with how their own procedures work than is this Court. The Court shows precious little deference to “administrative expertise” in its analysis of the facts.
It is not surprising that petitioners (who must practice before the Arizona Supreme Court) did not advance the theory on which this Court relies — that their challenged conduct is actually conduct of the Arizona Supreme Court. They surely understand that they are not the court, but rather its subordinate.
See also New Motor Vehicle Board of California v. Orrin W. Fox Co., 489 U. S. 96, 109 (1978); Cantor v. Detroit Edison Co., 428 U. S., at 593-595.
In response to this dissent, the Court has advanced the theory that the relevant “state action” was the State Supreme Court’s rejection of an original complaint filed in that court containing a “plethora of charges, including the substance of the complaint in this case.” Ante, at 576-577. See also ante, at 582. Presumably, that complaint was simply deficient as a matter of state law; if the allegations of respondent’s current complaint are taken as true then the fact that respondent failed the bar examination would have provided an adequate ground for the dismissal of respondent’s complaint without any review of respondent’s allegations. Even if it were the case *592that the Arizona Supreme Court reviewed petitioner’s complaint on its merits, all that would indicate is that the court has declined to exercise its power of revision with respect to petitioners’ alleged anticompetitive policies. That is far different from having required petitioners to adopt those policies in the first place, which is what Goldfarb requires.
The Court argues that “[o]nly the Arizona Supreme Court had the authority to grant or deny admission to practice in the State,” ante, at 575-576 (footnote omitted), and therefore concludes that the challenged conduct is that of the court. But there is no allegation that the challenged policy was adopted by the court; at most the court has permitted it by accepting the recommendations of petitioners. Yet as Bates and Goldfarb make clear, the challenged policy must be required by the sovereign. The fact that the court retained the power to disapprove of the examination procedure adopted by petitioners is no different from the fact that the Virginia Supreme Court retained the power to disapprove of the fee schedules set by the bar association in Goldfarb. Similar powers of revision were held insufficient to justify immunity in Lafayette, Cantor, and Midcal.
While the majority’s disavowal in its note 30 is quite unequivocal, at other points in its opinion, see ante, at 576-577, and in its ultimate statement of its holding, see ante, at 582, it does seem to rely on the denial of respondent’s petition for review. If that truly is critical for the majority, then it would follow that an individual in respondent’s position who did not file a petition for review would be able to mount an antitrust challenge free from the immunity barrier the majority erects. If it indeed is that easy to escape the majority’s holding, then that holding will not protect bar examiners against the parade of horribles discussed by the majority ante, at 580, and n. 34.
The cases the Court cites ante, at 577, n. 30, 581, all involve instances in which an applicant who had passed the bar examination was nevertheless not recommended for admission. If the applicant seeks judicial review, those eases indicate that the court will decide for itself whether to admit the applicant. However, none of those cases indicates that the court makes an independent decision, or indeed any decision at all, to deny the application of a person who has failed the bar examination.
See also 455 U. S., at 51-52, 54; Midcal, 445 U. S., at 104-105; New Motor Vehicle Board of California v. Orrin W. Fox Co., 439 U. S., at 109.
See also Cantor v. Detroit Edison Co., 428 U. S., at 604-605 (opinion of BURGER, C. J.). In Cantor, the Court wrote:
“Respondent could not maintain the lamp-exchange program without the approval of the Commission, and now may not abandon it without such approval. Nevertheless, there can be no doubt that the option to have, or not to have, such a program is primarily respondent’s, not the Commission’s. Indeed, respondent initiated the program years before the regulatory agency was even created. There is nothing unjust in a conclusion *596that respondent’s participation in the decision is sufficiently significant to require that its conduct implementing the decision, like comparable conduct by unregulated businesses, conform to applicable federal law. Accordingly, even though there may be cases in which the State’s participation in a decision is so dominant that it would be unfair to hold a private party responsible for his conduct in implementing it, this record discloses no such unfairness.” Id., at 594-595 (footnotes omitted).
In this Court petitioners appear to have abandoned the argument, advanced for the first time in a petition for rehearing in the Court of Appeals, that the examination grading formula was actually approved by the State Supreme Court. Because the majority appears to revive this abandoned contention, ante, at 572-573, and n. 22, see also ante, at 576, it is necessary to address it, though that requires no more than brief reference to the Court of Appeals’ opinion:
“Defendants contend for the first time on rehearing that the Committee’s grading formula ‘was submitted to the Court, reviewed by the Court, and accepted by the Court.’ In response, Ronwin has tendered to this court what purports to be the letter the Committee filed with the Supreme Court on February 8,1974 pursuant to Rule 28(c) (VII)(B). If, as Ronwin alleges, the Committee scored the examination to admit a pre-determined number of applicants, the letter does not so advise the court. Accordingly, if the letter presented to us constitutes the submission to the Supreme Court, it cannot be the basis for a clearly articulated and affirmatively expressed state policy. Although dismissal might have been proper if the facts were as defendants now argue for the first time on rehearing, those facts were never brought to the district court’s attention. Dismissal *597was therefore improper on the basis of the information before the district court.” Ronwin v. State Bar of Arizona, 686 F. 2d 692, 697 (GA9 1981). It is, of course, equally improper for this Court to rely on evidence not presented to the District Court as a basis for holding that the complaint was not sufficient to withstand a motion to dismiss. See Adickes v. S. H. Kress & Co., 398 U. S. 144, 157-158, n. 16 (1970).
See generally Arizona v. Maricopa County Medical Society, 457 U. S. 332, 348-349 (1982); National Society of Professional Engineers v. United States, 435 U. S. 679, 696 (1978).
The majority makes the rather surprising suggestion that under the well-settled principles I have discussed, those who advise state legislatures on legislation which restrains competition could be sued under the Sherman Act. Ante, at 580. Such persons of course would have a complete defense since in such a case they would have been delegated no power which could be used to restrain competition and hence cannot be liable for a restraint they did not impose. Moreover, the Sherman Act protects the right to seek favorable legislation, even if the reason for doing so is to injure competitors. See California Motor Transport Co. v. Trucking Unlimited, 404 U. S. 508 (1972); Eastern Railroad Presidents Conference *598v. Noerr Motor Freight, Inc., 365 U. S. 127 (1961). The majority’s focus on cases not before the Court surely reflects the weakness of its position with respect to the case that is here.
In order to preserve the secrecy of bar examination questions, the test must vary from year to year; after a test has been given, it may become apparent that the anticipated passing grade should be adjusted in order to provide roughly the same measure of competence as was used in prior years. Thus respondent’s burden of proving the conspiracy he has alleged requires far more than evidence that petitioners exercised discretion in setting the passing grade after the results were known.
The conspiracy respondent has alleged, if proved, would have no procompetitive justification at all; it would be plainly inconsistent with the goals of the Sherman Act. Thus petitioners’ claim of antitrust immunity arises in the least defensible context:
“[A]s a general proposition . . . state-sanctioned anticompetitive activity must fall like any other if its potential harms outweigh its benefits. This does not mean that state-sanctioned and private activity are to be treated alike. The former is different because the fact of state sanction figures powerfully in the calculus of harm and benefit. If, for example, the justification for the scheme lies in the protection of health or safety, the strength of that justification is forcefully attested to by the existence of a state enactment. ... A particularly strong justification exists for a state-sanctioned scheme if the State in effect has substituted itself for the forces of competition, and regulates private activity to the same ends sought to be achieved by the Sherman Act. Thus, an anticompetitive scheme which the State institutes on the plausible ground that it will improve the performance of the market in fostering efficient resource allocation and low prices can scarcely be assailed.” Cantor v. Detroit Edison Co., 428 U. S., at 610-611 (Blackmun, J., concurring in judgment).
See, e. g., Arizona v. Maricopa County Medical Society, 457 U. S. 332 (1982); American Society of Mechanical Engineers, Inc. v. Hydrolevel Corp., 456 U. S. 556 (1982); National Society of Professional Engineers v. United States, 435 U. S. 679 (1978); Silver v. New York Stock Exchange, 373 U. S. 341 (1963); American Medical Assn. v. United States, 317 U. S. 519 (1943); Fashion Originators’ Guild of America, Inc. v. FTC, 312 U. S. 457, 465-466 (1941).
If, as seems likely, respondent’s claim proves insubstantial, it should be dealt with in the same manner as other such claims — by means of summary judgment, perhaps coupled with an award of attorneys’ fees should it also develop that this case was “unreasonably and vexatiously” brought. See 28 U. S. C. § 1927.