concurring in the result.
The issue in this case is whether the Securities and Exchange Commission may invoke the injunctive remedies of the Investment Advisers Act, 15 U. S. C. §§ 80b — 1 to 80b-21, to prevent an unregistered adviser from publishing newsletters containing investment advice that is not specifically tailored to the needs of individual clients. The Court holds that it may not because the activities of petitioner Lowe (hereafter petitioner) do not make him an investment adviser covered by the Act. For the reasons that follow, I disagree with this improvident construction of the statute. In my view, petitioner is an investment adviser subject to regulation and sanction under the Act. I concur in the judgment, however, because to prevent petitioner from publishing at all is inconsistent with the First Amendment.
*212I
A
I have no quarrel with the principle that constitutional adjudication is to be avoided where it is fairly possible to do so without negating the intent of Congress. Due respect for the Legislative Branch requires that we exercise our power to strike down its enactments sparingly. For this reason, “[w]hen the validity of an act of the Congress is drawn in question, and even if a serious doubt of constitutionality is raised, it is a cardinal principle that this Court will first ascertain whether a construction of the statute is fairly possible by which the question may be avoided.” Crowell v. Benson, 285 U. S. 22, 62 (1932).
But our duty to avoid constitutional questions through statutory construction is not unlimited: it is subject to the condition that the construction adopted be “fairly possible.” As Chief Justice Taft warned, “amendment may not be substituted for construction, and ... a court may not exercise legislative functions to save the law from conflict with constitutional limitation.” Yu Cong Eng v. Trinidad, 271 U. S. 500, 518 (1926). Justice Brandeis, whose concurring opinion in Ashwander v. TVA, 297 U. S. 288, 341-356 (1936), is frequently cited as the definitive statement of the rule of “constitutional avoidance,” himself cautioned: “The court may not, in order to avoid holding a statute unconstitutional, engraft upon it an exception or other provision. . . . Neither may it do so to avoid having to resolve a constitutional doubt.” Crowell v. Benson, supra, at 76-77 (dissenting opinion). Adoption of a particular construction to avoid a constitutional ruling, Justice Brandéis stated, was appropriate only “where a statute is equally susceptible of two constructions, under one of which it is clearly valid and under the other of which it may be unconstitutional.” 285 U. S., at 76. These limits on our power to avoid constitutional issues through statutory construction flow from the same principle as does the policy of constitutional avoidance itself: that is, *213the principle of deference to the legislature’s exercise of its assigned role in our constitutional system. See Rescue Army v. Municipal Court, 331 U. S. 549, 571 (1947). The task of defining the objectives of public policy and weighing the relative merits of alternative means of reaching those objectives belongs to the legislature. The courts should not lightly take it upon themselves to state that the path chosen by Congress is an impermissible one; but neither are the courts free to redraft statutory schemes in ways not anticipated by Congress solely to avoid constitutional difficulties. The latter course may at times be a more drastic imposition on legislative authority than the former. When the choice facing a court is between finding a particular application of a statute unconstitutional and adopting a construction of the statute that avoids the difficulty but at the same time materially deviates from the legislative plan and frustrates permissible applications, the choice of constitutional adjudication may well be preferable.
With these guidelines in mind, I turn to consideration of the proper construction of the statute at hand.
B
The Investment Advisers Act of 1940, 54 Stat. 847, as amended, 15 U. S. C. §80b-l et seq., provides that persons doing business as “investment advisers” must (with certain exceptions) register with the SEC. §80b-3(a). The Act sets forth substantive grounds for the denial or revocation of an investment adviser’s registration. § 80b-3(e). It is unlawful for an adviser who has not registered or whose registration has been revoked, suspended, or denied to practice his trade; if he does so, he may be subject to criminal penalties, §80b-17, or to injunction, §80b-9(e). In addition to penalizing those who would offer investment advice without registering, the Act contains provisions applicable to all investment advisers, whether registered or not. Most notable among these are prohibitions on certain contracts between *214advisers and their clients, see §80b-5, recordkeeping requirements, see § 80b-4, and provisions that make it unlawful for advisers to engage in “fraudulent, deceptive, or manipulative” conduct, see § 80b-6.
There is no question but that if petitioner’s publishing activities bring him -within the statutory definition of an “investment adviser,” the Act subjects him to injunction (and, presumably, criminal penalties) if he persists in engaging in those activities. Thus, if petitioner is an “investment adviser,” the constitutional questions raised by the application of the Act’s enforcement provisions to his conduct must be faced.
The starting point, then, must be the definition itself:
“ 'Investment adviser’ means any person who, for compensation, engages in the business of advising others, either directly or through publications or writings, as to the value of securities or as to the advisability of investing in, purchasing, or selling securities, or who, for compensation and as part of a regular business, issues or promulgates analyses or reports concerning securities; but does not include . . . (D) the publisher of any bona fide newspaper, news magazine or business or financial publication of general and regular circulation.” 15 U. S. C. §80b-2(a)(ll).
Although petitioner does not offer his subscribers investment advice specifically tailored to their individual needs and engages in no direct communications with them, he undeniably “engages in the business of advising others . . . through publications ... as to the value of securities” and “issues or promulgates analyses or reports concerning securities.” Thus, he falls outside the definition of an “investment adviser” only if each of his publications qualifies as a “bona fide newspaper, news magazine or business or financial publication of general and regular circulation.” The question is whether the “bona fide publications” exception is to be con*215strued so broadly as to exclude from the definition all persons whose advisory activities are carried out solely through publications offering impersonal investment advice to their subscribers.
It is hardly crystal clear from the face of the statute how the primary definition and the “bona fide publications” exception are supposed to mesh, but the SEC has, since the Act’s inception, interpreted the statutory definition of “investment adviser” to cover persons whose activities are limited to the publication of investment advisory newsletters or reports such as those published by petitioner. At the conclusion of the Act’s first year of operation, the Commission reported that of the approximately 750 persons and firms registering under the Act, “165 firms indicated that their investment advisory service consisted only of the sale of uniform publications.” Seventh Annual Report of the Securities and Exchange Commission, Fiscal Year Ended June 30, 1941, p. 35 (1942).1 Since that time, it appears that the Commission has consistently and routinely applied the Act to the publishers of newsletters offering investment advice. See, e. g., SEC v. Capital Gains Research Bureau, Inc., 375 U. S. 180 (1963); In re Todd, 40 S. E. C. 303 (1960); see also Lovitch, The Investment Advisers Act of 1940 — Who Is an “Investment Adviser”?, 24 Kan. L. Rev. 67 (1975).2 The SEC’s *216longstanding position that publishers of newsletters offering investment advice are investment advisers for purposes of the Act reflects a construction of the “bona fide publications” exception as “applicable only where, based on the content, advertising material, readership, and other relevant factors, a publication is not primarily a vehicle for distributing investment advice.” Applicability of Investment Advisers Act to Certain Publications, SEC Release No. IA-563, 42 Fed. Reg. 2953 (1977), codified at 17 CFR §276 (1984); cf. SEC v. Suter, 732 F. 2d 1294 (CA7 1984); SEC v. Wall Street Transcript Corp., 422 F. 2d 1371 (CA2), cert. denied, 398 U. S. 958 (1970).
An agency’s construction of legislation that it is charged with enforcing is entitled to substantial weight, particularly when the construction is contemporaneous with the enactment of the statute. See Skidmore v. Swift & Co., 323 U. S. 134, 140 (1944). In cases where the policy of constitutional avoidance must be considered, however, the administrative construction cannot be decisive. See United States v. Clark, 445 U. S. 23, 33, n. 10 (1980). We must, therefore, turn to other guides to the meaning of the statute to determine whether a reasonable construction of the statute is available by which petitioner can be excluded from the category of investment advisers and the constitutional issues thereby be avoided.
Any construction that expands the “bona fide publications” exception beyond the bounds set by the SEC, however, poses great difficulties. If the exception is expanded to include more than just publications that are not primarily vehicles for distributing investment advice, it is difficult to imagine any workable definition that does not sweep in all publications that are not personally tailored to individual clients. Indeed, it appears that this is precisely the definition the Court *217adopts.3 But such an expansive definition of the exception renders superfluous certain key passages in the primary definition of an “investment adviser”: one who engages in the business of rendering investment advice “either directly or *218through -publications or writings” or who “issues or promulgates analyses or reports concerning securities.” Had Congress intended the “bona fide publications” exception to encompass all publications, it is difficult to imagine why the primary definition of “investment adviser” should have spoken in the disjunctive of those who rendered advice directly and those who rendered it through publications, analyses, or reports. Nor is it clear why Congress would have chosen the adjective “bona fide” had it not intended that the SEC look beyond the form of a publication in determining whether it fell within the exception.4 The construction of the Act *219that would exclude petitioner from the category of investment advisers because he offers his advice through publications thus conflicts with the fundamental axiom of statutory interpretation that a statute is to be construed so as to give effect to all its language. Connecticut Dept. of Income Maintenance v. Heckler, 471 U. S. 524, 530, and n. 15 (1985); Reiter v. Sonotone Corp., 442 U. S. 330, 339 (1979).
Nothing in the legislative history of the statute supports a construction of “investment adviser” that would exclude persons who offer investment advice only through such publications as newsletters and reports. Although there is very little discussion of the issue, it is significant that in the hearings on the proposed legislation, representatives of both the SEC and the investment advisers expressed their view that the Act would cover the publishers of investment newsletters. David Schenker, the Chief Counsel of the SEC Investment Trust Study and one of the primary architects of the proposed legislation, explained that the term “investment advisers” as used in the Act “encompasses that broad category ranging from people who are engaged in the profession of furnishing disinterested, impartial advice to a certain economic stratum of our population to the other extreme, individuals engaged in running tipster organizations, or sending through the mails stock market letters.” Hearings on S. 3580 before a Subcommittee of the Senate Committee on Banking and Currency, 76th Cong., 3d Sess., 47 (1940) (hereafter Senate Hearings). In the later House hearings, James White, a representative of a Boston investment counsel firm *220who was among the industry spokesmen who cooperated with the SEC in the later stages of the drafting of the bill, expressed the same view of the scope of the statutory definition in its final form: “the term includes people who send out bulletins from time to time on the advisability of buying or selling stocks, or even giving tips on cheap stocks, and goes all of the way from that to individuals and firms who undertake to give constant supervision to the entire investments of their clients on a personal basis and who even advise them on tax matters and other financial matters which essentially are not a question of choice of investments.”5 Hearings on H. R. 10065 before a Subcommittee of the House Committee on Interstate and Foreign Commerce, 76th Cong., 3d Sess., 87 (1940). Later in his testimony, White specifically explained to Representative Boren that persons whose advice was furnished solely through publications were not excepted from the class of investment advisers as defined in the Act. *221See id., at 90-91.6 And although the House and Senate Reports are in the main silent on the question of the extent to which advisers operating solely through publications are governed by the Act, the Senate Report does at least make clear that a personal relationship between adviser and client is not a sine qua non of an investment adviser under the statute: the Report states that the Act “recognizes that with respect to a certain class of investment advisers, a type of personalized relationship may exist with their clients.” S. Rep. No. 1775, 76th Cong., 3d Sess., 22 (1940) (emphasis added).7
*222The subsequent legislative history of the Act testifies to Congress’ continuing belief that the legislation it has enacted applies to publishers of investment advice as well as to per*223sons who offer personal investment counseling. In 1960, Congress substantially expanded the penalties available to the Commission for use against unregistered advisers and advisers engaged in fraudulent or manipulative activities. Pub. L. 86-750, 74 Stat. 885. In describing the scope of the legislation, the Senate Report explained that “[t]hose defined as investment advisers by the act range from investment counsel firms, brokers whose advice is not incidental to their business, financial publishing houses not of general circulation, tout sheets and others.” S. Rep. No. 1760, 86th Cong., 2d Sess., 2 (1960) (emphasis added). In 1970, Con*224gress again expanded the enforcement authority of the SEC, see Pub. L. 91-547, 84 Stat. 1430; and again, the Senate Report explained that the Act “regulates the activities of those who receive compensation for advising others with respect to investments in securities or who are in the business of issuing analyses or reports concerning securities.” S. Rep. No. 91-184, p. 43 (1969) (emphasis added).
A construction of the Act that excludes publishers of investment advisory newsletters from the definition of “investment adviser” not only runs counter to the statute’s language, legislative history, and administrative construction, but also frustrates the policy of the Act by preventing apparently legitimate applications of the statute. The SEC has long been concerned with the problem of fraudulent and manipulative practices by some investment advisory publishers — specifically, with the problem of “scalping,” whereby a person associated with an advisory service “purchases] shares of a security for his own account shortly before recommending that security for long-term investment and then immediately sell[s] the shares at a profit upon the rise in the market price following the recommendation.” SEC v. Capital Gains Research Bureau, Inc., 375 U. S. 180, 181 (1963). An SEC study issued in 1963 emphasized that this practice is most dangerous when engaged in by an “advisory service with a sizable circulation” — that is, a newsletter or other publication — whose recommendation “could have at least a short-term effect on a stock’s market price.” Report of Special Study of Securities Markets of the Securities and Exchange Commission, H. R. Doc. No. 95, 88th Cong., 1st Sess., pt. 1, p. 372 (1963). The SEC study concluded that scalping was a serious problem within the investment advisory industry. See id., at 371-373.
In SEC v. Capital Gains Research Bureau, Inc., supra, we held that the antifraud provisions of the Investment Advisers Act could be invoked against the publisher of an investment advisory newsletter who had engaged in scalping, and that such *225an adviser could be required “to make full and frank disclosure of his practice of trading on the effect of his recommendations.” Id., at 197. The Court’s construction of the Act, under which a publisher like petitioner is not an “investment adviser” and is therefore not subject to the Act’s antifraud provisions, effectively overrules Capital Gains and limits the SEC’s power to protect the public against a potentially serious form of fraud and manipulation. But there is no suggestion that the application of the antifraud provisions of the Act to require investment advisory publishers to disclose material facts would present serious First Amendment difficulties. See Zauderer v. Office of Disciplinary Counsel, 471 U. S. 626, 651 (1985); Village of Schaumburg v. Citizens for a Better Environment, 444 U. S. 620, 637-638 (1980); Schneider v. State, 308 U. S. 147, 164 (1939).8 Accordingly, the Court’s zeal to avoid the narrow constitutional issue presented by the case leads it to adopt a construction of the Act that, wholly unnecessarily, prevents what would seem to be desirable and constitutional applications of the Act — a result at odds with our longstanding policy of construing securities regulation enactments broadly and their exemptions narrowly in order to effectuate their remedial purposes. See, e. g., Tcherepnin v. Knight, 389 U. S. 332, 336 (1967).9
*226It is ironic that this construction, at odds with the language, history, and policies of the Act, is adopted in the name of constitutional avoidance. One does not have to read the Court’s opinion very closely to realize that its interpretation of the Act is in fact based on a thinly disguised conviction that the Act is unconstitutional as applied to prohibit publication of newsletters by unregistered advisers. Indeed, the Court tips its hand when it discusses the Court’s decisions in Lovell v. City of Griffin, 303 U. S. 444 (1938), and Near v. Minnesota ex rel. Olson, 283 U. S. 697 (1931). The Court reasons that given these decisions, which forbade certain forms of prior restraints on speech, the 76th Congress could not have intended to enact a licensing provision for investment advisers that would include persons whose advisory activities were limited to publishing. The implication is that the application of the Act’s penalties to unregistered publishers would violate the principles of Lovell and Near; and because Con*227gress is assumed to know the law, see ante, at 205, n. 50, the Court concludes that it must not have intended that result.
This reasoning begs the question. What we have been called on to decide in this case is precisely whether restraints on petitioner’s publication are unconstitutional in light of such decisions as Near and Lovell. While purporting not to decide the question, the Court bases its statutory holding in large measure on the assumption that Congress already knew the answer to it when the statute was enacted. The Court thus attributes to the 76th Congress a clairvoyance the Solicitor General and the Second Circuit apparently lack — that is, the ability to predict our constitutional holdings 45 years in advance of our declining to reach them. If the policy of constitutional avoidance amounts to no more than a preference for implicitly deciding constitutional questions without explaining our reasoning, and if the consequence of adopting the policy is a statutory decision more disruptive of the legislative framework than a decision on the narrow constitutional issue presented, the purposes underlying the policy have been ill-served. In light of the language, history, and purposes of the statute, I would read its definition of “investment adviser” to encompass publishers like petitioner, and turn to the constitutional question. In the words of Justice Cardozo:
“[Ajvoidance of a difficulty will not be pressed to the point of disingenuous evasion. Here the intention of the Congress is revealed too distinctly to permit us to ignore it because of mere misgivings as to power. The problem must be faced and answered.” George Moore Ice Cream Co. v. Rose, 289 U. S. 373, 379 (1933).
I — I I — I
Petitioner, an investment adviser whose registration has been revoked, seeks to continue the practice of his profession by publishing newsletters containing investment advice. *228The SEC, consistent with the terms of the Act as I read them, has attempted to enjoin petitioner from engaging in these activities. The question is whether the First Amendment permits the Federal Government so to prohibit petitioner’s publication of investment advice.
A
This issue involves a collision between the power of government to license and regulate those who would pursue a profession or vocation and the rights of freedom of speech and of the press guaranteed by the First Amendment. The Court determined long ago that although “[i]t is undoubtedly the right of every citizen of the United States to follow any lawful calling, business, or profession he may choose, . . . there is no arbitrary deprivation of such right where its exercise is not permitted because of a failure to comply with conditions imposed ... for the protection of society.” Dent v. West Virginia, 129 U. S. 114, 121-122 (1889). Regulations on entry into a profession, as a general matter, are constitutional if they “have a rational connection with the applicant’s fitness or capacity to practice” the profession. Schware v. Board of Bar Examiners, 353 U. S. 232, 239 (1957).
The power of government to regulate the professions is not lost whenever the practice of a profession entails speech. The underlying principle was expressed by the Court in Giboney v. Empire Storage & Ice Co., 336 U. S. 490, 502 (1949): “it has never been deemed an abridgment of freedom of speech or press to make a course of conduct illegal merely because the conduct was in part initiated, evidenced, or carried out by means of language, either spoken, written, or printed.”
Perhaps the most obvious example of a “speaking profession” that is subject to governmental licensing is the legal profession. Although a lawyer’s work is almost entirely devoted to the sort of communicative acts that, viewed in isolation, fall within the First Amendment’s protection, we *229have never doubted that “[a] State can require high standards of qualification, such as good moral character or proficiency in its law, before it admits an applicant to the bar... Schware v. Board of Bar Examiners, supra, at 239. The rationale for such limits was expressed by Justice Frankfurter:
“One does not have to inhale the self-adulatory bombast of after-dinner speeches to affirm that all the interests of man that are comprised under the constitutional guarantees given to ‘life, liberty and property’ are in the professional keeping of lawyers. It is a fair characterization of the lawyer’s responsibility in our society that he stands ‘as a shield,’ to quote Devlin, J., in defense of right and to ward off wrong. From a profession charged with such responsibilities there must be exacted those qualities of truth-speaking, of a high sense of honor, of granite discretion, of the strictest observance of fiduciary responsibility, that have, throughout the centuries, been compendiously described as ‘moral character.’” 353 U. S., at 247 (concurring opinion).
The Government’s position is that these same principles support the legitimacy of its regulation of the investment advisory profession, whether conducted through publications or through personal client-adviser relationships. Clients trust in investment advisers, if not for the protection of life and liberty, at least for the safekeeping and accumulation of property. Bad investment advice may be a cover for stock-market manipulations designed to bilk the client for the benefit of the adviser; worse, it may lead to ruinous losses for the client. To protect investors, the Government insists, it may require that investment advisers, like lawyers, evince the qualities of truth-speaking, honor, discretion, and fiduciary responsibility.
But the principle that the government may restrict entry into professions and vocations through licensing schemes has never been extended to encompass the licensing of speech *230per se or of the press. See Thomas v. Collins, 323 U. S. 516 (1945); Lovell v. City of Griffin, 303 U. S. 444 (1938); Schneider v. State, 308 U. S. 147 (1939); Near v. Minnesota ex rel. Olson, 283 U. S. 697 (1931); Cantwell v. Connecticut, 310 U. S. 296 (1940); Village of Schaumburg v. Citizens for a Better Environment, 444 U. S. 620 (1980); Jamison v. Texas, 318 U. S. 413 (1943). At some point, a measure is no longer a regulation of a profession but a regulation of speech or of the press; beyond that point, the statute must survive the level of scrutiny demanded by the First Amendment.
The Government submits that the location of the point at which professional regulation (with incidental effects on otherwise protected expression) becomes regulation of speech or the press is a matter that should be left to the legislature. In this case, the Government argues, Congress has determined that investment advisers — including publishers such as petitioner — are fiduciaries for their clients. Accordingly, Congress has the power to limit entry into the profession in order to ensure that only those who are suitable to fulfill their fiduciary responsibilities may engage in the profession.
I cannot accept this as a sufficient answer to petitioner’s constitutional objection. The question whether any given legislation restrains speech or is merely a permissible regulation of a profession is one that we ourselves must answer if we are to perform our proper function of reviewing legislation to ensure its conformity with the Constitution. “It is emphatically the province and duty of the judicial department to say what the law is.” Marbury v. Madison, 1 Cranch 137, 177 (1803). Although congressional enactments come to this Court with a presumption in favor of their validity, see Rostker v. Goldberg, 453 U. S. 57, 64 (1981), Congress’ characterization of its legislation cannot be decisive of the question of its constitutionality where individual rights are at issue. See Trop v. Dulles, 356 U. S. 86, 94-104 (1958) (plurality opinion of Warren, C. J.); cf. Buckley v. Valeo, *231424 U. S. 1, 14-24 (1976) (per curiam). Surely it cannot be said, for example, that if Congress were to declare editorial writers fiduciaries for their readers and establish a licensing scheme under which “unqualified” writers were forbidden to publish, this Court would be powerless to hold that the legislation violated the First Amendment. It is for us, then, to find some principle by which to answer the question whether the Investment Advisers Act as applied to petitioner operates as a regulation of speech or of professional conduct.
This is a problem Justice Jackson wrestled with in his concurring opinion in Thomas v. Collins, 323 U. S., at 544-548. His words are instructive:
“[A] rough distinction always exists, I think, which is more shortly illustrated than explained. A state may forbid one without its license to practice law as a vocation, but I think it could not stop an unlicensed person from making a speech about the rights of man or the rights of labor, or any other kind of right, including recommending that his hearers organize to support his views. Likewise, the state may prohibit the pursuit of medicine as an occupation without its license, but I do not think it could make it a crime publicly or privately to speak urging persons to follow or reject any school of medical thought. So the state to an extent not necessary now to determine may regulate one who makes a business or a livelihood of soliciting funds or memberships for unions. But I do not think it can prohibit one, even if he is a salaried labor leader, from making an address to a public meeting of workmen, telling them their rights as he sees them and urging them to unite in general or to join a specific union.” Id., at 544-545.
Justice Jackson concluded that the distinguishing factor was whether the speech in any particular case was “associated] . . . with some other factor which the state may regulate so as to bring the whole within official control.” Id., at 547. *232If “in a particular case the association or characterization is a proven and valid one,” he concluded, the regulation may-stand. Ibid.
These ideas help to locate the point where regulation of a profession leaves off and prohibitions on speech begin. One who takes the affairs of a client personally in hand and purports to exercise judgment on behalf of the client in the light of the client’s individual needs and circumstances is properly viewed as engaging in the practice of a profession. Just as offer and acceptance are communications incidental to the regulable transaction called a contract, the professional’s speech is incidental to the conduct of the profession. If the government enacts generally applicable licensing provisions limiting the class of persons who may practice the profession, it cannot be said to have enacted a limitation on freedom of speech or the press subject to First Amendment scrutiny.10 Where the personal nexus between professional and client does not exist, and a speaker does not purport to be exercising judgment on behalf of any particular individual with whose circumstances he is directly acquainted, government regulation ceases to function as legitimate regulation of professional practice with only incidental impact on speech; it becomes regulation of speaking or publishing as such, subject to the First Amendment’s command that “Congress shall make no law . . . abridging the freedom of speech, or of the press.”11
*233As applied to limit entry into the profession of providing investment advice tailored to the individual needs of each client, then, the Investment Advisers Act is not subject to scrutiny as a regulation of speech — it can be justified as a legitimate exercise of the power to license those who would practice a profession, and it is no more subject to constitutional attack than state-imposed limits on those who may practice the professions of law and medicine. The application of the Act’s enforcement provisions to prevent unregistered persons from engaging in the business of publishing investment advice for the benefit of any who would purchase their publications, however, is a direct restraint on freedom of speech and of the press subject to the searching scrutiny called for by the First Amendment.
B
The recognition that the prohibition on the publishing of investment advice by persons not registered under the Act is a restraint on speech does not end the inquiry. Not all restrictions on speech are impermissible. The Government contends that even if the statutory restraints on petitioner’s publishing activities are deemed to be restraints on speech rather than mere regulations of entry into a profession, petitioner’s speech is “expression related solely to the economic interests of the speaker and its audience,” Central Hudson Gas & Electric Corp. v. Public Service Comrn’n of New York, 447 U. S. 557, 561 (1980), and is therefore subject to the reduced protection afforded what we have come to describe as “commercial speech.” See Zauderer v. Office of Disciplinary Counsel, 471 U. S. 626 (1985). Under the commercial speech doctrine, restrictions on commercial speech that directly advance a substantial governmental interest may be upheld. See id., at 638. The prohibition on petitioner’s publishing activities, the Government suggests, is such a permissible restriction, as it directly advances the goal of protecting the investing public against unscrupulous advisers.
*234Petitioner, echoing the dissent below, argues that the expression contained in his newsletters is not commercial speech, as it does not propose a commercial transaction between the speaker and his audience. See Virginia Pharmacy Board v. Virginia Citizens Consumer Council, Inc., 425 U. S. 748, 762 (1976). Although petitioner concedes that his speech relates to economic subjects, he argues that it is not for that reason stripped of its status as fully protected speech. See Thomas v. Collins, 323 U. S., at 531. Accordingly, he argues, the prohibition on his speech can be upheld “only if the government can show that the regulation is a precisely drawn means of serving a compelling state interest.” Consolidated Edison Co. v. Public Service Comm’n of New York, 447 U. S. 530, 541 (1980).
I do not believe it is necessary to the resolution of this case to determine whether petitioner’s newsletters contain fully protected speech or commercial speech. The Act purports to make it unlawful for petitioner to publish newsletters containing investment advice and to authorize an injunction against such publication. The ban extends as well to legitimate, disinterested advice as to advice that is fraudulent, deceptive, or manipulative. Such a flat prohibition or prior restraint on speech is, as applied to fully protected speech, presumptively invalid and may be sustained only under the most extraordinary circumstances. See New York Times Co. v. United States, 403 U. S. 713 (1971); Schneider v. State, 308 U. S. 147 (1939); Near v. Minnesota ex rel. Olson, 283 U. S. 697 (1931). I do not understand the Government to argue that the circumstances that would justify a restraint on fully protected speech are remotely present in this case.
But even where mere “commercial speech” is concerned, the First Amendment permits restraints on speech only when they are narrowly tailored to advance a legitimate governmental interest. The interest here is certainly legitimate: the Government wants to prevent investors from falling into the hands of scoundrels and swindlers. The means *235chosen, however, is extreme. Based on petitioner’s past misconduct, the Government fears that he may in the future publish advice that is fraudulent or misleading; and it therefore seeks to prevent him from publishing any advice, regardless of whether it is actually objectionable. Our commercial speech cases have consistently rejected the proposition that such drastic prohibitions on speech may be justified by a mere possibility that the prohibited speech will be fraudulent. See Zauderer, supra; In re R. M. J., 455 U. S. 191, 203 (1982); Bates v. State Bar of Arizona, 433 U. S. 350 (1977). So also here. It cannot be plausibly maintained that investment advice from a person whose background indicates that he is unreliable is inherently misleading or deceptive,12 nor am I convinced that less drastic remedies than outright suppression (for example, application of the Act’s antifraud provisions) are not available to achieve the Government’s asserted purpose of protecting investors. Accordingly, I would hold that the Act, as applied to prevent petitioner from publishing investment advice altogether, is too blunt an instrument to survive even the reduced level of scrutiny called for by restrictions on commercial speech. The Court’s observation in Schneider v. State, supra, at 164, is applicable here as well:
“Frauds may be denounced as offenses and punished by law. ... If it is said that these means are less efficient and convenient than bestowal of power on police authorities to decide what information may be disseminated . . . and who may impart the information, the answer is that considerations of this sort do not empower [government] to abridge freedom of speech and press.”
*236III
I emphasize the narrowness of the constitutional basis on which I would decide this case. I see no infirmity in defining the term “investment adviser” to include a publisher like petitioner, and I would by no means foreclose the application of, for example, the Act’s antifraud or reporting provisions to investment advisers (registered or unregistered) who offer their advice through publications. Nor do I intend to suggest that it is unconstitutional to invoke the Act’s provisions for injunctive relief and criminal penalties against unregistered persons who, for compensation, offer personal investment advice to individual clients. I would hold only that the Act may not constitutionally be applied to prevent persons who are unregistered (including persons whose registration has been denied or revoked) from offering impersonal investment advice through publications such as the newsletters published by petitioner.
Although this constitutional holding, unlike the Court’s statutory holding, would not foreclose the SEC from treating petitioner as an “investment adviser” for some purposes, it would require reversal of the judgment of the Court of Appeals. I therefore concur in the result.
The Court argues that this fact is without significance, as it proves only that publishers found it to be to their own advantage to register. But the SEC’s matter-of-fact announcement of the number of publishers registering under the Act establishes something else: from the beginning, the SEC assumed the Act applied to such publishers.
In 1963, the Commission explained its view of the coverage of the Act as follows:
“The investment advisers who are required to register with the Commission under the Investment Advisers Act are certain firms (or individuals) engaged in the business of advising others for a fee on the value of the securities or the desirability of buying or selling securities. For the most part they fall into one of two groups: Those publishing advisory services and periodic market reports for subscribers, and those offering supervision *216of individual clients’ portfolios.” Report of Special Study of Securities Markets of the Securities and Exchange Commission, H. R. Doc. No. 95, 88th Cong., 1st Sess., pt. 1, p. 146 (1963).
The Court suggests that “tipsters” and “touts” might not qualify under its reading of the “bona fide publications” exception either because their publications are not sufficiently regular or because their advice is not sufficiently disinterested. Both suggestions seem implausible. As is evident from the Court’s conclusion that petitioner’s publications meet the regularity requirement, the Court’s construction of the requirement adopts the view of our major law reviews on the issue of regular publication: good intentions are enough. Thus, if a “tout” or “tipster” promised to publish his recommendations at more or less regular intervals, he, like petitioner, would meet the regularity requirement. Moreover, a truly “hit and run” practitioner — one who did not even claim an intention of issuing further recommendations — would not fall within the definition of an “investment adviser” because he would not be deemed to “engag[e] in the business” of advising others. See Applicability of Investment Advisers Act to Certain Publications, SEC Release No. 1A-563, 42 Fed. Reg. 2953 (1977), codified at 17 CFR § 276 (1984). As for the Court’s suggestion that “touts” and “tipsters” might not qualify under the exception if their advice was not disinterested, it appears completely unfounded: nowhere in the language or history of the Act is there any suggestion that whether a person is an investment adviser depends on whether his advice is disinterested. In addition, in suggesting that the character of the adviser’s advice determines whether he falls within the “bona fide publications” exception, the Court contradicts itself. At one point, it states that the exception is based on “objective” criteria, and it purports to eschew a content-based interpretation of the term “bona fide.” See ante, at, 207-208, n. 53. At another, the Court suggests that publications that offer advice that is not disinterested are not “bona fide.” See ante, at 207-209, and n. 55. It is hard to understand why the Court prefers its content-based reading to the SEC’s, particularly given that the SEC’s reading is much simpler to apply in practice: if a publication is primarily a device for offering investment advice, it is not a “bona fide” newspaper, news magazine, or business or financial publication. Under the Court’s reading, the SEC would have to force the publisher to disclose his own financial holdings and then compare his recommendations with his stock holdings in order to determine whether his publications were “bona fide.” This requirement would be self-defeating, since the SEC has no authority under the Act to order such disclosures by anyone whom it does not already know to be an investment adviser.
The Second Circuit’s explication of the use of the term “bona fide” in the statute is instructive:
“Section 202(a)(ll) of the Act lists a number of examples of persons or entities whose activities might fall within the broad definition of ‘investment adviser’ but whose customary practices would not place them in the special, otherwise unregulated, fiduciary role for which the law established standards. . . . The phrase ‘bona fide’ newspapers, in the context of this list, means those publications which do not deviate from customary newspaper activities to such an extent that there is a likelihood that the wrongdoing which the Act was designed to prevent has occurred. The determination of whether or not a given publication fits within this exclusion must depend upon the nature of its practices rather than upon the purely formal ‘indicia of a newspaper’ which it exhibits on its face and in the size and nature of its subscription list.” SEC v. Wall Street Transcript Corp., 422 F. 2d 1371, 1377, cert. denied, 398 U. S. 958 (1970).
The Second Circuit’s reasoning provides firm support for the SEC’s position that the point of the “bona fide publications” exception is to differentiate publications devoted solely or primarily to the provision of investment advice from publications that contain more diversified or general discussions of news events and business or financial topics. The aim of the Act is the protection of the investing public against fraud or manipulation on the part of advisers. Viewed in light of this purpose, a publication that is no more than a vehicle for investment advice is an obvious target for regulatory measures: it makes sense to treat the entire publication as an adviser and to impose liability on the publication itself in the case of fraud or manipulation. On the other hand, the publisher of a publication that presents diverse forms of information and is not narrowly focused on the provision of investment advice is not so likely to engage in abusive prac*219tices. Thus, it is logical to treat the publication itself as a “bona fide publication” and to exempt its publisher from classification as an investment adviser. Individual writers who make it their business to offer investment advice to the publication’s readers on a regular basis, however, may still be covered. See Lovitch, The Investment Advisers Act of 1940— Who Is an “Investment Adviser”?, 24 Kan. L. Rev. 67, 94, n. 222 (1975) (noting SEC staff’s position that columnists who offer investment advice in exempt publications are investment advisers).
The Court correctly points out that Mr. Schenker’s statement was made before the “bona fide publications” exception was in its final form and before the inclusion in the record of the Subcommittee hearings of the Illinois report that suggested that regulation of publishers might raise First Amendment problems. The Court neglects to acknowledge that Mr. White’s statement postdated both the submission of the report to the Senate Subcommittee and the amendment of the Act’s definition to its final form. White’s statement is a plain indication that the drafters of the bill had not changed their position since the inception of the Senate hearings: publishers were still viewed to be within the Act.
The Court also suggests that its interpretation of the scope of the exception is consistent with White’s statement that persons who “send out bulletins from time to time” offering investment advice are investment advisers. Such persons, the Court suggests, would not meet the “regularity” requirement of the “bona fide publications” exception. But the Court’s own loose construction of the requirement belies this argument: petitioner himself, at best, can be described as a person who sends out bulletins “from time to time.” If the timeliness of petitioner’s publications is sufficient to meet the Act’s regularity requirement, it is hard to imagine a publisher who could not qualify.
The Court argues that my interpretation of the exchange between Boren and White is incorrect. I am at a loss to understand this contention. To my mind, the colloquy, as reprinted by the Court, unambiguously supports my reading. Representative Boren asked Mr. White why persons who dispensed investment advice through publications should be excluded from the category of investment advisers. White answered the question by pointing out that its premise was incorrect: Boren was reading the wrong definition. The clear implication was that the correct definition did include such publishers, and Boren’s last remark — "that clarifies it for me” — indicates that he took the point.
In reaching the opposite conclusion, the Court relies on a hodgepodge of materials that are either completely irrelevant or reflect approaches that were explicitly rejected by the framers of the statute. For example, the Court correctly notes that the SEC Report that was in large measure the impetus for the Investment Advisers Act restricted its attention to “investment counsel” — that is, investment advisers maintaining a personal relationship with individual clients. See Investment Trusts and Investment Companies, Report of the Securities and Exchange Commission, Pursuant to Section 30 of the Public Utility Holding Company Act of 1935, Investment Counsel, Investment Management, Investment Supervisory, and Investment Advisory Services, H. R. Doc. No. 477, 76th Cong., 2d Sess. (1939). But imputing the narrow focus of the Report to the Act itself would be a serious mistake, for the Act explicitly covers investment advisers who cannot be described as “investment counsel.” This is evident from § 208(c) of the Act, which provides that no investment adviser may hold himself out as “investment counsel” unless “a substantial part of his . . . business consists of rendering investment supervisory services” — “investment supervisory services” being defined by § 202(a)(13) of the Act as “the giving of continuous advice as to the investment of funds *222ón the basis of the individual needs of each client.” The Act could not be clearer: not all “investment advisers” under the Act are “investment counsel.” The Act’s careful distinction between “investment counsel” and the other investment advisers subject to its provisions leaves no doubt that the framers of the Act intended it to cover advisers not engaged in personal investment counseling as well as “investment counsel.” For this reason, it can by no means be said that the SEC Report’s focus on “investment counsel” limits the scope of the Act.
The Court’s reliance on the self-serving statements of industry representatives regarding the importance of their personal relationships with their clients is similarly misplaced. First, it is abundantly clear that the investment counsel who testified before the Senate Subcommittee were not suggesting that only advisers with personal relationships with their clients should be covered by the Act — far from it. Rather, the import of their statements was that reputable “investment counsel” who had a personal fiduciary relationship with their clients did not require federal regulation (unlike the “touts and tipsters” whom these investment counselors unanimously reviled).
Second, it appears that the primary problem these “investment counsel” had with the Act was their fear that it would require them to disclose confidential communications with their clients. This concern was dealt with through the insertion into the Act of § 210(c), which provides that “[n]o provision of this subchapter shall be construed to require, or to authorize the Commission to require any investment adviser engaged in rendering investment supervisory services to disclose the identity, investments, or affairs of any client of such investment adviser, except insofar as such disclosure may be necessary or appropriate in a particular proceeding or investigation having as its object the enforcement of a provision or provisions of this subchapter.” 15 U. S. C. § 80b-10(c). The references in the House and Senate Reports to the “care [that] has been taken ... to respect this relationship between investment advisers and their clients,” see ante, at 201, obviously refer to this provision for confidentiality and to the provision restricting the class of investment advisers who may claim the title “investment counsel.” The Reports’ references to adviser-client relationships thus by no means suggest that the Act limited its definition of “investment advisers” to those who offered personalized services. Indeed, § 210(c) of the Act, in referring to “investment advisers engaged in rendering investment supervisory services” — that is, “the giving of con*223tinuous advice as to the investment of funds on the basis of the individual needs of each client” — makes quite clear that some persons defined as “investment advisers” under the Act do not offer such personalized services.
The Court also errs in relying on the Illinois report reprinted in the Senate Hearings as authority for the notion that Congress intended to exclude all publishers from the definition of “investment adviser” in order to avoid constitutional difficulties. See ante, at 197-199. This report cannot bear the weight the Court places on it. The discussion in the report — buried in a document placed into the record after weeks of hearings — contains the only mention in the legislative history of the Act of the potential First Amendment difficulties raised by including publications within the category of investment advisers. Still more significant is the definite rejection of the report’s recommended solution to the First Amendment problem by the drafters of the Act. The report’s recommendation was that any legislation regulating “investment counselors” should “carefully defin[e] the term ‘investment counselor’ so as to exclude ‘any person or organization which engages in the business of furnishing investment analysis, opinion, or advice solely through publications distributed to a list of subscribers and not furnishing specific advice to any client with respect to securities, and also persons or organizations furnishing only economic advice and not advice relating to the purchase or sale of securities.’” Senate Hearings, at 1009. This approach, the report noted, was “generally the same as that used by the [SEC] in limiting the scope of its report on investment counsel organizations.” Ibid. The Act, of course, did not carefully exclude persons who furnished advice through publications — it expressly included them in its definition. Moreover, the Act’s provisions make it quite clear that the definition of “investment adviser” in § 202(a)(ll) is more expansive than the definition of “investment counsel” used in the SEC study and in § 208(c) of the Act itself.
Similarly, the application of the Act’s reporting requirements, 15 U. S. C. § 80b-4, to investment advisers whose activities are restricted to publishing would not appear to raise serious First Amendment concerns. The reporting requirements would not inhibit such advisers from speaking, and it is well settled that “[t]he Amendment does not forbid . . . regulation which ends in no restraint upon expression or in any other evil outlawed by its terms and purposes.” Oklahoma Press Publishing Co. v. Walling, 327 U. S. 186, 193 (1946). See also Branzburg v. Hayes, 408 U. S. 665 (1972), in which we held that the press is not exempt from the generally applicable requirement that a citizen produce evidence in response to a subpoena.
The Court brushes aside the significance of this consequence by suggesting that alternative remedies — specifically, remedies under Rule 10b-5 — may be available. This may be so, although the requirement of Rule *22610b-5 that any nondisclosure violate an existing fiduciary duty, see Chiarella v. United States, 445 U. S. 222 (1980), leaves the matter in some doubt. The District Court in SEC v. Blavin, 557 F. Supp. 1304 (ED Mich. SD 1983), aff’d, 760 F. 2d 706 (CA6 1985), had little difficulty in finding a fiduciary duty, for it held that the defendant’s publishing activities brought him squarely within the Act’s definition of an “investment adviser,” and that “as [an investment adviser, he] had a duty to his clients and readers to undertake some reasonable investigation of the figures he was printing before he printed them.” 557 F. Supp., at 1314. The Court, of course, holds that publishers like petitioner (and Blavin) are not investment advisers and thus excludes the possibility that the Investment Advisers Act could supply the requisite fiduciary duty. The Court also hypothesizes that scalping by a publisher might constitute mail fraud, but again, as far as I am aware, that is no more than an open question. The certainty that the Investment Advisers Act provides a remedy against scalping thus remains, for me, a persuasive reason for not adopting a construction of the Act that would exclude petitioner. In addition, the antifraud provisions of the Act are supplemented by reporting requirements that may be used to aid the SEC in uncovering scalping. By taking petitioner outside the category of investment advisers, the Court places him beyond the reach of these additional tools for uncovering deceit.
Of course, it is possible that conditions the government might impose on entry into a profession would in some cases themselves violate the First Amendment. For example, denial of a license on the basis of the applicant’s beliefs or political statements he had made in the past could constitute a First Amendment violation. However, in such a case, the problem would not be that it was impermissible for the government to restrict entry into the profession because of the nature of the profession itself.
See Near v. Minnesota ex rel. Olson, 283 U. S. 697, 720 (1931) (“Characterizing the publication as a business, and the business as a nuisance, does not permit an invasion of the constitutional immunity against restraint”).
Cf. Near v. Minnesota ex rel. Olson, supra, in which the Court held that previous publication of defamatory material — unprotected speech— could not justify a prior restraint limited to further publication of defamatory matter. Here, the ban on petitioner’s future publishing activities extends to nondeceptive (that is, protected) as well as fraudulent speech.