concurring in part and concurring in the judgment.
I join the opinion of the Court (with the exception of Part III — B—1) because I agree that the Massachusetts cigarette advertising regulations are pre-empted by the Federal Cigarette Labeling and Advertising Act, 15 U. S. C. § 1831 et seq. I also agree with the Court’s disposition of the First Amendment challenges to the other regulations at issue here, and I share the Court’s view that the regulations fail even the intermediate scrutiny of Central Hudson Gas & Elec. Corp. v. Public Serv. Comm’n of N. Y, 447 U. S. 557 (1980). At the same time, I continue to believe that when the government seeks to restrict truthful speech in order to suppress the ideas it conveys, strict scrutiny is appropriate, whether or not the speech in question may be characterized as “commercial.” See .W Liquormart, Inc. v. Rhode Island, 517 U. S. 484, 518 (1996) (Thomas, J., concurring in part and concurring in judgment). I would subject all of the advertising restrictions to strict scrutiny and would hold that they violate the First Amendment.
I
At the heart of this litigation is a Massachusetts regulation that imposes a sweeping ban on speech about tobacco products. 940 Code of Mass. Regs. §21.04(5) (2000), which governs cigarettes and smokeless tobacco, and § 22.06(5), which governs cigars, prohibit all outdoor advertising, all indoor advertising that can be seen from outdoors, and all point-of-sale advertising (even if not visible from outdoors) that is *573lower than five feet from the floor.1 These restrictions are superficially limited in their geographic scope: They apply only within 1,000 feet of “any public playground, playground area in a public park, elementary school or secondary school.” §21.04(5)(a). But the Court of Appeals acknowledged that the zone of prohibition covers as much as 90 percent of the three largest cities in Massachusetts, Consolidated Cigar Corp. v. Reilly, 218 F. 3d 30, 50 (CA1 2000), so the practical effect is little different from that of a total ban. Cf. United States v. Playboy Entertainment Group, Inc., 529 U. S. 803, 812 (2000) (“The Government’s content-based burdens must satisfy the same rigorous scrutiny as its content-based bans”).
Respondents suggest in passing that the regulations are “zoning-type restrictions” that should receive “the intermediate level of scrutiny traditionally associated with various forms of ‘time, place, and manner’ regulations.” Brief for Respondents 31. We have indeed upheld time, place, and manner regulations that prohibited certain kinds of outdoor signs, see, e.g., Members of City Council of Los Angeles v. Taxpayers for Vincent, 466 U. S. 789 (1984), and we have similarly upheld zoning laws that had the effect of restricting certain kinds of sexually explicit expression, see, e. g., Renton v. Playtime Theatres, Inc., 475 U. S. 41 (1986). But the abiding characteristic of valid time, place, and manner regulations is their content neutrality. See Ward v. Rock Against Racism, 491 U. S. 781, 791-796 (1989). In Vincent the city prohibited all signs on public property, not to sup*574press the message conveyed by any of the signs, but simply to minimize the esthetic effect of visual clutter. Likewise, the ordinance in Renton was aimed not at expression, but at the “secondary effects” caused by adult businesses.
The regulations here are very different. Massachusetts is not concerned with any “secondary effects” of tobacco advertising — it is concerned with the advertising’s primary effect, which is to induce those who view the advertisements to purchase and use tobacco products. Cf. Boos v. Barry, 485 U. S. 312, 321 (1988) (“Listeners’ reactions to speech are not the type of ‘secondary effects’ we referred to in Renton”). In other words, it seeks to suppress speech about tobacco because it objects to the content of that speech. We have consistently applied strict scrutiny to such content-based regulations of speech. See, e. g., Turner Broadcasting System, Inc. v. FCC, 512 U. S. 622, 641-643 (1994).
A
There was once a time when this Court declined to give any First Amendment protection to commercial speech. In Valentine v. Chrestensen, 316 U. S. 52 (1942), the Court went so far as to say that “the Constitution imposes [no] restraint on government as respects purely commercial advertising.” Id., at 54. That position was repudiated in Virginia Bd. of Pharmacy v. Virginia Citizens Consumer Council, Inc., 425 U. S. 748 (1976), which explained that even speech “which does ‘no more than propose a commercial transaction’” is protected by the First Amendment. Id., at 762 (quoting Pittsburgh Press Co. v. Pittsburgh Comm’n on Human Relations, 413 U. S. 376, 385 (1973)). Since then, the Court has followed an uncertain course — much of the uncertainty being generated by the malleability of the four-part balancing test of Central Hudson. See 44 Liquormart, 517 U. S., at 520-522 (Thomas, J., concurring in part and concurring in judgment).
*575I have observed previously that there is no “philosophical or historical basis for asserting that ‘commercial’ speech is of ‘lower value’ than ‘noncommercial’ speech.” Id., at 522. Indeed, I doubt whether it is even possible to draw a coherent distinction between commercial and noncommercial speech. See id., at 523, n. 4 (citing Kozinski & Banner, Who’s Afraid of Commercial Speech, 76 Va. L. Rev. 627 (1990)).2
It should be clear that if these regulations targeted anything other than advertising for commercial products — if, for example, they were directed at billboards promoting political candidates — all would agree that the restrictions should be subjected to strict scrutiny. In my view, an asserted government interest in keeping people ignorant by suppressing expression “is per se illegitimate and can no more justify regulation of ‘commercial’ speech than it can justify regulation of ‘noncoinmercial’ speech.” 517 U. S., at 518 (Thomas, J., concurring in part and concurring in judgment). That is essentially the interest asserted here, and, adhering to the views I expressed in H Liquormart, I would subject the Massachusetts regulations to strict scrutiny.
B
Even if one accepts the premise that commercial speech generally is entitled to a lower level of constitutional protection than are other forms of speech, it does not follow that the regulations here deserve anything less than strict scrutiny. Although we have recognized several categories of *576speech that normally receive reduced First Amendment protection, or no First Amendment protection at all, we have never held that the government may regulate speech within those categories in any way that it wishes. Rather, we have said “that these areas of speech can, consistently with thé First Amendment, be regulated because of their constitutionally proscribable content.” R. A. V. v. St. Paul, 505 U. S. 377, 383 (1992). Even when speech falls into a category of reduced constitutional protection, the government may not engage in content discrimination for reasons unrelated to those characteristics of the speech that place it within the category. For example, a city may ban obscenity (because obscenity is an unprotected category, see, e.g., Roth v. United States, 354 U. S. 476 (1957)), but it may not ban “only those legally obscene works that contain criticism of the city government.” R. A. V, supra, at 384.
In explaining the distinction between commercial speech and other forms of speech, we have emphasized that commercial speech is both “more easily verifiable by its disseminator” and less likely to be “chilled by proper regulation.” Virginia Bd., 425 U. S., at 772, n. 24. These characteristics led us to conclude that, in the context of commercial speech, it is “less necessary to tolerate inaccurate statements for fear of silencing the speaker,” and also that it is more “appropriate to require that a commercial message appear in such a form, or include such additional information, warnings, and disclaimers, as are necessary to prevent its being deceptive.” Ibid. Whatever the validity of this reasoning, it is limited to the peculiarly commercial harms that commercial speech can threaten — i. e., the risk of deceptive or misleading advertising. As we observed in R. A. V:
“[A] State may choose to regulate price advertising in one industry but not in others, because the risk of fraud (one of the characteristics of commercial speech that justifies depriving it of full First Amendment protection) is in its view greater there. But a State may not prohibit *577only that commercial advertising that depicts men in a demeaning fashion.” 505 U. S., at 388-389 (citations omitted).
In 44 Liquormart, several Members of the Court said much the same thing:
“[W]hen a State entirely prohibits the dissemination of truthful, nonmisleading commercial messages for reasons unrelated to the preservation of a fair bargaining process, there is far less reason to depart from the rigorous review that the First Amendment generally demands.” 517 U. S., at 501 (opinion of Stevens, J., joined by Kennedy and Ginsburg, JJ.).
Whatever power the State may have to regulate commercial speech, it may not use that power to limit the content of commercial speech, as it has done here, “for reasons unrelated to the preservation of a fair bargaining process.” Such content-discriminatory regulation — like all other content-based regulation of speech — must be subjected to strict scrutiny.
C
In an effort to avoid the implications of these basic principles of First Amendment law, respondents make two principal claims. First, they argue that the regulations target deceptive and misleading speech. See Brief for Respondents 33 (“Petitioners’ advertising clearly engenders ‘the potential for deception or confusion’ that allows for regulation of commercial speech based on its content” (quoting Bolger v. Youngs Drug Products Corp., 463 U. S. 60, 65 (1983))). Second, they argue that the regulations restrict speech that promotes an illegal transaction — i e., the sale of tobacco to minors. See Brief for Respondents 15 (“The regulations . . . exhibit a close connection to a commercial transaction the State has prohibited”).
Neither theory is properly before the Court. For purposes of summary judgment, respondents were willing to as*578sume “that the tobacco advertisements at issue here are truthful, nonmisleading speech about a lawful activity.” 218 F. 3d, at 43. Although respondents now claim that they have not conceded this point, see Brief for Respondents 35, n. 17, the fact remains that they did not urge their theories in the lower courts, and in general, we do not consider arguments for affirmance that were not presented below. See, e. g., Glover v. United States, 531 U. S. 198, 205 (2001). These concessions should make this an easy case, one clearly controlled by 44 Liquormart and by Greater New Orleans Broadcasting Assn., Inc. v. United States, 527 U. S. 173 (1999). At all events, even if we were to entertain these arguments, neither is persuasive.
Respondents suggest that tobacco advertising is misleading because “its youthful imagery and . . . sheer ubiquity” leads children to believe “that tobacco use is desirable and pervasive.” Brief for Respondents 33; see also Brief for United States as Amicus Curiae 7 (“[S]o many children lack the maturity in judgment to resist the tobacco industry’s appeals to excitement, glamour, and independence”). This justification is belied, however, by the sweeping overinclusivity of the regulations. Massachusetts has done nothing to target its prohibition to advertisements appealing to “excitement, glamour, and independence”; the ban applies with equal force to appeals to torpor, homeliness, and servility. It has not focused on “youthful imagery”; smokers depicted on the sides of buildings may no more play shuffleboard than they may ride skateboards.
The regulations even prohibit a store from accurately stating the prices at which cigarettes are sold. Such a display could not possibly be misleading, unless one accepts the State’s apparent view that the simple existence of tobacco advertisements misleads people into believing that tobacco use is more pervasive than it actually is. The State misunderstands the purpose of advertising. Promoting a product that is not yet pervasively used (or a cause that is not yet *579widely supported) is a primary purpose of advertising. Tobacco advertisements would be no more misleading for suggesting pervasive use of tobacco products than are any other advertisements that attempt to expand a market for a product, or to rally support for a political movement. Any inference from the advertisements that businesses would like for tobacco use to be pervasive is entirely reasonable, and advertising that gives rise to that inference is in no way deceptive.
The State also contends that tobacco advertisements may be restricted because they propose an illegal sale of tobacco to minors. A direct solicitation of unlawful activity may of course be proscribed, whether or not it is commercial in nature. See Brandenburg v. Ohio, 395 U. S. 444 (1969) (per curiam). The State’s power to punish speech that solicits or incites crime has nothing to do with the commercial character of the speech. After all, it is often the case that solicitation to commit a crime is entirely noncommercial. The harm that the State seeks to prevent is the harm caused by the unlawful activity that is solicited; it is unrelated to the commercial transaction itself. Thus there is no reason to apply anything other than our usual rule for evaluating solicitation and incitement simply because the speech in question happens to be commercial. See Carey v. Population Services Int’l, 431 U. S. 678, 701-702 (1977).
Viewed as an effort to proscribe solicitation to unlawful conduct, these regulations clearly fail the Brandenburg test. A State may not “forbid or proscribe advocacy of the use of force or of law violation except where such advocacy is directed to inciting or producing imminent lawless action and is likely to incite or produce such action.” Brandenburg, supra, at 447. Even if Massachusetts could prohibit advertisements reading, “Hey kids, buy cigarettes here,” these regulations sweep much more broadly than that. They cover “any ... statement or representation ... the purpose or effect of which is to promote the use or sale” of tobacco products, whether or not the statement is directly or indi*580rectly addressed to minors. 940 Code of Mass. Regs. §21.03 (2000). On respondents’ theory, all tobacco advertising may be limited because some of its viewers may not legally act on it.
It is difficult to see any stopping point to a rule that would allow a State to prohibit all speech in favor of an activity in which it is illegal for minors to engage. Presumably, the State could ban car advertisements in an effort to enforce its restrictions on underage driving. It could regulate advertisements urging people to vote, because children are not permitted to vote. And, although the Solicitor General resisted this implication of her theory, see Tr. of Oral Arg. 55-56, the State could prohibit advertisements for adult businesses, which children are forbidden to patronize.
At bottom, respondents’ theory rests on the premise that an indirect solicitation is enough to empower the State to regulate speech, and that, as petitioners put it, even an advertisement directed at adults “will give any children who may happen to see it the wrong idea and therefore must be suppressed from public view.” Brief for Petitioners Lorillard Tobacco Co. et al. in No. 00-596, p. 36. This view is foreign to the First Amendment. “Every idea is an incitement,” Gitlow v. New York, 268 U. S. 652, 673 (1925) (Holmes, J., dissenting), and if speech may be suppressed whenever it might inspire someone to act unlawfully, then there is no limit to the State’s censorial power. Cf. American Booksellers Assn., Inc. v. Hudnut, 771 F. 2d 323 (CA7 1985), aff’d, 475 U. S. 1001 (1986).
There is a deeper flaw in the State’s argument. Even if Massachusetts has a valid interest in regulating speech directed at children — who, it argues, may be more easily misled, and to whom the sale of tobacco products is unlawful— it may not pursue that interest at the expense of the free speech rights of adults.
The theory that public debate should be limited in order to protect impressionable children has a long historical pedi*581gree: Socrates was condemned for being “a doer of evil, inasmuch as he corrupts the youth.” 1 Dialogues of Plato, Apology 348 (B. Jowett transí., 4th ed. 1953). But the theory has met with a less enthusiastic reception in this Court than it did in the Athenian assembly. In Butler v. Michigan, 352 U. S. 380 (1957), we struck down a statute restricting the sale of materials “ ‘tending to incite minors to violent or depraved or immoral acts.’ ” Id., at 381 (quoting then Mich. Penal Code § 343). The effect of the law, we observed, was “to reduce the adult population of Michigan to reading only what is fit for children.” 352 U. S., at 383. As Justice Frankfurter colorfully put it, “Surely, this is to burn the house to roast the pig.” Ibid.
We have held consistently that speech “cannot be suppressed solely to protect the young from ideas or images that a legislative body thinks unsuitable for them.” Erznoznik v. Jacksonville, 422 U. S. 205, 213-214 (1975); accord, Bolger, 463 U. S., at 74 (“The level of discourse reaching a mailbox simply cannot be limited to that which would be suitable for a sandbox”). To be sure, in FCC v. Pacifica Foundation, 438 U. S, 726 (1978), we upheld the Federal Communications Commission’s power to regulate indecent but nonobscene radio broadcasts. But Pacifica relied heavily on what it considered to be the “special justifications for regulation of the broadcast media that are not applicable to other speakers.” Reno v. American Civil Liberties Union, 521 U. S. 844, 868 (1997). It emphasized that radio is “uniquely pervasive” and “uniquely accessible to children, even those too young to read.” Pacifica, supra, at 748-749 (emphasis added).
Outside of the broadcasting context, we have adhered to the view that “the governmental interest in protecting children from harmful materials” does not “justify an unnecessarily broad suppression of speech addressed to adults.” Reno, supra, at 875; see also Playboy Entertainment, 529 U. S.) at 814 (“[Tjhe objective of shielding children does not *582suffice to support a blanket ban if the protection can be accomplished by a less restrictive alternative”). Massachusetts may not avoid the application of strict scrutiny simply because it seeks to protect children.
II
Under strict scrutiny, the advertising ban may be saved only if it is narrowly tailored to promote a compelling government interest. See, e.g., id., at 813. If that interest could be served by an alternative that is less restrictive of speech, then the State must use that alternative instead. See ibid.; Reno, supra, at 874. Applying this standard, the regulations here must fail.
A
Massachusetts asserts a compelling interest in reducing tobacco use among minors. Applied to adults, an interest in manipulating market choices by keeping people ignorant would not be legitimate, let alone compelling. See supra, at 575. But assuming that there is a compelling interest in reducing underage smoking, and that the ban on outdoor advertising promotes this interest, I doubt that the same is true of the ban on point-of-sale advertising below five feet. See 940 Code of Mass. Regs. §§21.04(5)(b), 22.06(5)(b) (2000). The Court of Appeals admitted to having “some misgivings about the effectiveness of a restriction that is based on the assumption that minors under five feet tall will not, or will less frequently, raise their view above eye-level,” 218 F. 3d, at 51, as well it might have, since respondents have produced no evidence to support this counterintuitive assumption. Obviously even short children can see objects that are taller than they are. Anyway, by the time they are 12V2 years old, both the median girl and the median boy are over five feet tall. See U. S. Centers for Disease Control and Prevention, Growth Charts (2000). Thus, there is no reason to believe that this regulation does anything to protect minors from *583exposure to tobacco advertising.3 Far from serving a compelling interest, the ban on displays below five feet seems to lack even a minimally rational relationship to any conceivable interest.
There is also considerable reason to doubt that the restrictions on cigar and smokeless tobacco outdoor advertising promote any state interest. Outdoor advertising for cigars, after all, is virtually nonexistent. Cigar makers use no billboards in Massachusetts, and in fact their nationwide outdoor advertising budget is only about $50,000 per year. See 218 F. 3d, at 49. To the extent outdoor advertising exists, there is no evidence that it is targeted at youth or has a significant effect on youth. The Court of Appeals focused on the State’s evidence of a relationship between “tobacco advertising and tobacco use,” id., at 48, thus eliding the dearth of evidence showing any relationship between cigar advertising and cigar use by minors. Respondents principally rely on a National Cancer Institute report on cigar smoking, see Brief for Respondents 39, n. 19. But that report contains only the conclusory assertion that cigars are being “heavily promoted in ways likely to influence adolescent use,” and it does not even discuss outdoor advertising, instead focusing on “[endorsements by celebrities,” “the re*584surgence of cigar smoking in movies,” and “cigar lifestyle magazines such as ‘Cigar Aficionado.’” National Cancer Institute, Cigars: Health Effects and Trends, Smoking and Tobacco Control Monograph No. 9, pp. 14-15 (1998), Record, Doc. No. 39, Exh. 67. The report candidly acknowledges that “[additional information is needed to better characterize marketing efforts for cigars” and “to learn the extent to which advertising and promotion for cigars . . . reaches and affects kids.” Id., at 216-217. In other words, respondents have adduced no evidence that a ban on cigar advertising will do anything to promote their asserted interest.
Much the same is true of smokeless tobacco. Here respondents place primary reliance on evidence that, in the late 1960’s, the U. S. Smokeless Tobacco Company increased its sales through advertising targeted at young males. See Brief for Respondents 39, n. 19. But this does nothing to show that advertising affecting minors is a problem today. The Court invokes the Food and Drug Administration’s findings, see ante, at 559-560, but the report it cites based its conclusions on the observed “very large increase in the use of smokeless tobacco products by young people.” 60 Fed. Reg. 41318 (1995). This premise is contradicted by one of respondents’ own studies, which reports a large, steady decrease in smokeless tobacco use among Massachusetts high school students during the 1990’s. See App. 292. This finding casts some doubt on whether the State’s interest in additional regulation is truly compelling. More importantly, because cigarette smoking among high school students has not exhibited such a trend, see ibid., it indicates that respondents’ effort to aggregate cigarettes and smokeless tobacco is misguided.
B
In any case, even assuming that the regulations advance a compelling state interest, they must be struck down because they are not narrowly tailored. The Court is correct, see ante, at 561-563, that the arbitrary 1,000-foot radius demon*585strates a lack of narrow tailoring, but the problem goes deeper than that. A prohibited zone defined solely by circles drawn around schools and playgrounds is necessarily overinclusive, regardless of the radii of the circles. Consider, for example, a billboard located within 1,000 feet of a school but visible only from an elevated freeway that runs nearby. Such a billboard would not threaten any of the interests respondents assert, but it would be banned anyway, because the regulations take no account of whether the advertisement could even be seen by children. The prohibited zone is even more suspect where, as here, it includes all but 10 percent of the area in the three largest cities in the State.
The loose tailoring of the advertising ban is displayed not only in its geographic scope but also in the nature of the advertisements it affects. The regulations define “advertisement” very broadly; the term includes any “written . . . statement or representation, made by” a person who sells tobacco products, “the purpose or effect of which is to promote the use or sale of the product.” 940 Code of Mass. Regs. § 21.03 (2000). Almost everything a business does has the purpose of promoting the sale of its products, so this definition would cover anything a tobacco retailer might say. Some of the prohibited speech would not even be commercial. If a store displayed a sign promoting a candidate for Attorney General who had promised to repeal the tobacco regulations if elected, it probably would be doing so with the long-term purpose of promoting sales, and the display of such a sign would be illegal.
Even if the definition of “advertisement” were read more narrowly so as to require a specific reference to tobacco products, it still would have Draconian effects. It would, for example, prohibit a tobacconist from displaying a sign reading “Joe’s Cigar Shop.” The effect of this rule is not to make cigars impossible to find; retailers are after all allowed to display a 576-square-inch black-and-white sign reading “Tobacco Products Sold Here.” §22.06(6). Rather, it is to *586make individual cigar retailers more difficult to identify by making them change their names. Respondents assert no interest in cigar retailer anonymity, and it is difficult to conceive of any other interest to which this rule could be said to be narrowly tailored.
The regulations fail the narrow tailoring inquiry for another, more fundamental reason. In addition to examining a narrower advertising ban, the State should have examined ways of advancing its interest that do not require limiting speech at all. Here, respondents had several alternatives. Most obviously, they could have directly regulated the conduct with which they were concerned. See, e. g., Rubin v. Coors Brewing Co., 514 U. S. 476, 490-491 (1995) (invalidating ban on disclosure of alcohol content on beer labels, in part because the Government could have pursued alternatives such as “directly limiting the alcohol content of beers”); see also 44 Liquormart, 517 U. S., at 524 (Thomas, J., concurring in part and concurring in judgment) (“[I]t would seem that directly banning a product (or ... otherwise restricting its sale in specific ways) would virtually always be at least as effective in discouraging consumption as merely restricting advertising”). Massachusetts already prohibits the sale of tobacco to minors, but it could take steps to enforce that prohibition more vigorously. It also could enact laws prohibiting the purchase, possession, or use of tobacco by minors. And, if its concern is that tobacco advertising communicates a message with which it disagrees, it could seek to counteract that message with “more speech, not enforced silence,” Whitney v. California, 274 U. S. 357, 377 (1927) (Brandeis, J., concurring).
Ill
Underlying many of the arguments of respondents and their amici is the idea that tobacco is in some sense sui gene-ris — that it is so special, so unlike any other object of regulation, that application of normal First Amendment principles should be suspended. See, e. g., Brief for Respondents 50 *587(referring to tobacco use as “one of the State’s — and indeed the Nation’s — most urgent problems”); Brief for United States as Amicus Curiae 19-20 (cataloging the prevalence and the effects of tobacco use); Brief for American Medical Association et al. as Amici Curiae 24 (advocating “the authority of governments to protect children from uniquely dangerous messages”). Smoking poses serious health risks, and advertising may induce children (who lack the judgment to make an intelligent decision about whether to smoke) to begin smoking, which can lead to addiction. The State’s assessment of the urgency of the problem posed by tobacco is a policy judgment, and it is not this Court’s place to second-guess it. Nevertheless, it seems appropriate to point out that to uphold the Massachusetts tobacco regulations would be to accept a line of reasoning that would permit restrictions on advertising for a host of other products.
Tobacco use is, we are told, “the single leading cause of preventable death in the United States.” Brief for United States as Amicus Curiae 19. The second largest contributor to mortality rates in the United States is obesity. Koplan & Dietz, Caloric Imbalance and Public Health Policy, 282 JAMA 1579 (1999). It is associated with increased incidence of diabetes, hypertension, and coronary artery disease, ibid., and it represents a public health problem that is rapidly growing worse. See Mokdad et al., The Spread of the Obesity Epidemic in the United States, 1991-1998, 282 JAMA 1519 (1999). Although the growth of obesity over the last few decades has had many causes, a significant factor has been the increased availability of large quantities of high-calorie, high-fat foods.' See Hill, Environmental Contributions to the Obesity Epidemic, 280 Science 1871 (1998). Such foods, of course, have been aggressively marketed and promoted by fast food companies. See Nestle & Jacobson, Halting the Obesity Epidemic, U. S. Dept. of Health and Human Services, 115 Public Health Reports 12, 18 (2000).
*588Respondents say that tobacco companies are covertly targeting children in their advertising. Fast food companies do so openly. See, e.g., Kramer, McB’s Steals Another Toy from BK, Advertising Age, Nov. 15, 1999, p. 1 (describing a McDonald’s promotional campaign); Lucas, BK Takes Choice Message to Kids, Adweek, June 29, 1998, p. 4 (describing a Burger King promotional campaign). Moreover, there is considerable evidence that they have been successful in changing children’s eating behavior. See Borzekowski & Robinson, The 30-Second Effect, 101 J. Am. Dietetic Assn. 42 (2001); Taras, Sallis, Patterson, Nader, & Nelson, Television’s Influence on Children’s Diet and Physical Activity, 10 J. Dev. & Behav. Pediatrics 176 (1989). The effect of advertising on children’s eating habits is significant for two reasons. First, childhood obesity is a serious health problem in its own right. Troiano & Flegal, Overweight Children and Adolescents, 101 Pediatrics 497 (1998). Second, eating preferences formed in childhood tend to persist in adulthood. Birch & Fisher, Development of Eating Behaviors Among Children and Adolescents, 101 Pediatrics 539 (1998). So even though fast food is not addictive in the same way tobacco is, children’s exposure to fast food advertising can have deleterious consequences that are difficult to reverse.
To take another example, the third largest cause of preventable deaths in the United States is alcohol. McGinnis & Foege, Actual Causes of Death in the United States, 270 JAMA 2207,2208 (1993). Alcohol use is associated with tens of thousands of deaths each year from cancers and digestive diseases. Id., at 2208-2209. And the victims of alcohol use are not limited to those who drink alcohol. In 1996, over 17,000 people were killed, and over 321,000 people were injured, in alcohol-related car accidents. U. S. Dept, of Justice, Alcohol and Crime 13 (1998). Each year, alcohol is involved in several million violent crimes,. including almost 200,000 sexual assaults. Id., at 3-4.
*589Although every State prohibits the sale of alcohol to those under age 21, much alcohol advertising is viewed by children. Federal Trade Commission, J. Evans & R. Kelly, Self-Regulation in the Alcohol Industry (Sept. 1999); Grube & Wallack, Television Beer Advertising and Drinking Knowledge, Beliefs, arid Intentions among Schoolchildren, 84 Am. J. Pub. Health 254 (1994). Not surprisingly, there is considerable evidence that exposure to alcohol advertising is associated with underage drinking. See Atkin, Survey and Experimental Research on Effects of Alcohol Advertising, in The Effects of the Mass Media on the Use and Abuse of Alcohol 39 (S. Martin ed. 1995); Madden & Grube, The Frequency and Nature of Alcohol and Tobacco Advertising in Televised Sports, 1990 through 1992, 84 Am. J. Pub. Health 297 (1994).
Like underage tobacco use, underage drinking has effects that cannot be undone later in life. Those who begin drinking early are much more likely to become .dependent on alcohol. Indeed, the probability of lifetime alcohol dependence decreases approximately 14 percent with each additional year of age at which alcohol is first used. Grant & Dawson, Age at Onset of Alcohol Use and its Association with DSM-IV Alcohol Abuse and Dependence, 9 J. Substance Abuse 103, 108 (1997). And obviously the effects of underage drinking are irreversible for the nearly 1,700 Americans killed each year by teenage drunk drivers. See National Highway Traffic Safety Administration, 1998 Youth Fatal Crash and Alcohol Facts.
Respondents have identified no principle of law or logic that would preclude the imposition of restrictions on fast food and alcohol advértising similar to those they seek to impose on tobacco advertising. Cf. Tr. of Oral Arg. 56-57. In effect, they seek a “vice” exception to the First Amendment. No such exception exists. See 44 Liquormart, 517 U. S., at 513-514 (opinion of Stevens, J., joined by Kennedy, Thomas, and Ginsburg, JJ.). If it did, it would have almost no limit, for “any product that poses some threat to *590public health or public morals might reasonably be characterized by a state legislature as relating to Vice activity.’ ” Id., at 514. That is why “a ‘vice’ label that is unaccompanied by a corresponding prohibition against the commercial behavior at issue fails to provide a principled justification for the regulation of commercial speech about that activity.” Ibid.
No legislature has ever sought to restrict speech about an activity it regarded as harmless and inoffensive. Calls for limits on expression always are made when the specter of some threatened harm is looming. The identity of the harm may vary People will be inspired by totalitarian dogmas and subvert the Republic. They will be inflamed by racial demagoguery and embrace hatred and bigotry. Or they will be enticed by cigarette advertisements and choose to smoke, risking disease. It is therefore no answer for the State to say that the makers of cigarettes are doing harm: perhaps they are. But in that respect they are no different from the purveyors of other harmful products, or the advocates of harmful ideas. When the State seeks to silence them, they are all entitled to the protection of the First Amendment.
Other regulations prohibit the sale of tobacco products “in any manner other than in a direct, face-to-face exchange,” forbid self-service displays, and require that tobacco products be accessible only to store personnel. See §§21.04(2)(a), (c)-(d), §§22.06(2)(a), (c)-(d). In addition, they prohibit sampling and promotional giveaways. See §§21.04(1), 22.06(1). I agree with the Court, see ante, at 567-570, that these regulations, which govern conduct rather than expression, should be upheld under the test of United States v. O’Brien, 391 U. S. 367'(1968).
Tobacco advertising provides a good illustration. The sale of tobacco products is the subject of considerable political controversy, and not surprisingly, some tobacco advertisements both promote a product and take a stand in this political debate. See Brief for National Association of Convenience Stores as Amicus Curiae 20-22. A recent cigarette advertisement, for example, displayed a brand logo next to text reading, “Why do politicians smoke cigars while taxing cigarettes?” App. to Brief for National Association of Convenience Stores as Amicus Curiae 2a.
This is not to say that the regulation does nothing at all. As the Court points out, see ante, at 565, security concerns require that convenience stores be designed so that the interior of the store is visible from the street. See also Occupational Safety and Health Administration, Recommendations for Workplace Violence Prevention Programs in Late-Night Retail Establishments 6 (1998) (“Shelves should be low enough to assure good visibility throughout the store”). The §21.04(5)(b) ban on displays below five feet and the §21.04(5)(a) ban on displays visible from outside the store, combined with these security concerns, would prevent many convenience stores from displaying any tobacco products at all. Thus, despite the State’s disclaimers, see Brief for Respondents 30 (“The State, quite clearly, is not trying to suppress altogether the communication of product information to interested consumers”), the restrictions effectively produce a total ban.