After affirming our prior judgment in part and reversing in part, the United States Supreme Court remanded this case for a determination of whether the California Dental Association’s advertising restrictions are anticompetitive under rule-of-reason analysis. Having closely examined the record under the rule of reason, we conclude that the Federal Trade Commission failed to prove that the restrictions are anticompetitive. We therefore vacate and remand with instruction that the Commission dismiss its case against the Association.

Petitioner, the California Dental Association (“CDA”), is a trade association for California dentists that is a part of the American Dental Association (“ADA”). CDA itself is composed of 32 local “component societies.” Individual dentists in California must be a member of a component society to belong to CDA, and must have CDA membership to join the ADA. CDA membership is not a condition to obtaining a dentist’s license from the State of California, but about 19,000 of the 26,000 licensed dentists in California belong to the association. CDA is organized as a nonprofit corporation under California law and qualifies for nonprofit status under I.R.C. § 501(c)(6).

CDA provides its members a variety of services, including lobbying, marketing and public relations on behalf of member dentists, seminars on practice management, assistance in compliance with OSHA and disability requirements, continuing education, placement services, an administrative procedure for handling patient complaints, and various publications. It also has several for-profit subsidiaries from which members can obtain liability insurance, financing for equipment and other purchases, discounts on long distance calling, auto leasing, and home mortgages.

Dentists must abide by the CDA Code of Ethics as a condition of membership. The Code provides that a dentist may be disciplined for unprofessional conduct as defined in the state’s Dental Practice Act and for violating any state law relating to the practice of dentistry. The Code goes on to provide more specific ethical standards in various areas of practice. Most relevant to this case is section 10, which governs advertising. It states:

To aid in interpreting this provision, CDA’s Judicial Council, which is responsible for enforcing the Code, has released several advisory opinions. According to the Code, these opinions are not binding on member dentists but “may be considered as persuasive by the trial body and any disciplinary proceedings under the CDA Bylaws.” The following are the advisory opinions most relevant here:

These guidelines substantially mirror parts of the California Business and Professions Code. See Cal. Bus. & Prof.Code §§ 651, 1680. CDA claims that its Code, along with the advisory opinions, is intended to ensure that dentists comply with these laws.

CDA has also issued a separate set of advertising guidelines intended to help members comply with the Code of Ethics and state law. According to the section on discount advertising, state law requires dentists offering discounts to list all of the following in the advertisement:

Both CDA and its component societies on their own enforce the advertising rules of the Code of Ethics. As a general matter, components undertake the initial investigation into a member’s advertising and, if possible, resolve the matter at the local level. Typically, if the component’s ethics committee concludes that a member’s advertising is false or misleading in violation of CDA’s Code of Ethics, it asks the member to discontinue or modify the advertisement. If the matter cannot be resolved or the component is unsure of how to apply the relevant standard under the Code, the case is referred to CDA’s Judicial Council, which holds a hearing. If a violation is found, and no settlement is reached, CDA can impose penalties ranging from censure to expulsion.

CDA and its components also review the advertisements of applicants for membership. If the applicant does not agree to discontinue noncomplying advertising and the component considers denying the application for that reason, it can refer the case to CDA’s Membership Application Review Subcommittee (known as “MARS”). After reviewing the advertising, MARS will recommend to the component that it grant or deny membership. In some cases, applicants with noncomplying advertising can be offered a form of conditional admission under which they must change their ads within a year.

In its complaint against CDA (the component societies were not made parties to this action), the Federal Trade Commission (“FTC”) staff alleges that the CDA applied these advertising guidelines in a way that impermissibly restricted truthful, nondeceptive advertising in violation of section 5 of the FTC Act, 15 U.S.C. § 45 . After a trial, Administrative Law Judge (“ALJ”) Lewis Parker determined that CDA had barred members from making representations of “low” or “reasonable” or “affordable” prices. He also found that CDA effectively prohibited across-the-board discounts by requiring dentists to post the nondiscounted price for all of the services subject to the discount. As enforced, he determined, these policies barred forms of price advertisement without regard to whether they were false or misleading. Finally, the ALJ found that CDA limited various forms of “quality” advertising regardless of truth or falsity. In particular, CDA objected to quality claims of any kind because they might be read to imply superiority over other dentists and were unverifiable. The association also deemed guarantees and attempts to allay patients’ fears, through such language as “gentle, quality care,” and “special care for cowards,” to be misleading, although the policy appears to have been relaxed on “gentleness” claims. The ALJ noted that CDA’s components had engaged in similar behavior, although they were not charged in the complaint with violating the FTC Act.

The ALJ concluded that the FTC had jurisdiction over CDA’s activities and CDA had conspired with its members and component societies to restrict advertising. Applying the FTC’s decision in In re Massachusetts Board of Registration in Optometry, 110 F.T.C. 549 (1988) (“Mass.Board ”), he held that the advertising restrictions were inherently suspect and had no plausible efficiency justification. Although he also found that CDA lacked market power, he held that a showing of market power was not necessary under Mass. Board. He concluded that CDA had unreasonably restrained competition in violation of section 5 of the FTC Act.

The majority of the Commission affirmed, on somewhat different reasoning. Chairman Pitofsky’s opinion held that the restrictions on price advertising were unlawful per se. It further held that the nonprice advertising guidelines were unlawful under an abbreviated rule-of-reason analysis. In so doing, it disagreed with the ALJ and found that CDA did possess market power. The Commission found as a matter of fact that CDA has applied its Code of Ethics so as to prohibit its members from making three kinds of advertising claims: (1) characterizations of a dentist’s prices as low, reasonable, or affordable; (2) across-the-board discounts on dental services; and (3) service quality claims by dentists. Commissioner Starek concurred in the result but would have applied the Mass. Board decision rather than a per se or abbreviated rule-of-reason analysis. Commissioner Azcuenaga dissented, finding no evidence of a sufficient pattern of anticompetitive acts to hold CDA liable and no evidence of market power.

On October 22, 1997, this Court affirmed the Commission. See California Dental Ass’n v. FTC, 128 F.3d 720 (9th Cir.1997). We held that the FTC properly exercised jurisdiction over CDA. While we held that the Commission erred by applying per se analysis to CDA’s restrictions, we nevertheless found CDA’s restrictions to be unreasonable under abbreviated rule-of-reason analysis. In choosing to apply abbreviated rule-of-reason analysis, we incorrectly stated that “the record provides no evidence that the [advertising restrictions have] in fact led to increased disclosure and transparency of dental pricing.” Id. at 728 . We also found, under abbreviated rule-of-reason analysis, that CDA exercises market power. The pivotal findings for the panel were the Commission’s conclusions that CDA prevented its members from engaging in truthful, non-misleading advertising offering across-the-board discounts or claims about service quality. Judge Real dissented from our opinion on the grounds that the FTC lacked jurisdiction over CDA and that the majority erred by applying abbreviated rule-of-reason analysis. Judge Real concluded that the panel should have applied rule-of-reason analysis, and indicated that he would have upheld the restrictions under the rule of reason.

The Supreme Court granted certiorari and, on May 24, 1999, affirmed in part and reversed in part. By a 9 to 0 vote, the Court affirmed the panel’s conclusion that the FTC properly exercised its jurisdiction over CDA. By a 5 to 4 vote, the Court held that the panel erred by applying abbreviated rule-of-reason analysis. The majority opinion held that CDA’s advertising restrictions are not such an obvious restraint on trade to render abbreviated rule-of- reason analysis appropriate. Specifically, it faulted us for failing to consider a number of theories under which the restrictions might well prove procompetitive and for our “adversión to empirical evidence” in favor of burden-shifting. The Court vacated the panel opinion and remanded “for a fuller consideration of the issue.”

On remand, we will apply a more extensive rule-of-reason analysis to determine whether substantial evidence supports the Commission’s conclusion that CDA’s restrictions are anticompetitive. See California Dental Ass’n v. FTC, 526 U.S. 756 , 769 n. 8, 119 S.Ct. 1604 , 143 L.Ed.2d 935 (1999).

II.

The Supreme Court held that we erred by applying “quick look”/abbreviated rule-of-reason analysis, as opposed to a more empirically rigorous rule-of-reason analysis. Id. at 781 , 119 S.Ct. 1604 . At the same time, the Court noted that the instant case does “not ... necessarily ... call for the fullest market analysis,” and called for a “sliding scale” formula that looks “to the circumstances, details, and logic of a restraint.” Id. at 779, 781 , 119 S.Ct. 1604 . Seeking to situate our inquiry somewhere on the rule-of-reáson continuum between abbreviated and full-blown, we employ a level of inquiry closer to the latter, bearing in mind that our ultimate task is to determine whether the challenged restraint enhances competition. See id. at 780 , 119 S.Ct. 1604 . In particular, we must determine whether, on balance, CDA’s restrictions on advertising are procompetitive or anticompetitive. See FTC v. Indiana Fed. of Dentists, 476 U.S. 447, 458 , 106 S.Ct. 2009 , 90 L.Ed.2d 445 (1986); Bhan v. NME Hosps., Inc., 929 F.2d 1404, 1413 (9th Cir.1991); CBS v. American Soc’y of Composers, Authors, & Publishers, 620 F.2d 930, 934 (2d Cir.1980). The restrictions qualify as anti-competitive only if they harm “both alloca-tive efficiency and raise[ ] the prices of goods above competitive levels or diminish[ ] their quality.” Rebel Oil Co. v. Atlantic Richfield Co., 51 F.3d 1421, 1433 (9th Cir.1995). Such analysis is rigorous, requiring “a detailed depiction of circumstances and the most careful weighing of alleged dangers and potential benefits.” U.S. Healthcare, Inc. v. Healthsource, Inc., 986 F.2d 589, 594 (1st Cir.1993). “The rule-of-reason analysis consists of three components: (1) the persons or entities to the agreement intend to harm or restrain competition; (2) an actual injury to competition occurs; and (3) the restraint is unreasonable as determined by balancing the restraint and any justifications or pro-competitive effects of the restraint.” American Ad Mgmt. v. GTE Corp., 92 F.3d 781 , 789 (9th Cir.1996).

Intent to Restrain Competition

Our prior opinion held that the CDA constituted an agreement among its members, see 128 F.3d at 728 . We did not dwell on the issue of intent, however, observing that “whatever its motivation, the point of the advertising policy was clearly to limit the types of advertising in which dentists could engage.” Id. at 729 . We then observed that good “motives will not validate an otherwise anticompetitive practice,” citing NCAA v. Board of Regents, 468 U.S. 85 , 101 n. 23, 104 S.Ct. 2948 , 82 L.Ed.2d 70 (1984). This truncated discussion of intent reflects the well-established pattern of the Supreme Court to examine intent only in those close cases where the plaintiff falls short of proving that the defendant’s actions were anticompetitive. See, e.g., Times Picayune Publ’g Co. v. United States, 345 U.S. 594, 614 , 73 S.Ct. 872 , 97 L.Ed. 1277 (1953); United States v. Griffith 334 U.S. 100, 105 , 68 S.Ct. 941 , 92 L.Ed. 1236 (1948). Even then, “an admitted intention to limit competition will not make illegal conduct that we know to be pro-competitive or otherwise immune from antitrust control.” 7 Phillip E. Areeda, Antitrust Law § 1506 (1986). And, while “smoking gun” evidence of an intent to restrain competition remains relevant to the court’s task of discerning the competitive consequences of a defendant’s actions, “ambiguous indications of intent do not help us ‘predict [the] consequences [of a defendant’s acts]’ ” and are therefore of no value to a court analyzing a restraint under the rule of reason, where the court’s ultimate role is to determine the net effects of those acts. Id. Under such circumstances, we apply the rule of reason without engaging in the relatively fruitless inquiry into a defendant’s intent. Cf. Bhan v. NME Hospitals, Inc., 929 F.2d 1404, 1410, 1412-13 (9th Cir.1991) (applying the rule of reason without delving into intent).

The primary piece of evidence marshaled by the FTC concerning the CDA’s intent is a statement made by a former CDA president in 1976. The statement in question reads as follows:

The last sentence in the paragraph does evince a desire on the part of CDA’s president to limit fee-based competition for patients. But as best the record indicates, CDS had nothing to do with advertising restrictions. Rather, it appears to have been a CDA-administered fee-setting arrangement between dentists and insurance companies, which presumably would have had a much more direct effect on price competition among dentists than advertising restrictions. The 1976 statement is therefore far too ambiguous to alter the rule of reason inquiry.

We do see substantial evidence that the CDA intended to restrict certain types of advertising, but in light of the CDA’s plausibly procompetitive justifications for the restrictions, such intent has no bearing on the question of whether those restrictions are in fact likely to prove anticompetitive or procompetitive. And the record reveals no unambiguous evidence that the CDA intended to restrain trade, so further analysis of CDA’s intent becomes superfluous. See generally 7 Areeda, supra, § 1506 (“Intention is often superfluous to the analysis of reasonableness, for it adds nothing to the conduct from which it is usually inferred.”). Under such circumstances, intent “drops out” of our rule-of-reason inquiry, just as it did in our prior opinion, and the case hinges on the actual economic consequences of the CDA’s restrictions.

Actual Injury to Competition

Our prior opinion held that substantial evidence supports a conclusion that the advertising restrictions at issue constituted an actual injury to competition. See 128 F.3d at 729 . Although we must now apply a more rigorous rule-of-reason analysis, that earlier determination should not be disturbed. As we explain below, although we are convinced that the restrictions’ pro-competitive benefits exceed their anticom-petitive harms, we are able to identify anticompetitive harms resulting from the restrictions.

Are the Restrictions Procompetitive?

The Supreme Court’s opinion focuses on the question of whether the presumptive economic benefits resulting from CDA’s advertising restrictions outweigh their economic harms. The Court noted that “it seems to us that the CDA’s advertising restrictions might plausibly be thought to havé?a net procompetitive ef-feet, or possibly no effect at all on competition.” 526 U.S. at 771 , 119 S.Ct. 1604 .

The Court’s opinion instructed us to consider with much greater care the potentially procompetitive justifications for CDA’s advertising policies. Specifically, the Court pointed to several aspects of the advertising restrictions that might cause them to have a net procompetitive effect:

See id. at 771-81 , 119 S.Ct. 1604 . We therefore must reevaluate in light of these considerations our earlier conclusion that the restrictions fail under abbreviated rule-of-reason analysis because they prevent truthful advertising concerning across-the-board discounts and service quality.

III.

We now consider the record evidence bearing on the six factors identified above.

Information Asymmetries

The parties have briefed the question of informational asymmetries extensively and the record contains a great deal of evidence bearing on that dynamic. We conclude from the record that dentists are far better at evaluating the quality of dental services than patients are, i.e., that there is an informational asymmetry in the market. See, e.g., ER5R:14 (“With respect to the asymmetry of information, dentists are clearly in a position to know more about the product or the service than consumers .... [Consumers are not in a position to evaluate in advance the nature of the service. And even after experiencing it [they] may have difficult[y] in evaluating [quality of care].”); ER5P:3 (“[W]hen it comes to quality of care most individuals are not able to discern what is or is not quality.”).

The record also contains evidence suggesting that there is an information asymmetry with respect to cost. A dentist can determine what his fellow dentists charge for services more easily than a consumer can. ER5R:16 (“The dentist certainly is in possession of the information required for verification, and could provide that information at the lower cost.”). The FTC concedes that “individual dentists can obtain information about the prices charged by dentists in their area.” The difficult question, then, is not whether information asymmetries exist, but whether CDA’s advertising restrictions mitigate the market inefficiencies that result from these asymmetries.

The FTC argues that even in markets for professional services, where informational asymmetries abound, broad advertising restrictions have raised prices for consumers. In so doing, the FTC relies on a number of scholarly articles, See Carolyn Cox & Susan Foster, The Costs and Benefits of Occupational Regulation (1990); Lee Benham, The Effect of Advertising on the Price of Eyeglasses, 15 J.L.

& Econ. 387 (1972); James A. Langenfeld & John R. Morris, Analyzing Agreements Among Competitors: What Does the Future Hold?, 36 Antitrust Bull. 651 (1991); Phillip Nelson, Advertising as Information, 82 J. Pol. Econ. 729 (1974); John R. Schroeter et al., Advertising and Competition in Routine Legal Service Markets: An Empirical Investigation, 36 J. Indus. Econ. 49 (1987). Both parties recognize that the Cox and Foster study, which summarizes and analyzes the empirical literature, is the scholarship most relevant to our inquiry. Indeed, Cox and Foster devote a fair amount of space to discussing all of the studies cited above except for the subsequently published Langenfeld and Morris article. Cox & Foster’s characterization of the existing research is as follows:

Cox & Foster, supra, at 25 (footnotes omitted) (emphasis in original). Thus, the authors concluded that the majority of the empirical evidence indicates that restrictions that are (broadly speaking) like those at issue here do not affect quality of care.

Subsequently, Cox and Foster describe the findings of particular studies. They note that in the optometry market, “restrictions [that] prevent both advertising and limit commercial practice” result in higher prices. Id. at 32. But it is undisputed that CDA does not prevent dentists from advertising; nor do they limit dentists’ commercial practice. And, as the Benham study, which Petitioners also cite, concluded, “[e]yeglasses may of course be a special case,” and further research was needed before results from the eyeglasses market could “be generalized to other goods.” Benham, supra, at 352. Thus, the optometry market evidence is of ex tremely limited value in helping us discern the economic effects of CDA’s restrictions.

Cox and Foster then discuss a study that found that the price of legal services was higher in cities that impose time, place, and manner restrictions on legal advertising. See Cox & Foster, supra, at 33. They also note that studies have found adverse effects on consumers stemming from restrictions on advertising in the legal profession. See id. But, as one of these studies observed, “states adopted widely different attorney advertising regulations,” with some states allowing only the most minimal advertising and others prohibiting only false or misleading advertising. Schroeter et ah, supra, at 54 n. 13. Presumably, such widely divergent regulations will engender widely different market consequences. Cox and Foster do not discuss any study that has looked at the effects of only those advertising restrictions that are substantially similar to CDA’s, and the FTC does not point us to any such study.

Later in their analysis, Cox and Foster turn to the empirical evidence concerning the dental services market. Their discussion cites no empirical evidence concerning dental advertising restrictions substantially similar to those enacted by CDA. Cox and Foster do note that “retail dentists attract customers by offering their services at lower prices,” and that “restrictions that prevent discounts and expanded hours may discourage some individuals to seek care who would otherwise do so.” Cox & Foster, supra, at 35-36. But neither conclusion helps us discern the effect of the advertising restrictions at issue here. While the record clearly indicates that some dentists compete by offering discounts, the restrictions at issue here by no means prevent dentists from doing so, nor do they regulate dentists’ hours of practice. In order to conclude that advertising restrictions are at all implicated by Cox and Foster’s conclusion, one has to first conclude that CDA’s advertising restrictions raise dental prices. But if prices rise, then the harm to consumers is clear. As applied to the instant litigation, therefore, Cox and Foster’s analysis of the dental services market simply begs the question.

Thus, the only relevant empirical evidence the FTC musters reveals that time, place, and manner restrictions on legal advertising raise the prices of legal services. But Cox and Foster themselves caution against generalizing from one professional industry to another:

Id. at 41 (emphasis added). So even the FTC’s strongest evidence that CDA’s advertising restrictions raise dental prices in California is significantly weakened by the authors of that evidence. While we reject CDA’s argument that the FTC must produce empirical evidence concerning the precise dental advertising restrictions at issue in California or a neighboring state, our rule-of-reason case law usually requires the antitrust plaintiff to show some relevant data from the precise market at issue in the litigation — dental services in this case. See American Ad Mgmt. v. GTE Corp., 92 F.3d 781 , 789-90 (“Proving injury to competition in a rule of reason case almost uniformly requires a claimant to prove the relevant market and to show the effects of competition within that market. ... Accordingly, the district court was correct in requiring proof of the relevant geographic and product markets, as well as proof on the effects within these markets.”) (quoting Oltz v. St. Peter’s Community Hosp., 861 F.2d 1440, 1445 (9th Cir.1988)). The Supreme Court has also cautioned against reaching general conclusions about the economics of advertising with respect to all professional markets based on data from only one professional market. See Bates v. State Bar of Ariz., 433 U.S. 350 , 366 n. 17, 97 S.Ct. 2691 , 53 L.Ed.2d 810 (1977) (noting that “the distinctions, historical and functional, between professions, may require consideration of quite different factors” when considering advertising regulation). Thus, we conclude that the social science evidence cited by the FTC does not constitute substantial evidence of the anticompetitive nature of CDA’s advertising restrictions.

The FTC offers an additional theory of how consumers might be harmed by the advertising restrictions. As the FTC states in its brief, “advertising can attract patients who might otherwise forego dental services if they were not aware of relevant price or quality information.... Thus, advertising restrictions could have the effect of reducing output, depriving consumers of services they would have purchased if they had known about them.” Our prior opinion accepted this theory. See 128 F.3d at 728 (“The restrictions may also affect output more directly, as quality and comfort advertising may induce some customers to obtain nonemergency care when they might not otherwise do so.”) But the Supreme Court unambiguously rejected our conclusion for incorrectly examining factors influencing the consumer demand curve. 526 U.S. at 776-77 , 119 S.Ct. 1604 . The FTC’s effort to characterize the Court’s analysis as something other than a rejection of the demand-side antitrust theory proposed again here is unpersuasive. Therefore, the FTC’s argument that the restrictions at issue here exacerbate informational asymmetries in the dental market is unconvincing.

CDA, on the other hand, has made a strong case that the advertising restrictions at issue here correct for some of the informational asymmetries inherent in the market for dental services. At trial before the ALJ, CDA called Professor Robert Knox, an expert in economics and industrial organization. Knox testified that he believes CDA’s policies are procompetitive in that they prevent “buyers from getting mistaken impressions about information contained in advertisements, and therefore arms them with more accurate and verifiable information; makes them better able to search for their particular value.” ER 5R:21. In other words, the policies correct for information asymmetries by requiring that dentists fully disclose information about price or quality. For example, the restrictions would force a dentist wishing to say “cleanings discounted to $75” to disclose whether the normal price charged for cleanings is $76 or $120, thereby giving the consumer a much better idea of how much he is saving. The policies might also restrict search costs in the following manner: Say one dentist offers a twenty dollar discount on bridge work for new patients and another advertises that her new patient discounts for bridge work are fifteen percent. It appears that CDA’s policies would require both dentists to disclose the regular and discounted rates, thereby allowing a price-conscious consumer to determine from the ads which of the two dentists is actually offering a lower fee: So conceived, the restrictions create a kind of network externality by mandating a common language to be used by those CDA members who advertise discounts. As a result, a consumer’s costs of searching for the less expensive service would be re duced. Lower search costs for consumers are generally understood to be procompet-itive. See, e.g., Langenfeld & Morris, supra, at 666; Note, Fixing the Price Fixing Confusion: A Rule of Reason Approach, 92 Yale. L.J. 706, 718 (1988). We are therefore persuaded that CDA’s restrictions do mitigate some of the informational asymmetries that exist in the market for dental services.

Consumer Loyalty

The market for professional care may be different from other markets in that consumers are more loyal to their professionals than they are to, say, gasoline stations. The Supreme Court conceived of this as a factor that further complicated the antitrust analysis that this Court must undertake. See 526 U.S. at 772-73 , 119 S.Ct. 1604 . We understand the Court’s .guidance as encouraging us to consider the view that restricting advertising in the dental market may be much less detrimental than restricting advertising in a market where consumers are much more likely to switch brands. It may be that the type of advertising that is barred by CDA (for example, unverifiable quality claims) would be ineffective at overcoming consumer inertia in selecting their dentists.

The only evidence the parties have brought to our attention that even tangentially concerns this point is the aforementioned article written by Professor Nelson. Nelson distinguished between “search goods,” for which most of a product’s qualities can be determined prior to purchase, and “experience goods,” for which most qualities usually cannot be determined pri- or to purchase. See Nelson, supra, at 780. Under Nelson’s dichotomy, dental services are almost purely experience goods, and a new dress is an example of a predominantly search good. Nelson concludes that advertising will give consumers more guidance in buying search goods than in purchasing experience goods. See id. at 747-48. By contrast, word of mouth will be more useful to consumers when making decisions about experience goods. See id. On the other hand, advertising will be more accurate where, as here, sellers depend on repeat business by customers. See id. at 730-31. Having pondered this evidence, we conclude that an analysis of the consequences of consumers’ heightened loyalty in the dental market therefore does not cut in either party’s favor.

Partial Versus Complete Advertising Ban

The Supreme Court faulted our earlier opinion for failing to recognize that “the restrictions at issue here are very far from a total ban on price or discount advertising” and for failing to consider “that the particular restrictions on professional advertising could have different effects from those ‘normally’ found in the commercial world, even to the point of promoting competition by reducing the occurrence of unverifiable and misleading across-the-board discount advertising.” 526 U.S. at 773-74 , 119 S.Ct. 1604 .

As was noted, supra, the bulk of the FTC’s economic analysis was based on empirical studies of complete bans on advertising within the relevant markets. While it seems clear that an absolute ban on advertising by dentists would indeed prove anticompetitive, that evidence may be of little use to us in analyzing the instant case under rule of reason. The Supreme Court’s analysis, as applicable to the case at bar, primarily seems to be an instruction that, in considering the case on remand, we avoid over-relying on economic analyses and presumptions formed from circumstances involving total advertising bans.

If CDA’s advertising restrictions amount to something approximating an absolute ban on price advertising it may be appropriate to treat them as an absolute ban. An examination of the testimonial evidence cited by the FTC in support of this point, however, leaves us convinced that substantial evidence does not support such a conclusion. By and large, the witnesses whose testimony the FTC used to argue that the restrictions amount to a complete ban did not actually testify to that effect. For example, the testimony cited from Dr. Cowan cited in the FTC’s brief is taken out of context. Cowan explicitly states that CDA does not require dentists to list fee schedules for each and every service, because doing so would be unreasonable. He also states that CDA would only ask the dentist to list a representative sample of his fees. SER 543. The testimony cited from Dr. Kinney likewise appears to have been quoted out of context in the FTC’s brief. Admittedly, the FTC does accurately present the testimony of Dr. Miley, a dentist practicing in California, who states that it is impossible to advertise any price discounts in a manner that complies with CDA’s guidelines. However, Dr. Miley’s statement is contradicted by a great deal of record evidence suggesting that CDA allows some discount-related advertising. Indeed, the FTC concedes as much in its brief, writing: “Of course, CDA’s rules would permit dentists to offer some discount information, such as a discount on a small number of services, which could practicably be listed in a single advertisement.”

The FTC also asserts that the advertising restrictions essentially bar dentists from advertising discounts for the most costly dental services, such as complex surgical procedures, which entail multiple services over a period of time. Revealingly, however, there are no record citations in this portion of the FTC’s brief. Indeed, we are hard pressed to imagine why a dentist might not be able to offer a discount deal for a specific service while remaining within CDA’s guidelines. For example, it appears that a dentist could advertise: “Root canals formerly $7500, now $6000” without violating the guidelines on their face or as enforced.

The FTC’s most plausible argument on this point appears to be that dentists who mention “low prices” or the like in advertisements signal consumers that the advertiser is sensitive to concerns about cost and invites consumers to inquire further about prices. But once again, the portion of the record cited by the FTC, consisting of John Christensen’s testimony, simply does not support the proposition for which it is being offered. Christensen testified that advertising a senior citizen discount will “cause seniors to come in” because they will think they are getting “a better than normal shake.” SER 519-20. Christensen also testified that advertisements that reveal a discounted price are informative, but that advertisements that reveal both the discounted and pre-discount prices, as is required by the CDA’s restrictions, are more useful to consumers. SER 521.

The FTC marshals an overwhelming quantum of evidence indicating that quality of service is a very important variable in patients’ decisions about which dentist to choose. But there is no evidence that dentists who advertise that their services are of high quality in fact offer high-quality service, an assumption upon which the assertion that CDA’s restrictions are anti-competitive at least implicitly relies. Nor do we see evidence indicating that when dentists advertise service quality, any other benefits to consumers result. The Supreme Court’s opinion pointed to language in a prior opinion indicating that claims about the quality of legal services “probably are not susceptible of precise measurement or verification and, under some circumstances, might well be deceptive or misleading to the public, or even false.” Bates, 433 U.S. at 366 , 97 S.Ct. 2691 . Indeed, nothing in the record contradicts our intuition that when a dentist advertises using an empty slogan like “The Best Dentist in the State,” a.reader learns more about that dentist’s lack of modesty than anything else. Notably, there is no evidence that CDA prevents members from making objectively verifiable quality claims in their advertisements. Thus, if an independent, respected publication such as Consumer Reports were to rank the quality of various dentists in a community, we see nothing that would prevent those CDA members who performed well in such a survey from trumpeting their high rankings.

The FTC argues that some objective claims about a dentist’s service quality are readily verifiable by the patient. For example, if a dentist guarantees that a filling or crown will be replaced within a certain amount of time, the consumer can readily evaluate the claim if he finds himself in need of such a replacement. But the FTC does not point to record evidence where a dentist was forbidden from advertising such verifiable guarantees. Indeed, such advertising claims do not appear on their face to fall within the category of “service quality claims” that the ALJ found proscribed by CDA’s restrictions.

Misleading Advertising Driving Consumers Away

The Supreme Court speculated that the use of across-the-board discounts in dental advertising could undermine its own effectiveness. Specifically, the Court noted:

526 U.S. at 774-75 , 119 S.Ct. 1604 . Under this theory, some consumers would become disenchanted after they came to realize that the claims of across-the-board discounts made in certain advertisements were somewhat misleading. For example, a patient who chooses a dentist because she offers 50% percent discounts to new patients might become disenchanted when she realizes that the dentist’s pre-discount rates are three times as high as those of her competitors. The consumer would then learn to disregard across-the-board discount advertising. In the long run, across-the-board advertising would become ineffective at attracting the by-then cynical population of dental patients. Accordingly, banning the use of across-the-board advertising would do no harm.

Although this anticipated dynamic is consistent with Professor Nelson’s aforementioned scholarship concerning experience goods and there is record evidence indicating that “consumers who feel that they have been misled by advertising might transfer their unhappiness to the dental profession as a whole,” ER5R:15, the FTC has not pointed to any evidence in the record that supports the proposition that this scenario has played out with respect to the advertising of across-the-board discounts. Indeed, a market test suggests just the opposite: If across-the-board discount advertising is so ineffective, then one presumes that few dentists would resort to it. But, as the numerous CDA enforcement actions cited by the FTC make clear, dentists do try to use across-the-board discounts to bring in patients, suggesting that they believe the method must be at least somewhat successful. This consideration therefore does not alter our view of the case in either party’s favor.

Do the Restrictions Prevent Consumers from Being Misled?

The Supreme Court also suggested that the use of across-the-board discount advertising might “continue to attract business indefinitely, but might work precisely because they were misleading customers.” 526 U.S. at 775 , 119 S.Ct. 1604 . This theory is the flip-side of the one mentioned above, that across-the-board discount advertising does not work.

The record does contain some evidence that in the absence of restrictions upon dental advertising, consumers can be misled into paying higher prices while thinking their dentist’s charges are a bargain. ER5O:10-ll. The responses of Dr. Curtis P. Hamann, who manages his wife’s extremely successful dental practice, to the following questions were illustrative:

ER5H:15; see also SER519 (quoting the testimony of a dental marketing officer that mentioning the words “senior citizen discount” in an advertisement “will cause seniors to come in, even though they don’t know specifically what the deal is”). These responses provide some support for CDA’s contention that across-the-board discount advertising, unaccompanied by the kind of full-disclosure requirements that CDA has established, may result in the anticompetitive hoodwinking of price-sensitive customers. On the other hand, we do not place too much credence in anecdotal evidence of this nature because we presume that in the long run, the market should self correct this type of failure. Presumably, less expensive dentists competing in the same community as Dr. Hamann’s wife would have an incentive to advertise the absolute prices of their services as a way of informing her customers that they are not getting as much of a bargain as they might think. In any event, to the extent that the restrictions do decrease the likelihood of consumers being fooled in the first instance, we believe the restrictions have some procompetitive benefits.

Relatedly, CDA argues that unverifiable claims in advertisements about dental care quality can alter patients’ perceptions of quality. The record evidence indicates that consumers are not able to evaluate the quality of a dental service, even one that they have just experienced. ER5R:14. Advertising that stresses how gentle a dentist is can actually cause the patient to perceive the dentist to be more gentle than he really is. ER 5E:3-5. Thus, unverifiable advertising may, by making a consumer less discerning, cause a patient to stay with a dentist longer than she otherwise might, thereby reducing the ability of other (possibly superior) dentists to compete for her business. It should be noted, however, that while this argument has some plausibility, CDA has not put forth rigorous psychological evidence supporting its assertion. Rather, it relies on the testimony of John Christensen, who runs a national business that assists dentists wishing to enhance their customer bases. Thus, we view this argument as slight additional support for the procom-petitive tendencies of CDA’s policies.

The Restrictions as a Ban on Puffery

The Supreme Court’s final concern with our prior opinion stems from more of a factual disagreement than a theoretical one. The Court noted that it is entirely “possible to understand the CDA’s restrictions on unverifiable quality and comfort advertising as nothing more than a procompetitive ban on puffery.” 526 U.S. at 778 , 119 S.Ct. 1604 . By puffery the Supreme Court presumably meant those advertising claims that are so obviously exaggerated that consumers would never accept them at face value. See generally Black’s Law Dictionary 1233 (6th ed.1990) (defining “puffing” as “[e]xaggeration by a salesperson concerning quality of goods (not considered a legally binding promise); usually concerns opinions rather than facts” and as “[a]dvertising which merely states in general terms that advertiser’s product is superior is only ‘puffing’ and is not actionable in action by competitor”).

While CDA argues that its policies only prohibit advertisements that it considers false or misleading, CDA does not argue that its policies are no more than a prohibition on puffery. Indeed, to the extent that puffery “would probably be discounted to some extent by consumers, and therefore would [be] likely to have little or no effect [on competition],” as CDA’s expert economist testified, ER5R:17, puffery is much less of a concern to CDA than misleading advertisements that consumers take seriously.

The record similarly belies the conclusion that CDA’s regulations are only an effort to restrict puffery. For example, the Commission found that CDA has prohibited the use of advertising terms such as “gentle, quality care,” and “fast and caring,” and “reasonable fees quoted in advance,” subjective claims that are obviously difficult to verify but which are not so exaggerated that an average consumer can be expected to discredit them.

IV.

In order to prevail under rule-of-reason analysis, the FTC must show that CDA’s restrictions engendered a net harm to competition in the California dental service market. After weighing the evidence summarized in the discussion above, we conclude that the FTC has not met that burden, and that substantial evidence therefore does not support the Commission’s determination.

To recap, the FTC’s strongest record evidence of the advertising restrictions’ anticompetitive nature is the empirical evidence that somewhat comparable time, place, and manner restrictions on legal advertising are thought to raise the price of legal services. But this evidence was held to be of limited cross-profession applicability by the authors of the chief report up on which the FTC relies. Moreover, our case law usually requires the antitrust plaintiff to show some relevant data from the precise market at issue in the litigation — dental services in this case. The FTC proved neither that dentists who advertise lower prices (through methods prohibited by the regulations) in fact offer below-average prices, nor that dentists who advertise the high quality of their services are qualitatively superior to those dentists who do not advertise the quality of their services. Finally, the FTC has never quantified any increase in price or reduction in output of dental services resulting from CDA’s restrictions.

By contrast, CDA’s strongest record evidence that the advertising restrictions’ are procompetitive consists of evidence that: (1) Full disclosure of prices corrects for informational asymmetries between dentists and patients over price; (2) Full disclosure of prices allows for easier comparative shopping by price-conscious consumers; (3) The ban on across-the-board discount advertisements prevents dentists from misleading their patients into believing that their services are a better bargain than they really are; and (4) The restrictions on quality advertising may make it more difficult for dentists to manipulate their patients' assessments of care quality. These empirical arguments are generally well-supported by expert testimony and anecdotal evidence from individual dentists practicing in California.

Such an analysis of the existing record makes it clear that CDA easily gets the better of the factual determination now central to this case — whether the advertising restrictions are net pro— or anticom-petitive. Under rule-of-reason analysis, then, because CDA’s advertising restrictions do not harm consumer welfare, there is no antitrust violation. In other words, the FTC has failed to demonstrate substantial evidence of a net anticompetitive effect.

In light of our holding, we must now consider whether to remand the case so that the Commission can consider additional testimony concerning whether CDA’s restrictions are anticompetitive under the rule of reason. Both parties contended in their briefs that the existing record is sufficient to allow us to evaluate the restrictions under the rule of reason. The FTC argued, however, that if we were to hold that the record does not support their conclusion that the restrictions are anticompetitive under the rule of reason, we should then remand so the Commission can consider additional empirical evidence. At oral argument, counsel for the FTC, with refreshing forthrightness, stated that the complaint counsel who litigated the case below did not feel it was necessary to supplement the existing economic literature cited by the FTC with the testimony of an expert economist witness that the FTC had designated during discovery. According to the FTC, the economic literature cited in the opinion below provides us with an adequate record to evaluate whether substantial evidence supports the Commission’s conclusion that CDA’s policies are anticompetitive. CDA counters that the FTC could have introduced empirical evidence into the record at an earlier stage in the proceedings and points to the inequity of allowing the FTC to have a “second bite at the apple.”

We have found ourselves in similar circumstances once before. In GTE Sylvania Inc. v. Continental T.V., Inc., 537 F.2d 980 (9th Cir.1976) (en banc), the Ninth Circuit en banc majority reversed the district court’s determination that a locations agreement violated the Sherman Act under a per se test. The en banc court held that rule-of-reason analysis should apply. In considering whether to remand or dismiss, the court chose to remand, noting:

Id. at 1000 . Thus, in a case where the antitrust plaintiff failed to prove a net anticompetitive effect, a remand was in order. Upon remand, (after the Supreme Court affirmed the Ninth Circuit), the district court refused to allow the plaintiffs to introduce a new theory of how the economic actions would violate the Sherman Act, excoriating the plaintiff for seeking to litigate the case as a horizontal restraints case after trying it as a vertical restraints case for 13 years of litigation. See Continental T.V., Inc. v. GTE Sylvania Inc., 461 F.Supp. 1046, 1052 (N.D.Cal.1978).

The Continental district court also granted the antitrust defendant summary judgment as a matter of law after applying rule-of-reason analysis to the restriction that had previously been stricken down under per se. See id. at 1052 . By granting summary judgment, the district court prevented the plaintiff from supplementing the existing record via trial testimony. The Ninth Circuit then affirmed the district court’s grant of summary judgment on the basis of the existing record, noting that the Supreme Court’s having required rule-of-reason analysis instead of per se did not entitle the plaintiff to a new trial:

Continental T.V., Inc. v. G.T.E. Sylvania Inc., 694 F.2d 1132, 1135-36 (9th Cir.1982) (footnote & citations omitted). The court brushed aside the plaintiffs argument that a new trial was required:

Id. at 1136 .

In the case at bar, FTC complaint counsel appears to have contemplated the possibility that the Commission would adjudicate the case under the rule of reason. But complaint counsel made a tactical decision not to call a previously designated expert economist to counter the testimony of CDA’s expert economist. Rather, complaint counsel focused on winning the case under per se or abbreviated rule of reason, evidently assuming that the economic literature would suffice to win the case under full-blown rule of reason if the Ninth Circuit or Supreme Court required a more onerous level of analysis.

Moreover, nothing in the proceedings before the FTC prevented complaint counsel from bringing forward additional empirical evidence at trial so that they might prevail even if CDA’s restrictions were evaluated under the rule of reason. The ALJ who tried the case placed no limits on the evidence that complaint counsel could offer. Trial counsel’s decision not to bring further empirical evidence to the FTC’s attention prompted Commissioner Azcue-naga to conclude: “Complaint counsel made no effort to try the case on a rule of reason theory and did not introduce testimony or documents to establish the elements of a rule of reason case.” In re California Dental Ass’n, 121 F.T.C. 190 , 333, 355 (1996) (Azcuenaga, dissenting). Under such circumstances, we agree with CDA that a remand for further factfinding would give the FTC an unwarranted second bite at the apple. Cf. Omni Outdoor Advertising, Inc. v. Columbia Outdoor Advertising, Inc., 974 F.2d 502, 505 (4th Cir.1992) (noting, in a monopolization suit, that “it is not fair for an adversary to have to defend the same lawsuit on appeal over and over.”) We therefore remand with direction that the FTC dismiss its case against CDA.

VI.

For the foregoing reasons, we vacate the judgment of the Commission and remand with instruction to dismiss the case.

VACATED and REMANDED.

. As we explain at length below, our earlier look at the voluminous record before the FTC was much too quick. Upon further review of the record we find substantial evidence that CDA’s restrictions have procompetitive attributes. We particularly regret our error in light of the Supreme Court’s reliance on our assessment of the record. See California Dental Ass’n v. FTC, 526 U.S. 756, 781, 787, 119 S.Ct. 1604 , 143 L.Ed.2d 935 (1999) (Breyer, J., dissenting). . Commentators analyzing the Supreme Court's opinion have emphasized that it does not absolutely require the fullest rule-of-reason analysis in the instant case. See, e.g., Stephen Calkins, California Dental Association: Not a Quick Look but Not the Full Monty, 67 Antitrust LJ. 495, 519 (2000) ("The Ninth Circuit was directed, in effect, to give those justifications a second hearing. But CDA did not win that which it really sought: invocation of the full-blown rule of reason, complete with merger-type measurement of market power.”); William J. Kolasky, California Dental Association v. FTC: The New Antitrust Empiricism, 14 Antitrust 68, 70 (Fall 1999) ("Many have expressed concern that California Dental may make summary disposition of antitrust cases, short of a full-blown rule of reason analysis, more difficult. The Court, however, goes out of its way to make clear that this is not its intent.... The point is simply that in California Dental itself, the Ninth Circuit needed to take a more careful ('more sedulous') look than it did.”) . In the instant case we opt for a particularly searching rule-of-reason inquiry in light of the plausibility and strength of the procompetitive justifications identified by the Supreme Court and CDA. See generally Kolasky, supra, at 70 ("[Cjourts must apply a sliding scale, in which the amount of proof demanded of the plaintiff depends both on how obvious the anticompetitive effects appear and how strong or weak the proffered justifications are.”). . The FTC’s opinion below also based its conclusion that the CDA intended to restrain trade on a number of other statements by various CDA component officers. See In re California Dental Ass'n, 121 F.T.C. 190 , 220-21 (1996). For example, the FTC opinion quoted a December 1987 statement by the executive director of one component, who forwarded a dentist's advertisement to the CDA, along with the following statement: This dentist is not in our area, Glendora is in the San Gabriel Valley component; however, if you wish me to handle this, I would be happy to do so, Italian style!!! Just let me know. These Drug Store Ads make me sick. Id. at 220 . It is entirely unclear to us how this crude complaint about one ad registered by a single local officer evinces an intent on the CDA’s part to restrain competition. Rather, the statement indicates the intensity of the author's opposition to certain kinds of advertisements without illuminating to any degree the basis for that opposition. The other statements quoted by the FTC opinion, while less colorful, are equally ambiguous with respect to the purpose of the CDA’s restrictions on advertising. In its brief to us, the FTC wisely de-emphasized these supposed indicia of intent. By contrast, there is relatively unambiguous, albeit self-serving, evidence in the record indicating that CDA’s motives in adopting advertising restrictions were benevolent. See infra note 5. . See, e.g., ER5D:7 (Dr. Lukens describing the function of the advertising restrictions as "to protect the public from anything that might be false or misleading in any material way”); ER5F:7 (Dr. Hoo testifying that "having a Code of Ethics and the Dental Practice Act protects the public more from advertising they might see that is misleading to them. And if you don't have that, then consumers, you know, will be misled by, you know, some unscrupulous people out there.”); ER5I:1 (Dr. Abrahams testifying that the dental advertising restrictions are intended to mimic the requirements of state law). . The FTC also points us to a sentence and footnote in the Langenfeld and Morris article. This sentence reads: "Studies of the price effects of advertising restrictions in professional occupations consistently have found that restrictions raise prices." Langenfeld & Morris, supra. The sentence is followed by a footnote citation to ten studies, none of which analyzes the dental services market in detail, save Cox and Foster’s. This single sentence hardly helps the FTC’s cause. As we explain above (and as the Court above explained), the record before us does not support this kind of blanket generalization about the effect of advertising restrictions as such in professional occupations writ large. Rather, different types of restrictions will have very different procompetitive or anticompetitive consequences, as will the same types of restrictions in different professional markets. . The FTC’s brief quotes Dr. Kinney's statement that "that kind of ad would probably take two pages in the telephone book. Nobody is going to really advertise in that fashion.” But the substance of Kinney’s response was that CDA would not require a dentist offering an across-the-board discount on all services to list the pre-discount and post-discount prices for every single discounted service because forcing a dentist to do so would be unreasonable. SER 541. In further testimony, Dr. Kinney clearly explained that CDA would not require such detailed disclosure: I can't imagine that there would ever be a circumstance in which you would have to have someone list all of their fees. Now, if they are offering a discount to senior citizens and this is an across the board discount for everything, it would be difficult to ask them to include all of these fees in this advertisement. In that case, I think you would have to be a little flexible and say okay, we are not going to require that you put every single fee that you are going to charge in [the advertisement], FTC Record (Testimony) 13:1373. Dr. Kinney’s testimony therefore contradicts the proposition for which the FTC cites it. . The ALJ found only that CDA’s restrictions effectively barred the use of across-the-board discounts, not the use of all price discount advertising. Having read the testimony of Drs. Cowan and Kinney, we believe that even this more limited conclusion by the ALJ was erroneous. . On the basis of this result, it is unnecessary for us to reach CDA’s alternative contention on appeal, which is that the FTC has failed to demonstrate under rule-of-reason analysis that CDA exercises market power. . It is not correct that only the Commission can determine in the first instance whether the restraints are unreasonable. That is why the Supreme Court remanded to the Court of Appeals here and directed the panel to consider "whether on remand it can effectively assess the Commission’s decision for substantial evidence on the record or whether it must remand to the Commission,” 526 U.S. at 769 n. 8., 119 S.Ct. 1604 , and why the Supreme Court has previously remanded to the Court of Appeals for initial applications of rule of reason. See, e.g., Broadcast Music, Inc. v. CBS, 441 U.S. 1 , 25 n. 44, 99 S.Ct. 1551 , 60 L.Ed.2d 1 (1979) ("The Court of Appeals did not address the rule-of-reason issue, and BMI insists that CBS did not preserve the question in that court. In any event, if the issue is open in the Court of Appeals, we prefer that that court first address the matter.”).