(dissenting, with whom Ireland and Cowin, JJ., join). This is a case about the propriety of a class certification decision. The issue is not, as the court puts it, “whether the marketing of Marlboro Lights as ‘light’ cigarettes that deliver ‘lower tar and nicotine’ may be challenged in a class action” under G. L. c. 93A, ante at 382; of course it may (emphasis added). The issue is whether this class of plaintiffs may bring that challenge. The proper focus in this case is, therefore, the constituency of the plaintiff class.
The plaintiff class certified by the motion judge consists of “purchasers of Marlboro Lights cigarettes in Massachusetts during the four years preceding the filing of [the plaintiffs’ original] complaint.” It is this description of the plaintiff class that must meet the requirements for certification under G. L. c. 93A. The procedure for these consumer class actions is governed by G. L. c. 93A, § 9 (2). That section provides in relevant part:
“Any persons entitled to bring such action[1] may, if the use or employment of the unfair or deceptive act or practice has caused similar injury to numerous other persons similarly situated and if the court finds in a preliminary hearing that he adequately and fairly represents such other persons, bring the action on behalf of himself and such other similarly injured and situated persons . . . .”
As the court states, ante at 391, and as the language of § 9 (2) itself makes clear, class certification depends on a showing that the defendants’ deceptive practice caused “similar injury” to the members of the plaintiff class. G. L. c. 93A, § 9 (2). Accord Fletcher v. Cape Cod Gas Co., 394 Mass. 595, 605 (1985) (plaintiffs must allege that they are “similarly situated” and have suffered “similar injury”).
*404The requirement that the plaintiffs demonstrate an “injury” may not be shrugged off lightly. Rather, as this court has consistently held, the injury requirement is a fundamental aspect of G. L. c. 93A for both individual plaintiffs and plaintiff classes. It is insufficient for a plaintiff to show merely that a trade practice is deceptive; rather, G. L. c. 93A “requires an injury before an award of even nominal damages is justified.” Leardi v. Brown, 394 Mass. 151, 165 (1985).2 The difficulty for the plaintiffs in this case is that some members of the plaintiff class, as the plaintiffs concede, actually did receive lower levels of tar and nicotine from Marlboro Lights. For simplicity, I will refer to the group of smokers who actually received lower levels of tar and nicotine as the “low-tar group,” and those smokers who actually received equal or higher levels of tar and nicotine as the “high-tar group.” In the words of the Appeals Court’s single justice, the low-tar group “got what the advertising promised,” and, as the court concludes, the members of the low-tar group “have not suffered the ‘injury’ of higher tar and nicotine.”3 Ante at 398 n.21. How, then, can a class meet the “similar injury” requirement when some of its members (the low-tar group) suffered no injury at all?
*405The court answers this question by pointing to the pragmatic considerations surrounding the class certification decision. Specifically, the court concludes that certification is appropriate in this case because “the members of the class who have not suffered the ‘injury’ of higher tar and nicotine are both very few in number and impossible to identify.” Ante at n.21. I agree with the court that, in principle, class certification should not be impeded by the presence of a de minimis number of uninjured class members who are difficult to identify with specificity. Nevertheless, I do not think that, on the record in this case, it is possible to form a reasonable judgment that the low-tar group is made up of “only a very few smokers,” who are “impossible to identify.” Ante at n.21.
On a motion for class certification, the “plaintiffs bear the burden of providing information sufficient to enable the motion judge to form a reasonable judgment that the class meets the requirements [for certification].” Weld v. Glaxo Wellcome Inc., 434 Mass. 81, 87 (2001). It is therefore the plaintiffs’ burden to provide information sufficient to permit the reasonable conclusion that the class as a whole is “similarly injured,” despite the presence of the uninjured members of the low-tar group; in other words, if the plaintiffs contend that the low-tar group is only a very few smokers, it is the plaintiffs’ burden to provide information concerning the size of the low-tar group, and if the plaintiffs claim that members of the low-tar group are difficult to identify, it is their burden to provide information to that effect.
The plaintiffs presented no evidence before the motion judge concerning how many members of the proposed plaintiff class may have received lower levels of tar and nicotine. Rather, the plaintiffs’ allegations and evidence focus almost exclusively on “the average smoker” (or, more generally, “the smoker”) and cigarettes “smoked under normal use.” The plaintiffs’ exhibits do make a single passing reference to the percentage of smokers who block cigarette ventilation holes — an action that can *406increase the tar and nicotine yield of a cigarette4 — but the study referenced, which concluded that only fifty-eight per cent of cigarettes showed signs of significant ventilation “hole-blocking,”5 actually suggests that the low-tar group is fairly large (i.e., those smokers who smoked the other forty-two per cent of cigarettes). In sum, there was not sufficient information in the record to permit the motion judge reasonably to conclude that the low-tar group was so small that the class could meet the “similar injury” requirement.6
Likewise, it is not possible reasonably to conclude from the information submitted by the plaintiffs that the members of the low-tar group are “impossible to identify.” The court may be correct to point out that no objective scientific test can measure how much tar and nicotine an individual smoker has ingested. See ante at 398. Nevertheless, the plaintiffs have plainly alleged that the increased levels of tar and nicotine ingested by class members were in large part the result of a number of behavioral factors, namely smokers’ tendency to cover ventilation holes near the cigarette’s filter and smokers’ tendency to increase the frequency and volume of puffs to compensate for lower nicotine levels.7 Plainly, smokers will be able to provide information *407concerning how they smoked Marlboro Lights cigarettes. It is therefore not impossible to identify smokers in the uninjured low-tar group and to exclude them from the class.8
In sum, the crux of my disagreement with the court concerns the sufficiency of the information presented by the plaintiffs in this case. Limited to the record before the motion judge, I do not think it is possible reasonably to conclude that the low-tar group is either so small as to be de minimis, or so unidentifiable as to permit class certification. By certifying a class that includes uninjured members, the motion judge effectively permitted precisely what we have criticized: a “purely ‘vicarious suit[] by self-constituted private attomeys-general. ’ ” Leardi v. Brown, 394 Mass. 151, 161 (1985), quoting Baldassari v. Public Fin. Trust, 369 Mass. 33, 46 (1975).
I turn briefly to the court’s treatment of the theory of injury actually pressed by the plaintiffs below, primarily to register a point of agreement with the court. The motion judge described the plaintiffs’ theory as follows: “Plaintiffs may show a causal relationship between what they paid for Marlboro Lights cigarettes and what they would have paid if the defendants had *408not incurred the expense of false advertising.”9 The single justice restated the plaintiff’s theory: “because of [Philip Morris’s misrepresentations], cigarette purchasers were misled to a willingness to pay more for Marlboro Lights than would have been so, had the cigarette purchasers known [the truth].” Insofar as this theory purports to be a theory of injury, the theory is essentially a variant of the “fraud on the market” theory recognized in Federal securities cases,10 under which the purchaser of a stock is entitled to the presumption that the purchase was made in reliance on all material information known to the market as a whole. See, e.g., Basic, Inc. v. Levinson, 485 U.S. 224, 242 (1988). The court explicitly states that its opinion in this case does not adopt a “ ‘fraud on the market’ type theory,” ante at n.23, and I agree that it would be imprudent to adopt the fraud on the market doctrine in a consumer products case, when no other State has done so,11 and in the face of numerous critiques of the theory’s validity.12
Nevertheless, because, like the Appeals Court’s single justice, *409I would conclude that class certification in this case was improper where the plaintiffs failed to meet their “burden of providing information sufficient to enable the motion judge to form a reasonable judgment that the class meets the requirements [for certification],” Weld v. Glaxo Wellcome Inc., 434 Mass. 81, 87 (2001), I respectfully dissent.'
“[S]uch action” refers to the action described in the previous paragraph, G. L. c. 93A, § 9 (1), which provides:
“Any person, other than a person entitled to bring action under section eleven of this chapter, who has been injured by another person’s use or employment of any method, act or practice declared to be unlawful by section two or any rule or regulation issued thereunder . . . may bring an action ....’’
This court and the Appeals Court have consistently made clear that a defendant’s deceptive act must adversely affect the plaintiff before recovery under G. L. c. 93A, § 9, is permitted. See, e.g., Gurnack v. John Hancock Mut. Life Ins. Co., 406 Mass. 748, 753 n.5 (1990), citing Van Dyke v. St. Paul Fire & Marine Ins. Co., 388 Mass. 671, 678 (1983) (denial of claim under insurance policy does not warrant liability unless plaintiff is harmed by that denial); Lord v. Commercial Union Ins. Co., 60 Mass. App. Ct. 309, 321 (2004) (“G. L. c. 93A, § 9 [3], [requires] actual injury or loss before even minimum damages of $25 can be awarded”); id. at 321-322 (“the Legislature, in enacting and amending G. L. c. 93A, § 9, did not intend to confer on plaintiffs who have suffered no harm the right to receive a nominal damage award”); Schwartz v. Travelers Indem Co., 50 Mass. App. Ct. 672, 676 n.5 (2001) (“Plaintiffs would not be allowed to recover on [a G. L. c. 93A, § 9,] action where they have suffered no harm at all; the case law indicates that some kind of harm is required before damages can be recovered”); Abdella v. United States Fid. & Guar. Co., 47 Mass. App. Ct. 148, 153 (1999) (affirming dismissal “[bjecause [plaintiff] was not injured by [defendant’s] violation. . .”).
The court “rejectfs] the proposition that the purchase of an intentionally falsely represented product cannot be, by itself, an ascertainable injury under our consumer protection statute.” Ante at 394.1 take the court to mean simply that purchase of a deceptively advertised product may in certain circumstances be alone sufficient to meet the injury requirement of G. L. c. 93A, § 9 (2), not *405that the mere purchase of a deceptively advertised product necessarily constitutes per se injury.
Ventilation holes are designed to allow smokers to draw in extra air while smoking, thereby reducing the relative levels of tar and nicotine per inhalation. Marlboro Lights contain ventilation holes, but those holes are unmarked and are in the area of the cigarette filter that may be covered by the smoker’s lips or fingers.
The exhibit, entitled “Health Warning: Low Tar Cigarettes Are a Deliberate Con,” references a 1988 study analyzing smoked cigarette butts. According to the exhibit, that study concluded that “58% of low tar cigarettes showed signs of significant hole-blocking.”
It is not surprising that the plaintiffs made no serious attempt to present evidence demonstrating that the low-tar group made up only a small portion of the certified class. The plaintiffs never argued the point because their theory of injury was that both groups were adversely affected not merely because they purchased a deceptively advertised product, but because the deceptive advertising inflated the market price of Marlboro Lights for all purchasers.
Despite the conceded fact that at least some smokers did receive lower levels of tar and nicotine from Marlboro Lights, the court nonetheless speculates that “it is probable that no smoker received the promised benefit of lowered tar and nicotine,” because “it may be unlikely that any individual would smoke a cigarette the exact same way twice.” Ante at 398 n.20. This *407speculation runs in the face of an uncontested fact found on the record by both the motion judge and the single justice — namely, that whether a smoker received lower levels of tar and nicotine was dependent on the way that the smoker smokes. If the plaintiffs had intended to rest their certification motion on the argument that no one “smoke[s] a cigarette the exact same way twice,” they would have made this argument below and provided “information sufficient” to permit the motion judge to make a finding on that issue. Weld v. Glaxo Wellcome Inc., 434 Mass. 81, 87 (2001). They did neither.
Importantly, the requirement that the plaintiffs present evidence concerning the manner in which individual class members smoke cigarettes would not necessarily preclude class certification under G. L. c. 93A, § 9 (2). While class certification under Mass. R. Civ. P. 23, 365 Mass. 767 (1974), requires compliance with both the prerequisites of rule 23 (a) — numerosity, commonality, typicality, and fairness and adequacy of representation — and the requirements of rule 23 (b) — predominance of common issues and superiority to other available methods of litigation — class certification under § 9 (2) does not require a showing of predominance of common issues or superiority to other available methods of litigation. Weld v. Glaxo Wellcome Inc., supra at 86. Baldassari v. Public Fin. Trust, 369 Mass. 33, 39-40 (1975). Because the plaintiff class in this case does not need to demonstrate the predominance of common issues of fact or law, the necessity of individual proof concerning smoking behavior would not foreclose certification.
I do not understand the motion judge to mean merely that the advertising costs incurred by the defendants increased the cost of Marlboro Lights. While the amount of advertising undertaken by the defendants may have affected the price at which it was willing to sell Marlboro Lights, the expense of advertising (and, correspondingly, its effect on price) does not depend on whether the advertising was misleading. An honest advertisement is just as expensive as a dishonest one. Rather, as the single justice did, I understand the motion judge to mean that the plaintiffs may show relationship between the content of the deceptive advertising and the market-determined price of Marlboro Lights.
The court notes, and I agree, that insofar as the plaintiffs’ market-based theory relates only to proof of the amount of damages, not injury, the analogy to the “fraud on the market” theory of causation and injury is inapt. See ante at n.23.
“Since the Supreme Court accepted fraud on the market in [Basic Inc. v. Levinson, 485 U.S. 224 (1988)]... no [S]tote court with the authority to consider whether Basic is persuasive has chosen to apply it to claims arising under its own [S]tate’s laws.” Kaufman v. i-Stat Corp., 165 N.J. 94, 113 (2000). Accord Recent Case, 114 Harv. L. Rev. 2550, 2550 (2001) (“no [S]tote appellate court has incorporated [the ‘fraud on the market doctrine’] into the common law”). But see Hurley v. Federal Deposit Ins. Corp., 719 F. Supp. 27, 34 n.4 (D. Mass. 1989) (speculating that “if the Supreme Judicial Court were faced with the question, it would recognize the fraud on the market theory as a substitute for pleading reliance where a company’s stock is traded in an efficient market”).
When the Supreme Court of the United States first adopted the fraud on the market theory, it did so over lingering doubts about both the validity of *409the economic theory itself and the ability of courts to understand, evaluate, and apply it. See Basic Inc. v. Levinson, 485 U.S. 224, 255-257 (1988) (White, J., concurring in part and dissenting in part) (outlining economic critique of fraud on the market theory); id. at 253 (White, J., concurring in part and dissenting in part) (“with no staff economists, no experts schooled in the ‘efficient-capital-market hypothesis,’ no ability to test the validity of empirical market studies, we are not well equipped to embrace novel constructions of a statute based on contemporary microeconomic theory”); id. at 253 n.4 (White, J., concurring in part and dissenting in part), quoting Gilson & Kraakman, The Mechanisms of Market Efficiency, 70 Va. L. Rev. 549, 549-550 (1984) (“legal culture’s remarkably rapid and broad acceptance of an economic concept that did not exist twenty years ago is not matched by an equivalent degree of understanding” [emphasis in original]). In the sixteen years since the Basic case, doubts about the economic validity of the “fraud on the market” theory remain. See Bernard, Challenges to the Efficient Market Hypothesis: Limits to the Applicability of the Fraud-on-the-Market Theory, 73 Neb. L. Rev. 781, 782-783 (1994), quoting Jensen, Some Anomalous Evidence Regarding Market Efficiency, 6 J. Fin. Econ. 95 (1978) (“Since Basic, there has been an explosion of literature in financial economics casting doubt on the efficiency of at least some segments of the stock market. The theory once characterized in 1978 by Professor Michael Jensen as having ‘more solid empirical evidence supporting it. . . [than any] other proposition in economics’ has undergone so much questioning that leading researchers are now creating new theories to explain how, in equilibrium, market prices could reflect random factors that have nothing to do with firms’ underlying fundamental values”). See also Perrillo, The “Less Than” Efficient Capital Markets Hypothesis: Requiring More Proof from Plaintiffs in Fraud-on-the-Market Cases, 78 St. John’s L. Rev. 81 (2004) (critiquing fraud-on-the-market theory); Gordon, Efficient Markets, Costly Information, and Securities Research, 60 N.Y.U. L. Rev. 761 (1985) (same).