Regents of the University of California v. Credit Suisse First Boston (USA), Inc.

DENNIS, Circuit Judge,

concurring in the judgment:

I concur in the judgment reversing the district court’s certification order, but I do so on grounds different from those assigned by the majority. I respectfully disagree with the majority as to the issues upon which it decides the case. Although I ultimately agree that the certification order must be reversed, I do not believe that the law necessarily prevents the plaintiffs from prosecuting this case as a class action, and, as I explain below, I would remand the case to the district court for further consideration of whether the criteria for certification have been satisfied.

The majority today holds that secondary actors (such as the investment banks involved in this case) who act in concert with issuers of publicly-traded securities in schemes to defraud the investing public cannot be held liable as primary violators of Section 10(b) or Rule 10b-5 unless they (1) directly make public misrepresentations; (2) owe the issuer’s shareholders a duty to disclose; or (3) directly “manipulate” the market for the issuer’s securities through practices such as wash sales or matched orders. In doing so, the majority aligns this court with the Eighth Circuit 1 and immunizes a broad array of undeniably fraudulent conduct from civil liability under Section 10(b), effectively giving secondary actors license to scheme with impunity, as long as they keep quiet. 2

Although, as I explain below, I cannot agree with the majority’s cramped interpretation of the statutory language of section 10(b), in my view, the majority commits a significant error by even reaching this issue. Because the issue on which the majority opinion bases its decision today— a significant and unsettled question about the scope of primarily liability under Section 10(b) — is unnecessary to a determination of whether the plaintiffs have satisfied the prerequisites for maintaining a class action under Federal Rule of Civil Procedure 23, we should not consider it on this interlocutory appeal of class certification.

The investment banks have, however, raised two substantial issues that are related to the district court’s Rule 23 inquiry. The banks argue that the plaintiffs are not entitled to the fraud-on-the-market presumption of reliance because they have not satisfied the requirements of this court’s decision in Greenberg v. Crossroads Sys *395 tems, Inc., 364 F.3d 657 (5th Cir.2004), 3 and that the district court erred when it concluded that any defendant found to have knowingly violated the securities laws could be held jointly and severally liable for all of the plaintiffs’ losses in connection with Enron’s multi-year fraudulent scheme. Greenberg , in my view, is inconsistent with prior precedents of the Supreme Court and this court insofar as it purports to relieve securities defendants of the burden of rebutting the fraud-on-the-market presumption. On the latter point, however, I conclude that the district court erred by construing too broadly the joint and several liability provision of the Private Securities Litigation Reform Act of 1995 (the “PSLRA”), 15 U.S.C. § 78u-4. I would remand the case to the district court to determine whether, applying the correct legal standard, common damages issues continue to predominate over individual issues and whether the case can be tried in a manageable fashion.

I.

Our inquiry on this interlocutory appeal under Rule 23(f) 4 is limited to determining whether the district court erred in certifying the case as a class action. See Bell v. Ascendant Solutions, Inc., 422 F.3d 307, 314 (5th Cir.2005) (stating that Rule 23(f) permits a party to “ ‘appeal only the issue of class certification; no other issues may be raised’ ”) (quoting Bertulli v. Indep. Ass’n of Cont’l Pilots, 242 F.3d 290, 294 (5th Cir.2001)). We are not permitted to go beyond the issues necessary to class certification and rule on the merits of plaintiffs’ claims. See Unger v. Amedisys, Inc., 401 F.3d 316, 321 (5th Cir.2005) (“Class certification hearings should not be mini-trials on the merits of the class or individual claims.”); see also Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 177 , 94 S.Ct. 2140 , 40 L.Ed.2d 732 (1974) (noting that courts cannot “conduct a preliminary inquiry into the merits of a suit” on class certification).

It is clear, though, that a district court cannot certify a class action unless it finds that the plaintiffs have satisfied all of the requirements of Rule 23(a) and of one of the three subsections of Rule 23(b). To fulfill that obligation, it is often necessary for the district court to go “beyond the pleadings” and “understand the claims, defenses, relevant facts, and applicable substantive law in order to make a meaningful determination of the certification issues.” Castano v. Am. Tobacco Co., 84 F.3d 734, 744 (5th Cir.1996). In addition, to the extent that issues relevant to the ultimate merits of the case are also necessary to the district court’s determination of one or more of the requirements of Rule 23, the district court can, and must, consider those issues at the class certification stage. See Bell, 422 F.3d at 311-12 ;, see also In re Initial Pub. Offering Sec. Litig., 471 F.3d 24, 41 (2d Cir.2006) (“[Tjhere is no reason to lessen a district court’s obligation to make a determination that every Rule 23 requirement is met before certifying a class just because of some or even full overlap of that requirement with a merits *396 issue.”)- 5 By the same token, any such issues are also necessarily within the scope of our review on an interlocutory appeal of a district court’s decision to certify a class. Like the district court, however, this court can consider issues relevant to the merits of the plaintiffs’ claims only to the extent that such consideration is necessary to determine whether the proposed class satisfies the requirements of Rule 23. See Initial Pub. Offering, 471 F.3d at 41 (“[A] district judge should not assess any aspect of the merits unrelated to a Rule 23 requirement.”) (emphasis added).

In the majority’s view, one of the issues that this court can review on this interlocutory appeal is the district court’s conclusion that a secondary actor can be held liable as a primary violator of Section 10(b) and Rule 10b-5 for its participation in a scheme to defraud, even though it does not make a direct public misrepresentation or have a duty to speak. 6 According to the majority, the district court’s determination that secondary actors can violate Section 10(b) and Rule 10b-5 by engaging in “deceptive” acts without making a public misrepresentation or having a duty to speak implicates Rule 23(b)(3)’s requirement that “questions of law or fact common to the members of the class predominate over any questions affecting only individual members.” Fed.R.Civ.P. 23(b)(3). Without that finding, the majority states, the district court could not have concluded that the plaintiffs were entitled to a class-wide presumption of reliance, and individual issues of reliance would overwhelm the common, classwide issues, rendering class treatment inappropriate. Thus, the majority variously characterizes the district court’s ruling on the scope of Section 10(b) liability as “integral” or “critical” to its class certification decision.

The majority opinion labors to create the impression of a relationship between the district court’s decision that securities plaintiffs can state a Section 10(b) claim against a secondary actor who did not make any affirmative misstatements and the issue of whether the plaintiffs are entitled to rely on the fraud-on-the-market presumption of Basic Inc. v. Levinson, 485 U.S. 224 , 108 S.Ct. 978 , 99 L.Ed.2d 194 (1988). According to the majority, because the banks did not make public misstatements and had no duty to disclose vis-a-vis *397 Enron’s shareholders, their participation with Enron in fraudulent transactions that lacked any independent business purpose is beyond the reach of Section 10(b); because the banks’ conduct is not actionable under Section 10(b), the plaintiffs cannot invoke a classwide presumption of reliance; and because reliance cannot be presumed on a classwide basis, individual issues of reliance predominate over common issues. In sum, the upshot of the majority’s reasoning is that plaintiffs are not entitled to maintain a class action because the conduct for which they seek to recover' — to take the majority’s example, Merrill Lynch’s alleged conduct in connection with the so-called “Nigerian Barges Transaction” — is not actionable under Section 10(b).

With the reasoning that underlies the majority’s view set out in this manner, it becomes apparent that any link between the district court’s liability ruling and its application of the fraud-on-the-market presumption is tangential at best. The question of whether the banks can be subject to Section 10(b) liability without making public misrepresentations is by no means necessarily related to the applicability of the fraud-on-the-market presumption of reliance. Although it is true that the fraud-on-the-market presumption requires that there be a public misrepresentation upon which the market can rely, in this case, there were certainly public misrepresentations that arose out of the banks’ allegedly fraudulent transactions with Enron; the rub, of course — and the banks’ primary argument for why they are not subject to Section 10(b) liability — is that Enron, not the banks, conveyed the misrepresentations to the market. According to the banks, because they made no public statements, they are, at worst, aiders and abettors of Enron’s fraud. The plaintiffs counter by asserting, among other things, that the banks’ allegedly fraudulent conduct is not immunized simply because their joint scheme to defraud affected the market only through Enron’s public statements.

The majority’s leap to reach and resolve this dispute — which is strictly a question about the substantive reach of Section 10(b) — at the certification stage overlooks a key fact: Regardless of whether this court ultimately agrees with the district court that the banks’ alleged actions are “deceptive” acts within the meaning of Section 10(b), those actions affected the market via Enron’s public misrepresentations. Thus, this court can determine whether the plaintiffs are entitled to the fraud-on-the-market presumption without delving into the district court’s decision that the banks’ conduct is covered by Section 10(b). Viewed in this light, there is little doubt that in this case the element of transaction causation, or reliance, can be satisfied by the market’s reliance on Enron’s public representations of its financial health and/or its statements about the transactions in question. See Simpson v. AOL Time Warner Inc., 452 F.3d 1040, 1052 (9th Cir.2006) (“[A] plaintiff may be presumed to have relied on th[e] scheme to defraud if a misrepresentation, which necessarily resulted from the scheme and the defendant’s conduct therein, was disseminated into an efficient market and was reflected in the market price.”); In re Parmalat Sec. Litig., 376 F.Supp.2d 472, 509 (S.D.N.Y.2005) (applying fraud-on-the-market presumption to claims against secondary actors even though only issuer made public misrepresentations to the market). Accordingly, this court can assess whether a classwide presumption of reliance applies in this case without first considering the district court’s merits ruling on the scope of Section 10(b) liability. I would therefore find that the latter issue is beyond the permissible scope of our *398 limited, interlocutory review under Rule 23(f).

The majority is, of course, correct in some sense — if the banks engaged in no conduct within the reach of Section 10(b), then the plaintiffs cannot prevail against them in a class action. But the plaintiffs’ inability to proceed under such circumstances would have nothing to do with the need to prove reliance on an individual basis. When this court decides, on a common, classwide basis, as the majority does today, that the banks’ alleged conduct is non-actionable as a matter of law, it is dubious to then claim that we are actually finding only that individual issues of reliance predominate over common issues. Under the majority’s reasoning, individual questions of reliance do not predominate; rather, reliance is simply irrelevant, because no plaintiff can, on an individual or a class basis, establish that the banks engaged in any actionable conduct.

The mere fact that the resolution of a merits issue against a putative class of plaintiffs would, by definition, preclude the maintenance of a class action simply cannot be sufficient to warrant review of that issue on an interlocutory appeal. Under such a rule, the resolution of any Rule 12(b)(6) issue would then become fair game for Rule 23(f) review. The D.C. Circuit has rejected just such a “but-for” approach to Rule 23(f) appeals. In In re: Lorazepam & Clorazepate Antitrust Litigation, 289 F.3d 98, 107 (D.C.Cir.2002), the court refused to grant a Rule 23(f) appeal despite the petitioner’s assertions that the plaintiff class members lacked antitrust standing to maintain the suit. In a passage that is particularly pertinent to this case, the court explained that Rule 23(f) does not permit review of every issue that, if resolved against the plaintiffs, would destroy the class action:

Mylan’s effort to recast its Rule 12(b)(6) arguments as a challenge to class certification on the ground that a class of direct purchasers lacks antitrust standing, is to no avail. That Mylan’s argument as to antitrust standing may dispose of the class as a whole and thereby preclude a lawsuit by direct purchasers goes well beyond the purpose of Rule 23(f) review because it is unrelated to the Rule 23 requirements. The fact that Mylan’s challenge would be disposi-tive of the class action is not unlike a variety of issues of law on the merits of a class action because of the very nature of commonality; review of such issues would expand Rule 23(f) interlocutory review to include review of any question raised in a motion to dismiss that may potentially dispose of a lawsuit as to the class as a whole. This result would inappropriately mix the issue of class certification with the merits of a case, which do not warrant interlocutory review pursuant to Rule 23(f). What matters for purposes of Rule 23(f) is whether the issue is related to class certification itself ....

Id. (internal citation omitted).

The relationship between the banks’ potential Section 10(b) liability and class certification in this case is no closer than the relationship between antitrust standing and class certification described in Lora-zepam. Despite the majority’s claims to the contrary, as I explained above, whether the banks’ conduct can give rise to a Section 10(b) action under any circumstances need not be decided in order to determine whether the plaintiffs are entitled to invoke Basic’s presumption of reliance, and it is therefore not sufficiently related to any of Rule 23’s class action requirements to warrant interlocutory review.

*399 II.

Even were it appropriate for this court to consider whether the banks’ alleged conduct can constitute a primary violation of Section 10(b) and Rule 10b-5, the majority errs by defining the term “deceptive” in Section 10(b) in an unduly restrictive fashion.

Based on language gathered from inap-posite Supreme Court decisions, the majority opinion concludes that the Supreme Court has defined “deceptive” in a manner that both departs from the plain meaning of the word and reduces Section 10(b)’s flexible prohibition on “directly or indirectly” using or employing any “deceptive device or contrivance” 7 to a much more circumscribed prohibition that applies only to specific misrepresentations or omissions in breach of an affirmative duty to speak. Having arrived at this narrow definition of deceptive acts, the majority opinion then finds that, because the banks did not make misrepresentations or have a duty to speak, Section 10(b) does not reach their conduct, and imposing liability on them would be indistinguishable from the type of aiding and abetting liability barred by Central Bank of Denver, N.A. v. First Interstate Bank of Denver, N.A., 511 U.S. 164 , 114 S.Ct. 1439 , 128 L.Ed.2d 119 (1994). 8

Such a narrow interpretation of Section 10(b) is neither compelled nor justified by Supreme Court precedent. The majority’s conclusion hinges almost entirely on its determination that a court considering the reach of Section 10(b) cannot give the term “deceptive” (in the phrase “manipulative or deceptive device or contrivance”) its commonly understood, or dictionary, meaning, 9 because the Supreme Court has told us that a defendant acts deceptively only if it makes a misrepresentation or remains silent in the face of a duty to disclose. The majority opinion relies primarily on Chiarella v. United States, 445 U.S. 222 , 100 S.Ct. 1108 , 63 L.Ed.2d 348 (1980), and United States v. O’Hagan, 521 U.S. 642 , 117 S.Ct. 2199 , 138 L.Ed.2d 724 (1997), to support this narrow view of the statutory language. The majority’s view is not implausible — some statements in Chiarella, O’Hagan, and other cases can be read together to support the majority’s position 10 — but neither is it compelling, and the passages upon which the majority relies *400 fall far short of establishing that the Supreme Court has limited Section 10(b)’s prohibition on “deceptive” practices exclusively to misrepresentations or omissions.

Neither Chiarella nor O’Hagan purported to hold that a person can never engage in “deceptive” action through conduct, rather than speech or nondisclosure. In those cases, the Court held that a person who trades on non-public information violates Section 10(b) only if he breaches a duty to disclose, either to the source of the information or to the other party to the trade. See O’Hagan, 521 U.S. at 654-56 , 117 S.Ct. 2199 ; Chiarella, 445 U.S. at 235 , 100 S.Ct. 1108 . 11 Those cases, however, dealt only with insider trading — the defendants in those cases were prosecuted only for their silence, i.e., their failure to disclose that they possessed material information that other investors in the market did not possess. Neither of those cases involved allegations of a multi-party scheme to defraud. Here, by contrast, the plaintiffs assert that the banks engaged in a scheme with Enron whereby they structured and entered into wholly fraudulent transactions that were designed for the sole purpose of falsifying Enron’s financial results. Nothing in Chiarella or O’Hagan forecloses the conclusion that Section 10(b) can reach “deceptive” conduct that is not in the form of a misrepresentation or omission in cases, like this one, that involve large-scale schemes to defraud.

Nor do any of the Supreme Court’s other decisions establish that fraudulent conduct is beyond the reach of Section 10(b) simply because it affects the market only through the misrepresentations of another participant in the fraudulent scheme. As noted above, Central Bank itself did not reach this question because the plaintiffs in that case asserted only that the defendant’s conduct amounted to aiding and abetting. Moreover, language from other of the Court’s opinions affirmatively indicates that “deceptive” conduct need not always be in the form of a misrepresentation or an omission. See Santa Fe Indus., Inc. v. Green, 430 U.S. 462, 475-76 , 97 S.Ct. 1292 , 51 L.Ed.2d 480 (1977) (explaining that the Court’s prior cases all “included some element of deception,” and did not “support the proposition ... that a breach of fiduciary duty by majority stockholders, without any deception, misrepresentation, or nondisclosure, violates the statute and the Rule”) (emphasis added).

Because the Supreme Court has not, as the majority opinion maintains, narrowly defined the term “deceptive” to capture only direct misrepresentations or omissions, this court must construe the disputed statutory language “not technically and restrictively, but flexibly to effectuate its remedial purposes.” SEC v. Zandford, 535 U.S. 813, 819 , 122 S.Ct. 1899 , 153 L.Ed.2d 1 (2002) (internal quotation marks omitted). In light of this canon of interpretation, I see no basis for the majority opinion’s strict, narrow reading, and I agree with the district court, the Ninth Circuit, Judge Kaplan, and the SEC that Section 10(b)’s prohibition on directly or indirectly employing any “deceptive device or contrivance” can reach secondary actors who, with scienter, engage in fraudulent transactions that are used to inflate an issuer’s financial results. See Simpson, 452 F.3d at 1050 (“If a defendant’s conduct *401 or role in an illegitimate transaction has the principal purpose and effect of creating a false appearance of fact in the furtherance of a scheme to defraud, then the defendant is using or employing a deceptive device within the meaning of § 10(b).”); Enron, 2006 WL 4381143, at *42-44 , 2006 U.S. Dist. LEXIS 43146, at *167-74 (adopting SEC view “that a deceptive act includes a transaction whose principal purpose and effect is to create a false appearance of revenues, which can be accomplished by acts as well as by words”) (internal quotation marks omitted); Parmalat, 376 F.Supp.2d at 502-03 . 12

III.

The investment banks also assert that the district court erred by failing to apply this court’s decision in Greenberg v. Crossroads Systems, Inc., 364 F.3d 657 (5th Cir.2004), to determine whether the plaintiffs could proceed under the fraud-on-the-market theory. In Greenberg , a panel of this court held that plaintiffs who seek to invoke the fraud-on-the-market presumption of reliance must show both that the misrepresentations made to the market were “non-confirmatory,” i.e., that they did not simply confirm the market’s expectations, and that the misrepresentations actually affected the market price of the securities in question. See id. at 665-66 . 13

In its June 5, 2006 opinion, the district court declined to apply Greenberg to this case. The district court concluded that Greenberg applies only to cases under Rule 10b-5(b) involving misrepresentations, not to cases like this one involving a scheme to defraud under Rule 10b-5(a) or (c). See Enron, 2006 WL 4381143, at *71 , 2006 U.S. Dist. LEXIS 43146, at *287-88 . On appeal, the plaintiffs proffer several additional reasons why Greenberg does not mandate reversal. Plaintiffs assert that (1) Greenberg does not apply to cases like this one, where the alleged scheme to defraud stems from the defendants’ fraudulent efforts to conceal from the market information that would show that the issuer is not actually meeting the market’s expectations; (2) Greenberg should not be read to saddle plaintiffs with the burden of showing that the alleged misrepresentations actually changed the market price of the issuer’s securities; and (3) Greenberg’s requirements have, in any event, been satisfied in this case.

I agree with the plaintiffs that this court cannot use Greenberg to relieve the defendants of the burden, allocated to them in Basic and in subsequent decisions of this court, of rebutting the fraud-on-the-market presumption. In Basic itself, the Supreme Court was unmistakably clear that the defendant has the burden of rebutting the presumption of reliance:

Any showing that severs the link between the alleged misrepresentation and either the price received (or paid) by the plaintiff, or his decision to trade at a fair market price, will be sufficient to rebut the presumption of reliance. For example, if [defendants] could show that the “market makers” were privy to the truth *402 about the merger discussions here with Combustion, and thus that the market price would not have been affected by their misrepresentations, the causal connection could be broken: the basis for finding that the fraud had been transmitted through market price would be gone.

Basic, 485 U.S. at 248 , 108 S.Ct. 978 ; see also id. at 245 , 108 S.Ct. 978 (“Arising out of considerations of fairness, public policy, and probability, as well as judicial economy, presumptions are also useful devices for allocating the burdens of proof between parties.”). The clear import of Basic was not lost on this court. In Fine v. American Solar King Corp., 919 F.2d 290, 299 (5th Cir.1990), this court recognized that the defendant could rebut Basic’s presumption of reliance only by showing: “(1) that the nondisclosures did not affect the market price, or (2) that the Plaintiffs would have purchased the stock at the same price had they known the information that was not disclosed; or (3) that the Plaintiffs actually knew the information that was not disclosed to the market.”

This court’s more recent decisions, including Greenberg , have at least professed fidelity to Basic’s burden-shifting approach. In Greenberg , the court described the fraud-on-the-market presumption as follows:

Under this theory, reliance on the statement is rebuttably presumed if the plaintiffs can show that (1) the defendant made public material misrepresentations, (2) the defendant’s shares were traded in an efficient market, and (3) the plaintiffs traded shares between the time the misrepresentations were made and the time the truth was revealed. The Defendants may rebut this presumption by “[a]ny showing that severs the link between the alleged misrepresentation and either the price received (or paid) by the plaintiff, or his decision to trade at fair market price[.]”

Greenberg, 364 F.3d at 661 -62 (quoting Basic, 485 U.S. at 247 , 108 S.Ct. 978 ) (alterations in original) (internal citations and footnote omitted). In parts IV and V of its opinion, however, the Greenberg panel changed course and found that it is actually the plaintiffs’ affirmative burden to show, as a prerequisite to the application of the presumption, that the defendant’s misrepresentation actually moved the market price of the security in question:

We are satisfied that plaintiffs cannot trigger the presumption of reliance by simply offering evidence of any decrease in price following the release of negative information. Such evidence does not raise an inference that the stock’s price was actually affected by an earlier release of positive information. To raise an inference through a decline in stock price that an earlier false, positive statement actually affected a stock’s price, the plaintiffs must show that the false statement causing the increase was related to the statement causing the decrease.

Id. at 665 ; see also id. at 663 (referring to the plaintiffs’ “burden in a fraud-on-the-market case to show that a stock’s price was actually affected by an allegedly false statement”).

Greenberg appears to have mistakenly relied on this court’s earlier decision in Nathenson v. Zonagen Inc., 267 F.3d 400 (5th Cir.2001), as the authority for its decision to relieve securities defendants of the burden of rebutting the fraud-on-the-market presumption. In Nathenson , a panel of this court held that “where the facts properly considered by the district court reflect that the information in question did not affect the price of the stock then the district court may properly deny fraud-on- *403 the-market based recovery.” Nathenson, 267 F.3d at 415 . Nathenson did not, however, expressly purport to convert the defendant’s burden of rebutting the fraud-on-the-market presumption' — by, for example, showing that the alleged misrepresentation had no effect on the market price of the security — into a burden on the plaintiff to show that the misrepresentation did affect the price of the security. Rather, the panel in Nathenson simply determined that a district court did not err when it found that the allegations of the plaintiffs’ complaint affirmatively demonstrated that the misrepresentations in question did not affect the price of the issuer’s stock. Id. at 414, 417-18 . In other words, the plaintiffs in Nathenson affirmatively pleaded themselves out of the fraud-on-the-market presumption. Thus, Nathenson lends no support to the view that securities plaintiffs can invoke the fraud-on-the-market presumption only if they first affirmatively demonstrate that the market moved in response to the alleged misrepresentation.

Because the Greenberg panel’s decision to reallocate the burdens in fraud-on-the-market cases conflicts not only with Basic , but also with earlier decisions of this court, such as Fine , I would follow those decisions and hold that the defendant retains the burden of rebutting Basic’s, presumption of reliance. See, e.g., Modica v. Taylor, 465 F.3d 174, 183 (5th Cir.2006) (“ ‘When panel opinions appear to conflict, we are bound to follow the earlier opinion.’ ”) (quoting H&D Tire & Auto. Hardware, Inc. v. Pitney Bowes Inc., 227 F.3d 326, 330 (5th Cir.2001)); cf. Unger, 401 F.3d at 322 n. 4 (“[I]t is the Supreme Court’s job to overrule Basic , in the absence of outright conflict with the Private Securities Litigation Reform Act.”) (citation omitted). The banks do not appear to have satisfied that burden on the record before us. 14

*404 IV.

The banks argue that the district court erroneously determined that any defendant found to have knowingly violated the securities laws could be held jointly and severally liable for all of the losses caused by Enron’s entire overarching fraudulent scheme. The banks assert that without this erroneous legal conclusion, the district court could not have found that the proposed class satisfied Rule 23(b)(3)’s predominance requirement or the manageability aspect of Rule 23(b)(3)’s superiority requirement. 15

Before the enactment of the PSLRA, the general rule in Section 10(b) actions was that defendants found to have violated Section 10(b) or Rule 10b-5 were jointly and severally liable for all of the plaintiffs damages. See, e.g., Musick, Peeler & Garrett v. Employers Ins. of Wausau, 508 U.S. 286, 292 , 113 S.Ct. 2085 , 124 L.Ed.2d 194 (1993) (noting that violators “share joint liability for that wrong under a remedial scheme established by the federal courts”); TBG, Inc. v. Bendis, 36 F.3d 916, 927 (10th Cir.1994); G.A. Thompson & Co. v. Partridge, 636 F.2d 945, 963 (5th Cir.1981); Ross v. Licht, 263 F.Supp. 395, 411 (S.D.N.Y.1967). 16 The legislative history of the PSLRA suggests that Congress was concerned about the unfairness that could result from the application of. the traditional joint and several liability rule in many cases. See H.R.Rep. No. 104-369, at 37 (1995) (Conf.Rep.), 1995 U.S.C.C.A.N. 730, 736 (“Under current law, a single defendant who has been found to be 1% liable may be forced to pay 100% of the damages in the case.”); S.Rep. No. 104-98, at 20 (1995), 1995 U.S.C.C.A.N. 679, 699 (“Under joint and several liability, each defendant is liable for all of the damages awarded to the plaintiff. Thus, a defendant found responsible for only 1% of the harm could be required to pay 100% of the damages.”). To combat this perceived unfairness, as part of the PSLRA, Congress enacted 15 U.S.C. § 78u-4(f), which replaced the existing joint and several liability regime with one of proportionate liability and limited joint and several liability to defendants who knowingly violate the securities laws. 15 U.S.C. § 78u-4(f)(2)(A) provides that any defendant “against whom a final judgment is entered in a private action shall be liable for damages jointly and severally only if the trier of fact specifically determines that such covered person knowingly committed a violation of *405 the securities laws.” If no knowing violation is found, the statute provides that the defendant “shall be liable solely for the portion of the judgment that corresponds to the percentage of responsibility of that [defendant].” Id. § 78u-4(f)(2)(B).

The banks assert that, under these statutory provisions, a defendant who knowingly violates Section 10(b) can be jointly and severally liable only for the damages caused by conduct in which that defendant knowingly participated. Anything more, the banks argue, would run afoul of the PSLRA’s requirement that the plaintiff must “prov[e] that the act or omission of the defendant alleged to violate this chapter caused the loss for which the plaintiff seeks to recover damages,” id. § 78u-4(b)(4), and would be tantamount to imposing liability for conspiracy to violate Section 10(b). See Dinsmore v. Squadron, Ellenojf, Plesent, Sheinfeld & Sorkin, 135 F.3d 837 , 841 (2d Cir.1998) (stating that, post-Central Bank, there is no cause of action for conspiracy to violate Section 10(b)). As the banks would have it, then, even if they were found to have knowingly violated Section 10(b), they could be held jointly and severally liable for only those damages that the plaintiffs suffered specifically as a result of the transactions in which the banks participated with Enron. The plaintiffs, on the other hand, assert that because the alleged conduct of the banks, Enron, and others was all part of a single fraudulent scheme, any knowing violator can be held jointly and severally liable for harm caused by the other scheme participants.

In considering whether the proposed class satisfied the requirements of Rule 23(b)(3), the district court rejected the banks’ argument and determined that the PSLRA permits courts to broadly impose on knowing violators joint and several liability for all of the damages caused by a fraudulent scheme as a whole:

The Court finds that a reasonable argument can be made that where a defendant knowingly engaged in a primary violation of the federal securities law that was in furtherance of a larger scheme, it should be jointly and severally liable for the loss caused by the entire overarching scheme, including conduct of other scheme participants about which it kneio nothing. Indeed, express joint and several liability in the statute is a meaningless concept if it is limited to a defendant’s own wrongdoing. This Court acknowledges that it has previously questioned whether liability for conduct caused by all the scheme participants is compatible with the “knowing” requirement under § 78u-4(f)(2)(A). Nevertheless, the Court observes that the PSLRA not only replaced joint and several liability with proportionate liability except when the conduct was “knowing”, but established a right to contribution under § 78u-4(f)(8) to provide a remedy for unfairness, and, with a similar result, the judgment reduction formula embodied in § 78u-4(f)(2)(A) [sic]. Accordingly this Court concludes that Lead Plaintiff may pursue its claims for joint and several liability against those Defendants found to be primary violators in the scheme, as a whole.

Enron, 2006 WL 4381143, at *55-56 , 2006 U.S. Dist. LEXIS 43146, at *222-23 (emphasis added) (internal citation omitted). 17

The text of the PSLRA’s joint and several liability provision does not, on its own, resolve this issue. Section 78u-4(f)(2)(A) does not purport to define the scope of joint and several liability; rather, that pro *406 vision simply places limits on who can be subject to joint and several liability. The statute’s legislative history, however, indicates that Congress intended that the potential scope of joint and several liability would remain the same as it was under the pre-PSLRA law. See H.R.Rep. No. 104-369, at 38, 1995 U.S.C.C.A.N. at 737 (“The Conference Report imposes full joint and several liability, as under current law, on defendants who engage in knowing violations of the securities laws.”); S.Rep. No. 104-98, at 22, 1995 U.S.C.C.A.N. at 701 (same).

Under the pre-PSLRA practice described above, any defendant found to have violated Section 10(b) or Rule 10b-5 was jointly and severally liable for all of the damages suffered by the plaintiff; no distinction was made based on what portion of the plaintiffs damages were caused by, or traceable to, any specific defendant’s conduct. See TBG, Inc., 36 F.3d at 927 (“Liability in Rule 10b-5 cases is strictly joint and several and is never allocated among individual defendants in deciding the plaintiffs claim.”); Ross, 263 F.Supp. at 411 (finding joint and several liability for Section 10(b) claim because “the participation of each defendant was essential to the success of the scheme and there is no way to apportion guilt”); see also 5 Alan R. Bromberg & Lewis D. Lowenfels, Bromberg & Lowenfels on Securities Fraud and Commodities Fraud § 8.48 (2d ed.) (“[A]ny person found to have any liability whatsoever, no matter how insignificant, could be liable to plaintiffs for the total damages caused by all the misconduct.”). Accordingly, there appears to be no support for the banks’ assertion that any joint and several liability must be limited to joint and several liability for only the damages caused directly by the specific transactions in which they participated with Enron. Therefore, if the plaintiffs could succeed in proving at trial that the conduct alleged amounted to a single, overarching fraudulent scheme (as opposed to, as the banks assert, a number of separate and distinct fraudulent schemes), they should be permitted to recover damages jointly and severally from any knowing violator, and the scope of that joint and several liability should not be limited to the damages caused directly by that defendant’s participation in the scheme. 18

The banks are correct, however, that the scope of joint and several liability for a knowing violation of Section 10(b) must be informed by the reach of Section 10(b) itself. In particular, it seems fundamental that a defendant cannot be jointly and severally liable to a plaintiff unless that defendant is, in fact, primarily liable to that plaintiff. In other words, assuming that a jury finds that the banks knowingly violated Section 10(b) through their alleged participation with Enron in a scheme to defraud Enron’s investors, they can be held jointly and severally liable for all of the scheme-related losses suffered by any investor who was harmed in some way by their conduct, but the banks cannot be held responsible for the losses of any investors to whom they are not primarily liable under Section 10(b).

In a multi-defendant securities class action such as this one, where presumably thousands of investors were harmed by a number of different acts committed by different defendants over a period of several years, not every plaintiff will have been *407 harmed by every defendant. For example, if a particular defendant, who was previously uninvolved with the scheme, first structured or participated in a fraudulent transaction to falsely inflate Enron’s financial results near the end of the class period, that defendant could not be held liable under Section 10(b) to any investors who purchased Enron stock (the price of which may have already been inflated by the fraudulent acts of other defendants) before that transaction. Since those investors purchased their stock before the defendant engaged in any fraudulent conduct, they could not state a Section 10(b) claim against it, because they would be unable to show either that the defendant’s conduct caused them to purchase Enron stock at an inflated price (the element of reliance, or transaction causation) or that it caused them any harm (the element of loss causation). To make the defendant jointly and severally liable for the damages of those investors would, therefore, effectively expand the defendant’s underlying Section 10(b) liability to encompass plaintiffs who could not otherwise state a claim against it. 19 It would simply be inconsistent with the elements of a Section 10(b) claim to hold a knowing violator jointly and severally liable for the damages of any plaintiff to whom it is not primarily liable under Section 10(b). 20

That the fraud in this case is alleged to have been the result of a single, overarching scheme to defraud does not alter this conclusion. After Central Bank, a defendant can be liable under Section 10(b) only if it commits a primary violation of the statute. See Central Bank, 511 U.S. at 191 , 114 S.Ct. 1439 . Under the district court’s open-ended interpretation of the PSLRA’s joint and several liability provision, however, a defendant’s knowing violation of the securities laws could not only increase the damages for which it can be liable, but could also make that defendant responsible for the damages of plaintiffs who were harmed exclusively by the conduct of others, and to whom that defendant could not otherwise be liable at all. This, in my view, exceeds the permissible bounds of primary liability under Section 10(b) and amounts to the impermissible imposition of conspiracy liability. See Dinsmore, 135 F.3d at 841.

Because the district court’s class certification decision was based, in part, on this legal error, I would reverse the decision to certify the class on this ground only and remand the case to the district court to consider whether, in light of the proper interpretation of the PSLRA’s joint and several liability provision, the proposed class still satisfies the predominance and superiority requirements of Rule 23(b)(3).

CONCLUSION

Consequently, I concur in the judgment reversing the district court’s certification order, but I do so only for the reasons assigned herein. I would remand the case to that court for additional consideration of whether, in light of this opinion, this case meets Rule 23’s requirements for class certification.

. See In re Charter Commc’ns, Inc. Sec. Litig., 443 F.3d 987 (8th Cir.2006).

. The majority notes that the investment banks' alleged conduct was "hardly praiseworthy,” but it brushes the significance of its decision to immunize their conduct aside by noting that "[ujltimately, ... the rule of liability must be either overinclusive or underinclu-sive.” See supra, at 392.

. Plaintiffs also seek to rely on the Affiliated Ute presumption of reliance. See Affiliated Ute Citizens of Utah v. United States, 406 U.S. 128 , 92 S.Ct. 1456 , 31 L.Ed.2d 741 (1972). I do not disagree with the majority's conclusion that the Affiliated Ute presumption does not apply in this case. See supra, at 383-85.

. Fed.R.Civ.P. 23(f) provides:

A court of appeals may in its discretion permit an appeal from an order of a district court granting or denying class action certification under this rule if application is made to it within ten days after entry of the order. An appeal does not stay proceedings in the district court unless the district judge or the court of appeals so orders.

. It is important to note that the factual findings made by the district court at the class certification stage are not binding on the trier of fact at trial. See Initial Pub. Offering, 471 F.3d at 41 (''[T]he determination as to a Rule 23 requirement is made only for purposes of class certification and is not binding on the trier of facts, even if that trier is the class certification judge.”); Unger, 401 F.3d at 323 (stating that "the court's determination for class certification purposes may be revised (or wholly rejected) by the ultimate factfinder”); Gariety v. Grant Thornton, LLP, 368 F.3d 356, 366 (4th Cir.2004) ("The findings made for resolving a class action certification motion serve the court only in its determination of whether the requirements of Rule 23 have been demonstrated.”).

. The district court initially held that secondary actors can be liable under Section 10(b) and Rule 10b-5 based on deceptive acts not involving a direct public misstatement or a duty to speak when it denied certain defendants’ motions to dismiss the Section 10(b) claims against them. See In re Enron Corp. Sec., Derivative & ERISA Litig., 235 F.Supp.2d 549, 590-94 (S.D.Tex.2002). In its opinion on class certification, the court reconsidered (and ultimately adhered to) its earlier views in light of recent developments in the case law, noting that it had "the power to reconsider such interlocutory decisions.” See In re Enron Corp. Sec., Derivative & ERISA Litig., No. H-01-3624, 2006 U.S. Dist. LEXIS 43146 , at *155 n. 84 (S.D. Tex. June 5, 2006). The district court did not, however, indicate at any point that it believed that its decision on that issue was relevant to any specific requirement of Rule 23.

. Section 10(b) also prohibits any "manipulative” device or contrivance, but I do not take issue with the majority opinion's conclusion that the banks’ alleged conduct in this case was not "manipulative,” as that term is used in Section 10(b). See supra, at 390.

. Although this issue is commonly framed as whether liability can be imposed consistent with the Central Bank decision, Central Bank itself establishes relatively little about the reach of Section 10(b). The plaintiffs in that case alleged only that Central Bank aided and abetted a violation of Section 10(b); the plaintiffs did not claim that the bank was liable as a primary violator of the statute. See Central Bank, 511 U.S. at 191 , 114 S.Ct. 1439 ("Respondents concede that Central Bank did not commit a manipulative or deceptive act within the meaning of § 10(b).”).

. Reference to the common, dictionary definition is, incidentally, the approach that the Supreme Court has used to define the words "manipulative,” "device,” and "contrivance.” See Ernst & Ernst v. Hochfelder, 425 U.S. 185 , 199 & nn. 20-21, 96 S.Ct. 1375 , 47 L.Ed.2d 668 (1976).

. It is unsurprising that the Supreme Court’s decisions generally speak of misrepresentations or omissions, as they are typically the means through which fraudulent conduct reaches the market. As Judge Kaplan remarked in Parmalat, "[A]ny deceptive device or practice, other than one involving manipulative trading activity, logically requires that somebody misrepresent or omit something at some point, even though the device could entail more than the misrepresentation.” Parmalat, 376 F.Supp.2d at 497 .

. Also central to those cases, however, was the Chiarella court's observation that entering into a transaction on the basis of unequal information, while perhaps unfair, is not inherently fraudulent or deceptive in any sense of those words. See Chiarella, 445 U.S. at 232 , 100 S.Ct. 1108 ("[N]ot eveiy instance of financial unfairness constitutes fraudulent activity under § 10(b).”); id. at 233, 100 S.Ct. 1108 ("[N]either the Congress nor the Commission ever has adopted a parity-of-information rule.”).

. Although those courts that have found that Section 10(b) can reach conduct of the type alleged here have developed different formulations of the standard for liability, see Simpson, 452 F.3d at 1048 & n. 5, I agree with the majority that any such distinctions are not relevant to this interlocutory appeal. See supra, at 386-87 n. 25.

. This required showing is in addition to Basic's requirements that plaintiffs show that (1) there were material, public misrepresentations; (2) the securities in question traded in an efficient market; and (3) the plaintiffs traded in the securities in question between the date of the misrepresentations and the date on which the truth was disclosed to the market. See Greenberg, 364 F.3d at 661 .

. Greenberg also purports to require that the plaintiffs establish that the defendant’s false statement was "non-confirmatory” — i.e., that it did not simply confirm the market's preexisting expectations about, for example, the size of the issuer’s quarterly earnings — before the fraud-on-the-market presumption can apply. See Greenberg, 364 F.3d at 665-66 . According to the Greenberg panel, the market cannot rely on allegedly false "confirmatory” statements because "confirmatory information has already been digested by the market and will not cause a change in the stock price.” Id..; see also id. at 666 ("Because the presumption of reliance is based upon actual movement of the stock price, confirmatory information cannot be the basis for a fraud-on-the-market claim.”).

This requirement from Greenberg appears to be based on a misinterpretation of this court’s earlier decision in Nathenson . The Nathen-son panel stated that, in certain "special circumstances,” such as when an issuer provides false information that confirms the market’s expectations, the market can rely on those false statements, even though the market price may not change at the time of the false "confirmatory” statement. See Nathenson, 267 F.3d at 419 . Rather, the statement’s effect on the market price will show only when the falsity of the statement is later disclosed and the market price declines. See id.

This makes sense: if the market expects earnings of $1.00 per, share, then the share price might not move in response to a false public statement confirming that the issuer earned $1.00 per share (even though the issuer in fact lost $.50 per share). The lack of movement does not, however, mean that the false statement had no actual effect on the share price. Had the issuer truthfully disclosed its loss of $.50 per share to a market that expected earnings of $1.00 per share, the share price would have declined, rather than remaining steady; the false "confirmatory” statement actually affected the share price by keeping it artificially high in a situation where a truthful statement would have caused the share price to decline. As the Nathenson court suggested, the effect that the false statement had on the share price in such a case can be shown when the falsity of the statement is disclosed and the share price declines. See id.

Thus, there appears to be no basis in Nath-enson or otherwise for Greenberg's conclusion *404 that false “confirmatory” statements can never support a claim proceeding under a fraud-on-the-market theory.

. Among the factors to be considered in determining whether Rule 23(b)(3)’s superiority requirement is satisfied are "the difficulties likely to be encountered in the management of a class action.” Fed.R.Civ.P. 23(b)(3).

. See also Amy J. St. Eve & Bryce C. Pilz, The Fault Allocation Provisions of the Private Securities Litigation Reform Act of 1995 — A Roadmap for Litigants and Courts, 3 N.Y.U. J.L. & Bus. 187, 194 (2006) (stating that prior to the PSLRA, "[c]ourts recognized an implied theory of joint and several liability in Section 10(b) and Rule 10b-5 claims”); Marc I. Steinberg & Christopher D. Olive, Contribution and Proportionate Liability Under the Federal Securities Laws in Multidefendant Securities Litigation After the Private Securities Litigation Reform Act of 1995, 50 SMU L.Rev. 337, 339-40 (1996) ("[I]f co-defendants were adjudged liable in a federal securities action, plaintiffs were entitled to recover the total judgment from any of the subject defendants.”); Stuart M. Grant et al., The Devil Is in the Details: Application of the PSLRA’s Proportionate Liability Provisions Is so Fraught With Uncertainty that They May Be Void for Vagueness, 1505 PLI/Corp. 83, 85 (2005) ("[Until 1995], each defendant found to have violated the federal securities laws could be required to pay the full amount of any judgment to the plaintiff, regardless of whether that defendant was the primary violator or merely one of many violators.”).

. The district court also correctly noted that there is a paucity of authority addressing this issue. See Enron, 2006 WL 4381143, at *55 , 2006 U.S. Dist. LEXIS 43146, at *222 .

. This court does not need to decide whether any defendant could be jointly and severally liable for damages caused by the actions of others if the plaintiffs could not prove that all of the conduct they allege was part of a single fraudulent scheme. In their brief, the plaintiffs concede that a defendant can be jointly and severally liable only for the damages caused by a scheme in which it participates.

. The plaintiffs expressly concede this point in their briefs, as they state that no defendant can be liable for damages from before the date on which it violated Section 10(b).

. In the example given, if the defendant knowingly violated Section 10(b), it could be jointly and severally liable to any investors who purchased Enron stock after its fraudulent conduct and before the disclosure of the truth, even if those investors were also harmed by the conduct of other participants in the scheme.