259 F.3d 154 (3rd Cir. 2001)
KENNETH E. NEWTON; MLPF&S CUST. FPO, BRUCE ZAKHEIM IRA FBO BRUCE ZAKHEIM
v.
MERRILL LYNCH, PIERCE, FENNER & SMITH, INC.; PAINEWEBBER, INC.
JEFFREY PHILLIP KRAVITZ,
V.
DEAN WITTER REYNOLDS, INC.
MLPF&S CUST. FPO, BRUCE ZAKHEIM IRA FBO BRUCE ZAKHEIM, JEFFREY PHILLIP KRAVITZ, GLORIA BINDER, APPELLANTS
UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT
No. 00-1586
Argued December 14, 2000
Filed August 6, 2001
On Appeal from the United States District Court for the District of New Jersey (D.C. Civil Action Nos. 94-CV-05343 & 95-CV-00213) Honorable Dickinson R. Debevoise
Karen L. Morris, Esquire (argued) Morris & Morris 1105 North Market Street, Suite 1600 Wilmington, Delaware 19801 Attorney for Appellants
Stephen M. Shapiro, Esquire (argued) Mayer, Brown & Platt 190 South LaSalle Street Chicago, Illinois 60603 Attorney for Appellees, Merrill Lynch, Pierce, Fenner & Smith, Inc., PaineWebber, Inc., and Dean Witter Reynolds, Inc.
David A. Brownlee, Esquire Kirkpatrick & Lockhart Henry W. Oliver Building 535 Smithfield Street Pittsburgh, Pennsylvania 15222 Attorney for Appellee, Merrill Lynch, Pierce, Fenner & Smith, Inc.
Paul J. Fishman, Esquire Friedman, Kaplan & Seiler One Gateway Center, 25th Floor Newark, New Jersey 07102 Robert B. McCAW, Esquire Wilmer, Cutler & Pickering 520 Madison Avenue New York, New York 10022 Attorneys for Appellee, PaineWebber, Inc.
William H. Pratt, Esquire Kirkland & Ellis Citigroup Center 153 East 53rd Street New York, New York 10022 Attorney for Appellee, Dean Witter Reynolds, Inc.
Karl A. Groskaufmanis, Esquire Fried, Frank, Harris, Shriver & Jacobson 1001 Pennsylvania Avenue, N.W., Suite 800 Washington, D.C. 20004 Attorney for Amicus Curiae-Appellees, Securities Industry Association
Before: Scirica, Fuentes and Garth, Circuit Judges
OPINION OF THE COURT
Scirica, Circuit Judge.
In this putative class action under S 10(b) of the Securities Exchange Act of 1934 and Rule 10b-5, thousands of investors sued their broker-dealers, who traded on the National Association of Securities Dealers Automated Quotation System (NASDAQ), for breaching their duty of best execution. Despite the broker-dealers' duty to execute trades under the most "favorable terms reasonably available," the investors charge the defendants executed orders at the price offered on the central National Best Bid and Offer system (NBBO), failing to investigate other feasible alternatives that potentially offered better prices. With hundreds of thousands of investors in the putative class, this alleged practice affected hundreds of millions of transactions.
The crux of this interlocutory appeal under Fed. R. Civ. P. 23(f) is whether plaintiffs' securities fraud claims satisfy the requirements for class certification under Fed. R. Civ. P. 23. The District Court denied plaintiffs' petition for class certification. We will affirm.
I.
The District Court had jurisdiction over the federal claims arising under the Securities Exchange Act of 1934, 15 U.S.C. S 78j(b), and 28 U.S.C. S 1331, as well as supplemental jurisdiction over the state law claims under 28 U.S.C. S 1367. Plaintiffs filed a petition for permission to appeal the denial of class certification under Fed. R. Civ. P. 23(f) which we granted. As an interlocutory appeal, we have jurisdiction under 28 U.S.C. S 1292(e).
II.
In 1998, the Supreme Court responded to the risk of improvident and largely unreviewable class certification decisions by amending Fed. R. Civ. P. 23 to provide for interlocutory appeal by permission of the court of appeals.1 Recognizing that denying or granting class certification is often the defining moment in class actions (for it may sound the "death knell" of the litigation on the part of plaintiffs, or create unwarranted pressure to settle non-meritorious claims on the part of defendants), the Rule acknowledges the extraordinary nature of class actions and permits the appellate courts to develop a coherent body of jurisprudence in this area.2
The new Rule provides that "[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." Fed. R. Civ. P. 23(f). Before its adoption, courts were hesitant to invoke an alternative grant of appellate jurisdictional authority under 28 U.S.C.S 2072(c), which enabled the Supreme Court by rule to "define when a ruling of a district court is final for the purposes of appeal under section 1291." 28 U.S.C. S 2072(c); see also Blair v. Equifax Checking Servs., Inc., 181 F.3d 832, 833 (7th Cir. 1999) (noting this authority "had gone unused, in part because it invites the question whether a particular rule truly `defines' or instead expands appellate jurisdiction"); 7B Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, Federal Practice and Procedure S 1802, pp. 105-06 (West Supp. 2000) (hereinafter Wright, Miller & Kane) ("[Rule 23(f)] is modelled on Section 1292(b), but differs in significant respects from that device in that it requires only appellate court approval of the appeal and it does not require that the district court's decision involve `a controlling question of law' about which the courts are divided."). On occasion, courts granted writs of mandamus to review certification decisions but with an uneasiness that their actions stretched the writ's traditionally restrictive parameters. See 5 James Wm. Moore et al., Moore's Federal Practice S 23.61[9][c] (discussing standard and cases); see also, e.g., In re Rhone-Poulenc Rorer Inc. , 51 F.3d 1293 (7th Cir. 1995) (granting order of mandamus to rescind class certification). Although we have issued rulings on Rule 23(f) motions, we have yet to articulate standards for granting or denying permission to appeal.3
The Committee Note is always a good starting point. It emphasizes that "[t]he court of appeals is given unfettered discretion whether to permit the [interlocutory] appeal, akin to the discretion exercised by the Supreme Court in acting on a petition for certiorari." Comm. Note, Fed. R. Civ. P. 23(f). The Note also sketches a rough outline of the types of cases courts of appeals should review: "Permission is most likely to be granted when the certification decision turns on a novel or unsettled question of law, or when, as a practical matter, the decision of certification is likely dispositive of the litigation."4 Id.; see also 5 Moore's Federal Practice S 23.61[9][b]. To provide further guidance on how to separate the wheat from the chaff, the Note instructs that
several concerns justify expansion of present opportunities to appeal. An order denying certification may confront the plaintiff with a situation in which the only sure path to appellate review is by proceeding to final judgment on the merits of an individual claim that, standing alone, is far smaller than the costs of litigation. An order granting certification, on the other hand, may force a defendant to settle rather than incur the costs of defending a class action and run the risk of potentially ruinous liability.
Comm. Note, Fed. R. Civ. P. 23(f). We can glean from the Note, therefore, at least three principles to guide the appellate courts in their exercise of discretionary jurisdiction: (1) when denial of certification effectively terminates the litigation because the value of each plaintiff's claim is outweighed by the costs of stand-alone litigation; (2) when class certification places inordinate or hydraulic pressure on defendants to settle, avoiding the risk, however small, of potentially ruinous liability; and (3) when an appeal implicates novel or unsettled questions of law; in this situation, early resolution through interlocutory appeal may facilitate the orderly development of the law.5
But interlocutory review is not cabined by these circumstances. The Note signals that the new Rule gives appellate courts broad discretion. For example, an error in the class certification decision that does not implicate novel or unsettled legal questions may still merit interlocutory review given the consequences likely to ensue. To put it another way, if the appellant demonstrates that the ruling on class certification is likely erroneous, " `taking into account the discretion the district judge possesses in implementing Rule 23, and the correspondingly deferential standard of appellate review,' " Mowbray, 208 F.3d at 293 (quoting Blair, 181 F.3d at 835), interlocutory review may be proper.
Furthermore, as explained in the Note, interlocutory review is not constrained by the potentially limiting requirement of 28 U.S.C. S 1292(b) that the district court order "involve[ ] a controlling question of law as to which there is a substantial ground for difference of opinion and that an immediate appeal from the order may materially advance the ultimate termination of the litigation." Yet if allowing the litigation to follow its natural course would provide the moving party with an adequate remedy, interlocutory review will generally prove unnecessary. In the end, however, the courts of appeals are afforded wide latitude as "[p]ermission to appeal may be granted or denied on the basis of any consideration that the courts of appeals finds persuasive." Comm. Note, Fed. R. Civ. P. 23(f).
We believe these principles provide a useful template for courts to work from when evaluating petitions under Rule 23(f). It is, of course, difficult to foresee all the permutations to which this rule will apply, and courts will have the task of exercising their best judgment in making these decisions. See Lienhart, 255 F.3d at 144-45 (rejecting "stringent standards" for review of Rule 23(f) petitions); Blair, 181 F.3d at 834 ("[I]t would be a mistake for us to draw up a list that determines how the power under Rule 23(f) [should] be exercised. Neither a bright-line approach nor a catalog of factors would serve well--especially at the outset, when courts necessarily must experiment with the new class of appeal."); see also Comm. Note, Fed. R. Civ. P. 23(f) ("The courts of appeals will develop standards for granting review that reflect the changing areas of uncertainty in class litigation."). Further, as the Committee Note mentions, class certification decisions often involve "familiar and almost routine issues" that do not necessitate interlocutory appeal. If granting the appeal, however, would permit us to address (1) the possible case-ending effect of an imprudent class certification decision (the decision is likely dispositive of the litigation); (2) an erroneous ruling; or (3) facilitate development of the law on class certification, then granting the motion would be appropriate. But these instances should not circumscribe our discretion; there may also be other valid reasons for the exercise of interlocutory review. Again, we emphasize that the courts of appeals have been afforded the authority to grant or deny these petitions "on the basis of any consideration that the court of appeals finds persuasive."6 Comm. Note, Fed. R. Civ. P. 23(f).
The claims here touch on several reasons justifying interlocutory appeal. On the one hand, some of the securities claims pressed by the putative class members may be too small to survive as individual claims. On the other, certifying the class may place unwarranted or hydraulic pressure to settle on defendants. Either way, an adverse certification decision will likely have a dispositive impact on the course and outcome of the litigation. Moreover, this case raises fundamental questions about what type of private securities claims merit class certification. For these reasons, the motion was properly granted.
III.
We review a decision granting or denying class certification for abuse of discretion. In re LifeUSA Holding Inc., 242 F.3d 136, 143 (3d Cir. 2001); Holmes v. Pension Plan of Bethlehem Steel Corp., 213 F.3d 124, 136 (3d Cir. 2000). The district court abused its discretion if its decision " `rests upon a clearly erroneous finding of fact, an errant conclusion of law or an improper application of law to fact.' "7 In re General Motors Corp. Pick-Up Truck Fuel Tank Prods. Liab. Litig., 55 F.3d 768, 783 (3d Cir. 1995) (hereinafter "G.M. Trucks") (quoting Int'l Union, UAW v. Mack Trucks, Inc., 820 F.2d 91, 95 (3d Cir. 1987)). A class certification decision requires a thorough examination of the factual and legal allegations. Barnes v. Am. Tobacco Co., 161 F.3d 127, 140 (3d Cir. 1998), cert. denied, 526 U.S. 1114 (1999). For this purpose, "it may be necessary for the court to probe behind the pleadings before coming to rest on the certification question." General Tel. Co. of Southwest v. Falcon, 457 U.S. 147, 160 (1982); see also Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 634-35 (1997) (Breyer, J., concurring in part and dissenting in part); 7B Wright, Miller & Kane, S 1785, p. 16 (West Supp. 2000). "Before deciding whether to allow a case to proceed as a class action, . . . [courts] should make whatever factual and legal inquiries are necessary under Rule 23." Szabo v. Bridgeport Machs. Inc., 249 F.3d 672, 676 (7th Cir. 2001); see also 5 Moore's Federal Practice S 23.46[4] ("[B]ecause the determination of a certification request invariably involves some examination of factual and legal issues underlying the plaintiffs' cause of action, a court may consider the substantive elements of the plaintiffs' case in order to envision the form that a trial on those issues would take.") (footnotes omitted).
Over twenty-five years ago in Eisen v. Carlisle & Jacquelin, the Supreme Court cautioned against going beyond the pleadings in class certification decisions. 417 U.S. 156, 177 (1974) ("[N]othing in either the language or history of Rule 23 . . . gives a court any authority to conduct a preliminary inquiry into the merits of a suit in order to determine whether it may be maintained as a class action."). But this admonition must be examined in context. At the time, it was ancillary to the principal issue of whether Fed. R. Civ. P. 23 required a class representative in a securities class action to provide notice to all class members. With a claim that amounted to no more than seventy dollars, the plaintiff in Eisen sought to shift his notice burden to the defendant because providing notice to the 2.25 million potential class members was extraordinarily costly (roughly $225,000). The district court held that the defendant should bear 90% of the cost, because the plaintiff was "more than likely" to "prevail on his claims." Holding this burden could not be shifted, the Supreme Court affirmed the reversal by the court of appeals.
Not long after Eisen, the Court stepped away from this bright-line declaration in Coopers & Lybrand v. Livesay, when it held that
[e]valuation of many of the questions entering into determination of class action questions is intimately involved with the merits of the claims. The typicality of the representative's claims or defenses, the adequacy of the representative, and the presence of common questions of law or fact are obvious examples. The more complex determinations required in Rule 23(b)(3) class actions entail even greater entanglement with the merits . . . .
437 U.S. 463, 469 n.12 (1978) (quotation and citation omitted). Subsequently, in General Tel. Co. of Southwest v. Falcon, the Court appeared to move even further away from Eisen, recognizing that
[s]ometimes the issues are plain enough from the pleadings to determine whether the interests of the absent parties are fairly encompassed within the named plaintiff's claim, and sometimes it may be necessary for the court to probe behind the pleadings before coming to rest on the certification question . . . . [A]ctual, not presumed conformance with Rule 23(a) remains . . . indispensable.
Falcon, 457 U.S. at 160. This reasoning applies with equal force to certification questions surrounding Fed. R. Civ. P. 23(b)(3). Szabo, 249 F.3d at 677. As the Court concluded in Livesay, class certification may require courts to answer questions that are often " `enmeshed in the factual and legal issues comprising the plaintiff's cause of action.' " 437 U.S. at 469 (quoting Mercantile Nat'l Bank v. Langdeau, 371 U.S. 555, 558 (1963)). To address these questions, courts may "delve beyond the pleadings to determine whether the requirements for class certification are satisfied." 5 Moore's Federal Practice S 23.61[5]; Szabo, 249 F.3d at 677 (holding courts may "look[ ] beneath the surface of a complaint" to "make a preliminary inquiry into the merits"); see also Amchem, 521 U.S. at 615 (Fed. R. Civ. P. 23(b)(3) invites a "close look" before determining class certification); 7B Wright, Miller & Kane, S 1785, p.16 (West Supp. 2000) (courts not precluded from "necessary inquiry into the underlying elements of the case in order to evaluate whether Rule 23 has been met"); Moore's Federal Practice, Manual For Complex Litigation (Third) S 30.1 ("The decision on whether or not to certify a class, therefore, can be as important as decisions on the merits of the action and should be made only after consideration of all relevant evidence and arguments presented by the parties.").
Since Eisen was decided, the nature of class actions and how they are litigated have undergone a sea change. Irrespective of the merits, certification decisions may have a decisive effect on litigation. As mentioned, if individual claims are small, then plaintiffs may not have the incentive or resources to pursue their claims if certification is denied --sounding the "death knell" to the litigation.8 On the other hand, granting certification may generate unwarranted pressure to settle non-meritorious or marginal claims. Rhone-Poulenc, 51 F.3d at 1299-1300 (granting order of mandamus to rescind certification based in part on the "the demonstrated great likelihood that the plaintiffs' claims, despite their human appeal, lack legal merit"); see also G.M. Trucks, 55 F.3d at 784-85 (vacating class certification for settlement and remanding for further development on the record). In a similar vein, the Court of Appeals for the Fifth Circuit has concluded that "[g]oing beyond the pleadings is necessary, as a court must understand the claims, defenses, relevant facts, and applicable substantive law in order to make a meaningful determination of the certification issues." Castano, 84 F.3d at 744 (decertifying class that sued tobacco manufacturers for nicotine addiction). In Castano, the court held that
a mass tort cannot be properly certified without a prior track record of trials from which the district court can draw the information necessary to make the predominance and superiority analysis required by rule 23. This is because certification of an immature tort results in a higher than normal risk that the class action may not be superior to individual adjudication.
Id. at 747. Other courts have followed similar approaches. Szabo, 249 F.3d at 675-78; see also Rutstein v. Avis Rent-A-Car Sys., Inc., 211 F.3d 1228, 1234 (11th Cir. 2000); Hanon v. Dataproducts Corp., 976 F.2d 497, 508-09 (9th Cir. 1992).
In reviewing a motion for class certification, a preliminary inquiry into the merits is sometimes necessary to determine whether the alleged claims can be properly resolved as a class action.9 This is such an instance. We must probe beyond the surface of plaintiffs' allegations in performing our review to assess whether plaintiffs' securities claims satisfy Fed. R. Civ. P. 23's requirements.)
IV.
A.
This case is before us for the second time. We have already provided a succinct description of the facts, including the operation of the NASDAQ market and defendants' role.10 Newton v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 135 F.3d 266 (3d Cir. 1998) (en banc) (hereinafter "Newton").
Plaintiff-Appellants are investors who purchased and sold securities on the NASDAQ market, the major electronic market for "over-the-counter" securities, during the . . . period from November 4, 1992 to [August 28, 1996] ("the class period"). The defendants are NASDAQ market makers. NASDAQ is a self-regulating market owned by the National Association of Securities Dealers ("NASD"), subject to oversight by the Securities and Exchange Commission ("SEC").
An "over-the-counter" market like NASDAQ differs in important respects from the more familiar auction markets, like the New York and American Stock Exchanges. The NYSE and AMEX markets are distinguished by a physical exchange floor where buy and sell orders actually "meet," with prices set by the interaction of those orders under the supervision of a market "specialist." In a dealer market like NASDAQ, the market exists electronically, in the form of a communications system which constantly receives and reports the prices at which geographically dispersed market makers are willing to buy and sell different securities. These market makers compete with one another to buy and sell the same securities using the electronic system; NASDAQ is, then, an electronic inter-dealer quotation system.
In a dealer market, market makers create liquidity by being continuously willing to buy and sell the security in which they are making a market. In this way, an individual who wishes to buy or sell a security does not have to wait until someone is found who wishes to take the opposite side in the desired transaction. To account for the effort and risk required to maintain liquidity, market makers are allowed to set the prices at which they are prepared to buy and sell a particular security; the difference between the listed "ask" and "bid" prices is the "spread" that market makers capture as compensation.
The electronic quotation system ties together the numerous market makers for all over-the-counter securities available on NASDAQ. All NASDAQ market makers are required to input their bid and offer prices to the NASD computer, which collects the information and transmits, for each security, the highest bid price and lowest ask price currently available. These prices are called the "National Best Bid and Offer," or NBBO. The NASD computer, publicly available to all NASDAQ market makers, brokers and dealers, displays and continuously updates the NBBO for each offered security.
Plaintiffs allege that technological advances made it feasible during the class period for the defendant market makers to execute orders at prices quoted on private on-line services like SelectNet and Instinet and that those prices were frequently more favorable to their investor clients than the NBBO price. According to plaintiffs, the defendants regularly used these services and knew that prices better than NBBO were often available through them. Even though they knew that their investor clients expected them to secure the best reasonably available price, plaintiffs say, the defendants executed plaintiffs' orders at the NBBO price when they knew that price was inferior and when they, at the same time, were trading at the more favorable price for their own accounts. In this way, they were able to inflate their own profit margins at the expense of their investor clients. This practice is alleged to violate section 10 of the Securities [Exchange] Act of 1934, 15 U.S.C. S 78j, and Rule 10b-5 promulgated thereunder, 17 C.F.R. S 240.10b-5.
The plaintiffs also charge defendants with two other violations of section 10 and Rule 10b-5. Market makers who simultaneously hold a market order for both sides of a transaction may obtain more favorable prices than the NBBO by "crossing" these in-house orders. Transactions handled in this way are executed within the spread, giving both the purchaser and the seller a better price. Similarly, a customer order can be matched by a market maker with an in-house limit order on the other side of the transaction. Since a limit order specifies a particular price at which to execute a transaction, matching another customer order at that price may beat the currently displayed NBBO quote for that security. Plaintiffs allege that the failure of the defendants to execute orders of their clients in these ways when feasible constitutes a fraudulent practice because, by executing at the NBBO rather than matching customer orders, the defendants capture the full market "spread" as a fee for their services without incurring any actual risk in the transaction.11
Newton, 135 F.3d at 268-69.12
Since the initiation of this action, the Securities and Exchange Commission has promulgated new rules that effectively end the alleged improper practice by the defendants. See Order Execution Obligations, Exchange Act. Rel. No. 34-37619A, 61 Fed. Reg. 48290, 48306-16, 48322-23 (Sept. 12, 1996) ("While in the past quote-based executions in OTC [over the counter] securities were generally recognized as satisfying best execution obligations, the development of efficient new facilities has altered what broker dealers must consider in seeking execution of customer orders."); see also Newton, 135 F.3d at 271. The new regulations require the NBBO to incorporate prices displayed on Instinet and SelectNet as well as other sources of liquidity. See 17 C.F.R. SS 240.11Ac1-1 to -4 (2000).
B.
Defendants initially moved to dismiss this action under Fed. R. Civ. P. 12(b)(6) for failure to state a claim upon which relief could be granted. At the request of the District Court, defendants converted their motion into one for summary judgment which was subsequently granted. The District Court held that plaintiffs' claims failed to satisfy two requirements necessary to maintain a Rule 10b-5 securities violation--misrepresentation or omission of a material fact, and scienter--because defendants' "duty of best execution" remained ill-defined during the class period. In re Merrill Lynch, et al. Sec. Litig., 911 F. Supp. at 769-71. Without a clear standard to apply against defendants' employment of an industry-wide practice, the court found the nondisclosure of their trading execution practice, along with their implied representation to obtain best execution, did not constitute a misrepresentation or omission of material fact. Id. Even if the practice constituted a material misrepresentation, the district court held defendants had not formed the requisite scienter, or intent to deceive, because they were not aware their practice actually violated the securities laws. Id. at 771-72.
On appeal, a divided panel of this court affirmed. We granted rehearing en banc. The en banc court unanimously reversed the district court and remanded, holding the execution of trades at the NBBO, albeit the industry standard, could still be considered fraudulent behavior violating the standards of Rule 10b-5. Newton , 135 F.3d at 274. Noting this practice could constitute a material misrepresentation with scienter when better prices were reasonably available, we expressed no opinion on defendants' liability, only whether defendants' practice could be actionable under Rule 10b-5. Id. at 272-74. On remand, plaintiffs amended their complaint, extending the class period to the time new securities regulations took effect outlawing the defendants' alleged tortious practice. Plaintiffs then moved for class certification which the District Court denied. An interlocutory appeal under Fed. R. Civ. P. 23(f) was then granted.
C.
Plaintiffs contend defendants' behavior in this case was unvarying, alleging it was their established practice to execute trades at prices displayed solely on the NBBO without investigating other sources. They claim this "common scheme" provides a uniform course of unlawful conduct well-suited for adjudication as a class action. Plaintiffs also argue that during the class period defendants capitalized on their access to alternative trading sources to find better prices when trading on their own accounts. As noted in Newton, an SEC study reported that a "two-tiered market" existed during the class period where market makers exploited these services to garner better prices for themselves while simultaneously denying them to their customers.13 Id. at 273. For their part, defendants argue that without examining each transaction, their past ability to obtain a better price for a particular trade is purely speculative. On appeal, the availability of better prices remains hotly contested.
In this interlocutory appeal, we do not decide whether defendants' alleged practice constitutes a Rule 10b-5 securities violation with respect to each individual member of the putative class. Our inquiry only addresses whether the federal securities claims alleged by the investors satisfy the requirements demanded by Fed. R. Civ. P. 23.
V.
To determine whether the claims alleged by the putative class meet the requirements for class certification, we must first examine the underlying cause of action--in this case, a Rule 10b-5 private securities fraud claim. See Barnes, 161 F.3d at 138; McCarthy v. Kleindienst, 741 F.2d 1406, 1412 (D.C. Cir. 1984). This analysis is critical because class certification under Fed. R. Civ. P. 23(b)(3) is permissible only when "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). For the elements of the Rule 10b-5 claim which remain in dispute, "[r]equiring proof of individualized reliance [and injury] from each member of the proposed plaintiff class effectively would . . . prevent[ ] [plaintiffs] from proceeding with a class action, since individual issues then would . . . overwhelm[ ] the common ones." Basic Inc. v. Levinson, 485 U.S. 224, 242 (1988). On the other hand, presuming these elements would resolve "the problem of balancing the substantive requirement of proof of reliance [and injury] in securities cases against the procedural requisites of [Federal Rule of Civil Procedure] 23." Id. (quotation and citation omitted); see also Peil v. Speiser, 806 F.2d 1154 (3d Cir. 1986) (upholding presumption of reliance in Rule 10b-5 claims based on fraud-on-the-market theory); Hoxworth v. Blinder, Robinson & Co., 980 F.2d 912 (3d Cir. 1992) (affirming class certification where reliance presumed) (hereinafter "Hoxworth II"); William Rubenstein, A Transnational Model of Adjudication, 89 Geo. L.J. 371, 391-92 (2001) (discussing effect of presuming reliance in securities class actions). If proof of the essential elements of the cause of action requires individual treatment, then class certification is unsuitable. See Binder v. Gillespie, 184 F.3d 1059, 1063-66 (9th Cir. 1999) (upholding class decertification where presumption of reliance and loss unavailable), cert. denied, 528 U.S. 1154 (2000).
Under Rule 10b-5 causation is two-pronged. Huddleston v. Herman & MacLean, 640 F.2d 534, 549 n.24 (5th Cir. 1981), aff'd in part, rev'd in part on other grounds, 459 U.S. 375 (1983); see also James D. Cox, Robert W. Hillman & Donald C. Langevoort, Securities Regulation: Cases and Materials 769-71 (3d ed. 2001); 5 A. Jacobs, The Impact of Rule 10b-5 S 64.01[a], at 3-221 to 3-222 (Supp. 1980). Reliance, or transaction causation, establishes that but for the fraudulent misrepresentation, the investor would not have purchased or sold the security. Suez Equity Investors, L.P. and SEI Assocs. v. Toronto-Dominion Bank, 250 F.3d 87, 95-96 (2d Cir. 2001); Weiner v. Quaker Oats Co., 129 F.3d 310, 315 (3d Cir. 1997); Robbins v. Koger Props., Inc., 116 F.3d 1441, 1447 (11th Cir. 1997). Loss causation demonstrates that the fraudulent misrepresentation actually caused the loss suffered. Suez Equity Investors, 250 F.3d at 95-96; EP MedSystems, Inc. v. EchoCath, Inc., 235 F.3d 865, 883-84 (3d Cir. 2000). We must first address whether plaintiffs' claims are entitled to class-wide presumptions of reliance and economic loss before turning to the requirements for certification under Fed. R. Civ. P. 23(b)(3).
A.
Section 10(b) of the Securities Exchange Act of 1934 makes it unlawful "[t]o use or employ, in connection with the purchase or sale of any security registered on a national securities exchange . . . any manipulative or deceptive device or contrivance in contravention of such rules and regulations as the [Securities and Exchange] Commission may prescribe." 15 U.S.C. S 78j(b). Under this statute, the Securities and Exchange Commission promulgated Rule 10b-5 which provides:
It shall be unlawful for any person, directly or indirectly, by the use of any means or instrumentality of interstate commerce, or of the mails or of any facility of any national securities exchange,
(a) to employ any device, scheme, or artifice to defraud,
(b) to make any untrue statement of a material fact or to omit to state a material fact necessary in order to make the statements made, in the light of the circumstances under which they were made, not misleading, or
(c) To engage in any act, practice, or course of business which operates or would operate as a fraud or deceit upon any person, in connection with the purchase or sale of any security.
17 C.F.R. S 240.10b-5. In Newton, we set forth the necessary elements of a Rule 10b-5 violation:
To state a claim for securities fraud under S 10 of the Securities [Exchange] Act of 1934 and Rule 10b-5, plaintiffs must demonstrate: (1) a misrepresentation or omission of a material fact in connection with the purchase or sale of a security; (2) scienter on the part of the defendant; (3) reliance on the misrepresentation; and (4) damage resulting from the misrepresentation.
Newton, 135 F.3d at 269; see also Semerenko v. Cendant Corp., 223 F.3d 165, 174 (3d Cir. 2000).
It is important to recognize that the facts of this case do not resonate with those typical of securities violations under Rule 10b-5. Customarily those claims involve a fraudulent material misrepresentation or omission that affects a security's value. See EchoCath, 235 F.3d at 884 (citing typical cases: Semerenko, 223 F.3d at 171 (financial statement); Weiner, 129 F.3d at 311-12 (corporation's financial condition); In re Burlington Coat Factory Sec. Litig., 114 F.3d 1410, 1415-17 (3d Cir. 1997) (projected future earnings); In re Westinghouse Sec. Litig., 90 F.3d 696, 700-01 (3d Cir. 1996) (fraudulent representation of company's state of affairs)).
The alleged material nondisclosure here consisted of a broker-dealer accepting an investor's order under the implied representation of the duty of best execution. This duty requires a broker-dealer to "use reasonable efforts to maximize the economic benefit to the client in each transaction." Newton, 135 F.3d at 270. A broker-dealer who "accepts such an order while intending to breach that duty makes a misrepresentation that is material to the purchase or sale [of a security]." Id. at 269. If the order was executed in a manner inconsistent with this duty, it was also performed with scienter. Id. at 273-74. Despite defendants' claim that execution at the NBBO price represented an acceptable industry-wide practice, we held in Newton that plaintiffs had alleged a claim that at least satisfied Rule 10b-5's material misrepresentation and scienter requirements. Id. at 274-75. We did not examine the other elements of a Rule 10b-5 claim--specifically, reliance and loss causation--because these elements were not relevant to the duty of best execution. On remand, the District Court picked up where we left off. Although the court found no issues affecting misrepresentation and scienter that would preclude class certification, it held that individual questions on reliance and economic loss presented formidable obstacles to class certification.
The parties disagree whether evidence of reliance and economic loss are consistent with each trade or would require individual treatment at trial. Defendants argue that reliance and economic loss cannot be presumed across the class for the hundreds of millions of trades at issue. Because only class members who detrimentally relied on a defendants' execution practice would have a cause of action, they maintain the individual inquiry necessary to establish reliance and economic loss renders plaintiffs' claims unfit for class certification. Whether proof of reliance and economic loss are unique to each investor, necessitating a trade-by-trade examination, remains contested.
B.
In Rule 10b-5 securities class actions, a plaintiff must prove reliance on a fraudulent material misrepresentation or omission. Kline v. First W. Gov't Sec., Inc. , 24 F.3d 480, 487 88 (3d Cir. 1994); Peil, 806 F.2d at 1160 (discussing reliance). "It is axiomatic that a private action for securities fraud must be dismissed when a plaintiff fails to plead that he or she reasonably and justifiably relied on an alleged misrepresentation." Semerenko, 223 F.3d at 178 (citing Weiner, 129 F.3d at 315; In re Burlington Coat Factory, 114 F.3d at 1417). This burden requires a plaintiff to demonstrate that defendants' conduct caused him" `to engage in the transaction in question.' " Robbins, 116 F.3d at 1447 (quoting Currie v. Cayman Res. Corp., 835 F.2d 780, 785 (11th Cir. 1988)); see also Peil, 806 F.2d at 1160. "Recognizing that the requirement of showing direct reliance presents an unreasonable evidentiary burden in a securities market where face-to-face transactions are rare and where lawsuits are brought by classes of investors . . . this court has adopted a rule that creates a presumption of reliance in certain cases." Semerenko, 223 F.3d at 178. "The reason for shifting the burden on the reliance issue has been an assumption that the plaintiff is generally incapable of proving that he relied on a material [nondisclosure]." Sharp v. Coopers & Lybrand, 649 F.2d 175, 188 (3d Cir. 1981), overruled in part on other grounds, In re Data Access Sys. Sec. Litig., 843 F.2d 1537 (3d Cir.1988) (en banc).
The seminal opinion on the presumption of reliance in securities fraud cases is Affiliated Ute Citizens of Utah v. United States, 406 U.S. 128, 153-54 (1970). Affiliated Ute involved an effort by some members of the Ute tribe to distribute its assets among its members. For this purpose, the tribe placed its assets in a corporation and issued each member ten shares of stock that were subsequently deposited in a local bank. As fiduciary, the bank assumed responsibility for enforcing the stocks' restrictions. For its own benefit and unknown to the Utes, the bank facilitated sales of the stock to outside investors at costs below its fair market value. When the Utes discovered the bank's practice, they sued under Rule 10b-5. In defense, the bank claimed the Utes failed to establish reliance on a misrepresentation of material fact. But the Supreme Court held the plaintiffs were entitled to a presumption of reliance, holding that in cases "involving primarily a failure to disclose [material facts], positive proof of reliance is not a prerequisite to recovery." Id. at 153. Applying this precept, we have held that "the proper approach to the problem of reliance is to analyze the plaintiff's allegations, in light of the likely proof at trial, and determine the most reasonable placement of the burden of proof of reliance." Sharp, 649 F.2d at 188; see also Joseph v. Wiles, 223 F.3d 1155, 1162 (10th Cir. 2000) (recognizing that " Affiliated Ute presumption of reliance exists in the first place to aid plaintiffs when reliance on a negative would be practically impossible"); Blackie v. Barrack, 524 F.2d 891, 907 (9th Cir. 1975) (same).
We extended the Affiliated Ute presumption of reliance to investors when securities dealers failed to disclose a pricing policy that overcharged investors in the purchase and sale of "penny stocks."14 In Hoxworth II, investors alleged they were systematically defrauded by a securities dealer's failure to disclose its pricing policy of excessive markups or markdowns on different securities. 980 F.2d 912. Because of this uniform, material nondisclosure, we concluded that the "plaintiffs [were] entitled to the presumption of reliance set forth in Affiliated Ute." Id. at 924; cf. Eisenberg v. Gagnon, 766 F.2d 770, 786-87 (3d Cir. 1985) (holding individual questions of reliance in securities class action involving investment in tax shelters did not preclude certification). In analogous cases, reliance has not been a hurdle to class certification. See Grandon v. Merrill Lynch & Co., Inc., 147 F.3d 184, 190 (2d Cir. 1998) ("A broker-dealer commits fraud (in violation of S 10(b) and Rule 10b-5) by charging customers excessive markups without proper disclosure."); Bank of Lexington & Trust Co. v. Vining-Sparks Sec., Inc., 959 F.2d 606, 613 (6th Cir. 1992) ("[T]he failure to disclose exorbitant mark-ups violates section 78j(b) and Rule 10b-5."); Angelastro v. Prudential-Bache Sec., Inc., 764 F.2d 939, 942-46 (3d Cir. 1985) (reliance does not bar private securities fraud action involving nondisclosure of fraudulent credit terms on margin accounts); Ettinger v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 122 F.R.D. 177, 180, 182 (E.D. Pa. 1988) (reliance not an issue in securities class action alleging securities dealer failed to disclose improper markups on bonds).
Investors may also be entitled to a rebuttable presumption of reliance under the "fraud-on-the-market theory." This is because "in an efficient market the misinformation directly affects the stock prices at which the investor trades and thus, through the inflated or deflated price, causes injury even in the absence of direct reliance." In re Burlington Coat Factory, 114 F.3d at 1419 n.8 (citing Basic Inc., 485 U.S. at 241-42). Reliance may be presumed when a fraudulent misrepresentation or omission impairs the value of a security traded in an efficient market. Basic Inc., 485 U.S. at 241-42; Semerenko, 223 F.3d at 178; In re Burlington Coat Factory, 114 F.3d at 1419 n.8. Here plaintiffs' claims do not involve an omission or misrepresentation that affected the value of a security in an efficient market. Therefore, a presumption of reliance based on this theory would be inappropriate.
The District Court did not explicitly rule on whether reliance could be presumed. Instead the court observed that the investors' trades "involved multiple circumstances which bear decisively upon the existence of reliance." In re Merrill Lynch, et al. Sec. Litig., 191 F.R.D. 391, 395 (D.N.J. 1999) (hereinafter "Merrill Lynch"). On this point, the court found that some plaintiffs may have known about the defendants' practice, belying their argument. Id. ("The degree of sophistication of the putative class members varies widely. Some, no doubt, were new to the world of NASDAQ trading; some were institutional investors.").
Plaintiffs contend that defendants' uniform practice of executing trades at the NBBO price, even if better prices were reasonably available from alternative sources, and their failure to disclose the practice to their customers warrant a presumption of reliance. Defendants respond that at least some plaintiffs knew of the execution practice which nullifies their reliance. In support, they cite several news articles describing the practice15 as well as an SEC report noting that institutional investors (who fall within the putative class's broad definition)16 used alternative electronic trading sources to obtain better prices for their trades. Br. of Appellees at 56-57. Because some plaintiffs knew or should have known of their practice, defendants assert that reasonable reliance on the alleged nondisclosure did not occur class-wide. For this reason, a presumption of reliance is arguably unavailable. See Straub v. Vaisman & Co., Inc., 540 F.2d 591, 595-98 (3d Cir. 1976).
While it seems apparent that some class members likely knew of defendants' practice, this knowledge does not necessarily invalidate the presumption. When defendants fail to disclose material information about a uniform practice involving the purchase or sale of securities, plaintiffs may be entitled to a presumption of reliance which defendants may rebut. See, e.g., Affiliated Ute, 406 U.S. at 153-54; Hoxworth II, 980 F.2d at 924; Blackie, 524 F.2d at 905-06; see also In re Prudential Ins. Co. of Am. Sales Practice Litig., 148 F.3d 283, 314 (3d Cir. 1998) (approvingly noting conclusion that "because plaintiffs' fraud-based claims stem largely from misleading omissions, reliance can be presumed") (quotation and citation omitted) (hereinafter "Prudential"). Presuming reliance class-wide is proper when the material nondisclosure is part of a common course of conduct.17 Hoxworth II, 980 F.2d at 924 (holding class entitled to presumption of reliance against securities dealer for failure to disclose exorbitant pricing policy for securities); see also Prudential, 148 F.3d at 314; Ettinger, 122 F.R.D. at 180.
To reiterate, the investors have alleged that the broker-dealers failed to disclose their policy of executing NASDAQ trades at the NBBO price. Like a securities dealer's failure to disclose its policy of overcharging investors, defendants' execution of investors' trades at the NBBO price, when better prices may have been available from alternative services, constitutes a potentially fraudulent common course of conduct from which reliance can be presumed. See Hoxworth II, 980 F.2d at 924 (holding plaintiffs entitled to presumption of reliance because of defendants' nondisclosure of pricing policy);18 see also Prudential, 148 F.3d at 314. We will not require each plaintiff to prove he relied on a practice which defendants did not affirmatively disclose. See Sharp, 649 F.2d at 188-89; Ettinger, 122 F.R.D. at 180. Because their allegations of a uniform nondisclosure would make it impractical for investors to affirmatively prove their lack of knowledge of defendants' practice, the burden of rebutting a presumption of reliance is properly placed on defendants here. Therefore, under this set of facts, we hold presuming reliance would be appropriate because defendants allegedly failed to disclose their trade execution practice.
C.
1.
Under Rule 10b-5, a plaintiff must also establish that he suffered an economic loss that was caused by defendant's fraudulent conduct. Semerenko, 223 F.3d at 185; Huddleston, 640 F.2d at 549. If economic loss is evident, then plaintiff must prove a "sufficient causal nexus between the loss and the alleged [nondisclosure]." Semerenko, 223 F.3d at 184. Loss causation derives its function from the "standard rule of tort law that the plaintiff must allege and prove that, but for the defendant's wrongdoing, the plaintiff would not have incurred the harm of which he complains." Bastian v. Petren Res. Corp., 892 F.2d 680, 685 (7th Cir. 1990).
Initially, loss causation was a requirement established by the courts. In re Phillips Petroleum Sec. Litig. , 881 F.2d 1236, 1244 (3d Cir. 1989) (citing Huddleston, 640 F.2d at 549); Angelastro, 764 F.2d at 942-43; see also Bastian, 892 F.2d at 685. But under the Private Securities Litigation Reform Act of 1995, it became a statutory element of private securities fraud claims under Rule 10b-5. The Act provides that "[i]n any private action arising under this chapter, the plaintiff shall have the burden of proving that the act or omission of the defendant alleged to violate this chapter caused the loss for which the plaintiff seeks to recover damages." 15 U.S.C. S 78u-4(b)(4).
In any event, it is necessary here to separate the concept of economic loss from the issue of loss causation. Of particular importance is whether plaintiffs have, in fact, suffered an economic loss. "[F]ailure to show actual damages is a fatal defect in a Rule 10b-5 cause of action." Feldman v. Pioneer Petroleum, Inc., 813 F.2d 296, 302 (10th Cir. 1987); see also 2 T. Hazen, Law of Securities Regulation S 13.7, p. 553 (3d ed. 1995) ("Failure to allege or prove actual damages will result in dismissal of any 10b-5 damage claim."). For this reason, "[i]nvestors cannot complain about a fraud that did not cause them any harm." Latigo Ventures v. Laventhol & Horwath, 876 F.2d 1322, 1325 (7th Cir. 1989); Huddleston, 640 F.2d at 555. The economic loss that plaintiffs claim would be the difference between the price at which their trades were executed and the "better" price allegedly available from an alternative trading source.19 Therefore, to show economic loss, plaintiffs must establish that a "better" price was obtainable for each executed trade. If a "better" price was unavailable for a particular trade, then a class member could not have suffered injury and cannot maintain a Rule 10b-5 claim.
2.
The District Court held that economic loss could neither be established nor presumed class-wide. Merrill Lynch, 191 F.R.D. at 397. Finding that defendants' practice did not detrimentally affect the value of plaintiffs' securities across the entire market nor did it necessarily result in overcharging, the District Court found no resemblance to cases where economic injury naturally flowed from defendant's alleged conduct. Id. at 396. Irrespective of reliance, the District Court found that, after reviewing the record, many investors received the best available price when defendants executed their trades at the NBBO listed price. Id. ("The record as it is presently constituted requires the conclusion that in a large number of transactions there were no better prices from other sources."). Drawing on the summary judgment record where it determined from a sample analysis of twelve trades executed by defendants that only one resulted in actual economic injury to a class representative, the District Court concluded that an undefined number of class members sustained no economic loss whatsoever, necessitating the conclusion that damages were not susceptible to class-wide proof.20 Id. at 396; see also In re Merrill Lynch et al., Sec. Litig., 911 F. Supp. at 766. Based on this reasoning, the District Court found the question of economic loss remained unique to each investor. Plaintiffs argue against extrapolating the improbability of class-wide damages from twelve trades and contend the District Court erred in finding that "many" class members were uninjured. But we agree with the District Court's finding that plaintiffs' claims would require individual treatment to determine actual injury.
In fraud-on-the-market or overcharging cases that warrant a presumption of reliance, plaintiffs satisfy their initial burden because they sustain economic loss by reason of the alleged conduct.21 In fraud-on-the-market cases, the price at which a stock is traded is presumably affected by the fraudulent information, thus injuring every investor who trades in the security. In re Burlington Coat Factory, 114 F.3d at 1419 n.8 (citing Basic Inc., 485 U.S. at 241-42). Nor was economic loss a question in those securities claims where defendants failed to disclose a fraudulent pricing policy that overcharged investors. Accordingly, presuming economic loss was the ineluctable by-product of the alleged fraud. The same does not hold true here. The execution of plaintiffs' trades at the NBBO listed price did not necessarily injure each class member. Plaintiffs may be entitled to a presumption of economic loss only when it is clear each class member has in fact sustained economic injury. See Semerenko, 223 F.3d at 185. In a securities class action, a putative class may presumptively establish economic loss on a common basis only if the evidence adequately demonstrates some loss to each individual plaintiff.
Because securities claims may take on several forms, proving economic loss on a common basis is a fact-specific inquiry. See EchoCath, 235 F.3d at 884; Grandon, 147 F.3d at 190; see also Bogosian, 561 F.2d at 454 (evaluating loss in antitrust class actions). We find no support in the case law for presuming economic injury for purposes of class certification in Rule 10b-5 claims22 absent indication that each plaintiff has suffered an economic loss. See Semerenko, 223 F.3d at 184-85 ("[W]here the claimed loss involves the purchase of a security at a price that is inflated due to an alleged misrepresentation, there is a sufficient causal nexus between the loss and the alleged misrepresentation to satisfy the loss causation requirement."); see also Robbins, 116 F.3d at 1448 ("Our decisions explicitly require proof of a causal connection between the misrepresentation and the investment's subsequent decline in value."); Hayes v. Gross, 982 F.2d 104, 107 (3d Cir. 1992) (injury assumed when security purchased at price inflated by fraudulent misrepresentation); Scattergood v. Perelman, 945 F.2d 618, 624 (3d Cir. 1991) (same).
In assessing the question of economic loss, it is important to bear in mind how the facts here differ from those in a typical securities action. Unlike a "fraud-on-the-market" claim, this case does not involve a misrepresentation or omission that decreased the value of a security. Furthermore, unlike excessive over-pricing policy claims, this case does not involve a practice that necessarily harmed investors across the class.23 See generally Grandon, 147 F.3d 184; Hoxworth II , 980 F.2d 912; Vining-Sparks, 959 F.2d 606; Angelastro, 764 F.2d 939; Ettinger, 122 F.R.D. 177. In this case, defendants allegedly executed trades solely at the NBBO price. Depending on the facts of each trade, the NBBO listed price may or may not have provided a class member with the best price. Therefore, economic loss to the plaintiffs cannot be presumed by the purchase or sale of a security at the NBBO price, and we will not presume it across the class.
In sum, we conclude that the putative class would be entitled to a rebuttable presumption of reliance but not of economic loss. Therefore, their claims do not warrant a rebuttable presumption of class-wide injury.24
VI.
Turning to the test for class certification, we must examine the elements of a Rule 10b-5 claim through the prism of Fed. R. Civ. P. 23 to determine whether the District Court abused its discretion in failing to certify the class. A putative class must satisfy the four conjunctive criteria of Fed. R. Civ. P. 23(a): numerosity, commonality, typicality, and adequacy of representation.25 A class seeking money damages must also satisfy the (b)(3) requirements of predominance and superiority--namely, whether common questions of law or fact predominate and whether the class action represents the superior method for adjudicating the case.26
Denying class certification, the District Court found that plaintiffs' claims were atypical and the class representatives inadequate to represent the class. On related grounds, the court also held that common issues did not predominate and the class action device was neither superior nor manageable. Merrill Lynch, 191 F.R.D. at 397-98. As noted, we review the District Court's decision denying class certification for abuse of discretion. See supra p. 165.
A.
The certification requirements of Fed. R. Civ. P. 23(a) embrace two rudimentary principles: 1) the necessity and efficiency of adjudicating the claims as a class and 2) the assurance of protecting the interests of absentee members.27 Baby Neal v. Casey, 43 F.3d 48, 55 (3d Cir. 1994); G.M. Trucks, 55 F.3d at 785. If the class does not satisfy each of the 23(a) criteria, the suit cannot be maintained as a class action. For this reason, we will address each criterion in turn.
1.
Numerosity requires a finding that the putative class is "so numerous that joinder of all members is impracticable." Fed. R. Civ. P. 23(a)(1). It is clear the size of putative class satisfies this criterion. There are hundreds of thousands of class members and joinder would be impracticable. Id.; Merrill Lynch, 191 F.R.D. at 394.
2.
" `The concepts of commonality and typicality are broadly defined and tend to merge,' " because they focus on similar aspects of the alleged claims. Barnes, 161 F.3d at 141 (quoting Baby Neal, 43 F.3d at 56). Commonality requires the presence of "questions of law or fact common to the class," Fed. R. Civ. P. 23(a)(2), and typicality demands that "the claims or defenses of the representative parties are typical of the claims or defenses of the class." Fed. R. Civ. P. 23(a)(3). The significance of commonality is self-evident: it provides the necessary glue among class members to make adjudicating the case as a class worthwhile. 1 Herbert B. Newberg & Alba Conte, Newberg on Class Actions S 3.01, p. 3-4 (3d ed. 1992). Typicality ensures the interests of the class and the class representatives are aligned "so that the latter will work to benefit the entire class through the pursuit of their own goals." Barnes, 161 F.3d at 141. "The typicality criterion is intended to preclude certification of those cases where the legal theories of the named plaintiffs potentially conflict with those of the absentees by requiring that the common claims are comparably central to the claims of the named plaintiffs as to the claims of the absentees." Baby Neal , 43 F.3d at 57 (citing Weiss v. York Hosp., 745 F.2d 786, 810 (3d Cir. 1984)).
We have set a low threshold for satisfying both requirements. See Barnes, 162 F.3d at 141 (noting claims based on common course of conduct satisfy typicality); In re Sch. Asbestos Litig., 789 F.2d 996, 1010 (3d Cir. 1986) (highlighting that the " `threshold of commonality is not high' ") (quoting Jenkins v. Raymark Indus., Inc., 782 F.2d 468, 472 (5th Cir. 1986)). That is, "Rule 23(a) does not require that class members share every factual and legal predicate to meet the commonality and typicality standards." G.M. Trucks, 55 F.3d at 817." `[N]either of these requirements mandates that all putative class members share identical claims.' " Prudential, 148 F.3d at 311 (quoting Baby Neal, 43 F.3d at 56). Nevertheless, we require courts to examine them separately because the criteria remain distinct. Baby Neal, 43 F.3d at 56.
a.
As noted, commonality does not require an identity of claims or facts among class members. Prudential , 148 F.3d at 310 (citing Baby Neal, 43 F.3d at 56)." `The commonality requirement will be satisfied if the named plaintiffs share at least one question of fact or law with the grievances of the prospective class.' " Id. (quoting Baby Neal, 43 F.3d at 56); Georgine v. Amchem Prods., Inc., 83 F.3d 610, 627 (3d Cir. 1996), aff'd sub nom., Amchem Prods., Inc. v. Windsor, 521 U.S. 591 (1997).
The District Court found, and it is not seriously contested on appeal, that common questions of law and fact are present. Merrill Lynch, 191 F.R.D. at 394. Whether the defendants' execution of their customers' trades at prices quoted on the NBBO violates Rule 10b-5 constitutes a factual and legal claim that is common to the entire class. In fact, plaintiffs' claims raise several common issues including: 1) did defendants intentionally execute the plaintiffs' orders at the NBBO price without examining other alternatives; 2) did defendants fail to disclose their practice in violation of their duty to their customers; 3) were defendants technologically capable of providing superior prices to those offered on the NBBO; and 4) did defendants' conduct violate S 10(b) of the Securities Exchange Act of 1934. See Br. of Appellants at 31. The District Court properly held the putative class satisfied this requirement.
b.
The typicality inquiry here centers on whether " `the named plaintiff[s'] individual circumstances are markedly different or . . . the legal theory upon which the claims are based differs from that upon which the claims of other class members will perforce be based.' " Eisenberg, 766 F.2d at 786 (quoting Weiss, 745 F.2d at 809 n.36). The criterion acts as a bar to class certification only when "the legal theories of the named representatives potentially conflict with those of the absentees." Georgine, 83 F.3d at 631. If the claims of the named plaintiffs and putative class members involve the same conduct by the defendant, typicality is established regardless of factual differences.28 Barnes, 161 F.3d at 141; see also 1 Newberg on Class Actions S 3.15, p. 3-78 ("Factual differences will not render a claim atypical if the claim arises from the same event or practice or course of conduct that gives rise to the claims of the class members, and if it is based on the same legal theory."). Our jurisprudence "assures that a claim framed as a violative practice can support a class action embracing a variety of injuries so long as those injuries can all be linked to the practice." Baby Neal, 43 F.3d at 63 (discussing Falcon, 457 U.S. 147). As a result, we have concluded that the requirement "does not mandate that all putative class members share identical claims," Barnes, 161 F.3d at 141, because " `even relatively pronounced factual differences will generally not preclude a finding of typicality where there is a strong similarity of legal theories' or where the claim arises from the same practice or course of conduct." Prudential, 148 F.3d at 311 (quoting Baby Neal, 43 F.3d at 58); Hoxworth II, 980 F.2d at 923.
The District Court found that the different circumstances surrounding each trade over the class period rendered the claims of the named representatives "[a]typical of those members of the huge class." Merrill Lynch, 191 F.R.D. at 397. It reasoned that "[i]f proof of the representatives' claims would not necessarily prove all the proposed class members['] claims, the representatives['] claims are not typical of the proposed members' claims." Id. The District Court also believed that individual questions on reliance and injury buttressed its finding. Id. But typicality does not require similarity of individual questions concerning reliance or damages on the part of the class representatives and class members in a securities fraud action. Blackie, 524 F.2d at 905-06. In fact, whether the class representatives' claims prove the claims of the entire class highlights important issues of individual reliance and damages that are more properly considered and relevant under the predominance and superiority analysis.
In Hoxworth II, we found a putative class of securities investors, who had purchased or sold excessively marked-up securities, satisfied the typicality requirement. 980 F.2d at 923. Although the class members may have purchased or sold different securities at varying prices, all their claims stemmed from defendant's "course of conduct in failing to advise purchasers of its excessive markup policy." Id.; see also Ettinger, 122 F.R.D. at 180-81 (holding typicality satisfied where "[p]laintiff 's claims and those of the class arise from the same conduct of defendant and are based on the same legal theory"). Similarly, in Eisenberg v. Gagnon, we held that securities claims involving fraudulent inducement to invest in worthless tax shelters satisfied typicality. 766 F.2d 770 (3d Cir. 1985). Although the named plaintiffs invested in different tax shelters, their investments were "prepared by the same defendants, and contain[ed] the same alleged omissions and misrepresentations." Id. at 786. The typicality in their legal claims was sufficient to meet this criterion.
The named plaintiffs here, like the members of the putative class, are purchasers and sellers of securities on the NASDAQ. They allege that defendants violated their duty of best execution by automatically executing each investor's trade at prices listed on the NBBO without consulting alternative sources that may have provided better value. Plaintiffs' claims rest solely on a single legal theory--a Rule 10b-5 violation-- and allege a uniform course of conduct--automatic execution of their trades at the NBBO listed price. Any differences then among class members are factual--which security, at what price, under what circumstances, etc. The alleged cause of their injuries, however, remains typical throughout the class. Because each class member "would need to demonstrate the existence of this scheme, their interests are sufficiently aligned that the class representatives can be expected to adequately pursue the interests of the absentee class members." Prudential, 148 F.3d. at 312. The inability of a class representative to prove every other class members' claim does not necessarily result in failure of the typicality requirement. The District Court erred in finding potential factual differences rendered plaintiffs' claims atypical.
3.
Class representatives must "fairly and adequately protect the interests of the class." Fed. R. Civ. P. 23(a)(4). This requires a determination of (1) whether the representatives' interests conflict with those of the class and (2) whether the class attorney is capable of representing the class. Falcon, 457 U.S. at 157 & n.13; Barnes, 161 F.3d at 141. The Supreme Court has counseled that this element "serves to uncover conflicts of interest between named parties and the class they seek to represent." Amchem, 521 U.S. at 625. It also functions as a catch-all requirement that "tend[s] to merge with the commonality and typicality criteria of Rule 23(a)." Id. at 626 n.20.29
After determining economic loss could not be presumed class-wide,30 the District Court found the class's employment of a statistical formula to calculate and allocate damages would create conflicts between plaintiffs who were actually injured and uninjured plaintiffs in search of a windfall. Merrill Lynch, 191 F.R.D. at 398. Questioning plaintiffs' ability to resolve this conflict, the court refrained from deciding whether the class satisfied the adequacy of representation requirement.31
Following the Supreme Court's observation that adequacy of representation is an admixture of commonality and typicality, the District Court's reservation appears to be based on its earlier conclusion that the class did not satisfy the typicality requirement. As noted, the District Court's typicality concerns reflect inquiries better addressed under our review of predominance and superiority. See supra p. 185. While the commonality and typicality criteria tend to merge into an analysis of adequacy of representation under Fed. R. Civ. P. 23(a), the standards for measuring the predominance of common issues under Fed. R. Civ. P. 23(b)(3) should not be imputed to adequacy of representation.32 That is to say, the reasons for denying certification under the predominance standard may not necessarily compel denying certification under 23(a)(4), because the predominance requirement is more stringent than commonality and typicality. See supra note 32. On these facts, we hold that counsel would suitably represent the class and that there are no foreseeable conflicts between the named representatives and the class they seek to represent that would bar certification. Therefore, the putative class would satisfy the adequacy of representation requirement.
B.
Class certification under Fed. R. Civ. P. 23(b)(3) must also satisfy the twin requirements of predominance and superiority. The predominance inquiry demands "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). Superiority calls for a determination that a class action is the best method of achieving a "fair and efficient adjudication of the controversy." Id. The Supreme Court has explained that these elements were adopted
to cover cases in which a class action would achieve economies of time, effort, and expense, and promote . . . uniformity of decision as to persons similarly situated, without sacrificing procedural fairness or bringing about other undesirable results . . . [which] . . . invite[ ] a close look at the case before it is accepted as a class action . . . .
Amchem, 521 U.S. at 615 (internal quotations and citations omitted). To assist in this "close look," Fed. R. Civ. P. 23(b)(3) includes a nonexclusive list of relevant factors:
(A) the interest of members of the class in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already commenced by or against members of the class; (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; (D) the difficulties likely to be encountered in the management of a class action.
Fed. R. Civ. P. 23(b)(3). Simply, "[i]ssues common to the class must predominate over individual issues, and the class action device must be superior to other means of handling the litigation." Prudential, 148 F.3d at 313-14.
1.
Predominance measures whether the class is sufficiently cohesive to warrant certification. Amchem, 521 U.S. at 623. Unlike commonality, predominance is significantly more demanding, requiring more than a common claim. Id. at 623-24. After holding the class was not entitled to a presumption of class-wide loss, the District Court found that individual questions of whether each class member sustained economic injury presented insurmountable obstacles to certification. Merrill Lynch, 191 F.R.D. at 396-97 ("[A]bsent proof of classwide pecuniary loss resulting from that reliance, there can be no classwide claim for securities fraud."). Examining millions of trades to ascertain whether or not there was injury, said the court, meant that individual issues overwhelmed common questions among the class. Id. at 397-98. We agree.
Because the automated execution of orders at the NBBO listed price did not necessarily injure each class member, the District Court found that "whether a class member suffered damages would have to be determined on a trade by trade basis," because "some class members would have suffered damages; while some would not." Id. at 396. This individual inquiry is complicated by several factors. Assessing economic injury to a class member would first require examining whether a particular trade provided an investor with "the best reasonably available price." Newton, 135 F.3d at 270. The comparison between the price at which a particular trade was executed on the NBBO with the prices and trades available at the same time on alternative electronic sources would only begin to answer this question. As the Newton court recognized:
[A]scertaining what prices are reasonably available in any particular situation may require a factual inquiry into all of the surrounding circumstances . . . .
* * * * * *
Other terms in addition to price are also relevant to best execution. In determining how to execute a client's order, a broker-dealer must take into account order size, trading characteristics of the security, speed of execution, clearing costs, and the cost and difficulty of executing an order in a particular market. When the plaintiffs state that better "prices" were reasonably available from sources other than the NBBO, we understand that to mean that, given an evaluation of price as well as all of the other relevant terms, the trade would be better executed through a source of liquidity other than the NBBO (e.g. SelectNet, Instinet, in-house limit orders or market orders held by the defendants, or limit orders place by the public in the Small Order Execution System).
Id. at 270 & n.2 (internal citations omitted). These factors would appear to vary from class member to class member and, for each class member, from trade to trade. Whether a class member suffered economic loss from a given securities transaction would require proof of the circumstances surrounding each trade, the available alternative prices, and the state of mind of each investor at the time the trade was requested. This Herculean task, involving hundreds of millions of transactions, counsels against finding predominance.
In an effort to gloss over this requirement, plaintiffs suggest their expert could calculate the amount of damages each class member sustained thereby removing proof of injury as an obstacle to certification. In a sworn declaration, plaintiffs' expert provided no model formula, but instead projected that he could devise a formula that would measure damages among the class and serve as a plan for allocation.33 We are not convinced. But even if plaintiffs could present a viable formula for calculating damages (which they have not), defendants could still require individualized proof of economic loss. See, e.g., Kline, 508 F.2d at 236 & n.9.
The District Court rejected plaintiffs' arguments. Drawing guidance from antitrust jurisprudence, the court concluded that "[p]roof of damage . . . must be distinguished from the mere calculation of damages." Merrill Lynch , 191 F.R.D. at 396. As the Court of Appeals for the Eighth Circuit recognized after reviewing Supreme Court jurisprudence, " `an antitrust plaintiff must prove that his damages were caused by the unlawful acts of the defendant. . . [before] the amount of damages may be determined.' " Amerinet v. Xerox Corp., 972 F.2d 1483, 1494 (8th Cir. 1992) (quoting MCI Communications Corp. v. Am. Tel. & Tel. Co., 708 F.2d 1081, 1161 (7th Cir. 1983)). On this basis, the District Court reasoned that
[c]lass treatment of damages issues, however, presumes the ability to prove the fact of damage without becoming enmeshed in individual questions of actual damage . . . [.] Where proof of fact of damage requires evidence concerning individual class members, the common questions of fact become subordinate to the individual issues, thereby rendering class certification problematic.
Merrill Lynch, 191 F.R.D. at 396 (quotation and citation omitted). Proof of injury (whether or not an injury occurred at all) must be distinguished from calculation of damages (which determines the actual value of the injury). Even assuming plaintiffs' ability to calculate damages, the District Could held this did not exempt them from proving each class member suffered economic injury. Therefore, the court found that determining actual economic loss on the part of each investor would involve individual questions that predominate over common ones.
The District Court's analogy to antitrust class actions is well-taken. In a Rule 10b-5 securities claim that "impacts upon a class of persons . . . there is no reason . .. why proof of the impact cannot be made on a common basis so long as the common proof adequately demonstrates some damage to each individual." Bogosian, 561 F.2d at 454 (discussing injury in antitrust cases). The ability to calculate the aggregate amount of damages does not absolve plaintiffs from the duty to prove each investor was harmed by the defendants' practice. In class actions based on a "fraud-on-the-market," an excessive pricing policy for securities, or an antitrust violation, the alleged conduct itself causes economic injury. But only those class members whose trades could have been executed at better prices sustained economic injury here. Determining which class members were economically harmed would require an individual analysis into each trade and its alternatives. The individual questions, therefore, are overpowering. As we held in Georgine:
Even if we were to assume that some issues common to the class beyond the essentially settled question of the harmfulness of asbestos exposure remain, the huge number of important individualized issues overwhelm any common questions. Given the multiplicity of individualized factual and legal issues, . . . we can by no means conclude that the questions of law or fact common to the members of the class predominate over any questions affecting only individual members.
83 F.3d at 630 (internal quotes omitted); Barnes , 161 F.3d at 146 ("Because nicotine addiction must be determined on an individual basis and remains an essential part of plaintiffs' . . . claim . . . class treatment is inappropriate."). While obstacles to calculating damages may not preclude class certification, the putative class must first demonstrate economic loss on a common basis. As noted, the issue is not the calculation of damages but whether or not class members have any claims at all.
The District Court was also guided by our decision in Georgine. Merrill Lynch, 191 F.R.D. at 396 ("Although in Georgine, as in the present case, there were several common questions, the Court held that class treatment was inappropriate because `each individual plaintiff's claim raises radically different factual and legal issues from those of other plaintiffs . . . [.] In such circumstances, the predominance requirement of Rule 23(b)(3) cannot be met.' ") (quoting Georgine, 83 F.3d at 618). In Amchem, the Supreme Court affirmed our determination that a settlement class of individuals exposed to asbestos products failed the predominance prong because of significant individual issues surrounding each claim. The plaintiffs had been exposed to "different asbestos containing products, for different amounts of time, in different ways, and over different periods." Georgine, 83 F.3d at 626. There were also different classes of plaintiffs--some who were presently injured and some who had been exposed but whose future injury remained speculative. Id. The individualized differences as to amount of asbestos exposure and future injury were significant because they would "lead to disparate applications of legal rules, including matters of causation, comparative fault, and the types of damages available to each plaintiff." Id. at 627. For these reasons, the constellation of individual issues eclipsed common questions.
Citing the Supreme Court's guidance that "[p]redominance is a test readily met in certain cases alleging consumer or securities fraud or violations of the antitrust laws," Amchem, 521 U.S. at 625 (citing Comm. Note, Fed. R. Civ. P. 23), plaintiffs contend Amchem, a mass tort action, may be distinguished, and argue that individual economic injury need not be factored into the predominance calculus. See Hoxworth II, 980 F.2d at 924 (holding that "uniform scheme to defraud investors in the class securities . . . would support class action treatment"); Ettinger, 122 F.R.D. at 182 (common question of exorbitant markups on securities predominates over individual damage calculations).
Although the securities claims are unlike the mass tort claims in Amchem, the obstacles to satisfying the predominance requirement are comparable. In Amchem, the Court found that individual questions on the varying degrees and effects of asbestos exposure overpowered common ones. The breadth of the claims here may be different from Amchem, but the sheer number of claims raising individual questions of injury is strikingly similar. In Georgine, we recognized "individualized issues can become overwhelming in actions involving long-term mass torts (i.e., those which do not rise out of a single accident)." 161 F.3d at 628. The alleged injuries in Newton arise out of the execution of hundreds of millions of trades, not a single act of fraudulent conduct. The distinct facts among the hundreds of thousands of plaintiffs involving hundreds of millions of trades will determine whether securities violations occurred. Because plaintiffs' claims will require an economic injury determination for each trade, we hold the putative class fails to satisfy the predominance requirement.
Moreover, as we have noted in securities cases involving fraud-on-the-market or excessive markups, injury necessarily flowed from defendant's conduct and reliance and injury could be presumed. In those cases, if defendant's conduct was held fraudulent, a claim of loss naturally followed. Here it remains contested whether defendants' conduct in each trade was fraudulent as well as whether the investors suffered a loss as a result. Because it is clear that at least some of the plaintiffs have not suffered economic injury, individual questions remain that would have to be adjudicated separately. For these reasons, we hold this case does not fall within the scope of those "certain cases alleging . . . securities fraud," Amchem, 521 U.S. at 625 (emphasis supplied), in which predominance may be readily established.34
Because economic loss cannot be presumed, ascertaining which class members have sustained injury means individual issues predominate over common ones. Therefore, the District Court exercised its sound discretion in finding the putative class did not satisfy the predominance requirement.
2.
Even if reliance and damages could be presumed or determined in separate proceedings after certification, this action fails to satisfy Fed. R. Civ. P. 23(b)(3)'s superiority requirement.
A class action must represent the best "available method[ ] for the fair and efficient adjudication of the controversy." Fed. R. Civ. P. 23(b)(3). Here we must address "the difficulties likely to be encountered in the management of a class action." Fed. R. Civ. P. 23(b)(3)(D). According to the District Court, the need for individualized inquiry into actual injury transformed the "[e]xploration of each and every customer's NASDAQ transactions with defendants during the period from November 4, 1992 to August 28, 1996 [into] a mind-boggling undertaking." Merrill Lynch, 191 F.R.D. at 398. We agree. With hundreds of millions of trades, it is difficult to imagine how this case can be tried.35
Contending each individual claim is so small that only a class action will provide a remedy, plaintiffs maintain that denying certification will absolve defendants from wrongdoing. The District Court rejected this rationale as a "basis for excusing plaintiffs from proving the essential elements of their cause of action." Merrill Lynch, 191 F.R.D. at 398. We agree. Recently we held this factor "by itself is insufficient to overcome the hurdles of predominance and superiority and efficient and fair management of a trial, which Rule 23(b) requires." In re LifeUSA Holding Inc., 242 F.3d at 148 n.13. We also recognize that some class members, such as large institutional investors who fall within the class definition, arguably would have a significant financial stake to raise stand-alone claims.
Turning to manageability, the District Court's evaluation must be "granted a wide range of discretion." Link v. Mercedes Benz of N. Am., Inc., 550 F.2d 860, 864 (3d Cir. 1977). "Manageability is a practical problem, one with which a district court generally has a greater degree of expertise and familiarity than does the appellate court." In re Sch. Asbestos Litig., 789 F.2d at 1011. It encompasses "the whole range of practical problems that may render the class action format inappropriate for a particular suit." Eisen, 417 U.S. at 164.
Here there are hundreds of millions of transactions executed over several years. Plaintiffs maintain their expert can devise a formula for calculating injury and damages that will allay manageability concerns. Yet we are hesitant to rely on a formulaic nostrum given the consequences if it fails to meet expectations. See Windham, 565 F.2d at 70 ("But where the court finds, on the basis of substantial evidence as here, that there are serious problems now appearing, it should not certify the class merely on the assurance of counsel that some solution will be found."). As noted, actual injury cannot be presumed, and defendants have the right to raise individual defenses against each class member. See In re Sch. Asbestos Litig., 789 F.2d at 1011 ("The potential for individual defenses . . . clearly poses significant case management concerns."). We hold that establishing proof of the plaintiffs' injuries and litigating the defenses available to the defendants would present insurmountable manageability problems for the District Court.
The superiority requirement also casts serious doubt on the efficiency and manageability of certifying this class for trial. "In terms of efficiency, a class of this magnitude and complexity could not be tried. There are simply too many uncommon issues, and the number of class members is surely too large. Considered as a litigation class, then, the difficulties likely to be encountered in the management of this action are insurmountable." Georgine, 83 F.3d at 632-33.36 Although plaintiffs attempt to fit this case under Prudential, that case raised different issues because the class was certified for the purpose of settlement. This is significant because "the settlement approval inquiry is far different from the certification inquiry. In settlement situations, the superiority requirement arguably translates into the question whether the settlement is a more desirable outcome for the class than individualized litigation, and may assure that the settlement has not grossly undervalued plaintiffs' interests." G.M. Trucks, 55 F.3d at 796. Significantly, in Prudential we did not have to "inquire whether the case, if tried, would present intractable management problems . . . for the proposal [was] that there be no trial." Amchem , 521 U.S. at 620. Additionally we have recognized that adjudicating Rule 10b-5 securities claims as a class action satisfies superiority only if the litigation results in fewer individual actions. Because injury determinations must be made on an individual basis in this case, adjudicating the claims as a class will not reduce litigation or save scarce judicial resources. Under these circumstances, plaintiffs fail to satisfy the superiority standard. See G.M. Trucks, 55 F.3d at 783 ("One of the paramount values in [class actions] is efficiency. Class certification enables courts to treat common claims together, obviating the need for repeated adjudications on the same issue.").
We are also mindful that Amchem and Prudential involved mature claims. The class settlements were the result of verdicts on established liability and damages awards. This case does not share a similar track record. Of course, many securities fraud claims do not generally implicate maturity concerns because they do not raise complex issues of causation and injury. Furthermore, the divergent outcomes in Amchem and Prudential make it clear that maturity alone is neither necessary nor sufficient for certification, but it may help to ensure that class certification is "superior to individual adjudication." Castano, 84 F.3d at 747.
The specter of adjudicating plaintiffs' claims at trial is, at the very least, daunting. Individual questions of economic loss present insurmountable manageability problems. Moreover, class certification would place hydraulic pressure on defendants to settle which weighs in the superiority analysis. See supra note 8. At trial, determining actual injury would require hundreds of millions of individual assessments. For these reasons, the District Court was clearly within its sound discretion to hold this case failed to satisfy the superiority requirement.
VII.
In sum, although the putative class satisfies the requirements of Fed. R. Civ. P. 23(a), it cannot meet the requirements of Fed. R. Civ. P. 23(b)(3). In particular, the investors' claims fail the predominance and superiority requirements under Fed. R. Civ. P. 23(b)(3). For these reasons, we will affirm the judgment of the District Court.
Notes:
The permissive interlocutory appeal provision was adopted under the power conferred by 28 U.S.C. S 1292(e).
Before Rule 23(f) was promulgated, the Supreme Court rejected the "death knell" doctrine as a justification for circumventing the federal-appellate-jurisdiction precondition that a district court decision " `end[ ] the litigation on the merits and leave[ ] nothing for the court to do but execute the judgment.' " Coopers & Lybrand v. Livesay, 437 U.S. 463, 467 (1978) (quoting Catlin v. United States, 324 U.S. 229, 233 (1945)). In these instances, appellate jurisdiction was limited by 28 U.S.C. S 1291, which provided that the "courts of appeals shall have jurisdiction of appeals from all decisions of the district courts of the United States . . . except where a direct review may be had in the Supreme Court." 28 U.S.C. S 1291 (1978). Because plaintiffs had the opportunity to pursue litigation individually if class certification was denied, a district court decision decertifying a putative class did not constitute a final decision under 28 U.S.C. S 1291. Livesay, 437 U.S. at 467. At the time, there existed no special rules on appealing class certification decisions. Reasoning that a "death knell" exception would have to apply with equal force to all forms of litigation, the Court rejected this proposition. While the Court recognized several policy arguments in favor of permitting appeals of certification decisions which effectively put an end to litigation, without legislative guidance or authority, it ultimately found the arguments against such a rule more persuasive. Id. at 470-77. The new Rule 23(f) provides the authority as well as the guidance for these appeals which was previously wanting.
Fortunately, four of our sister circuits have written thoughtful opinions on the new rule. Lienhart v. Dryvit Sys., Inc., No. 00-908, 255 F.3d 138 142-46 (4th Cir. June 26, 2001); Prado-Steiman v. Bush, 221 F.3d 1266, 1271-77 (11th Cir. 2000); Waste Mgmt. Holdings, Inc. v. Mowbray, 208 F.3d 288, 292-95 (1st Cir. 2000); Blair v. Equifax Checking Servs., Inc., 181 F.3d 832, 833-36 (7th Cir. 1999).
In effect, the Rule authorizes appellate courts to "restore equilibrium when a doubtful class certification ruling would virtually compel a party to abandon a potentially meritorious claim or defense before trial." Mowbray, 208 F.3d at 293.
Other courts of appeals have adopted a taxonomy based on these principles. In the first case examining the standards for interlocutory appeal, Blair v. Equifax Checking Servs., Inc. , 181 F.3d 832, 833-36 (7th Cir. 1999), the Court of Appeals for the Seventh Circuit provided an in-depth description of the three examples mentioned above that would merit exercise of interlocutory review. Taking its cue from the Committee Note, the court held that cases where certification tolled the "death knell" of litigation for plaintiffs or placed irresistible pressure to settle on defendants presented circumstances ripe for review. The court also held that appeals which would help develop the law similarly invited the exercise of this review. In Waste Mgmt. Holdings, Inc. v. Mowbray, 208 F.3d 288, 292-95 (1st Cir. 2000), the Court of Appeals for the First Circuit largely adopted the Seventh Circuit's methodology with one restriction. To prevent fecund legal minds from framing every legal issue as an important question of fundamental law, the court narrowed this review to cases in which "an appeal will permit the resolution of an unsettled legal issue that is important to the particular litigation as well as important in itself and likely to escape effective review if left hanging until the end of the case." Id. at 294. Thereafter, the Court of Appeals for the Eleventh Circuit discussed the standards for reviewing petitions in Prado-Steiman v. Bush, 221 F.3d 1266, 1271-77 (11th Cir. 2000). Adding other factors, the court elaborated on the principles set forth previously by the Court of Appeals for the Seventh Circuit. In addition to those already mentioned, the court looked to (1) whether the certification decision is likely dispositive of the litigation; (2) whether the decision involved a novel or unsettled legal question; (3) the strength of the district court's reasoning; (4) the status of the case before the district court; and (5) the "likelihood that future events may make immediate appellate review more or less appropriate." Id. at 1276. Recently, the Court of Appeals for the Fourth Circuit reviewed the standards for granting a motion under Fed. R. Civ. P. 23(f) and adopted the analysis enunciated in Prado-Steiman. Lienhart v. Dryvit Sys., Inc., No. 00-908, 255 F.3d 138,142-46 (4th Cir. June 26, 2001).
As set forth in the Note, the district courts "having walked through the certification decision, can provide cogent advice on the factors" animating their decisions. Comm. Note, Fed. R. Civ. P. 23(f). Furthermore, because permission to appeal does not stay trial court proceedings, any stay should be sought first from the trial court. Id.
In its amicus brief, the Securities Industry Association contends we should be wary of extending class certification to cases where the court will in effect set market standards (such as "best execution") and, by doing so, affect the certainty of capital markets. Generally, it is desirable for these types of changes to occur through rule making by the appropriate agency. But courts should not hesitate to provide remedies for litigants injured by unlawful conduct that may not clearly violate regulatory standards. Newton v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 135 F.3d 266, 274 (3d Cir. 1998) (en banc) ("[T]here is no statute, rule, regulation, or interpretation, by the SEC or by a court, that authoritatively establishes that, for all trades, the NBBO exhausted the category of `reasonably available prices' during the class period. This absence of precedent did not, however, absolve the district court of the duty to resolve the plaintiffs' securities fraud claim once it was presented in this suit."); see also Deposit Guaranty Nat'l Bank v. Roper, 445 U.S. 326, 339 (1980) ("The aggregation of individual claims in the context of a classwide suit is an evolutionary response to the existence of injuries unremedied by the regulatory action of government.").
Even the Securities and Exchange Commission has argued that "the mere fact that the [Securities and Exchange] Commission was considering (and has now adopted) rules that prospectively affect . . . broker-dealers' order handling obligations would not make it appropriate for the court to abstain from deciding whether the defendants committed fraud with respect to [their duty of best execution] as it existed during the period at issue in this case." Br. of Amicus Curiae the Securities and Exchange Commission, at 12 n.14, in Newton, 135 F.3d 266.
Trial of plaintiffs' claims here touches on concerns raised by the Court of Appeals for the Seventh Circuit in In re Rhone-Poulenc Rorer Inc., 51 F.3d 1293 (7th Cir. 1995). In Rhone-Poulenc, the court granted a writ of mandamus to reverse the certification of a class of hemophiliacs who received blood contaminated with HIV. The class involved only a few hundred parties, but each individual had claims possibly worth millions of dollars. The court reasoned that the enormous size of the potential liability would impose an "intense pressure to settle," id. at 1298, because "[t]he risk of facing an all-or-nothing verdict presents too high a risk, even when the probability of an adverse judgment is low." Castano v. Am. Tobacco Co., 84 F.3d 734, 746 (5th Cir. 1996) (citing Rhone-Poulenc, 51 F.3d at 1298). It considered settlements forcibly induced by the small probability of an immense judgment "blackmail settlements." Rhone-Poulenc, 51 F.3d at 1298; see also Castano, 84 F.3d at 746; G.M. Trucks, 55 F.3d at 784-85. Although finding the hydraulic pressure to settle should not dispositively affect a certification decision, the court suggested that it should be balanced against the benefits of a class action. Rhone-Poulenc, 51 F.3d at 1299 ("We do not want to be misunderstood as saying that class actions are bad because they place pressure on defendants to settle. That pressure is a reality, but it must be balanced against the undoubted benefits of the class action that have made it an authorized procedure for employment by federal courts."); see also Kline v. Coldwell, Banker & Co. 508 F.2d 226, 238 (9th Cir. 1974) ("I doubt that plaintiffs' counsel expect the immense and unmanageable case that they seek to create to be tried. What they seek to create will become (whether they intend this result or not) an overwhelmingly costly and potent engine for the compulsion of settlements, whether just or unjust.") (Duniway, J., concurring). The Supreme Court has also recognized the dynamic pressure certification sets in motion. The Court has observed that "[c]ertification of a large class may so increase the defendant's potential damages liability and litigation costs that he may find it economically prudent to settle and to abandon a meritorious defense." Livesay, 437 U.S. at 476. Certifying this class raises a similar concern because the size of the class and number of claims may place acute and unwarranted pressure on defendants to settle. It is a factor we weigh in our certification calculus.
See Geoffrey C. Hazard, Jr., Class Certification Based on Merits of the Claims, 69 Tenn. L. Rev. (forthcoming Fall 2001).
The defendants who executed the plaintiffs' orders are Merrill Lynch, Pierce, Fenner & Smith, Inc.; Dean Witter Reynolds, Inc.; and PaineWebber, Inc. Each defendant is an "integrated broker/dealer" brokerage company that executed trades both as an agent and a principal. For certification, plaintiffs are divided into three subclasses. Each subclass consists of all the class members that placed market orders with a particular defendant.
For a more detailed description of the alternative avenues for executing trades on the NASDAQ (e.g., Instinet, SelectNet) see the district court opinion granting defendants summary judgment, In re Merrill Lynch, et al. Sec. Litig., 911 F. Supp. 745, 759-60 (E.D. Pa. 1995), rev'd, Newton, 135 F.3d 266.
Defendants' duty to provide best execution remained consistent whether they were acting as agents of the trade or principals. Newton, 135 F.3d at 270 n.1.
As noted in Newton, a three month study of prices within the class period indicated that "85% of the bids and offers displayed by market makers on Instinet and 90% of the bids and offers displayed on SelectNet were at better prices than those posted publicly on NASDAQ." 135 F.3d at 272 (citing Order Execution Obligations, Exchange Act. Rel. No. 34-37619A, 61 Fed. Reg. 48290, 48307-08 (Sept. 12, 1996)). These apparently incriminating percentages may be mitigated, however, by the fact that during a full year of the class period (1993), for example, the electronic communication networks (including services like SelectNet and Instinet) "accounted for only 13% of share volume in Nasdaq securities and only 1.4% of listed share volume." Securities and Exchange Commission, Div. of Market Regulation, Electronic Communication Networks and After-Hours Trading 6 (June 2000) (citing Market 2000 Report: Study II, Structure of the U.S. Equity Markets, 1994 SEC Lexis 133, at *43-44 (Jan. 1994)), available at http://www.sec.gov.news/studies/ecnafter.htm. Based on these figures, defendants contend plaintiffs' trades could have been executed at superior prices from alternative sources only about 30% of the time during the class period. See Br. of Appellees at 8 n.7.
"Penny stocks are low-priced, high-risk equity securities for which there is frequently no well-developed market." Hoxworth II, 980 F.2d at 914 n.1.
See, e.g., Floyd Norris, The S.E.C. Tries to Insure that Investors Get Better Stock Prices, N.Y. Times, Sept. 28, 1995, at D8; Daniel Kadlec, Young Traders Can Be Gamble for Street, USA Today, Sept. 27, 1995, at 3B; Warren Getler, Reuter's Instinet is Biting Off Chunks of Nasdaq's Territory, Wall St. J., Oct. 4, 1994, at C1; Gretchen Morgenson, Fun and Games on Nasdaq, Forbes, Aug. 16, 1993, at 74; Craig Torres, How Street Turns Your Stock Trades to Gold, Wall St. J., Feb. 16, 1993, at C1.
Plaintiffs define the class as consisting "of all persons who placed market orders with Merrill Lynch or PaineWebber or Dean Witter to purchase or sell shares of OTC stock between November 4, 1992 and August 28, 1996."
Further explaining the justification for presuming reliance from material nondisclosures, we noted in Sharp that " `[s]ince nothing is affirmatively represented in a nondisclosure case, demanding proof of reliance would require the plaintiff to demonstrate that he had in mind the converse of the omitted facts, which would be virtually impossible to demonstrate in most cases.' " 649 F.2d at 188 n.18 (quoting Note, The Reliance Requirement in Private Actions Under SEC Rule 10b-5, 88 Harv. L. Rev. 584, 590 (1975) (footnote omitted)). The justifications for creating presumptions in general are explored in greater depth in 2 J. Strong, McCormick on Evidence S 343, p. 437-43 (5th ed. 1999).
This was based on an earlier decision by our circuit that the "burden shifting rationale of Affiliated Ute was fully applicable" to the nondisclosure of an exorbitant markup pricing policy. Hoxworth v. Blinder, Robinson & Co., Inc., 903 F.2d 186, 202 (3d Cir. 1990) (hereinafter "Hoxworth I").
The Court of Appeals for the Tenth Circuit has explained how the measure of damages should be calculated under Rule 10b-5:
Although neither Section 10(b) of the [Securities Exchange] Act nor Rule 10b-5 contains explicit provisions for determining damages, courts have applied the "actual damages" standard of Section 28 of the Securities Exchange Act of 1934, 15 U.S.C. Sec. 78bb(a), to Rule 10b-5 claims. Under Section 28 of the Act, the "correct measure of damages . . . is the difference between the fair value of all that the (plaintiff) received and the fair value of what he would have received had there been no fraudulent conduct." Randall v. Loftsgaarden, 478 U.S. 647, 661-62 (1986).
Feldman, 813 F.2d at 301-02 (footnotes omitted).
In defendants' motion for summary judgment, the District Court examined twelve of the class representatives' trades with the defendants during the initial class period (November 4, 1992 to November 4, 1994). Although the court's opinion mentions thirteen trades, only twelve were evaluated. While the court was provided pricing information from SelectNet to compare the prices at which the trades were executed, "[n]o information on either Instinet prices, SOES limit order file prices or payment for order flow arrangements was supplied for any of the transactions." In re Merrill Lynch, et al. Sec. Litig., 911 F. Supp. at 766 (footnote omitted).
In the only trade where the District Court found a plaintiff had clearly sustained economic injury, plaintiff Binder purchased 1000 shares of Optical Radiation through PaineWebber on April 21, 1994. Six minutes after receiving the order, PaineWebber executed it at a price of $20/share. However, earlier that morning an offer was sent over SelectNet that remained open for the entire day to sell 2000 shares of Optical Radiation in blocks of 1000 at $19 3/4. Had PaineWebber executed the trade at the price offered on SelectNet, the plaintiff would have saved $25.
In two other trades, the court also found inferential and speculative evidence that better prices may have been available. On the same day that plaintiff Binder placed an order to buy 7000 shares of Hydron Technologies at 2 9/16, which PaineWebber executed through a market-maker in the security, an offer restricted to Lehman Bros. to sell up to 3000 shares of Hydron Technologies at 2 was broadcast on SelectNet. Based on this restricted offer, plaintiffs contend the lower price indicated a better price for the stock would have been available from other sources, potentially Instinet. Additionally, plaintiffs assert Merrill Lynch's execution of plaintiff Zakheim's purchase order for 120 shares of U.S. Healthcare at 42 3/4 was, on average, $0.16/share more than the price at which Merrill Lynch executed trades in the stock throughout the day. For this reason, plaintiffs reasoned a better price was more than likely available on Instinet that day.
The District Court found SelectNet would not have provided superior prices in six other transactions. Id. While no information from an alternative source was provided for the remaining three transactions, the court noted it was still possible that superior prices may have been available for them. Id.
It bears noting that class actions alleging antitrust injury often raise similar concerns. In antitrust class actions, injury may be presumed when it is clear the violation results in harm to the entire class. Bogosian v. Gulf Oil Corp., 561 F.2d 434, 454 (3d Cir. 1977); In re Nasdaq Market-Makers Antitrust Litig., 169 F.R.D. 493, 526 (S.D.N.Y. 1996). In Bogosian, a class of service station dealers sued their lessors, major oil companies, for forcing them to purchase gasoline at controlled prices in violation of the antitrust laws. 561 F.2d 434. We recognized that
when an antitrust violation impacts upon a class of persons . . . , there is no reason why proof of the impact cannot be made on a common basis so long as the common proof adequately demonstrates some damage to each individual. Whether or not fact of damage can be proven on a common basis therefore depends upon the circumstances of each case.
Id. at 454. Likewise, in In re Nasdaq Market-Makers, a class was certified to pursue allegations that market makers of NASDAQ traded securities conspired to charge supra-competitive prices on the securities they traded for investors. 169 F.R.D. 493. The court noted that plaintiffs' claim of antitrust injury was "susceptible [to] . . . common classwide proof [because] . . . an illegal price-fixing scheme presumptively damages all purchasers of a price-fixed product in an affected market." Id. at 526. Nevertheless, antitrust cases still require proof of injury to each individual for common questions to predominate in a class action. Windham v. Am. Brands, Inc., 565 F.2d 59, 65-66 (4th Cir. 1977) (en banc); see also Broussard v. Meineke Disc. Muffler Shops, Inc., 155 F.3d 331, 342-43 (4th Cir. 1998) (citing Windham, 565 F.2d at 66); Kline v. Coldwell, Banker & Co., 508 F.2d 226, 233 (9th Cir. 1974).
The cases on which we rely generally discuss loss causation in the context of applying the requirements of a Rule 10b-5 claim, not class certification. Many of these cases arise as class actions, but they typically involve motions to dismiss for failure to state a claim or motions for summary judgment rather than class certification.
Plaintiffs also contend that the claims in this case are similar to those in Prudential, where we certified a settlement class alleging fraud. 148 F.3d 283. The claims here are easily distinguished from the securities fraud claims in Prudential. In Prudential, the federal securities claims involved the sale of vanishing premium life insurance policies which the insurance company fraudulently claimed would become self-funding. 148 F.3d at 300 (citing In re Prudential Ins. Co. of Am. Sales Practices Litig., 962 F. Supp. at 500). By purchasing the policies, plaintiffs risked economic injury because the instruments were structurally incapable of meeting the financial expectations Prudential had promised. Whether or not class members actually suffered economic injury was immaterial to the viability of the class's claims because the insurance company did not contest liability. See Prudential, 148 F.3d at 296-97.
Citing AUSA Life Ins. Co. v. Ernst and Young, 206 F.3d 202 (2d Cir. 2000), plaintiffs contend that defendants' trading practice established loss causation throughout the class because it was foreseeable the practice would cause economic harm to class members. See 206 F.3d at 217-20 (remanding dismissal of securities action for reconsideration of loss causation in terms of foreseeability that defendant's conduct would have caused alleged economic harm). By focusing on whether or not the loss was simply foreseeable, plaintiffs have put an improper gloss on the court's opinion. The Court of Appeals for the Second Circuit has explained that "loss causation . . . examines how directly . . . [the fraudulent conduct] caused the loss, and whether the resulting loss was a foreseeable outcome of the fraudulent [conduct]." Suez Equity Investors, 250 F.3d at 96 (summarizing the Court of Appeals for the Second Circuit's definition of loss causation). Whether or not the loss is foreseeable becomes a factor only if direct causation has been demonstrated. Id.
Our test for loss causation is framed somewhat differently. As noted, a viable Rule 10b-5 securities claim must show a "sufficient causal nexus between the loss and the alleged [nondisclosure]." Semerenko, 223 F.3d at 184; see also In re Phillips Petroleum , 881 F.2d at 1244 (holding fraudulent misrepresentation "must touch upon the reasons for the investment's decline in value") (citing Huddleston, 640 F.2d at 549). In other words, to establish loss causation, a claim must demonstrate that the fraudulent conduct proximately caused or substantially contributed to causing plaintiff's economic loss. See EchoCath, 235 F.3d at 883-84; Semerenko, 223 F.3d at 186. Whether there are differences between these standards for loss causation, it is far from certain in this case that each plaintiff has sustained a loss, unlike the insurance companies in AUSA.
In the end, we need not address here whether their claims establish loss causation because we find that plaintiffs' claims do not warrant a class-wide presumption of economic loss. For those investors who did not receive the best available price and suffered a loss as a result, establishing loss causation would not be an issue. See Hoxworth I, 903 F.2d at 203 n.24 (rejecting defendants' argument that excessive markups or markdowns may not have been the cause of plaintiffs' injuries).
Fed. R. Civ. P. 23(a) provides:
Prerequisites to a Class Action. One or more members of a class may sue or be sued as representative parties on behalf of all only if (1) the class is so numerous that joinder of all members is impracticable, (2) there are questions of law or fact common to the class, (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class, and (4) the representative parties will fairly and adequately protect the interests of the class.
Fed. R. Civ. P. 23(b) provides:
Class Actions Maintainable. An action may be maintained as a class action if the prerequisites of subdivision (a) are satisfied, and in addition:
* * *
(3) the court finds that the questions of law or fact common to the members of the class predominate over any questions affecting only individual members, and that a class action is superior to other available methods for the fair and efficient adjudication of the controversy. The matters pertinent to the findings include: (A) the interest of members of the class in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already commenced by or against members of the class; (C) the desirability or undesirability of concentrating litigation of the claims in the particular forum; (D) the difficulties likely to be encountered in the management of a class action.
We have explained in greater detail that
[t]he drafters designed the procedural requirements of Rule 23, especially the requisites of subsection (a), so that the court can assure, to the greatest extent possible, that the actions are prosecuted on behalf of the actual class members in a way that makes it fair to bind their interests. The rule thus represents a measured response to the issues of how the due process rights of absentee interests can be protected and how absentees' represented status can be reconciled with a litigation system premised on traditional bipolar litigation.
***
The Rule 23(a) class inquiries (numerosity, commonality, typicality, and adequacy of representation) constitute a multipart attempt to safeguard the due process rights of absentees. Thus, the ultimate focus falls on the appropriateness of the class device to assert and vindicate class interests.
G.M. Trucks, 55 F.3d at 785, 796.
One treatise describes this standard as met"[w]hen it is alleged that the same unlawful conduct was directed at or affected both the named plaintiffs and the class sought to be represented." 1 Newberg on Class Actions S 3.13, p. 3-77; see also Prudential, 148 F.3d at 311 (" `Commentators have noted that cases challenging the same unlawful conduct which affects both the named plaintiffs and the putative class usually satisfy the typicality requirement irrespective of the varying fact patterns underlying the individual claims.' ") (quoting Baby Neal, 43 F.3d at 58); In re NASDAQ Market-Makers, 169 F.R.D. at 510-11 ("[The typicality requirement is satisfied] where the claims are based on the same legal theory and where the class members have allegedly been injured by the same course of conduct as that which allegedly injured the proposed representatives.").
Although in a different context, both the Supreme Court and our Court found the potential harm to absentee members of the class particularly significant in denying certification in the Amchem litigation. Amchem, 521 U.S. at 626-28 ("As the Third Circuit pointed out, named parties with diverse medical conditions sought to act on behalf of a single giant class rather than on behalf of discrete subclasses. In significant respects, the interests of those within the single class are not aligned. Most saliently, for the currently injured, the critical goal is generous immediate payments. That goal tugs against the interest of exposure-only plaintiffs in ensuring an ample, inflation-protected fund for the future."). This concern is not implicated in the case before us.
"Turning to the other elements of a Rule 10b-5 claim, reliance and damages, it is evident that common questions of fact do not prevail." Merrill Lynch, 191 F.R.D. at 395.
Defendants assert that plaintiffs did not properly preserve this issue on appeal because it is only mentioned in a footnote in their brief. While plaintiffs do not discuss adequacy of representation, they do contest the underlying factors motivating the District Court's conclusion. Because the issue was present in their opening brief and is implicit in their claims, it was sufficiently raised on appeal.
We have interpreted the "predominance requirement [as] incorporat[ing] the commonality requirement." Georgine, 83 F.3d at 626. The Supreme Court has also noted that "the predominance requirement of Rule 23(b)(3) is similar to the requirement of Rule 23(a)(3) that `claims or defenses' of the named representatives must be `typical of the claims or defenses of the class.' " Amchem, 521 U.S. at 623 n.18. Notwithstanding, the Court has counseled that "the predominance requirement is far more demanding [than commonality]." Id. at 623-24.
On this point, the testimony by plaintiffs' expert was limited to the averment that
based on my work and my familiarity with statistical relationships which can be powerfully applied to relevant market data, it is my opinion a reliable measure of damages can be developed in this case based on the application of well-established statistical techniques. Based upon an analysis of the types of data set forth in Plaintiffs' Damage Submission, I can devise a formula which measures class- wide damages and from which a plan of allocation can be constructed. I will develop the formula using explanatory variables that have been widely-used in published studies analyzing transaction costs and the bid-asked spread. I will test the formula against actual transaction data to make any necessary adjustments. The methodology described herein will, in my opinion, yield a reliable measure of damages suffered as a result of the practices challenged in this lawsuit.
In the alternative, the putative class argues that under Prudential, 148 F.3d 283, the predominance inquiry will not be undermined if each class member has not suffered economic injury. Br. of Appellants at 47-49. But plaintiffs' reliance on Prudential is misplaced. First, the lion's share of the fraud claims in Prudential were largely unrelated to federal securities law. Prudential, 148 F.3d at 314. In describing the claims, the district court observed:
[M]ost of the plaintiffs' claims [did] not even involve a reliance element. Plaintiffs' claims for breach of contract, breach of implied obligation of good faith and fair dealing, negligence, negligent training and supervision, and unjust enrichment do not involve reliance. An individual issue with respect to one element of a small portion of plaintiffs' claims does not outweigh the multitude of issues common to the remaining elements and claims.
In re Prudential Ins. Co. of Am. Sales Practice Litig., 962 F. Supp. at 516. The claims in this appeal fall squarely under Rule 10b-5. Second, plaintiffs note that the class in Prudential contained "both injured and uninjured policyholders" without barring certification. 148 F.3d at 306. Based on this, plaintiffs contend that even if some class members did not suffer economic loss, class certification should not be prohibited in this case either. We disagree. In Prudential, the presence of injured and uninjured class members was evaluated in the context of standing, not class certification. Id. at 306-07. But satisfaction of constitutional standing does not answer the predominance inquiry. Furthermore, there was no dispute over the individual question of economic loss in Prudential because the defendant insurance company did not contest liability and waived all defenses. See Prudential, 148 F.3d at 296-97.
Defendants contend that plaintiffs have not preserved this issue on appeal because they did not address superiority in their initial brief. See Br. of Appellees at 69. The plaintiffs' waiver of a dispositive issue, they argue, provides sufficient grounds for affirming the District Court. See Nagle v. Alspach, 8 F.3d 141, 143-44 (3d Cir. 1993) (holding issues to be appealed must appear in briefing to be preserved). Although the plaintiffs do not address superiority directly in their brief, they raise the issue specifically in their reply brief, and the facts and arguments on superiority are present throughout their brief. We have concluded that " `[a]n issue is waived unless a party raises it in its opening brief.' " Reform Party of Allegheny County v. Allegheny County Dept. of Elections, 174 F.3d 305, 316 n.11 (3d Cir. 1999) (quoting Laborers' Int'l Union of N. Am. v. Foster Wheeler Corp., 26 F.3d 375, 398 (3d Cir. 1994)). We also reasoned that "absent extraordinary circumstances, briefs must contain statements of all issues presented for appeal, together with supporting arguments and citations." Simmons v. City of Philadelphia, 947 F.2d 1042, 1065 (3d Cir. 1991). But we believe the issue of superiority was implicit in the plaintiffs' opening brief and was adequately raised on appeal.
Of course, one of the central concerns behind denying certification in Georgine was the potential harm to absentee members. This is not a factor here.