Loss causation and class certification.
TABLE OF CONTENTS INTRODUCTION I. LOSS CAUSATION UNDER THE FRAUD-ON-THE-MARKET THEORY A. Defining Loss Causation B. The Fraud-on-the-Market Presumption and Loss Causation 1. Materiality and Loss Causation 2. Efficient Markets and Loss Causation 3. Fraud-on-the-Market Rebuttal and Loss Causation II. Loss CAUSATION AT THE CLASS CERTIFICATION STAGE THROUGH THE CLASS DEFINITION A. The Class Period at Class Certification B. Requirements of Rule 23 C. Economic Cost of Deferring the Class-Period Definition D. The Added Benefits and Reduced Costs of Early Loss Causation Consideration CONCLUSION
Rule 10b-5, (1) promulgated under the catchall provision of the Securities Act of 1934, (2) serves as a bulwark against knowing misrepresentations by securities issuers by providing shareholders a private cause of action. (3) Individuals, however, are likely to find it prohibitively costly to initiate suits under this provision. Enter the class action, which multiplies the damages over a mass of shareholders, eliciting substantial settlements from risk-averse corporations and officers. The tenuous connection to the merits (4) and the considerable legal fees characteristic of these suits have engendered the longstanding perception that securities class actions are prone to abuse. (5)
Not surprisingly, class certification--where a court examines the propriety of the class under Rule 23(b)(3) of the Federal Rules of Civil Procedure, the typical vehicle for a 10b-5 claim--is a watershed in securities fraud class actions. (6) Failure at this stage usually means plaintiffs will drop the claim, (7) while success leverages the threat of a potentially devastating judgment to impel defendants to settle. (8) As a result, securities-fraud class actions rarely reach trial. (9) Therefore, the need for certainty at the class certification stage is immense. These stakes are an important counterpoise to the dogma that certification should not turn on the merits of a claim. (10)
To attain class certification under Rule 23(b)(3) a plaintiff must establish six requirements: numerosity of plaintiffs, typicality of the lead plaintiff's claim, commonality of the lead plaintiff's claim, adequacy of lead plaintiff's representation, predominance of common claims, and superiority of a class action as a method of adjudication. (11) Before granting certification, the court must be satisfied by a preponderance of evidence that the plaintiff has met each requirement. (12) The most important class criterion in the context of securities fraud is predominance, which requires that "common issues ... predominate over any individual issues." (13)
Securities-fraud claims under 10b-5 generally satisfy the predominance requirement of 23(b)(3) with one exception: reliance. The majority of the 10b-5 elements (14) focus on the behavior of the defendant, rendering the issues common across all plaintiffs. For instance, the "purchase or sale of the security" and loss causation requirements are common to the class. (15) Reliance, however, requires the plaintiff to prove that she bought or sold a security in response to the misrepresentation. (16) Requiting this proof at the individual level would violate the predominance requirement, (17) presenting a roadblock to class certification.
The fraud-on-the-market presumption, most famously articulated in Basic v. Levinson, is a means of circumventing this individualized requirement for reliance. (18) A plaintiff can invoke the fraud-on-the-market presumption by proving that (i) the defendant made public material misrepresentations regarding the stock, (2) the shares were traded on an efficient market, and (3) the plaintiff traded the shares between the time of the misrepresentation and the time the truth emerged. (19) By purchasing or selling stock, an investor automatically relies on the integrity of the market and therefore indirectly relies on any fraud perpetrated against the market. (20) This transforms an individual inquiry into a common one, satisfying the predominance requirement. (21) The defendant, however, can rebut (hereinafter "the rebuttal") the presumption with "any showing that severs the link between the alleged misrepresentation and either the price received (or paid) by the plaintiff, or his decision to trade at a fair market price." (22)
Courts contemplate the fraud-on-the-market presumption at class certification notwithstanding the proscription on merits adjudication at that stage. The Supreme Court's holding in Eisen v. Carlisle & Jacquelin instructs courts not to "conduct a preliminary inquiry into the merits" (23) at class certification. But because the fraud-on-the-market presumption--whose elements overlap substantially with the merits--makes reliance a common inquiry, plaintiffs must prove it at class certification to establish predominance and satisfy Rule 23. Fortunately, this has coincided with the liberalization of the Eisen rule, starting soon after with the Court's decision in General Telephone Co. of Southwest v. Falcon. (24) The circuits now nearly uniformly allow an examination of the merits at the class certification stage insofar as it is necessary to make findings for the purposes of Rule 23. (25)
Starting with Oscar Private Equity Investments v. Allegiance Telecom, Inc., (26) the Fifth Circuit has capitalized on liberalized merits consideration at class certification and dramatically expanded plaintiffs' burden at this stage by requiring them to show loss causation--the proximate cause element of 10b-5--to establish the fraud-on-the-market presumption. The court justified the decision by casting loss causation as an ineluctable indicator of market efficiency, one of the factual predicates of the Basic fraud-on-the-market presumption. (27) Subsequent commentaries on Oscar have reached a similar conclusion with respect to loss causation and class certification but justified that conclusion on the grounds of the Basic predicate of materiality or its rebuttal. (28) At the same time, Oscar spawned a wave of academic criticism and split the circuits, leaving the Fifth Circuit alone in requiring plaintiffs to show loss causation at the class certification stage. (29)
This Note uses this split as its point of departure and provides a new basis for considering loss causation at the 10b-5 class certification stage. It argues that plaintiffs must establish loss causation at the class certification stage, not to prove reliance, but to properly define the class. Part I evaluates the prevailing justifications--the fraud-on-the-market prerequisites of materiality and efficiency, and the Basic rebuttal--for requiring plaintiffs to show loss causation at the class certification stage. Part I concludes that loss causation is conceptually incongruous with reliance, and while loss causation-type evidence may help adduce the predicates and rebuttal of the fraud-on-the-market theory, the plaintiff does not bear the burden of raising it. Part II then obviates the split created by Oscar by proffering the class definition as an alternate justification for requiring the plaintiff to show loss causation. The identification of a corrective disclosure, which marks the end of the class period, requires proof of loss causation. The statutory requirements and sound policy dictate that courts must use loss causation to accurately define 10b-5 classes at the class certification stage.
I. LOSS CAUSATION UNDER THE FRAUD-ON-THE-MARKET THEORY
Liberalized merits consideration at the certification stage and continuing concerns over the irrelevance of merits in securities class actions have increasingly spurred commentators to call for plaintiffs to show loss causation at the class certification stage. (30) In Oscar Private Equity v. Allegiance Telecom, Inc., (31) the Fifth Circuit became the first U.S. Court of Appeals to reach this conclusion. (32) This Part examines the element of loss causation and explains why the existing justifications for consideration of loss causation at the class certification stage, based on the fraud-on-the-market theory, are inapposite. Section I.A defines loss causation and differentiates it from the related elements of reliance and economic loss. Section I.B demonstrates that neither the elements nor the rebuttal of the fraud-on-the-market presumption requires the plaintiff to show loss causation at class certification.
A. Defining Loss Causation
Causation under 10b-5 requires both a misrepresentation and corrective disclosure. (33) Reliance, or but-for causation, centers on the misrepresentation. (34) On the other hand, loss causation--or proximate causation--focuses on the relationship between the misrepresentation and the subsequent corrective disclosure. (35) The term corrective disclosure refers to the event or series of events that cured the misrepresentation. A corrective disclosure does not necessarily have to identify the misrepresentation and refute it; instead, it can be a series of partially corrective disclosures or some other type of market dissemination of the correct information. (36) To show loss causation, the plaintiff must prove first that the disclosures she alleges were in fact corrective. (37) The plaintiff can then show that she suffered a loss as a result of the misrepresentation by selling after the corrective disclosure. On the other hand, if a purchaser sells her shares before the corrective disclosure, she suffers no economic loss, and hence there is no loss causation. (38)
B. The Fraud-on-the-Market Presumption and Loss Causation
The Basic fraud-on-the-market presumption is a substitute for a traditional reliance showing and ameliorates the prohibitive proof difficulties that would otherwise result under Rule 23. It dispenses with the problematic subjective mental-state analysis of reliance and leaves an objective market standard (39) founded on the premises of market efficiency and integrity. (40) Unlike a face-to-face transaction, where parties can bargain over price, in a market transaction the market performs the valuation. (41) The efficient market serves as the "unpaid agent of the investor, informing him that given all the information available to it, the value of the stock is worth the market price." (42) The individual implicitly relies on the market, which under efficient capital markets hypothesis in its semistrong form, (43) reflects all publicly available information. (44)
Requiring loss causation to establish the fraud-on-the-market presumption, and by extension any of its elements, conflates loss causation and reliance. Whether or not the market actually responded after the corrective disclosure is not direct evidence of the mental state of the buyer at the time of his purchase--the central inquiry of reliance. (45) The remaining Sections in this Part extend this analysis to the fraud-on-the-market elements of materiality and efficient markets as well as the rebuttal, all of which have been used by courts and commentators to place the loss causation burden on plaintiffs at class certification and conclude that none of these rationales supports imposing the loss causation requirement on plaintiffs at class certification.
1. Materiality and Loss Causation
Some courts and commentators require a plaintiff to show loss causation-type evidence to establish the materiality of the misrepresentation--an element of the fraud-on-the-market presumption. (46) The reasoning states that for a misrepresentation to be material, there must be market movement after either the misrepresentation or the disclosure, or both--possible antecedents of loss causation. (47)
The problem with this is that, unlike loss causation, materiality is an ex ante standard. Materiality asks whether there is "a substantial likelihood that a reasonable shareholder would consider [the information] important." (48) Some courts, including the Third Circuit, have understood this to require stock-price movement following the alleged corrective disclosure. (49) This view misconstrues materiality. Materiality, like reliance under the fraud-on-the-market theory, asks what the reasonable investor would have done. Stated differently, the court must place itself in the shoes of a reasonable investor at the time of the misrepresentation, (50) rather than engaging in an indirect empirical analysis of materiality based on hindsight and colored by intervening events. (51)
The overwhelming weight of authority therefore recognizes that market movement following a misrepresentation--while certainly probative and perhaps even sufficient--is not necessary in determining materiality. (52) Market movement can merely help adduce what a reasonable investor would have done, using the market as a proxy. It is by no means perfectly suited for even this task, however, because the market may be moved by immaterial information (53) or extrinsic distortions (54) that would not induce the reasonable person to deal. As a result, the prevailing standard for materiality is a qualitative approach that considers both the type of information (55) and the extent of the misrepresentation. (56) Materiality can be shown independent of market movement, and therefore the plaintiff should not bear the burden of establishing this loss causation-type evidence at class certification.
2. Efficient Markets and Loss Causation
Another justification for requiring plaintiffs to show loss causation at class certification is the efficient market element of the fraud-on-the-market theory. Like materiality, market efficiency in its strictest form may be shown through loss causation because, the reasoning goes, an efficient market would respond to material information. (57) This highly individualized understanding of market efficiency is attractive in light of the heterogeneity of market behavior across different types of information and securities. (58) But like materiality, market movement, and by extension loss causation, are not the sine qua non of efficient markets under Basic.
Basic implicitly takes the view that an efficient market under the fraud-on-the-market theory must be semistrong form efficient. (59) Semistrong form efficiency requires that the market absorb and reflect publicly available information. (60) Unfortunately, Basic does not specify means for making this showing. (61)
In defining efficiency, the approaches can be divided into two major types: particular and general. The particularized approach requires a showing that the market behaved efficiently in this instance--i.e., a strict "move the market" type test. Provided the analysis focuses on both the misrepresentation and disclosure (or market dissemination), this is equivalent to loss causation. (62) Meanwhile, the general approach requires a showing that the market is semistrong efficient as a whole. Within this general approach there is one further level of abstraction. At its broadest, it may be sufficient to show that the disputed security traded on a market--for instance, the New York Stock Exchange--is efficient. At its most narrow, it may require a court to establish that the market is semistrong efficient with respect to this particular type of information (63) or this particular security. (64)
Basic demands only a general understanding of market efficiency. The Court continually referred to an "open and developed" securities market in a general sense; (65) the majority of courts have developed a definition on the basis of this understanding. (66) For instance, the First Circuit has adopted a general approach that examines "the structure of the market and the speed with which all publicly available information is impounded in [the] price" within the market. (67)
This is not to say that market movement is not probative in determining whether a market is generally efficient. (68) An ostensibly efficient market may, in fact, be inefficient, or strong form efficient (i.e., the market is aware of nonpublic information), across different types of information. (69) Notwithstanding the potential for overlap with the materiality element, it could be important to have a narrower scope of efficiency, with respect to the particular type of information and particular security, to establish market efficiency.
Yet this normative conclusion does not mandate a move-the-market test. The market movement would be just one example of the particular security's reaction to the particular information, which is not any more meaningful in determining efficiency than any other instance where the security responded to this type of information. (70) Furthermore, overreliance on market movement can lead to false positives. The market can incorporate information, the sole purpose of the efficiency determination, without a change in price. For example, where there is a simultaneous release of positive and negative information, the market may respond without a price change. Also, an adherence to a bright line market test provides potential defendants a means of evading subsequent liability through the use of countervailing statements--one true and one false, combined to eliminate market movement--to conceal misrepresentations or disclosures. The best course is therefore to permit a plaintiff to raise loss causation-type evidence to show general market efficiency while understanding that it is not absolutely necessary and better evidence may in fact exist.
3. Fraud-on-the-Market Rebuttal and Loss Causation
The fraud-on-the-market rebuttal, which explicitly allows defendants to introduce loss causation-type evidence at the class certification stage, (71) does not require the plaintiff to make an affirmative showing of loss causation to utilize the presumption. Specifically, the rebuttal allows the defendant to rebut the Basic presumption with "[a]ny showing that severs the link between the alleged misrepresentation and either the price received (or paid) by the plaintiff, or his decision to trade at a fair market price." (72) In a case where there is no market movement after the misrepresentation, the rebuttal could be as manifest as merely presenting this fact. (73) Thus, some commentators have argued that this results in a loss causation-type burden for which the plaintiff must present counterevidence. (74)
First, this position assumes, without support, that Basic endorses the bursting-bubble approach to presumptions. There are three possible approaches to presumptions. The first states that a presumption places both the burden of persuasion and production to counter the presumed fact on the opposing party. While this view still finds support among courts and commentators, (75) Rule 301 of the Federal Rules of Evidence states that a presumption shifts only the burden of production and not the burden of per. suasion. (76) The second alternative is the bursting-bubble approach, which eliminates the presumption as soon as there is contrary evidence, requiring some form of counterevidence to reestablish the once-presumed fact. (77) The third is an "intermediate" approach, which allows the once-presumed fact to be found without any counterevidence. (78) There is no indication that the Basic court intended to adopt the bursting-bubble approach. Even assuming this much, an absence of market movement does not necessarily establish the rebuttal by a preponderance of evidence. (79) So while the court may have to ignore the presumed fact in the event of the market-movement rebuttal, the court could still conclude that the balance of evidence supports the fraud-on-the-market presumption.
Second, the counterevidence required to reestablish the presumed fact after a rebuttal based on an absence of market movement is not tantamount to loss causation. A showing of loss causation would be a sufficient response to a successful rebuttal, but it is certainly not necessary. (80) A plaintiff would merely have to present some countervailing information or force that could have been responsible for the nonmovement (rather than make an affirmative case for loss causation), putting the defendant's rebuttal into doubt. (81) Therefore, while loss causation is probative with respect to its predicates of materiality and efficient markets and its rebuttal, Basic does not require plaintiffs to establish loss causation at class certification to benefit from the fraud-on-the-market presumption. There is, however, another basis in Rule 23 for just such a requirement.
II. Loss CAUSATION AT THE CLASS CERTIFICATION STAGE THROUGH THE CLASS DEFINITION
Loss causation must be shown by the plaintiff at class certification to define the class with the accuracy that Rule 23 demands. This Part shows that both existing law and sound policy evince the requirement of a timely and precise class definition, which can be shown only with loss causation. Section II.A asserts that loss causation is an indispensable precursor to an accurate class definition. Sections II.B and II.C demonstrate that modem courts' practice of deferring the issue violates the Rule 23(b)(3) definiteness requirement and contravenes the Private Securities Litigation Reform Act's ("PSLRA") goal of deterring strike suits. Finally, Section II.D shows that the early consideration of loss causation, generally, furthers social welfare.
A. The Class Period at Class Certification
According to Rule 23(c)(1)(B), "[a]n order that certifies a class action must define the class." (82) Rule 10b-5 defines the affected class as all individuals who bought (or sold, depending on the nature of the misrepresentation) the security on a date between the misrepresentation and the corrective disclosure. (83) The start date of the class, or the date of the misrepresentation, is usually easily ascertainable--it is the date of the purported misrepresentation. The class then includes all purchasers of the security up until the day of the corrective disclosure or, in the instances of multiple disclosures, the final corrective disclosure. (84) The class period therefore ends, and liability under the securities laws terminates, when fully "curative information is publicly announced or otherwise effectively disseminated." (85)
While a class definition should generally not use terminology that depends on resolving the merits, (86) the 10b-5 class definition is inextricably laden with fact analysis. The term "curative" presupposes that a court has knowledge of which disclosures, if any, were corrective. Disclosures can range from fully corrective to noncorrective, with a complete gradation in between.
The class definition requires two separate analyses--the timing and effect of the alleged misrepresentation and the corrective disclosure(s)--that together compose loss causation. (87) The effect of the misrepresentation is germane because it marks the start of the class period and it identifies the harm the disclosure must correct. (88) And the corrective disclosure determines the end of the class period. The complexity of the loss causation analysis will depend on the specifics of the case. In many instances, there will be a perfectly symmetrical disclosure--i.e., a disclosure that is a direct refutation of misrepresentation and reveals the truth--with no possible intervening factors. Here, the loss causation showing should be trivial and the class period should be easily defined. (89) But in cases where the disclosures are several or diffuse, the loss causation analysis, which identifies the corrective disclosure or market dissemination, will be more involved. Without loss causation, the court will have no idea which disclosures, if any, are corrective and therefore cannot define the end of the class period. Because loss causation defines the class, and the plaintiff bears the burden of showing that a class satisfies Rule 23, (90) the plaintiff must demonstrate loss causation at class certification.
This type of merits analysis is permissible at the class certification stage. Eisen v. Carlisle & Jacquelin held that a court may not "conduct a preliminary inquiry into the merits" at class certification. (91) This, however, is a bar only to considering the likelihood of success on the merits as a basis for class certification. (92) Class certification requires rigorous analysis and proof, and courts should look beyond the mere pleadings to the factual evidence to make certain the plaintiff has met the requirements of Rule 23, including the class definition. (93) And modern courts uniformly apply an exacting standard in assessing whether plaintiffs have met the Rule 23 requirements. (94)
Nonetheless, the spirit of Eisen presents courts with a paradox. Either they must sacrifice accuracy in the class definition or they must engage loss causation more fully. Most courts have adopted the former approach. (95) When dealing with disclosures that are partial and several, courts often defer the decision by setting the class end date to its latest possible point and making adjustments as necessary when it undertakes the merits. (96) This approach, however, is irreconcilable with sound policy and congressional intent. Instead, courts must zealously undertake the class issue, not only to ensure fairness in securities litigation, but to satisfy the definiteness and lead-plaintiff requirements of Rule 23 and further the animating purpose of the PSLRA to deter strike suits.
B. Requirements of Rule 23
Precision in the class period, which requires loss causation analysis, inheres in the definiteness requirement of Rule 23. (97) The purpose of an adequate class definition is "to determine the scope of the class and the propriety of permitting plaintiffs to represent ... it." (98) A class cannot be "amorphous, vague, or indeterminate." (99) While the determination of the actual number of individuals in a class is not essential at the class certification stage, the definition must allow a court to determine, at any given time, whether an individual is a member of the class. (100)
To illustrate how amorphous a class definition can be under the current popular regime, imagine a fact pattern similar to that of Dura. (101) Company A distributes a false earnings statement. Over a period of months, the officials of A make false statements regarding earnings (X) at [T.sub.0], and future products (Y) at [T.sub.1] (both material). At [T.sub.2], A discloses that statement Y was false and A's stock price decreases by 25 percent. At [T.sub.3], A discloses that statement X was false, leading to another 25 percent stock price decrease. To define the class, the court would have to determine the date of the corrective disclosure. Under these facts, plaintiffs would likely argue that the primary corrective disclosure was [T.sub.3] to enlarge the class, while defendants would favor [T.sub.2]. If [T.sub.2] and [T.sub.3] are several months apart and A is a heavily traded stock (as it would have to be to establish the fraud-on-the-market presumption), or if there were no disclosures at [T.sub.2] and [T.sub.3] but a market correction nonetheless, a very large number of plaintiffs would be uncertain of their membership in the class. By deferring accurate class definitions to beyond the class certification stage and proceeding with vague and amorphous classes, courts are violating this important precept of definiteness.
Beyond the general definiteness requirement, this uncertainty over the end of the class period can raise concerns over the adequacy of the lead plaintiff. For instance, if in the above hypothetical the lead plaintiff acted between [T.sub.1] and [T.sub.3], the court would be unable to determine whether the class representative is adequate or representative because he may turn out not to even be a member of the class. Furthermore, in this time, the lead plaintiff could exacerbate agency costs by acting in an interested manner to advance his interests at the expense of the class--e.g., by diminishing the causal effect of factors that may serve the majority of class members.
C. Economic Cost of Deferring the Class-Period Definition
An overbroad class will be sticky in the parties' perception of the expected judgment, and therefore inflate the expected settlement. This will have two adverse effects: distorting the parties' settlement decision and encouraging negative-value suits. Congress's concern with negative-value strike suits, coerced settlements, and opportunistic plaintiff's attorneys filing lawsuits every time there was a stock price drop motivated the passage of the PSLRA in 1995. (102)
Deferring the class definition amplifies these threats by anchoring the parties' expectations. In a securities class action, the expected judgment is a function of the harm caused by the fraud and the number of plaintiffs. While the court will resolve both these factors when it adjudicates the merits, the class definition actually bounds the number of plaintiffs. Naturally, plaintiffs will plead the greatest number of plaintiffs; and, similarly, courts have consistently shown a preference to adopt the broadest possible class definition at class certification. (103) By granting class certification on these classes, the court places its imprimatur on the class definition, (104) and this definition will anchor the parties' expected judgments. Therefore, both parties' expected judgments at trial will be higher, and more accurate, after class certification. (105) Even if the court were to inform the parties that, by not considering loss causation, the size of the class could very likely change, this effect would persist. (106) Because the decision to settle almost always occurs after class certification, (107) the parties will almost always be under the influence of this bias when facing the settlement decision. (108)
This increased expected judgment will have two negative effects. First, it will result in larger, skewed settlements, which effectuate an inefficient wealth transfer from defendants to plaintiffs and distort the plaintiffs' decision to bring suit. If, instead, the court were to determine the true class size at class certification, by examining loss causation, settlements after class certification would be more efficient.
Second, this increased expected judgment will also increase the success of harmful negative-value suits, or suits where the costs exceed the judgment. Negative-value suits are socially undesirable in instances where they derive from a low probability of success on the merits but a large possible judgment ("bad type"); they are desirable, however, where there is a high probability of success on the merits but a small possible judgment ("good type"). (109) Securities class actions are more frequently of the bad type. (110) Because of the high costs of litigation--which increase the likelihood of settlement (111)--and potentially enormous judgments--which trigger risk aversion among defendants (112)--securities class actions are especially prone to settlement, and this makes them an attractive choice for negative-value suits. (113) An accurate class definition will reduce the expected judgment after class certification, the time when most parties settle. This will lessen the effects of risk aversion on the decisionmaking of defendants, as well as lessen the costs of trial, and will therefore make plaintiffs less likely to settle negative-value suits at this stage. The reduced likelihood of settlement will diminish the likelihood that these suits will be brought, which will produce a net societal benefit. (114)
D. The Added Benefits and Reduced Costs of Early Loss Causation Consideration
This Section contemplates the ancillary effects of considering loss causation to form precise class definitions at the certification stage and shows that the benefits far outweigh the costs.
Moving loss causation to class certification increases the likelihood that the merits will be heard at all, which, in turn, will reduce the prevalence of negative-value strike suits. Congress has a long-stated interest in making the merits matter more in securities litigation. (115) The stronger the case for loss causation, the more likely the plaintiff will be able to extract a pre-class certification settlement. (116) Next, because settlement occurs most frequently after class certification, and attaining class certification requires a stronger claim on the merits, the parties will be settling claims that are more likely to be meritorious. (117) This additional hurdle eliminates more of the bad-type negative-value strike suits because they, by definition, have a low probability of success on the merits, including loss causation. Considering loss causation at class certification thus makes the merits matter sooner and more. This is a socially desirable outcome and furthers the purpose of the PSLRA. (118)
There will, however, be some costs associated with this increased scrutiny prior to trial. (119) It is possible that the rule may deter the good-type negative value suit--i.e., those where the likelihood of success on the merits is high but the expected judgment is low. Meritorious suits will face increased difficulty and cost to reach a stage where they are likely to settle. Three factors mitigate this effect, however. First, as stated earlier, very few securities class actions are actually good-type suits. Because of risk aversion, bad-type suits have higher expected payouts and therefore plaintiffs' lawyers are more like to pursue them. Good-type suits are less attractive to plaintiffs as the defendant is more likely to call the plaintiff's bluff--i.e., his noncredible threat to go to trial--because the smaller potential judgments are less likely to trigger the defendant's risk-averse tendency to settle. Second, because the parties' expected costs at the outset will increase, the likelihood of a pre-class certification settlement will also increase. (120) Third, proving loss causation and attaining class certification--a trivial hurdle in good-type cases because of their strength on the merits--will increase the expected settlement value of these cases. (121) So while good-type negative value cases may be slightly more expensive to bring, they are already very infrequent in the 10b-5 context and may now settle more frequently before certification and at higher amounts after certification, offsetting the costs.
The other potential concern is the prejudice, or error cost, that could result without the procedural safeguards of trial. Trial offers a lengthier period for discovery, closer scrutiny by the trier of fact, and procedural safeguards that are unavailable at class certification. (122) Therefore, the likelihood of an error in the loss causation determination would theoretically decrease.
In fact, loss causation at class certification would have minimal error costs. As an initial matter, considering loss causation at the class certification stage does not bind the trial court in any manner. (123) Therefore, the risk is limited to the cost of an incorrect determination in the time between class certification and trial. There have been changes in the law that should limit this cost. Under Rule 23, courts historically made the class certification decision "as soon as practicable," (124) but there has been a recent push, both in law and practice, toward later certification hearings. (125) In 2003, Congress amended Rule 23 to allow a ruling "at an early practicable time." (126) The Advisory Committee's note to the amendment states that "[a] court that is not satisfied that the requirements of Rule 23 have been met should refuse certification until they have been met." (127) This development allows judges to control both the timing--to allow further discovery--and nature--to allow closer scrutiny--of the class certification hearing, which should minimize the cost of incorrect determinations. In addition, district court judges are familiar and competent with pretrial hearings. For instance, judges in Markman hearings for claim construction in patent law make similar, but binding, merits adjudications early in a case. (128) There is therefore no reason to expect a substantial diminution in the quality of the loss causation determination.
In sum, many commentators have presumed that the general effect of considering loss causation at the class certification stage will be to harm plaintiffs. (129) But taking away the class definition inaccuracy, plaintiffs will have the same expected judgment (and therefore settlement value) before trial and a higher expected judgment after class certification, when most class actions settle. And while perhaps fewer total claims will reach this stage, the meritorious ones should clear the class certification hurdle and attain these greater settlements. Therefore, it is only the meritless strike suit that will lose value--a welfare-maximizing outcome.
Loss causation is at the crux of the class definition, an outcome determinant in securities class actions. Courts and commentators have long endorsed the confluence of loss causation and class certification but, while often bearing the right instincts, have struggled to find a convincing justification. Existing explanations mistakenly focus on the fraud-on-the-market presumption for reliance and its predicates, efficient markets and materiality and its rebuttal--which while nebulous in certain regards, do not require market movement. Instead, plaintiffs must show loss causation at class certification because loss causation is necessary to define the class with the precision demanded by Rule 23 and the PSLRA. To do any less would burden the efficiency and substance of the fraud provisions of the Securities Act.
(1.) Rule 10b-5 makes it unlawful for any person to "make any untrue statement [or omission] of a material fact ... in connection with the purchase or sale of any security." 17 C.F.R. [section] 240.10b-5 (2007).
(2.) Securities Exchange Act of 1934 [section] 10(b), 15 U.S.C. [section] 78j(b) (2006).
(3.) The Court first recognized this implied remedy in Superintendent of Insurance v. Bankers Life & Casualty Co., 404 U.S. 6 (1971), later acknowledging that it had become "a judicial oak which [grew] from little more than a legislative acorn." Blue Chip Stamps v. Manor Drug Stores, 421 U.S. 723, 737 (1975).
(4.) Stephen J. Choi & Robert B. Thompson, Securities Litigation and Its Lawyers: Changes During The First Decade After The PSLRA, 106 COLUM. L. REV. 1489, 1490 (2006).
(5.) See Karen Donovan, Class-Action Cases Rise, Fueled by Subprime Troubles, N.Y. TIMES, Dec. 21, 2007, at C7 ("Class-action filings, excluding subprime cases and those stemming from the backdating of stock options, have increased almost 40 percent from 2006. Average settlements have also jumped, to $33.2 million from $22.7 million.").
(6.) 4 ALAN R. BROMBERG & LEWIS D. LOWENFELS, BROMBERG AND LOWENFELS ON SECUPITIES FRAUD & COMMODITIES FRAUD [section] 7:469, at 7-911 (2d ed. 2008).
(8.) FED. R. CIV. P. 23(f) advisory committee's note to 1993 amendments; BROMBERG & LOWENFELS, supra note 6. Even a small diminution in stock price spread over a mass of shareholders can result in an enormous judgment against a defendant.
(9.) In re Polymedica Corp. Sec. Litig., 432 F.3d l, 17 n.20 (1st Cir. 2005).
(10.) Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 178 (1974) (reasoning that merits adjudication is best addressed under the "traditional rules and procedures applicable to civil trials" and not in the "absence of established safeguards").
(11.) FED. R. CIV. P. 23.
(12.) In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 320 (3d Cir. 2008); Teamsters Local 445 Freight Div. Pension Fund v. Bombardier Inc., 546 F.3d 196, 202-03 (2d Cir. 2008).
(13.) See BROMBERG & LOWENFELS, supra note 6, [section] 7.462, at 7-898.54.
(14.) Courts, following the lead of common law fraud, have distilled 10b-5 into six elements: a material misrepresentation (or omission), scienter, a connection with the purchase or sale of a security, reliance, economic loss, and loss causation. Dura Pharms., Inc. v. Broudo, 544 U.S. 336, 341-42 (2005).
(15.) Loss causation asks whether the diminution in value can be attributed to the misrepresentation and subsequent corrective disclosure. 4 THOMAS LEE HAZEN, TREATISE ON THE LAW OF SECURITIES REGULATION [section] 12.11, at 160-61 (6th ed. 2009). In its simplest sense, this can be answered with respect to all stockholders with the same methodology.
(16.) BROMBERG & LOWENFELS, supra note 6, [section] 7:441, at 7-898.5.
(17.) Basic Inc. v. Levinson, 485 U.S. 224, 242 (1988). Thus, the court must ask "whether the element of reliance raises a common issue, and assuming that it does not whether the common issues still predominate." Kom v. Franchard Corp., 456 F.2d 1206, 1210 (2d Cir. 1972). Before Basic, courts often incorrectly classified reliance as a question of damages to be determined individually following resolution of the merits of the class, or even eliminated it as an element altogether, to avert conflict with the predominance requirement of Rule 23(b)(3). Comment, The Impact of Class Actions on Rule 10b-5, 38 U. CHI. L. REV. 337, 345-46 (1971). The typical response of courts was to have each shareholder sign an affidavit attesting their reliance, which the court would then have the discretion to accept. Frederick C. Dunbar & Dana Heller, Fraud on the Market Meets Behavioral Finance, 31 DEE J. CORP. L. 455, 461 (2006). This would be a costly and potentially perjurious exercise for plaintiffs. Courts went to such lengths because the alternative would have been to dismiss every single class action because of the reliance element. The Impact of Class Actions on Rule 10b-5, supra.
(18.) Fraud-on-the-market theory is "a practical resolution to the problem of balancing the substantive requirement of proof of reliance in securities cases against the procedural requisites of [Rule] 23." Basic, 485 U.S. at 242 (internal quotation marks omitted). Basic memorialized the theorem, which had been around for some time prior. See, e.g., In re LTV Sec. Litig., 88 F.R.D. 134, 143 (N.D. Tex. 1980).
(19.) Basic, 485 U.S. at 248.
(20.) Id. at 247.
(21.) Unger v. Amedisys Inc., 401 F.3d 316, 322 (5th Cir. 2005).
(22.) Basic, 485 U.S. at 248.
(23.) 417 U.S. 156, 177 (1974).
(24.) 457 U.S. 147 (1982).
(25.) E.g., In re Initial Pub. Offerings Sec. Litig., 471 F.3d 24, 41 (2d Cir. 2006) (holding that class certification "require[s] definitive assessment of Rule 23 requirements, notwithstanding their overlap with merits issues"); Gariety v. Grant Thornton, LLP, 368 F.3d 356, 366 (4th Cir. 2004). Nonetheless, courts attempting to vitiate the rigor demanded by Rule 23 continue to invoke Eisen. E.g., In re Hydrogen Peroxide Antitrust Litig., 240 F.R.D. 163, 169 (E.D. Pa. 2007), overruled by In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305 (3d Cir. 2009).
(26.) 487 F.3d 261, 265 (5th Cir. 2007). The Fifth Circuit reaffirmed Oscar in Luskin v. Intervoice-Brite Inc., 261 F. App'x 697 (5th Cir. 2008).
(27.) Oscar, 487 F.3d at 268-70.
(28.) See infra notes 46, 74.
(29.) For examples of courts in other circuits that have rejected Oscar, see In re Nature's Sunshine Product's Inc. Securities Litigation, 251 F.R.D. 656, 665 (D. Utah 2008) ("[T]be Fifth Circuit's decision in Oscar appears to be in conflict with Supreme Court and Tenth Circuit precedent which warn against determining the merits at the class certification stage."); Lapin v. Goldman Sachs & Co., 254 F.R.D. 168, 186 (S.D.N.Y. 2008) ("Oscar should be rejected as a misreading of Basic."); and In re Micron Technologies, Inc., Securities Litigation, 247 F.R.D. 627, 634 (D. Idaho 2007) ("It is unlikely that [Oscar's standard] would be adopted in this Circuit because it misreads Basic."). For examples of commentators, see Recent Case, Securities Litigation--Class Certification--Fifth Circuit Holds That Plaintiffs Must Prove Loss Causation Before Being Certified As A Class.--Oscar Private Equity Investments v. Allegiance Telecom, Inc., 487 F.3d 261 (5th Cir. 2007), 121 HARV. L. REV. 890, 895 (2008); and Tad E. Thompson, Recent Development, Messin' with Texas: How The Fifth Circuit's Decision in Oscar Private Equity Misinterprets The Fraud-on-the-Market Theory, 86 N.C.L. REV. 1086, 1100 (2008).
(30.) See, e.g., Jeffrey L. Oldham, Comment, Taking "Efficient Markets" Out of the Fraud-on-the-Market Doctrine After the Private Securities Litigation Reform Act, 97 Nw. U. L. REV. 995 (2003).
(31.) 487 F.3d 261 (5th Cir. 2007).
(32.) Id. at 272 (Dennis, J., dissenting).
(33.) See Dura Pharms., Inc. v. Broudo, 544 U.S. 336, 342-43 (2005); see also Glaser v. Enzo Biochem, Inc., 464 F.3d 474, 479 (4th Cir. 2006) ("It is only after the fraudulent conduct is disclosed to the investing public, followed by a drop in value of the stock, that the ... investor has suffered a "loss" that is actionable after the Supreme Court's decision in Dura.").
(34.) Transaction causation and reliance diverge in certain contexts--for example, in cases concerning proxy statements, where proof of reliance is not always necessary to show transaction causation, or in the case of omissions. 2 FEDERAL REGULATION OF SECURITIES, ANNOTATED [section] 1489, at 43:222 (1997); see Affiliated Ute Citizens of Utah v. United States, 406 U.S. 128, 153-54 (1972). However, for the purposes of 10b-5, transaction causation has long been equated to reliance. See Dura, 544 U.S. at 341-42. Prior to the PSLRA, some lower courts adopted a much more stringent measure of transaction causation that required more to prove reliance than mere but-for causation, see Herman & MacLean v. Huddleston, 459 U.S. 375, 387 (1983); however, these courts were unwittingly treating causation as a whole.
(35.) See Devonbrook, Inc. v. Lily Lynn, Inc., No. 70 Civ. 4687, 1974 WL 419, at *3-4 (S.D.N.Y. June 26, 1974). Professor Donald Langevoort characterizes the elements as follows: "[F]irst, the showing that the market was distorted by the fraud; second, that the emergence of the truth, corrective disclosure, caused a loss to some or all investors. The former, as just noted, is what Basic focused on as a predicate for the presumption of reliance. The latter, loss causation, is conceptually separate." Donald C. Langevoort, Basic at Twenty: Rethinking Fraud-on-the-Market 29 (Georgetown Law & Econ. Research Paper No. 1026316, 2007), available at http://ssrn.com/ abstractID=1026316.
(36.) Alaska Elec. Pension Fund v. Flowserve Corp., 572 F.3d 221, 230 (5th Cir. 2009); In re Motorola Sec. Litig., 505 F. Supp. 2d 501, 542 (N.D. Ill. 2007).
(37.) Loss causation requires "that the defendant's misrepresentation (or other fraudulent conduct) proximately caused the plaintiff's economic loss." Dura, 544 U.S. at 346. Implicit in the definition of a corrective disclosure is a causal link to the misrepresentation. Therefore, by showing that the disclosure is the corrective disclosure, and that it moved the market, the plaintiff establishes loss causation. See Oscar, 487 F.3d at 265, for a similar conclusion regarding market movement and loss causation.
(38.) Dura, 544 U.S. at 342; Semerenko v. Cendant Corp., 223 F.3d 165, 185 (3d Cir. 2000) (holding that at any time before the corrective disclosure, "the cost of the alleged misrepresentation is still incorporated into the value of the security and may be recovered at any time simply by reselling the security at the inflated price").
(39.) The public misrepresentation defrauds purchasers of stock even if they did not specifically rely on the bad information. Basic Inc. v. Levinson, 485 U.S. 224, 241-42 (1988).
(40.) Id. at 245; accord Bradford Cornell & James C. Rutten, Market Efficiency, Crashes, and Securities Litigation, 81 TUL L. REV. 443, 455-58 (2006) (noting that fully efficient markets are "not logically possible"). There is some dispute as to what Basic meant by continually referring to the "integrity" of the market. The correspondence between Justice Brennan and Justice Blackmun at the time of the opinion suggests that Justice Blackmun thought that reliance on the integrity of the market was necessary to establish the presumption--that is, "if there exists such a person who did not rely on the integrity of the market price to be accurate, that person was not defrauded by the misrepresentation." See STEPHEN J. CHOI & A.C. PRITCHARD, SECURITIES REGULATION: CASES AND ANALYSIS (2d ed. 2008). Justice Brennan's understanding did not view this belief in integrity to be necessary, and took the view that merely buying through the market was sufficient to establish the presumption. See id.
(41.) Basic, 485 U.S. at 244.
(42.) In re LTV Sec. Litig., 88 F.R.D. 134, 143 (N.D. Tex. 1980).
(43.) 4 HAZEN, supra note 15, [section] 12.10; Jonathan R. Macey & Geoffrey P. Miller, Good Finance, Bad Economics: An Analysis of the Fraud-on-the-Market Theory, 42 STAN. L. REV. 1059, 1078-79 (1990).
(44.) Macey & Miller, supra note 43, at 1077. There are three efficiency hypotheses: weak, semistrong, and strong. Weak form efficiency supposes that stock price is independent of past performance because the market's valuation of the security already includes all historical information. By contrast, strong form efficiency supposes that current market prices reflect all information, both public and private; therefore, even insider traders cannot outperform the market. Id. at 1077-78.
(45.) Cornell & Rutten, supra note 40, at 458-60; see infra Part II.
(46.) At least one commentator has suggested the materiality requirement as an alternate means to justify the conclusion of Oscar. See Langevoort, supra note 35, at 39-40. Similarly, the Fifth Circuit opinions in Nathenson v. Zonagen Inc., 267 F.3d 400 (5th Cir. 2001), and Greenberg v. Crossroads Systems, Inc., 364 F.3d 657 (5th Cir. 2004), can be read as attempts by the Fifth Circuit to impose a higher standard of matefiality. Given that the cases came before Dura, however, they could also be trying to impose some sort of loss causation pleading requirement.
(47.) See supra note 37.
(48.) TSC Indus., Inc. v. Northway, Inc., 426 U.S. 438, 449 (1976).
(49.) Oran v. Stafford, 226 F.3d 275, 282 (3d Cir. 2000) ("[I]f a company's disclosure of information has no effect on stock prices. 'it follows that the information disclosed ... was immaterial as a matter of law.'" (quoting In re Burlington Coat Factory Sec. Litig., 114 F.3d 1410, 1425 (3d Cir. 1997))).
(50.) Ganino v. Citizens Utils. Co., 228 F.3d 154, 165 (2d Cir. 2000) ("Materiality is determined in light of the circumstances existing at the time the alleged misstatement occurred.").
(51.) Spielman v. Gen. Host Corp., 402 F. Supp. 190, 194 (S.D.N.Y. 1975) ("The determination of materiality is to be made upon all the facts as of the time of the transaction and not upon a 20-20 hindsight view long after the event." (emphasis added)), aff'd, 538 F.2d 39 (2d Cir. 1976).
(52.) For example, the Second, Fourth, and Ninth circuits, as well as the SEC, espouse this view. In re Salomon Analyst Metromedia Litig., 544 F.3d 474, 482 (2d Cir. 2008) (holding, post-Oscar, that the concept of materiality in Basic does not require market movement and to require it would "misread" Basic); Greenhouse v. MCG Capital Corp., 392 F.3d 650, 660-61 (4th Cir. 2004) ("The majority rule seems to be that [a stock price drop] can be some evidence, but not, standing alone, dispositive evidence."); No. 84 Employer-Teamster Joint Council Pension Trust Fund v. Am. W. Holding Corp., 320 F.3d 920, 934 (9th Cir. 2003); SEC Staff Accounting Bulletin No. 99, 64 Fed. Reg. 45,150, 45,152 (1999) ("Consideration of potential market reaction to disclosure of a misstatement is by itself 'too blunt an instrument to be depended on' in considering whether a fact is material"); see also United States v. Bilzerian, 926 F.2d 1285, 1298 (2d Cir. 1991) ("[W]hether a public company's stock price moves up or down or stays the same after the filing of a Schedule 13D does not establish the materiality of the statements made, though stock movement is a factor the jury may consider relevant").
(53.) Dunbar & Heller, supra note 17, at 509-10 (explaining how immaterial information can affect stock prices).
(54.) No. 84 Employer-Teamster Joint Council Pension Trust Fund, 320 F.3d at 934 ("Because of the distortions [in 'efficient markets'], adoption of a bright-line rule assuming that the stock price will instantly react would fall to address the realities of the market.").
(55.) See 4 HAZEN, supra note 15, [section] 12.9, for a list of different categories of information that courts have deemed material.
(56.) SEC Staff Accounting Bulletin No. 99, 64 Fed. Reg. 45,150, 45,152. One commentator has argued that the standard should instead focus on the length of the misrepresentation, rather than the degree, because persistent misstatements are more likely to affect the market. James J. Park, Assessing the Materiality of Financial Misstatements (Brooklyn Law Sch., Legal Studies Paper No. 109), available at http://papers.ssm.com/sol3/papers.cfm?abstract_id=1158566. In any case, this is a general standard of materiality.
(57.) See Oscar Private Equity Invs. v. Allegiance Telecom, Inc., 487 F.3d 261, 269 (5th Cir. 2007).
(58.) See id. (citing Macey & Miller, supra note 43, at 1083). For a commentator's take on the same point, see, for example, Carol R. Goforth, The Efficient Capital Market Hypothesis--An Inadequate Justification for the Fraud-on-the-Market Presumption, 27 WAKE FOREST L. REV. 895 (1992).
(59.) Macey & Miller, supra note 43, at 1078. The Court did not apply either strong or weak-form efficiency. Strong form efficiency would preclude a misrepresentation claim because the market would know the truth regardless of the misrepresentations of the defendant. At the same time, Basic explicitly went beyond weak form efficiency, which focuses only on past stock prices, by noting that the market price of shares reflects all publicly available information--the definition of semistrong form efficiency. Id.; see Basic v. Levinson, 485 U.S. 224, 246 (1988).
(60.) Macey & Miller, supra note 43, at 1078. See generally RICHAD A. BREALEY ET AL., PRINCIPLES OF CORPORATE FINANCE 331-354 (8th ed. 2006) (describing the efficient markets hypothesis and its most common anomalies).
(61.) Gariety v. Grant Thomton, LLP, 368 F.3d 356, 368 (4th Cir. 2004) ("[Basic] offers little guidance for determining whether a market is efficient"). Courts have had a difficult time providing a definition for an efficient market. Brad M. Barber et al., The Fraud-on-the-Market Theory and the Indicators of Common Stocks' Efficiency, 19 J. CORP. L. 285, 310 (1994) ("[D]etermination of efficiency has to date been based on intuition, rules of thumb, or both").
(62.) See supra Section I.A.
(63.) The Oscar court was particularly concerned with this type of efficiency. Oscar Private Equity Invs. v. Allegiance Telecom, Inc., 487 F.3d 261, 269-70 (5th Cir. 2007).
(64.) The Cammer test, discussed infra note 66, predominantly focuses on security-specific factors. Cammer v. Bloom, 711 F. Supp. 1264, 1286-87 (D.N.J. 1989).
(65.) Basic Inc. v. Levinson, 485 U.S. 224, 241 (1988).
(66.) The most common test, first announced in Cammer, 711 F. Supp. at 1286-87, is a general approach that looks at a number of factors characteristic of efficiency: (1) a large weekly trading volume; (2) the existence of a significant number of reports by securities analysts; (3) the existence of market makers and arbitrageurs in the security; (4) the eligibility of the company to file an S-3 Registration Statement; and (5) a history of immediate movement of the stock price caused by unexpected corporate events or financial releases.
(67.) In re Polymedica Corp. Sec. Litig., 432 F.3d 1, 14 (1st Cir. 2005).
(68.) The case for market movement is stronger for materiality--but similarly unavailing--which by its definition focuses on the specific type of information. See infra Section II.C.
(69.) Oscar Private Equity Invs. v. Allegiance Telecom, Inc., 487 F.3d 261, 269-70 (5th Cir. 2007).
(70.) Like materiality, the specific instance may in fact be sufficient but not necessary to generally show efficiency because it would be the most probative example for factor five of the Cammer test. See supra note 66.
(71.) Basic Inc. v. Levinson, 485 U.S. 224, 248-49 (1988).
(72.) Id. at 248.
(73.) Oscar, 487 F.3d at 265 ("In Nathenson [v. Zonagen Inc., 267 F.3d 400 (5th Cir. 2001)], the link was severed by publicly available information that the misrepresentation didn't move the stock price."). It is worth noting at this point that market movement after the disclosure does not rebut the presumption. This evidence bears on loss causation and only incidentally on reliance, provided the defendant can show that the disclosure really was at the time when the market disseminated the information and no other changed circumstances accounted for the absence of a drop. Interestingly, this would effectively require the defendant to show all the causal elements of loss causation, but without a loss, to mount a rebuttal. See Greenberg v. Crossroads Sys., Inc., 364 F.3d 657, 666 (5th Cir. 2004) ("[P]laintiffs must demonstrate: (1) that the negative 'truthful' information causing the decrease in price is related to an allegedly false, non-confirmatory positive statement made earlier and (2) that it is more probable than not that it was this negative statement, and not other unrelated negative statements, that caused a significant amount of the decline."). In other words, loss causation provides the necessary link between the disclosure and the misrepresentation, and this, by extension, allows reliance--or but-for causation, which is focused on the misrepresentation--to be refuted through the disclosure.
(74.) E.g., Jaime A. Levitt & Michael Gerard, Loss Causation at Class Certification: Illusory Circuit Split, N.Y.L.J., Nov. 5, 2008, at para. 20.
(75.) Id. But see Oscar, 487 F.3d at 275 (Dennis, J., dissenting). Judge Dennis takes the position that Basic also shifts the burden of persuasion. Id. Although neither the language of Basic nor the Rules of Evidence explicidy supports this approach to presumptions, it does find support in the scholarly literature. Applying the approach would make the Basic rebuttal even more incongruous with a loss causation burden on plaintiffs.
(76.) FED. R. EVID. 301.
(77.) Oscar, 487 F.3d at 274 (Dennis, J., dissenting).
(78.) 1 CHRISTOPHER B. MUELLER & LAIRD C. KIRKPATRICK, FEDERAL EVIDENCE [section] 71, at 1336-39 (2d ed. 1994).
(79.) See supra Sections I.B.1 & I.B.2.
(80.) As explained in note 73, supra, an absence of a price movement after a disclosure is not in itself a rebuttal.
(81.) Alternatively, the plaintiff could abandon the fraud-on-the-market presumption and attempt to prove actual reliance for some or all class members. Zlotnick v. TIE Commc'ns, 836 F.2d 818, 824 (3d Cir. 1988).
(82.) FED. R. CIV. P. 23(c)(1)(B) (emphasis added).
(83.) Cf. Dura Pharms., Inc. v. Broudo, 544 U.S. 336, 342 (2005) ("[I]f, say, the purchaser sells the shares quickly before the relevant truth begins to leak out, the misrepresentation will not have led to any loss").
(84.) See Alien Ferrell & Atanu Saha, The Loss Causation Requirement for Rule 10b-5 Causes of Action: The Implications of Dura Pharmaceuticals, Inc. v. Broudo, 63 BUS. LAW. 163, 175-76 (2007).
(85.) In re Sun Microsystems, Inc. Sec. Litig., No. C-89-20351-RPA, 1990 WL 169140, at *8 (N.D. Cal. Aug. 20, 1990) (quoting In re Sunrise Sec. Litig., No. 655, 1987 WL 19343, at *2 (E.D. Pa. July 7, 1987)).
(86.) Colindres v. QuietFlex Mfg., 235 F.R.D. 347, 368 (S.D. Tex. 2006).
(87.) See Dura, 544 U.S. at 342. The definition of corrective disclosure assumes reflexivity with the initial misrepresentation. It may take one of many forms, ranging from an outright statement to a diffuse market dissemination of the truth. To understand the nature of this definition of disclosure is to understand, if possible, exactly what "proximately caused the loss" See supra note 37 and accompanying text for further discussion on this identity.
(88.) In re SciMed Securities Litigation, No. 3-91-575, 1993 WL 616692 (D. Minn. Sept. 29, 1993), held that "commencement and termination dates for a class period and concomitant class membership determinations can be resolved only by an inquiry to the merits of the suit." Id. at *7. The court then concluded, however, that Eisen prohibited such an exercise. Id.
(89.) But even in these instances, the corrective disclosure may actually turn out to be a market dissemination of the truth in advance of the apparent disclosure.
(90.) See In re Ikon Office Solutions, Inc., 191 F.R.D. 457, 467 (E.D. Pa. 2000).
(91.) 417 U.S. 156, 178 (1974).
(92.) 7AA CHARLES ALAN WRIGHT ET AL., FEDERAL PRACTICE AND PROCEDURE [section] 1785 (3d ed. 2005 & Supp. 2009).
(93.) Gen. Tel. Co. of Sw. v. Falcon, 457 U.S. 147, 160 (1982); In re Initial Pub. Offerings Sec. Litig., 471 F.3d 24, 41 (2d Cir. 2006) ("[Class certification] require[s] definitive assessment of Rule 23 requirements, notwithstanding their overlap with merits issues." (emphasis added)), reh'g denied, 483 F.3d 70 (2d Cir. 2007); Gariety v. Grant Thornton, LLP, 368 F.3d 356, 366 (4th Cir. 2004); Newton v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 259 F.3d 154, 166 (3d Cir. 2001); cf. Prof'l Adjusting Sys. of Am., Inc. v. Gen. Adjustment Bureau, Inc., 64 F.R.D. 35, 38 (S.D.N.Y. 1974) (holding it acceptable, in a pre-2003 amendment context, to conduct "discovery directed at an early stage of the litigation to the very purpose of defining the class or determining that it is too amorphous for judicial handling"). The Fifth Circuit has warned, however, that a class certification hearing should not digress into "mini-trials on the merits of the class or individual claims." Unger v. Amedisys Inc., 401 F.3d 316, 321 (5th Cir. 2005).
(94.) See, e.g., In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 320 (3d Cir. 2008) ("[A] district court errs as a matter of law when it fails to resolve a genuine legal or factual dispute relevant to determining the requirements [of Rule 23]."); Oscar Private Equity Invs. v. Allegiance Telecom Inc., 487 F.3d 261, 267-68 (5th Cir. 2007).
(95.) See, e.g., In re Boesky Sec. Litig., 120 F.R.D. 626, 628 (S.D.N.Y. 1988) (holding a mere determination of the "character or type" of the plaintiffs was necessary); Staffin v. Greenberg, No. 79-3157, 1980 WL 1404, at *2 (E.D. Pa. Feb. 27, 1980) ("It may well be that ... a number of class members may be shown to have suffered no damage. That fact does not and should not, prevent the court from certifying a class provided that the Rule 23 requirements are met." (quoting Tucker v. Arthur Anderson, 67 F.R.D. 468, 482 (S.D.N.Y. 1975))).
(96.) See FED. R. CIV. P. 23(c)(1) ("An order that grants or denies class certification may be altered or amended before final judgment,"); see also, e.g., In re Blech Sec. Litig., No. 94 Civ. 7696(RWS), 2003 WL 1610775, at *16 (S.D.N.Y. Mar. 26, 2003) (changing the class period to the period in which plaintiff actually alleged damages); Gilbert v. First Alert, Inc., 165 F.R.D. 81 (N.D. Ill. 1996).
(97.) See Alliance to End Repression v. Rochford, 565 F.2d 975, 977 (7th Cir. 1977) (listing cases that state a definiteness requirement).
(98.) Vernon J. Rockier & Co., Inc. v. Graphic Enters., Inc., 52 F.R.D. 335, 338 (D. Minn. 1971).
(99.) E.g., Mueller v. CBS, Inc., 200 F.R.D. 227, 233 (W.D. Pa. 2001) (quoting In re Tetracycline Cases, 107 F.R.D. 719, 728 (W.D. Mo. 1985)).
(100.) Bzdawka v. Milwaukee County, 238 F.R.D. 469, 474 (E.D. Wis. 2006). Courts usually make this determination on a case-by-case basis. E.g., Roman v. ESB, Inc., 550 F.2d 1343, 1348 (4th Cir. 1976).
(101.) For further precision, see Dura Pharmaceuticals, Inc. v. Broudo, 544 U.S. 336 (2005).
(102.) See Choi & Thompson, supra note 4, at 1490 ("[T]he PSLRA reflected several core assertions that had been part of the push for the legislation: plaintiffs' attorneys initiated and managed securities class actions; attorneys had incentives to pursue claims that were not optimal for corporations, shareholders, and the larger society; settlement was independent of the merits.").
(103.) See, e.g., In re Victor Techs. Sec. Litig., 102 F.R.D. 53, 58 (N.D. Cal. 1984), aff'd, 792 F.2d 862 (9th Cir. 1986) ("[S]hould the development of facts during the course of the litigation prove otherwise, this court can always fashion appropriate relief at that time."); Weinberger v. Thornton, 114 F.R.D. 599, 606 (S.D. Cal. 1986) ("[T]his court must, at this stage of the proceedings, accept the proposed time limits of the Class Period. They will be adjusted, if necessary, by the trier of fact."). Courts will then amend the class period or introduce subclasses to ensure the requirements of Rule 23 are met.
(104.) See Robert G. Bone & David S. Evans, Class Certification and the Substantive Merits, 51 DUKE L.J. 1251, 1292 (2002) ("[T]he class obtains substantial settlement leverage from a favorable certification decision.").
(105.) See Bryant G. Garth, Studying Civil Litigation Through the Class Action, 62 IND. L.J. 497, 501-02 (1987) ("[T]he failure to win certification reduces the bargaining power of the plaintiff and the will to continue the fight.... Class certification makes such a difference in the settlement value of the case that defendants will not take plaintiffs' claims seriously until certification is achieved.").
(106.) Studies have shown that even where the parties are experts or are informed of the potential framing effect, the effects of anchoring persist. Birte Englich & Thomas Mussweiler, Sentencing Under Uncertainty: Anchoring Effects in the Courtroom, 31 J. APPLIED SOC. PSYCHOL. 1535, 1536 (2001). For background discussion on the effects of anchoring in the courtroom, as well as the phenomenon in general, see id.
(107.) See Bone & Evans, supra note 104 ("[T]he vast majority of certified class actions settle, most soon after certification."); Garth, supra note 105.
(108.) The timing of the class period decision should have no effect on the expected settlement prior to the class certification stage. The probability of proving the affirmative case, which defines the expected judgment, is constant regardless of the timing.
(109.) STEVEN SHAVELL, FOUNDATIONS OF ECONOMIC ANALYSIS OF LAW 422-23 (2004).
(110.) See infra Section II.D for a discussion on why this is the case.
(111.) See H.R. REP. No. 104-369, at 31 (1995) (Conf. Rep.), as reprinted in 1995 U.S.C.C.A.N. 730, 730 (noting that the PSLRA was an attempt to curb "the abuse of the discovery process to impose costs so burdensome that it is often economical for the victimized party to settle"); Richard A. Posner, An Economic Approach to Legal Procedure and Judicial Administration, 2 J. LEGAL STUD. 399, 418 (1973).
(112.) Posner, supra note 111, at 418-19.
(113.) First, the potential judgments are exponentially large compared to the typical civil suits. Risk aversion has commonly been used as a justification for the frequency of settlement in securities class actions and the oft-noted irrelevance of the merits. Unsurprisingly, studies show that the greater the risk--i.e., the judgment--the greater the aversion--i.e., the desire to settle. Second, the costs of litigation are very high. The higher the costs of litigation, the greater the range of possible settlement values. See id. Because the cost of taking a securities suit to trial can be in the millions and span several years, it is more likely that the parties' respective expectations of prevailing (taking into account an optimism bias) will fail to overcome this range. This intrinsic high likelihood of settlement makes it more advantageous for plaintiffs to seek negative-value suits--suits with an expected value of less than zero at the outset. See Bone & Evans, supra note 104, at 1294-95, 1298 ("[F]irst, the class action magnifies the stakes through aggregation; second, it increases the defendant's risk-bearing, litigation, and reputation costs.... [C]orporate defendants tend to be risk-neutral, but they are likely to become more risk-averse as the potential class-wide liability represents a larger fraction of corporate assets.").
(114.) At the time of the PSLRA, some commentators expressed concern that the heightened pleading standards, an arguably analogous modification to the one at issue in this Note, would impose a societal cost because the increase in false negatives would outweigh the decrease in false positives. See Lynn A. Stout, Type 1 Error, Type H Error, and the Private Securities Litigation Reform Act, 38 ARIZ. L. REV. 711 (1996).
(115.) For a discussion concluding that Congress has succeeded in this intent to some extent thus far, see Marilyn F. Johnson et al., Do the Merits Matter More? The Impact of the Private Securities Litigation Reform Act, 23 J.L. ECON. & ORG. 627 (2007).
(116.) Increased expected trial costs for the defendant mean a greater likelihood of settlement. When the loss causation case is strong, requiring the plaintiff to prove it at class certification will not reduce his probability of attaining class certification (which makes the defendant less likely to settle) sufficiently to offset the increased cost of class certification.
(117.) Because class certification requires the parties to prove loss causation, those cases that make it past this point will be more likely to prevail at trial. This increases the expected judgment after class certification. It is worth noting that, again, the expected judgment before class certification will not change.
(118.) For a discussion of the motivations behind the PSLRA and how it has changed securities class actions, see Choi & Thompson, supra note 4.
(119.) These fears were noted by many as a byproduct of the PSLRA. See, e.g., Stephen J. Choi, Do the Merits Matter Less After the Private Securities Litigation Reform Act?, 23 J.L. ECON. & ORG. 598 (2007); Stout, supra note 114. There could be a similar effect here. A fundamental difference between a heightened standard at class certification rather than a motion to dismiss--which Professor Choi's paper addresses--is that the former allows more substantive discovery and careful examination and therefore courts are less likely to be influenced by superficial factors such as the "hard evidence" of a concurrent SEC investigation.
(120.) An increase in the expected costs of litigation makes settlement more likely. See Posner, supra note 111, at 417-20.
(121.) See supra note 117.
(122.) See Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 178 (1974). This is true despite the fact that class certification decisions can be appealed under FED. R. CIV. P. 23(f).
(123.) In re Initial Pub. Offerings Sec. Litig., 471 F.3d 24, 41 (2d Cir. 2006) ("[D]etermination as to a Rule 23 requirement is made only for purposes of class certification and is not binding on the trier of facts, even if that trier is the class certification judge.").
(124.) FED. R. CIV. P. 23(c)(1)(A) (1998) (amended 2003).
(125.) See Oscar Private Equity Invs. v. Allegiance Telecom, Inc., 487 F.3d 261, 267 (5th Cir. 2007).
(126.) FED. R. CIV. P. 23(c)(1)(A).
(127.) FED. R. CIV. P. 23(c)(1)(C) advisory committee's note to 2003 Amendment.
(128.) See generally Edward Brunet, Markman Hearings, Summary Judgment, and Judicial Discretion, 9 LEWIS & CLARK L. REV. 93 (2005) (discussing the timing and procedural ramifications of Markman hearings).
(129.) See, e.g., Thompson, supra note 29, at 1100 ("By increasing the standard for materiality, shifting the rebuttable presumption in favor of plaintiffs to a burden they must bear, and moving this determination to the class certification stage, the logical conclusion is that the Fifth Circuit's holding will make it much more difficult for plaintiffs alleging securities fraud to succeed in 10b-5 litigation") (emphasis added); Recent Case, supra note 29, at 896 ("Requiring loss causation at the certification stage reduces the likelihood that such attorneys can extract a quick settlement by alleging loss causation even though the case is unlikely to prevail at trial.").
Steven Serajeddini, J.D. and M.A. in Applied Economics, May 2009. I would particularly like to thank Professor Adam C. Pritchard for many very helpful discussions and suggestions. I would also like to thank Professor James J. Prescott for his structural feedback. Finally, I am grateful to my Note Editors, Colin Watson, Lance Phillips, and Sylvia Vannini, and the Volume 107 and 108 Notes Offices for their valuable comments and work in getting this Note to publication.
|Printer friendly Cite/link Email Feedback|
|Publication:||Michigan Law Review|
|Date:||Nov 1, 2009|
|Previous Article:||A sea of confusion: the Shipowner's Limitation of Liability Act as an independent basis for admiralty jurisdiction.|
|Next Article:||Negligence and insufficient activity: the missing paradigm in torts.|