Class Action Defense Review : Antitrust, Employment Discrimination, Securities Fraud and Unfair Competition Lawyer & Attorneys: Cozen O'Conner Law Firm: Philadelphia

The Supreme Court’s 5-4 decision last week to overturn the certification of a class of cable television subscribers is extraordinary—not because it continues the Court’s exhortations to lower courts to make merits determinations at the class certification stage, but because of the level of merits-related scrutiny the Court is now requiring.

In Comcast Corp. v. Behrend, the plaintiffs alleged that Comcast violated the federal antitrust laws through a “clustering scheme” that systematically increased Comcast’s share of subscribers in the Philadelphia-area through the targeted acquisition of cable systems within the area. The scheme allegedly eliminated competition among cable service providers in the Philadelphia area and resulted in artificially high prices.

The plaintiffs articulated four different methods in which the clustering alleged increased subscription rates to consumers. The district court, however, held that only one of these four theories was capable of classwide proof, and limited its certification order accordingly: the argument that Comcast’s clustering reduced competition from “overbuilders,” companies that build competing cable networks in areas where an incumbent cable company already operates. The district court further found that damages resulting from decreased overbuilder competition could be calculated on a classwide basis, relying on a regression model that established hypothetical prices that would have existed in a competitive market.

Although the Third Circuit affirmed the district court’s certification order, the Supreme Court found a fatal flaw in the regression model: it did not allow a court to distinguish between damages resulting from overbuilder deterrence and damages resulting from the three other forms of antitrust impact that the district court found incapable of classwide proof. Justice Scalia, writing for the majority, reasoned that because only one theory of damages was accepted for class action treatement, “a model purporting to serve as evidence of damages in this class action must measure only those damages attributable to that theory.” Otherwise, the Court held, the plaintiffs “cannot show Rule 23(b)(3) predominance: Questions of individual damage calculations will inevitably overwhelm questions common to the class.”

The Court’s opinion was largely a rebuke to the Third Circuit for what the Court perceived as a disregard for the Court’s opinions requiring review of merits issues relevant to class certification. The Court characterized the Third Circuit as setting too low a bar for the evaluation of an expert damages model in assessing predominance, criticizing the Third Circuit for failing to require the plaintiffs to tie their damage model to a particular theory of antitrust impact. Justice Scalia wrote that the Third Circuit’s reasoning “flatly contradicts our cases requiring a determination that Rule 23 is satisfied, even when that requires inquiry into the merits of the claim.”

What makes the majority opinion extraordinary, however, is the import of its holding in light of the underlying allegations and theories. Remember that the plaintiffs’ regression model established the prices that allegedly should have prevailed in a competitive market. Such a model, if admissible, is therefore evidence that the difference between its hypothetical prices and the actual prices is due entirely to Comcast’s anticompetitive behavior. Yet the Court held, in essence, that unless the plaintiffs can devise a damage model at the class certification stage that distinguishes one type of anticompetitive behavior from another, they cannot proceed as a class to challenge any of Comcast’s anticompetitive behavior.

The Court’s opinion acknowledged as much: “For all we know, cable subscribers in Gloucester County may have been overcharged because of [Comcast’s] elimination of satellite competition . . . ; while subscribers in Camden County may have paid elevated prices because of [Comcast’s] increased bargaining power vis-à-vis content providers . . . .” Although all such subscribers arguably would have paid higher prices as a result of antitrust violations by Comcast, the Court has placed the burden on plaintiffs at the class certification stage not only to articulate a theory of damages, but to come forward with a damage model that discounts all other potential lawful and unlawful causes of those damages. One is left to wonder whether the result would have been different if the plaintiffs had alleged only that Comcast’s clustering deterred overbuilders, and had not even raised the other three theories.

Justices Ginsburg and Breyer’s dissent, in addition to criticizing the majority for reaching factual conclusions about the significance of an expert report, made another point: In the district court, Comcast argued that the three “rejected” theories of injury had no impact on prices. “If Comcast was right,” argued the dissenters, then the damages identified in the regression model “must have stemmed exclusively” from conduct that deterred overbuilders.

The lesson to class counsel may be that in class actions, pleading or arguing in the alternative can have serious adverse consequences, as the Supreme Court has essentially instructed lower courts to flyspeck class certification evidence for inconsistencies with discarded liability theories. The lesson to defense counsel, conversely, is that any suggestion that an expert report—or any other piece of evidence—should be given anything less than full scrutiny is subject to attack.

• Email This
• Print
• Comments
• Trackbacks

Following up on my recent post on the case, I had the opportunity to speak with Colin O'Keefe of LXBN regarding the Supreme Court's ruling in Standard Fire v. Knowles. In the brief interview, I explain the background of the case and shed some light on its potential impact on class actions. 

• Email This
• Print
• Comments
• Trackbacks

While the Supreme Court’s recent jurisprudence has increasingly favored merits inquiries at the class certification stage, there is one area in which the Court has been reluctant to blur the distinction between certification and merits: the “fraud on the market” presumption in securities class actions. This reluctance is manifest in the Court’s recent opinion in Amgen Inc. v. Connecticut Retirement Plans and Trust Funds.

The fraud-on-the-market presumption was articulated by the Court in 1988 in Basic Inc. v. Levinson. In Basic, the Court held that securities fraud plaintiffs may invoke a rebuttable presumption of reliance based on the theory that the market price reflects all information, including the defendant’s alleged misrepresentations, and that an investor who buys stock at the market price therefore relies on the misrepresentations in doing so. This presumption is a step toward resolving the element of reliance on a classwide basis; by successfully invoking the presumption, class counsel avoids the need for individual adjudications of each class member’s knowledge and state of mind at the time of the transaction.

In 2011, the Court reinforced the validity of the fraud-on-the-market presumption by unanimously rejecting the notion that plaintiffs seeking to invoke the presumption at the class certification stage must prove a causal connection between the misrepresentation and the economic loss suffered by investors. The Court’s ruling effectively drew a line between commonality—a critical issue at the class certification stage—and the merits.

That line was reinforced in Amgen, a 6-3 decision in which the Court held that securities fraud plaintiffs need not prove materiality at the class certification stage in order to invoke the fraud-on-the-market presumption. At issue was whether a court could make a Rule 23(b)(3) finding that common questions predominate absent proof that the alleged misrepresentation had a material effect on stock price. Justice Ginsburg, writing for the Court, responded that “Rule 23(b)(3) requires a showing that questions common to the class predominate, not that those questions will be answered, on the merits, in favor of the class.” (Emphasis in original.)

Because materiality is judged according to an objective standard--i.e., the significance of the misrepresented or omitted fact to a reasonable investor—the Court held that materiality is a question common to all class members. The materiality of the alleged misrepresentations or omissions, or lack thereof, would be the same for all investors. Therefore, the Court reasoned, even if the class fails to prove materiality on the merits, the result would not be the predominace of individual questions, but rather the end of the case.

For the same reason, the Court held that it is unnecessary, at the class certification stage, to consider a defendant’s “rebuttal” evidence that the alleged misrepresentations or omissions did not affect the market price of the securities. The Court reasoned that such evidence goes to the materiality of the misrepresentation or omission, and thus is not relevant at the class certification stage.

In dissent, Justice Thomas accused the majority of “all but eliminating materiality as one of the predicates of the fraud-on-the-market theory.” According to Justice Thomas, while materiality may be a common question, reliance is not a common question—unless the fraud-on-the-market presumption, which requires a showing of materiality, is established. “Plaintiffs cannot be excused of their Rule 23 burden to show at certification that questions of reliance are common,” Justice Thomas wrote, “merely because they might lose later on the merits element of materiality.”

At first blush, Justice Thomas’ quibble appears at times to exalt form over substance. After all, why would it matter if a class is certified based on the fraud-on-the-market theory even if materiality is later found to be lacking? Either way, won’t the lawsuit stand or fall on the common, objective question of whether a misrepresentation or omission was material?

The answer, more practical than legal, lies in a footnote to Justice Thomas’ dissent: “Of course, the Court’s assertion that materiality will be resolved on the merits presumes that certification will not bring in terrorem settlement pressures to bear, foreclosing any materiality inquiry at all.” (Justice Scalia made a similar argument in a separate dissent.) While the timing of a materiality ruling may have little impact on the merits of a securities class action, it can nonetheless have a significant impact on the outcome. Nevertheless, the Court has again limited the scope of inquiry at the certification stage prior to acceptance of the fraud-on-the-market presumption. 

• Email This
• Print
• Comments
• Trackbacks

The Supreme Court held today that a named class plaintiff’s pre-certification stipulation that the class will seek less than $5 million in damages does not defeat federal diversity jurisdiction under CAFA.

In Standard Fire Insurance Co. v. Knowles, a unanimous Court reversed the order of a federal district court in Arkansas remanding a class action to state court based on the named plaintiff’s affidavit, filed with the complaint, stating that the plaintiff would not seek damages “in excess of $5,000,000 in the aggregate.” The district court, applying Eighth Circuit precedent, held that the affidavit was sufficient to establish the required “legal certainty” that the amount in controversy fell below $5 million. The Eighth Circuit denied Standard Fire’s permission to appeal.

Standard Fire then petitioned for rehearing en banc. While the petition was pending, the Eighth Circuit, in a separate case, affirmed an order of remand based on a stipulation similar to the affidavit submitted in Knowles. The Eighth Circuit subsequently denied rehearing en banc in Knowles.

In an opinion written by Justice Breyer, the Court stated that its reason for holding that the stipulation could not defeat jurisdiction “is a simple one: Stipulations must be binding.” Because a plaintiff who files a proposed class action cannot legally bind members of the proposed class until the class is certified, a stipulation filed with the complaint cannot reduce the value of the unnamed class members’ claims.

The Court further characterized the stipulation as “contingent” on future events and therefore impermissible as a juridicitional basis. For example, the Court noted, the state court, on remand, might order the stipulation excised as a condition of class certification, or find the named plaintiff an inadequate class representative based on his proposed recovery cap. Or another class member might intervene with an amended complaint and no stipulation, and the action might proceed with a new representative. Refusing to exercise federal jurisdiction based on the stipulation would require an unwarranted assumption that these scenarios are “farfetched.”

Justice Breyer described the plaintiff’s strongest argument as his contention that any such modification would effectively create a different case than the one filed, and that CAFA permits the district court to consider only the complaint that the plaintiff has filed. He responded, however, that the Court “do[es] not agree that CAFA forbids the federal courts to consider, for purposes of determining the amount in controversy, the very real possibility that a non-binding, amount-limiting separation may not survive the class certification process.” Such an outcome, Justice Breyer wrote, would not “result in the creation of a new case,” and to hold otherwise would “treat a nonbinding stipulation as if it were binding” and “exalt form over substance.”

The Court’s decision is a resounding victory for class action defendants seeking to ensure federal court jurisdiction. Standard Fire had argued that the plaintiff’s counsel in Knowles had filed numerous class actions in Arkansas state court, most of them in the same county, and had obtained orders deferring briefing on all dispositive motions until after discovery is complete and class certification was decided. The state court, according to Standard Fire, would then typically “force” settlement in the cases by permitting “staggeringly expensive discovery,” compliance with which was ordered “prior to briefing on certification in order to force massive nationwide settlements in cases in which the federal courts would never have certified a class.” Today’s opinion removes the damages stipulation as a means of circumventing CAFA jurisdiction.

• Email This
• Print
• Comments
• Trackbacks

Late last year, the Ninth Circuit affirmed an expansive use of cy pres remedies in class action settlements, refusing to second-guess either the parties’ selection of a cy pres recipient or the district court’s determination that the use of a cy pres remedy was appropriate. Last month, however, the Third Circuit set aside a settlement containing a cy pres remedy, holding that the district court based its approval of the settlement on an incomplete record.

In In re Baby Products Antitrust Litigation, the district court approved a $35.5 million settlement of consolidated antitrust class actions brought against several toy manufacturers and retailers. The settlement provided that after the payment of all claims and attorneys’ fees and costs, all remaining funds would go to one or more charitable organizations proposed by the parties and selected by the court for a purpose underlying the interests of the class—i.e., cy pres recipients.

Although the parties apparently contemplated that the cy pres remedy would account for a relatively small portion of the total settlement, the opposite proved to be the case. About $14 million was to go to attorneys’ fees, and of the remainder, it was projected that only about $3 million would be distributed to individual claimants, leaving a whopping $18.5 million for cy pres distribution.

The Third Circuit, holding for the first time that a cy pres remedy can be a permissible component of a class action settlement, nevertheless overturned the district court’s approval. In doing so, the panel emphasized that direct distributions to the class are preferred to cy pres distributions, and held that the degree of direct benefit to the class is a necessary inquiry in evaluating a settlement.

The basis for the court’s reversal, however, was not that the cy pres distribution was necessarily too high, but that the district court lacked a sufficient factual basis for approval--particularly with respect to the allocation of the funds. At the time the district court conducted its fairness hearing, the deadline for claim submissions had not yet passed, and the district court apparently overestimated the likely volume of claims. Following the conclusion of the claims period, the parties did not provide the actual allocation figures to the district court, and the district court failed to fulfill its obligation to “affirmatively seek out such information.”

Interestingly, even the district court’s estimate of the likely payments to class members, made before the claims deadline passed, was only $8.1 million at most—which still would have left more than $13 million for cy pres recipients. Thus, the district court approved the settlement with the expectation that once attorneys’ fees and costs were deducted, a substantial majority of the portion designated for the class would go to cy pres recipients rather than class members. Yet if the district court’s information had proven correct, it is hard to imagine that the Third Circuit have been satisfied with this breakdown.

The key to the opinion may lie in another piece of information the district court lacked. Under the settlement, claimants who provided proof of purchase were eligible to receive a refund of 20 percent of their purchase price. Claimants who did not provide proof of purchase were eligible to receive an award of $5. In the event settlement funds remained after payment of these claims, claimants who submitted the proof of purchase and received the 20 percent refund would obtain pro rata enhancements from the remainder, up to a maximum of three times the original award. The vast majority of claimants, however, fell into the $5 category and were not subject to an enhancement from the remainder.

In approving the settlement, the district court based its acceptance of the $5 cap on compensation for claimants without proof of purchase in part on a determination that the standard of proof required to receive the higher award was “fairly low.” The Third Circuit, however, held that the disproportionately large number of claimants who did not submit proof of purchase “cast[] doubt on this assumption.” In remanding, the court noted that the parties might wish to consider either increasing the $5 payment or lowering the evidentiary bar for receiving the higher award.

Despite taking great pains to couch its opinion in terms of the information considered by the district court, the Third Circuit appears to have to imposed substantive requirements on settlements containing cy pres awards. The court’s obvious concern about the percentage of the settlement going to cy pres recipients, even though the district court fully expected the percentage to be significant based on the information to it available at the time, is one example. The panel’s open suggestion of potential permissible modifications to the settlement agreement, while expressed in terms of voluntary action by the parties and the district court’s role in evaluating the fairness of any modified settlement, all but orders the parties to revise the payout parameters.

Officially, Baby Products holds that there is no numerical requirement regarding the distribution of class action settlement proceeds between class members and cy pres recipients. Unofficially, however, practitioners in the Third Circuit can assume that any settlement including a cy pres award had better ensure that class members receive at least as much as cy pres recipients, and probably substantially more. 

• Email This
• Print
• Comments
• Trackbacks

The 2012-13 Supreme Court term has been a hotbed of class action activity, with the justices set to decide at least half a dozen cases that will directly affect class action litigation. Although none of this term’s decisions is likely to have the impact of the Court’s recent decisions in Wal-Mart Stores v. Dukes or AT&T Mobility v. Concepcion, the sheer number of opinions expected this spring promises significant clarifications of some murky areas. These include:
 

  • Whether a plaintiff may defeat removal under CAFA by stipulating that he or she will not seek more than the $5 million jurisdictional threshold on behalf of the class. In Standard Fire Insurance v. Knowles, the Eighth Circuit denied permission to appeal a district court’s determination that such a stipulation was sufficient, after affirming the validity of a jurisdictional stipulation in a similar case. Is the Eighth Circuit correct, or does such a stipulation improperly bind members of a class that the named plaintiff does not yet represent?

• The extent to which the Supreme Court’s 2011 decision in Wal-Mart requires courts to delve into the merits of a lawsuit when considering class certification motions. In Comcast v. Behrend, the Third Circuit affirmed the grant of class certification in an antitrust action, despite the district court’s decision not to resolve disputes about the relevant market and the existence of classwide impact at the class certification stage. Instead, the Third Circuit held that it was sufficient for the court to determine that the class could establish the relevant market through common proof, that the element of antitrust impact was capable of proof through evidence common to the class, and that the plaintiffs had presented a common methodology to determine damages on a classwide basis. Does Wal-Mart, which included a footnote implying that merits inquiries are appropriate in applying Rule 23(b), require more?

• The extent to which the plaintiff in a securities fraud class action must establish that the alleged misrepresentation was material in order to obtain class certification based on a “fraud on the market” presumption. In Amgen v. Connecticut Retirement Plans and Trust Funds, the Ninth Circuit held that a plaintiff need only plausibly allege materiality at the class certification stage. Did the Ninth Circuit get it right, or does failure to establish materiality at the class certification stage preclude a finding that classwide issues predominate?

• Whether a defendant renders a class action moot by offering full relief to the named plaintiff prior to class certification. In Genesis HealthCare v. Symczyk, the Third Circuit held that in a collective action brought under the Fair Labor Standards Act—which, unlike a Rule 23 class action, requires “class” members to affirmatively consent to participation in the class—an offer of judgment to the named plaintiff, though made before any other plaintiffs had “opted in,” did not moot the lawsuit as to the “class.” Can a defendant thwart a class action pre-certification by settling with the putative class representative, or must the Article III “case or controversy” requirement be read more broadly in class actions? And is the answer different for FLSA actions than for Rule 23 class actions?

• Whether a class arbitration waiver can be held invalid if it prevents plaintiffs from enforcing their federal statutory rights. In In re American Express Merchants’ Litigation, the Second Circuit struck down a class arbitration waiver on the ground that the waiver had the practical effect of precluding potential class members from enforcing their Sherman Act claims. Does the Supreme Court’s 2011 opinion in AT&T Mobility, which held that the Federal Arbitration Act’s general protection of arbitration clauses preempted a state common law unconscionability doctrine, apply more broadly, or is there an exception where the waiver might interfere with enforcement of another federal statute?

• How specific an arbitration clause must be in order to support a finding that the parties consented to class arbitration. In Oxford Health Plans v. Sutter, the Third Circuit affirmed an arbitrator’s finding that the parties had agreed to class arbitration based on a contractual provision mandating simply that “all” disputes be submitted to arbitration. Is such language sufficient for an arbitrator to find consent, or must an arbitrator infer that the parties did not contemplate class proceedings absent an explicit reference to class arbitration?

With all of these cases pending before the Court, as well as several controversial issues percolating in the lower federal courts, the first half of 2013 could be a high-water mark for class action developments. Stay tuned.

• Email This
• Print
• Comments
• Trackbacks

Four years ago, the Fifth Circuit became one of the first courts to consider the application of CAFA’s “mass action” provision to a suit filed by a state attorney general on behalf of a subset of injured citizens. The court’s approach has since been rejected in other circuits. Now, a recent case in which the Fifth Circuit awkwardly applied its own precedent has created a confusing set of circumstances and led at least one Fifth Circuit judge to opine that the other courts probably have it right after all.

CAFA provides that removal of a case to federal court is proper if the suit is either a “class action”—defined with reference to Rule 23 and parallel state law provisions—or a “mass action.” A case is a “mass action” under CAFA if it proposes the joint trial, based on common questions of law or fact, of monetary relief claims of at least 100 persons with an amount in controversy of at least $75,000.
In 2008, the Fifth Circuit held in Louisiana ex rel. Caldwell v. Allstate Insurance Company that where removal of a state attorney general’s lawsuit is sought on “mass action” grounds, a court should “pierce the pleadings” to determine whether the “real party in interest” is the state or the injured citizens. In Caldwell, the Louisiana attorney general brought a parens patriae action on behalf of injured policyholders alleging that insurance companies conspired in violation of the state antitrust laws. The court held that the policyholders, not the state, were the real parties in interest, and therefore the lawsuit was removable as a mass action.

Courts that addressed the same issue after Caldwell took the opposite approach. In 2011, the Seventh Circuit affirmed the remand of a lawsuit brought by Illinois on behalf of injured purchasers of allegedly price-fixed LCD products, holding that whether a state is the real party in interest “is a question to be determined from the essential nature and effect of the proceeding.” In March 2012, the Ninth Circuit sided with the Seventh Circuit, requiring the remand of an action filed by the Nevada attorney general alleging that Bank of America misled Nevada consumers about the terms and operation of its home mortgage modification and foreclosure processes. The Ninth Circuit noted “the devastating effect of the foreclosure crisis on Nevada” and “Nevada’s sovereign interest in protecting its citizens and economy from deceptive mortgage practices” in determining that Nevada was the real party in interest.

The issue returned to the Fifth Circuit in Mississippi v. AU Optronics Corporation, another antitrust and consumer protection case brought against LCD manufacturers on behalf of purchasers. The Fifth Circuit noted that its approach had been rejected by other courts, but nonetheless applied Caldwell and held that the Mississippi attorney general must litigate his “mass action” in federal court.

The awkwardness of the Fifth Circuit’s adherence to Caldwell, however, was illustrated in a concurring opinion by Judge Jennifer Walker Elrod. Judge Elrod supported the denial of remand on the ground that the panel had accurately applied Caldwell, but then proceeded to explain why the Fifth Circuit might be wise to free itself of Caldwell’s constraints:

1. CAFA provides no textual support for piercing the complaint “to determine whether [the] plaintiff is the sole beneficiary of each basis for relief.”

2. The Supreme Court has repeatedly warned that removal statutes should be strictly construed, with ambiguities resolved in favor of remand.
 

3. Principles of comity and federalism dictate that this rule should apply “with particular force” where the plaintiff is a state suing in its own courts.

4. The Caldwell approach negates the “general public” exception to CAFA, which provides that a “mass action” does not include a civil action in which all of the claims “are asserted on behalf of the general public (and not on behalf of individual claimants or members of a purported class) pursuant to a State statute specifically authorizing such action.” Because Caldwell, as well as the majority in AU Optronics, hold the injured citizens on behalf of whom the state has sued are the real parties in interest, these cases also hold that the claims are not “asserted on behalf of the general public” for purposes of the exception. The result is that “a case cannot satisfy the criteria of both the mass action provision and the general public exception,” thereby vitiating the general public exception.

Based on this own critique from a concurring judge, it would seem unlikely that future circuits to consider the matter on first impression would follow the Fifth Circuit’s approach. The question is whether the Fifth Circuit, at its next opportunity, will accept Judge Elrod’s suggestion and reconsider, or whether the Fifth Circuit will stand as the only circuit in which a state’s parens patriae lawsuit is removable under CAFA. 

• Email This
• Print
• Comments
• Trackbacks

The question of Rule 23(b)(3) predominance has become an increasingly thorny one for courts ruling on class certification motions, in no small part due to the Supreme Court’s landmark 2011 opinion in Wal-Mart Stores, Inc. v. Dukes. It was thus a refreshing development when a recent Seventh Circuit opinion undertook to address predominance in strikingly simple terms.

“Predominance is a question of efficiency,” wrote Judge Richard Posner in Butler v. Sears, Roebuck and Co. “Is it more efficient, in terms both of economy of judicial resources and of the expense of litigation to the parties, to decide some issues on a class basis or all issues in separate trials?”

The underlying lawsuit involved a class action involving a pair of defects in Sears washing machines. With respect to one defect, which allegedly caused mold buildup in the machines, the district court denied class certification on the ground that the manufacturer made several design modifications, as a result of which different models were differently defective—and some might not be defective at all. With respect to the other defect, a control unit malfunction that allegedly caused the machines to shut down in mid-cycle, the district court granted class certification.

The Seventh Circuit held that both classes should have been certified. The fundamental question in the mold litigation, the panel held, was “were the machines defective in permitting mold to accumulate and generate noxious odors?” This question, the court said, was “common to the entire mold class, although the answer may vary with differences in design.” To the extent that, as Sears claimed, many class members did not experience a mold problem, the panel held, “that is an argument not for refusing to certify the class but for certifying it and then entering a judgment that will largely exonerate Sears.” For similar reasons, the court upheld class certification in the control unit litigation, holding that “[t]he principal issue is whether the control unit was indeed defective.”

Notably, the Seventh Circuit determined that the issue of whether the product was defective predominated not only over individual questions of damages, but also over differences in state laws that might govern the class’ claims. In particular, the panel noted that in some of the affected states, a defective product can be the subject of a successful suit for breach of warranty even if the defect has not yet caused harm—thus permitting plaintiffs in those states to recover for breach of warranty even if they have not yet encountered an odor or a control unit malfunction. Judge Posner noted only that the district court may want to consider whether the states’ differing laws warrant the creation of subclasses—as might the difference in designs in the various models involved in the mold litigation.

Much ink has been expended over the last couple of years in efforts to weigh common and individual questions for Rule 23(b)(3) purposes. The Butler opinion, short and to the point, is a welcome reminder that identifying the predominant issue in a lawsuit need not always be so difficult.

• Email This
• Print
• Comments
• Trackbacks

Three years ago, a class action alleging a violation of privacy rights by Facebook was settled prior to class certification. The settlement provided for no compensation to class members, no injunction against similar conduct by Facebook in the future, and a substantial expansion of the class beyond the definition proposed in the complaint. The plaintiffs’ attorneys received approximately $3 million in fees.

Yet a federal district court in San Francisco approved the settlement, and a divided Ninth Circuit panel recently affirmed. The case illustrates both the deference awarded a district court’s evaluation of a class action settlement and the limited scrutiny to be given cy pres remedies—i.e., settlements in which the plaintiffs received an “indirect” benefit when the award is distributed to beneficiaries other than those actually wronged.

The decision, Lane v. Facebook, Inc., arose out of Facebook’s short-lived Beacon program, which updated a Facebook member’s personal profile to reflect transactions the member had made with from other vendors who participated in Beacon—such as letting the member’s “friends” know that the member had just rented a particular movie from a Beacon participant. Beacon was initially designed as an “opt out” program, such that members’ purchase information was published without their affirmative consent. In the wake of consumer complaints, Facebook changed Beacon after a few weeks to an “opt in” program, so that information would be shared only with the user’s permission. Some users, however, apparently complained that even after the change, their information was shared without their “opting in.”

The lawsuit was filed only on behalf of users affected during the brief “opt out” period. The settlement agreement, however, expanded the class to include users with complaints arising during the “opt in” period. The settlement provided for a $9.5 million payout, of which approximately $3 million would consist of attorneys’ fees, administrative costs, and incentive payments to the class representatives, and the remainder would be used to set up a new charity organization that would fund and sponsor educational programs relating to online privacy. The charity organization would be run by a three-member board of directors, one of whom would be a Facebook executive. Although Facebook agreed to permanently terminate the Beacon program, the settlement did not expressly prohibit Facebook from reviving the program under another name.

After the district court approved the settlement, four objectors appealed to the Ninth Circuit, contending that the district court’s approval constituted an abuse of discretion. They challenged both the cy pres remedy, particularly the involvement of the Facebook executive on the new charity’s board, and the overall amount of the settlement.

The Ninth Circuit panel affirmed the settlement, holding that Facebook’s involvement on the charity’s board did not constitute a conflict of interest. The court state that while a cy pres remedy must provide the “next best distribution” other than a direct monetary payment to absent class members, “[w]e do not require . . . that settling parties select a cy pres recipient that the court or class members would find ideal.” Rather, the cy pres distribution need only bear “a substantial nexus to the interests of the class members.” Facebook’s insistence on a board seat to “preserv[e] its role in the process of selecting the organizations that would receive a share of that substantial settlement fund,” the court determined, was simply “the unremarkable result of the parties’ give-and-take negotiations.”

As to the settlement amount, the objectors complained that the value of the class’ claims was far greater than the $9.5 million settlement. In particular, the objectors argued that the district court had not properly considered the class’ claims under the Video Privacy Protection Act, which provides for liquidated damages. The panel not only found that the district court had considered the claim, it expressly rejected the notion that a district court is required to make a factual finding as to the potential recovery for each specific cause of action.

Judge Andrew J. Kleinfeld, in dissent, opined that the settlement “perverts the class action into a device for depriving victims of remedies for wrongs, while enriching both the wrongdoers and the lawyers purporting to represent the class.” Judge Kleinfeld noted that class counsel received a substantial fee award, Facebook and its Beacon partners obtained the benefits of an expanded class now barred from bringing claims, and Facebook was not prevented from reviving Beacon under a different name, yet the victims “obtained nothing.”

The majority opinion is firmly rooted in Supreme Court and Ninth Circuit precedent, and in isolation, none of the arguably offending factors—the expansion of the settlement class, Facebook’s representation on the cy pres board, or the lack of an injunction barring the Beacon program—would be unusual. But the convergence of all of these issues creates a decision that may not be legally flawed, but is troubling nonetheless.

• Email This
• Print
• Comments
• Trackbacks

The Class Action Fairness Act of 2005 creates federal diversity jurisdiction over class actions in which the amount in controversy exceeds $5 million. Suppose a defendant removes a putative class action to federal court, making a showing that the $5 million threshold is met, and the plaintiff signs a stipulation that he or she will not seek more than $5 million on behalf of the class. May the case be remanded on the basis of such a stipulation?

That question will be decided by the Supreme Court this term in Standard Fire Insurance Co. v. Knowles. In Knowles, the complaint was filed in an Arkansas state court and accompanied by the plaintiff’s signed affidavit, which stated that the plaintiff “will not at any time during this case . . . seek damages for the class . . . in excess of $5,000,000 in the aggregate (inclusive of costs and attorneys’ fees).” Standard Fire removed to federal court, and the plaintiff moved to remand.

The district court, applying Eighth Circuit precedent, held that Standard Fire had satisfied its burden of establishing, by a preponderance of the evidence, that the $5 million threshold was satisfied, shifting the burden to the plaintiff to prove to a “legal certainty” that the amount in controversy fell below $5 million. The district court further held, however, that the plaintiff’s affidavit was sufficient to establish that “legal certainty.” Standard Fire sought permission to appeal to the Eighth Circuit, which was denied.

Standard Fire then petitioned for rehearing en banc. While the petition was pending, the Eighth Circuit decided Rolwing v. Nestle Holdings, Inc., affirming an order of remand based on a stipulation similar to the affidavit submitted in Knowles. The Eighth Circuit subsequently denied rehearing en banc in Knowles.

The case raises important issues about due process and the proper role of a named plaintiff in a putative class action case. Standard Fire argues that a named plaintiff, prior to class certification, has no authority to “stipulate” away damages for, or otherwise bind, a class that he or she does not yet represent. Permitting such a stipulation, Standard Fire contends, violates the due process rights of absent class members. It also allows one party to effectively waive a jurisdictional requirement.

Standard Fire also claims that the underlying lawsuit illustrates precisely the type of abuses CAFA was intended to prevent. According to Standard Fire, the plaintiffs’ counsel in Knowles has filed numerous class actions in Arkansas state court, most of them in the same county, and has obtained orders deferring briefing on all dispositive motions until after discovery is complete and class certification is decided. The state court, according to Standard Fire, then typically “forces” settlement in the cases by permitting “staggeringly expensive discovery,” compliance with which is ordered “prior to briefing on certification in order to force massive nationwide settlements in cases in which the federal courts would never have certified a class.”

The plaintiff, in turn, contends that the stipulation is just one of a number of lawful methods in which class action plaintiffs structure their complaints to remain under the removal threshold—such as narrowly defining the class or otherwise limiting the scope of the claims asserted. No absent class member is required to “bound” by the stipulation, because any class member who objects to the $5 million limitation is free to opt out of the class.

The Supreme Court’s unusual decision to grant certiorari in a case that was not even heard or decided at the Eighth Circuit level may be a sign that some on the Court are sufficiently troubled by the “stipulation” tactic that they see an immediate need to dispense with it. On its face, the notion that a plaintiff can purport to bind absent members of an uncertified class seems inconsistent with the most fundamental principles of class actions, including the Court’s own recent decisions. If the “stipulation” approach is upheld, the Court may give class action plaintiffs a new path around CAFA.

• Email This
• Print
• Comments
• Trackbacks

A recent New Hampshire case provides yet another example of the difficulty of establishing predominance for class certification purposes in advertising/consumer deception cases. This time, the industry is one that is amply familiar with class action jurisprudence: the tobacco industry.

In Lawrence v. Philip Morris USA, the plaintiff alleged Philip Morris falsely represented that its “Marlboro Lights” cigarettes would deliver less tar and nicotine that other cigarettes. In reality, the plaintiff alleged, the “Lights” merely had filters with ventilation holes that would dilute the tar and nicotine per puff, allowing the cigarettes to “pass” smoking machine tests without delivering any less tar or nicotine to human smokers. The plaintiff did not claim personal injury as a result of the alleged deception; rather, she sought compensation for the difference in value between the cigarettes she purchased and the lower-tar cigarettes she was promised.

The trial court certified a class of “Lights” purchasers in New Hampshire. On appeal to the New Hampshire Supreme Court, however, the class ran into a roadblock that has become increasingly insurmountable in advertising class actions: the individual nature of each consumer’s purchasing decision, and the knowledge that went into that decision.

Here, Philip Morris submitted expert testimony documenting decades of various studies, news reports, and other publicly disseminated material about low-tar cigarettes. Much of this information reported that such cigarettes did not actually contain less tar or nicotine, and that many smokers of low-tar cigarettes “compensate” for the ventilation holes by smoking more cigarettes, inhaling more deeply, or covering the holes with their fingertips. As a result of “the volume of information available to consumers,” the court concluded that “the number of class members exposed to this information was not de minimis.” Accordingly, “determining the information about Lights to which individual class members were exposed and what they believed are individual issues that will predominate over common ones.”

Given the recent consistency with which courts have been rejecting class certification motions in advertising cases, one wonders whether the studies and reports proferred by Philip Morris were even necessary. Arguably, even if little public information existed about the perils of “light” cigarettes, wouldn’t each purchaser’s state of mind remain an individual question? Presumably some consumers placed a greater value than others on the “light” nature of the cigarettes; would a court be required to determine on an individual basis the extent to which the desire to consume less tar and nicotine drove each consumer’s purchases? Moroever, wouldn’t a court be obligated to consider, individually, the extent to which each purchaser actually did “compensate” for the ventilation holes, in order to determine whether that consumer received the benefits promised?

The reality is that few courts are willing to find predominance in advertising cases. Where there is significant information in the product about the public domain, thereby diluting the potential impact of the advertising on the decision to purchase the product, courts point to this information to show the varying amounts of information available to purchasers in support of a denial of class certification. When there is no such information, courts can nonetheless rely on the complicated decision-making process that goes into each consumer’s purchase of a product, and on the differing values to each consumer of the product “promised” and the product received. Given this trend, successful advertising class actions are likely to be few and far between.

• Email This
• Print
• Comments
• Trackbacks

The Pennsylvania Supreme Court last week upheld the decertification of a class of H&R Block customers challenging the tax preparer’s “Rapid Refund” program as deceptive, holding that the existence of a confidential relationship between H&R Block and each class member—a prerequisite to the plaintiff’s claim for breach of fiduciary duty—cannot be determined on a classwide basis.

In Basile v. H & R Block, Inc., the complaint—originally filed 19 years ago—claims that Block, which loans customers the amount of their anticipated tax refund through a short-term bank loan, deliberately confused customers about the nature of the loan. According to the plaintiff, customers did not know that the payment was actually a loan, nor did they understand the high rate of interest involved.

In 2001, after certifying a class of 600,000 customers, the trial court granted summary judgment in Block’s favor on the ground that Block was not the plaintiffs’ agent, no confidential relationship existed between Block and its customers, and therefore Block owed no fiduciary duty. An intermediate court, however, reversed the summary judgment ruling, determining that the class had proferred sufficient evidence for a factfinder to find a confidential relationship.

The common pleas court, based on this ruling, decertified the class, finding that it would be necessary to consider “the unique qualities of each class member” in order for a factfinder to determine whether each class member placed “complete trust in the defendant’s expertise.” The intermediate court overturned this ruling as well, noting that the class would be relying primarily on Block’s internal documents, which showed that Block understood its customers to enter the relationship “in a position of pronounced economic and intellectual weakness” and intentionally provided only minimal information about the refund loans.

The Supreme Court reversed, holding that the common pleas court correctly decertified the class. The court faulted the intermediate court for relying on factual assertions accepted as true in the summary judgment context, given the differing evidentiary standards governing summary judgment and class certification. The court further held that notwithstanding the common reliance on Block’s own documents, “it is not appropriate to presume that Block’s marketing and customer relations strategies had the same impact on each and every putative class member.”

On the one hand, the Basile decision is yet another example of the difficulty of certifying a class where the complaint alleges false or deceptive advertising. Class certification is frequently denied in such cases on the ground that each consumer’s decision-making process requires an individualized inquiry to determine the extent to which the consumer was affected by the alleged deception.

On the other hand, Basile is not a deceptive advertising case, but a breach of fiduciary duty case. It seems counterintuitive that the existence of a fiduciary duty flowing from Block to its customers would be an individualized issue. Did Block have a confidential relationship, or owe a fiduciary duty, only to those individuals who assumed Block was not being deceptive about the character of the loan?

Presumably, the court would have reached a different result if the defendant had been a law firm, a financial advisor, or a similar professional whose relationships with clients fall more clearly into “confidential relationship” territory. A court presumably would not look at the personal circumstances of each of an attorney’s clients to determine whether the attorney and the client had a confidential relationship. For this reason, and because a class action is an unusual avenue for a breach of fiduciary duty claim, Basile may not have a significant impact on class certification jurisprudence. But it is a victory for companies, like Block, whose business falls closer to the line, wherever it may be, that separates the confidential from the non-confidential.
 

• Email This
• Print
• Comments
• Trackbacks

A recent New York federal court opinion illustrates the difficulty in establishing predominance where the primary injury alleged is overpayment for a defective product based on misrepresentation or concealment of the defect.

In Oscar v. BMW of North America, the plaintiff attempted to sue BMW on behalf of a class of New York purchasers of MINI vehicles with Goodyear run-flat-tires whose tires had “gone flat and been replaced.” The plaintiffs alleged that BMW had violated New York state law by making defective disclosures regarding the reliability and replacement cost of the RFTs. Specifically, the lawsuit alleged that BMW had improperly failed to disclose RFTs’ increased risk of failure; the lack of a spare tire; that RFTs cannot be repaired; that RFTs cost more than normal tires to replace; and that replacing RFTs is more inconvenient and time-consuming than replacing normal tires.

In support of his argument that classwide issues predominated, the plaintiff asserted a classwide injury in the form of a portion of the MINI purchase price paid by class members that could be attributed to the RFTs. To illustrate this claim, the plaintiff noted that the suggested retail price of a 2006 MINI S with RFTs was $20,600, while the price of a 2006 MINI Cooper without RFTs was $18,000. The plaintiff argued that the $2,600 price difference could be attributed to the presence of RFTs. The court rejected this theory, in part because the MINI S included many pieces of equipment not on the standard MINI. Thus, it would be impossible to determine on a classwide basis whether the presence of RFTs “caused” consumers to pay a higher sum for the MINI S: “For some consumers, the RFTs may have been an important factor; for others, not at all; for others, somewhere in between; and others, perhaps most others, may never have thought to isolate the relative contribution of each of the differences between these models. . . .”

Moreover, the court found it would be “impossible” to determine, on a classwide basis, the extent to which class members’ decisions to purchase the vehicles would have been different if the RFT disclosures had been made. The court cited both “the inherently independent nature of the purchase decision” and “the conjectural nature of the post hoc injury into how a consumer’s purchase decision might have been affected.” While some consumers might have found the undisclosed defects relevant, “there are simply too many other variables in play to permit that conclusion as to the entire class of consumers . . . .” The court also noted potential other variables adversely impacting a classwide determination of injury, such as the extent to which some class members might have overvalued some other feature of the vehicle (and thus been willing to pay the higher purchase price even had they known of the RFT flaws) and individualized factors affecting each class member’s negotiation of the purchase price with the dealer.

Demonstrating that common questions predominate over individual questions is always difficult where nondisclosure of a product’s features will be alleged, because such lawsuits typically require an examination of each consumer’s knowledge and decision-making process in purchasing the product. Where the product is an automobile, which usually is not even sold to consumers at a uniform price, the predominance hurdle will be nearly insurmountable.

• Email This
• Print
• Comments
• Trackbacks

How much effort must parties to a class action settlement expend in order to ensure that all class members have notice of the settlement and an opportunity to object? Rule 23(c)(2)(B) requires “reasonable effort,” but when does ensuring individual notice become “unreasonable”?

A recent Third Circuit opinion underscores that compliance with Rule 23(c)(2)(B) is not to be taken lightly. In Larson v. AT&T Mobility, LLC, the Third Circuit reversed a district court’s determination that notice by publication would be sufficient.

The plaintiffs filed suit against cellular phone service providers who included a flat-rate early termination fee (ETF) in their cellular contracts, alleging that the fees were penalties that violated the Federal Communications Act and state consumer protection laws. The plaintiffs then entered into a $17.5 million settlement agreement with Sprint.

After preliminarily approving the settlement, the court held an initial fairness hearing and ruled that the parties’ initial notice plan did not comply with Rule 23(c)(2). The court determined that while it would be unreasonable to require Sprint to compile a full list of class members over the 10-year class period because it would require 6-12 months of work and cost at least $1 million, Sprint’s records showed that Sprint could potentially identify subsets of customers who were members of the class.

The parties subsequently filed an amended notice plan, which asserted that it would be unreasonable to search any of Sprint’s billing records for class members who were charged a flat-rate ETF. This assertion was supported by a declaration from a Sprint executive, who testified that it would take 1-2 months to capture information for a three-month segment of the class period, at a cost of $20,000, and 4-5 months to capture information for an additional two-year segment, at a cost of $80,000. The court agreed and approved the amended notice plan, ruling that it would be “unreasonable to require Sprint to engage in further efforts to identify class members” beyond the 285,000 individuals who initially received notice of the Settlement Agreement.

Agreeing with the objectors that the district court abused its discretion, the Third Circuit reversed and remanded the matter for further proceedings. The panel characterized the district court’s acceptance of the amended notice plan as an “about face,” holding that the court “did not provide any support for its new and very different determination that Sprint did not need to conduct a search of its billing records to provide individual notice to a larger group of class members.”The Third Circuit looked to the Supreme Court’s previous opinion on “reasonable effort” in Eisen v. Carlisle & Jacquelin, a class action involving a prospective class consisting of nearly six million individuals who had engaged in odd-lot stock purchases. In the opinion, the Supreme Court stated that “the names and addresses of 2,250,000 class members [were] easily ascertainable, and there [was] nothing to show that individual notice [could not] be mailed to each.” The Court also stated that notice by publication had “long been recognized as a poor substitute for actual notice.”

Given Sprint’s acknowledgement that a search through billing records could potentially identify millions of class members who were charged a flat-rate ETF, the court determined that while this search “cannot be made with push-button ease,” it is clear it would “bring the effort required within the range of reasonableness.” While the cost of providing notice is an appropriate factor to be taken into account, the panel noted that if only two million people were identified through the additional search, the search would cost only 5 cents per class member identified—43 cents per class member when the cost of sending notice to each was included. Moreover, the court noted that “[e]ven if the costs had been higher . . . that would not automatically mean they were unreasonable.”

The court thus remanded the case for the district court to reassess the amended notice plan “on a more complete record and with a fuller explanation.” In particular, the panel noted “the availability of statistical sampling of Sprint’s billing records” as one way to provide “a better grounded estimate of the number of class members” who could be identified through a search of those records.

The Third Circuit’s view as to the outcome it expects from the district court’s reconciliation was abundantly clear, as is the instruction to be drawn from Larson: circumventing actual notice to absent class members will not be condoned absent a showing that the effort required to provide such notice would be extraordinary.

• Email This
• Print
• Comments
• Trackbacks