Modest decision

In Comcast Corp. v. Behrend, 133 S. Ct. 1426 (2013), a 5-4 majority — over an extraordinary joint dissent by Justices Ginsburg and Breyer — had to work hard to make a modest ruling. The Court held that plaintiffs seeking class treatment under Rule 23(b)(3) sometimes may have to plausibly link their theory of liability (the misconduct that caused damages) to the theory of class-wide damages (the estimate of the damages flowing from the misconduct) in order to obtain class certification.

I say emphatically that the Court did not hold that any plaintiff class seeking certification under Rule 23(b)(3) must prove damages on a class-wide basis. It said only that if a class cannot obtain class certification without establishing class-wide damages, then by golly it must show that it can establish class-wide damages.

I should know; I briefed and argued the case for the plaintiff class.

It gets worse

As the joint dissenters stressed, the majority actually changed the question it answered. The change occurred, without notice, between the time of oral argument, on November 5, 2012, and the announcement of the decision, on March 27, 2013. The issue on which the Court had granted review concerned whether or not the expert evidence that plaintiffs offer in support of class certification must past muster under Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993).

The switch allowed the majority to sidestep the fundamental problem that, in the trial court and court of appeals, Comcast had never raised a Daubert challenge to the plaintiffs’ damages expert. Comcast had thus forfeited any complaint about admissibility under Daubert.

The majority thus could focus on whether it believed the plaintiffs had linked the theory of liability — that Comcast thwarted competition from a cable company that wanted to “overbuild” parts of Comcast’s large footprint in the Philadelphia area — with the expert’s damages model — which showed that the suppression of competition forced cable subscribers to pay supra-competitive rates.

Plaintiffs’ liability expert linked anti-competitive conduct to damages

The plaintiffs had provided the causal link but had done so not through the damages expert but through the liability expert. Because the entire case had until issuance of the Court’s opinion dealt only with admissibility of the damages expert’s opinions under Daubert, the briefing and argument paid scant attention to the linkage. The majority’s consequently incomplete understanding of the record facilitated, if it did not produce, its failure to locate the causal link.

Comcast doesn’t require class-wide proof of damages

The joint dissent went on to emphasize the limits of the majority’s opinion:

• The predominance of common questions requirement in Rule 23(b)(3) “scarcely demands commonality as to all questions”. Comcast, 133 S. Ct. at 1436.
• “[W]hen adjudication of questions of liability common to the class will achieve economies of time and expense, the predominance standard is generally satisfied even if damages are not provable in the aggregate.” Id. at 1437.
• “Recognition that individual damages calculations do not preclude class certification under Rule 23(b)(3) is well nigh universal.” Id.
• “In the mine run of cases, it remains the ‘black letter rule’ that a class may obtain certification under Rule 23(b)(3) when liability questions common to the class predominate over damages questions unique to class members.” Id.

And my favorite: “The Court’s ruling is good for this day and case only.” Id. (emphasis added).

Fifth Circuit decision

But the Fifth Circuit thought just the opposite. It said, in reviewing an order certifying a securities fraud class action against British Petroleum, the following:

• “[I]n order to certify a class, the damages methodology . . . must ‘produce[] commonality of damages.'” Ludlow v. BP, P.L.C., No. 14-20420, slip op. at 11 (5th Cir. Sept. 8, 2015) (emphasis added) (quoting Comcast, 133 S. Ct. at 1434).
• In Comcast, “the Court emphasized the necessity of establishing, before a class could be certified, congruence between theories of damages and liability.” Id. at 16 (emphasis in original).
• Comcast requires “a damages model [to be] ‘susceptible of measurement across the entire class for purposes of Rule 23(b)(3),'” id. at 22 (quoting Comcast), and to “be applied uniformly across the class”, id. at 25.

The panel went further, explaining in a footnote that it believed Comcast compelled it to “ask whether in operation the commonality [of liability] is undone by the damages theory”. Id. at 11 n.36 (emphasis in original). But the court did not, in fact, ask that question or venture to answer it.

The panel instead ruled that the district court properly certified a class of people who bought BP stock after the Macondo spill in 2010 and that the court correctly refused to certify a class of people who bought before the spill, in each instance basing its decision on whether or not it believed the damages model adequately meshed with the theory of liability for securities fraud. The court gave no explanation for how or why the problems with the damages model “undid” the predominance of liability issues. Nor of course does the footnote undo the unqualified statements about Comcast‘s holding in the text.

The court thus misconstrued Comcast as requiring proof of damages on a class-wide basis in all Rule 23(b)(3) cases that seek damages.

The correct outcome

Although the panel got the holding of Comcast wrong and even though the confusion will vex class certification decisions in Louisiana, Mississippi, and Texas for so long as the Ludlow decision stands, the court did at least reach the right conclusion.

The plaintiffs in the post-spill class did present a damages model that linked the inflation of BP share prices to misrepresentations about the severity of the spill (the “flow rate” of hydrocarbons from the well into the Gulf of Mexico). The ones in the pre-spill class, on the other hand, could not distinguish between class members who would have bought BP shares even if BP had not misrepresented the quality of its safety programs and those less adventurous souls who would have spent their money on something else. Plus the theory of liability focused on non-price factors, undercutting the rationale for presuming the reliance element of securities fraud under Basic Inc. v. Levinson, 485 U.S. 224 (1988).

Those flaws relate more to liability than to damages and thus could have doomed certification even if Comcast did not exist.

An ending

The case that produced the Comcast decision reached an important milestone on the day after the Fifth Circuit’s ruling in Ludlow. United States District Judge John Padova, who has presided over the case since its filing in December 2003, granted final approval to a $50 million settlement.