The predominance requirement under Rule 23(b)(3) has always been the hardest certification hurdle to clear in complex commercial cases. Then came Comcast Corp. v. Behrend, 569 U.S. 27 (2013), and suddenly defense counsel had a new weapon and plaintiffs’ lawyers had a new headache. The circuits have spent the last decade-plus working out what Comcast actually requires, and they have not landed in the same place. If you’re litigating a class certification motion anywhere in the federal system, you need to know which version of the rule you’re playing under.
What the Rule Actually Says
(b) Types of Class Actions. A class action may be maintained if Rule 23(a) is satisfied and if:
(3) the court finds that the questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy. The matters pertinent to these findings include:
(A) the class members’ interests in individually controlling the prosecution or defense of separate actions;
(B) the extent and nature of any litigation concerning the controversy already begun by or against class members;
(C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and
(D) the likely difficulties in managing a class action.
The text says nothing about damages models. That silence is where all the disagreement lives.
What Comcast Actually Held (And What It Didn’t)
Comcast reversed class certification in an antitrust case because the plaintiffs’ damages model measured harm from four theories of liability but only one theory survived to the certification stage. The Court held that a damages model must actually measure damages attributable to the plaintiff’s specific theory of liability. Straightforward enough.
The splintered nature of the opinion is what created the mess. The dissent argued the majority was reaching a merits question under the guise of predominance review. The majority insisted it was not requiring damages to be susceptible to classwide proof in every case, only that the model fit the theory. Courts have been arguing about that distinction ever since.
Where the Circuits Agree
On the basic framework, there’s genuine consensus. Every circuit agrees that Comcast requires some linkage between the damages methodology and the liability theory. You cannot certify a class where individual damages calculations would require so many mini-trials that they swamp the common questions. That much is settled law across the board.
Circuits also agree that Comcast does not bar certification simply because damages require individual calculations. The existence of individualized damages issues does not automatically defeat predominance. This principle predates Comcast and has survived it intact in every circuit.
Where They Split: The Damages Model Question
Here’s where practitioners need to pay close attention, because the circuits diverge in ways that matter at the pleading and expert stage.
The Third Circuit read Comcast narrowly, essentially confining it to its facts. In antitrust cases with multiple liability theories, your model has to fit your surviving theory. But the Third Circuit has been reluctant to extend that requirement aggressively beyond that context. Plaintiffs’ counsel in Delaware, New Jersey, and Pennsylvania have more room to present a damages methodology that doesn’t achieve pixel-perfect precision at the certification stage.
The Seventh Circuit has taken a similarly plaintiff-friendly view. Judge Posner, before his retirement, wrote opinions emphasizing that predominance is about whether common questions predominate at the liability stage, with damages being a secondary consideration. The Seventh Circuit has generally held that if liability can be established classwide, the presence of individualized damages questions does not defeat certification so long as there’s a workable methodology for calculating them, even if the methodology isn’t yet fully developed.
The Ninth Circuit occupies middle ground but leans toward requiring more from plaintiffs at the certification stage. The court has read Comcast to require that plaintiffs present a damages model capable of measuring classwide harm using common proof. Vague assurances that damages can be calculated class-wide are not enough. You need a credible, specific methodology, and the district court needs to actually engage with it.
The Second Circuit has been strict. Courts in the Southern and Eastern Districts of New York, operating under Second Circuit guidance, have required that plaintiffs affirmatively demonstrate that their damages model is capable of producing common proof of damages on a classwide basis. If your expert cannot explain at the certification stage how individual class member damages will be calculated without individual inquiry, you have a real problem. This matters enormously in securities fraud cases and antitrust litigation concentrated in New York.
The Fifth Circuit applies what is arguably the most defendant-friendly reading. Texas, Louisiana, and Mississippi practitioners on the defense side have the strongest Comcast arguments. The Fifth Circuit has required plaintiffs to show at certification that damages are capable of measurement on a classwide basis through common evidence, and courts there have been willing to engage deeply with competing expert testimony at the certification stage in ways that resemble a mini-trial on damages.
The Common Mistake Plaintiffs’ Counsel Make
The single biggest error I see is treating the damages model as something you can develop later. You cannot. If you are filing a class certification motion in any circuit, you need a fully developed damages model from your expert before the motion is briefed, not just a roadmap. Comcast at minimum requires that. What varies by circuit is how rigorous the court will be in scrutinizing that model, but no circuit will let you wave your hands and promise the damages calculation will be worked out post-certification.
The corollary mistake is failing to tie your expert’s methodology explicitly to your liability theory. If your antitrust case survives on a single theory of harm, your damages model needs to measure exactly that harm, and your expert’s report needs to say so explicitly. Generic regression analyses that could apply to any theory are exactly what Comcast targeted.
Practical Takeaways by Circuit
- Second and Fifth Circuits: Budget for a full Daubert fight at certification. Your damages expert will be attacked, and the court will engage with those attacks seriously. Plan for a full evidentiary hearing.
- Seventh Circuit: Lead with liability-side common questions. If you can show that liability is genuinely common, courts there give more latitude on damages methodology.
- Third Circuit: Know your liability theory cold before you finalize your damages model. Comcast’s core requirement, fitting the model to the theory, is non-negotiable everywhere, but Third Circuit courts will at least not layer additional requirements on top of it.
- Ninth Circuit: Budget for a detailed expert report that explains the methodology specifically. “We will calculate damages using a regression analysis” is not enough. Explain what variables, why those variables, and how individual class member damages flow from the classwide model.
For related background on how these expert battles play out in discovery, see Rule 26. The intersection of expert disclosure requirements and the certification timeline is its own trap, and getting your damages expert disclosed on a schedule that allows for full briefing at certification is a logistics problem that bites practitioners constantly.
The circuit split here is real and it is outcome-determinative. Where you file, or where the defendant is located, shapes the certification landscape before you write a single word of your complaint.