Antitrust Law Daily Wrap Up, ANTITRUST—7th Cir.: Elite universities can appeal certification of antitrust class for purported misapplication of Daubert, (Aug 26, 2026)

By Justin Marcus Smith, J.D.

Five non-settling elite universities argue the district court ignored the Seventh Circuit instruction to consider whether admissible defendant expert evidence, if credited, would mean that individual questions would predominate.

The U.S. Court of Appeals for the Seventh Circuit has granted five elite universities leave to appeal certification of a class of need-based financial aid recipients in a pending class action. With up to $1.7 billion in liability at stake, the five non-settling universities in the appeal are Cornell; Georgetown; Massachusetts Institute of Technology; Notre Dame; and the University of Pennsylvania. The Seventh Circuit also granted the defendant universities leave to file a reply brief addressing the plaintiffs’ opposition to appeal. The reply brief mainly argued that the district court erred in applying earlier Daubert reasoning about admissibility to the class certification decision when it should have, instead, applied more rigorous Arandell analysis. The reply brief was highly critical of the plaintiffs’ expert model. Among other purported issues, the model extrapolated from one school, Yale, and ostensibly would have shown no harm at all at three other schools. The defendants objected to numerous uninjured persons in the class. Aside from those objections, the reply brief suggested the district court should have automatically denied class certification on a prior district court finding that one of the plaintiffs’ counsels engaged in misconduct. The misconduct ostensibly conflicted with and was prejudicial to the proposed class (Corzo v. Cornell University, No. 26-8014 (7th Cir. Aug. 19, 2026)).

Background. Current and former students sued 17 universities, alleging that their common financial-aid methodology, developed through the 568 Presidents Group, constituted a horizontal price-fixing conspiracy under Section 1 of the Sherman Act. After twelve universities settled for nearly $320 million, the federal district court in Chicago certified a class of need-based financial aid recipients in March 2026.

In June 2026, the five remaining defendants petitioned the U.S. Court of Appeals for the Seventh Circuit for permission, pursuant to Fed. R. Civ. P. 23(f), to appeal the March class certification decision. Counsel for the plaintiff-respondents opposed, and the defendant-petitioners moved for leave to file a reply pursuant to Fed. R. App. P. 27. The defendants’ motion said they sought to file their reply as an opportunity to “address the mischaracterizations of law and facts in [plaintiffs’] opposition.” The defendants said the plaintiffs had “doubl[ed] down on the district court’s mistakes by attempting to circumvent the ‘rigorous analysis’ required under Rule 23 with the district court’s prior Daubert analysis in violation of Arandell Corporation v. Xcel Energy Inc., 149 F.4th 883 (7th Cir. 2025).” In other words, the defendant-petitioners contend the plaintiffs attempted to defend the district court’s “deficient” class certification decision by “hiding behind the court’s earlier Daubert and summary judgment rulings.”

On August 24, the Seventh Circuit issued an order granting permission to appeal limited to the sole issue of whether the district court adequately analyzed expert testimony related to the question of common proof of antitrust impact. The order specified the scope of review might include the merits of the district court analysis under Fed. R. Evid. 702. The Seventh Circuit also granted the defendants leave to file a reply to the plaintiffs’ opposition to appeal.

Reply brief. The defendants’ reply brief contended the plaintiffs’ opposition to appeal made the same mistake as the district court. In the defendants’ view, Arandell requires rigorous analysis for class certification, whereas Daubert analysis or summary judgment rulings on an entirely different matter cannot supplant that rigorous analysis. The defendants said interlocutory review was warranted given the immense damages at stake. The class has about 220,000 members posing potential liability of around $1.7 billion. The defendants said it was irrelevant that other defendants already settled.

Expert opinion. The defendants began their reply brief by arguing the district court should not have relied solely on the plaintiff expert’s regression analysis as evidence of common antitrust impact without mentioning the defendants’ rebuttal experts.

The defendants argued the court had a duty to resolve factual disputes at the class certification stage because Daubert admissibility is a mere threshold and may not justify class certification. The district court Daubert decision only considered relevance and reliability. The district court should have made a factual finding whether all schools participated in the alleged conspiracy in exactly the same way over the entire class period, but it did not do so in applying only Daubert.

Second, although the district court was not required to resolve every expert dispute on class certification, that ignored the Seventh Circuit instruction to consider whether admissible defendant evidence, if credited, would mean that individual questions would predominate. In that event, the district court must investigate the “realism” of the expert evidence pursuant to Arandell.

Here, according to the defendants, the plaintiffs did not dispute that the district court’s certification decision did not address how their expert model could prove common antitrust injury against “extensive, unrebutted expert and record evidence” that the defendant schools’ respective conduct varied widely. This was crucial because simply allowing the model to vary impact by school showed no common impact and even no impact at several of the defendant schools. Among other things, the same model estimated “wild and inexplicable year-to-year swings in individual students’ purported overcharges and undercharges.” According to the defendants, the second step in-sample model also “improbably” calculated the precisely identical overcharge of $1,202 for 85% of class members.

Third, the so-called Yale Model the district court relied on for its Daubert opinion directly contradicted the uncontroverted record. According to the defendants, the same method, if the expert had used it to create a model for three other schools, would have shown no harm at all.

Fourth, whether the in-sample prediction is “broadly accepted” is no substitute for judicial evaluation, and that acceptance was fiction in any event. In the defendants’ view, the expert’s two-step approach is widely criticized.

Fifth, the plaintiffs’ assertion that the expert’s model accurately reflected the plaintiffs’ theory that the defendants violated the Sherman Act by agreeing to suppress competition over financial aid awards was so general as to obscure the “blatant contradiction” between the model and the plaintiffs’ working theory of liability. What matters is how the plaintiffs claim the defendants suppressed competition, whereas the plaintiffs “concede[d]” that 568 Group never developed a single formula for the schools to use.

Uninjured persons. The plaintiffs continued that the district court should not have certified the class because it includes a great many who suffered no injury. The plaintiffs again cited the district court’s Daubert opinion which analyzed the separate question of the reliability of the regression analysis. The district court failed to address the prevalence of uninjured class members.

Second, common sense suggested that the expert implausibly premised that a student whose award is unimpacted in one year is at least as likely as any other student to have an impacted award in another year. The defendants contended there was nothing in the record to support that.

Third, the defendants contended the 93% injury rate of the Yale Model was “conjured from thin air.” In their view, the expert explained he did not measure impact by calculating the share of class members ever impacted, but even if he did, there was no basis for extrapolation to the whole class. The defendants urged here that a single school was not representative.

Fourth, plaintiffs who were not direct purchasers, insofar as their parents paid their tuition, lacked antitrust standing.

Alleged counsel conflict. Next, the defendants argued denial of class certification was required because plaintiffs’ counsel had engaged in “ethical misconduct.” As previously reported, the court found that the proposed class counsel’s lack of candor about a litigation-funding arrangement raised serious adequacy concerns. However, the court nevertheless declined to hold that counsel misconduct automatically barred class certification. The misconduct had not prejudiced the class in the litigation to date, and the district court further reasoned that this was not a typical pre-certification case involving an anonymous or purely hypothetical class (see Antitrust, Jun. 2, 2026).

According to the defendants, the court “unambiguously found” that plaintiffs’ counsel made “material” misstatements capable of influencing the court. In the defendants’ view, this not only prejudiced the class but created a direct conflict between counsel and the class. The problem, according to the defendants, was that all then-existing counsel had “submitted false statements to the court to maximize their share of the class’s recovery at the expense of the class.”

The Case is No. 26-8014.

Judge: Kennelly, M.

Attorneys: Norman Armstrong, Jr. (Cooley LLP) for Cornell University. Steven Francis Molo (Molo Lamken LLP) for Andrew Corzo.

Companies: Cornell University

Cases: Antitrust IllinoisNews IndianaNews WisconsinNews GCNNews