Written by Chris Walton, JDKroger ran a Rx Savings Club offering generic drugs at $3, $6, or $9 for a $36 annual membership. Most customers enrolled. But when insured customers filled prescriptions, Kroger reported the higher retail price — not the Savings Club price — as its “usual and customary” price to the pharmacy benefit managers. The PBMs used that inflated price to set copays. Insured customers paid more than uninsured cash-paying members for the same drugs.
Judy Kirkbride paid $56.28 for two generic prescriptions. The same drugs would have cost $51.82 through the Savings Club, including the membership fee. The per-transaction overcharge was small. Multiplied across 158 million pharmacy transactions, the proposed class-wide damages were not.
In Kirkbride v. The Kroger Co. (S.D. Ohio Apr. 9, 2025), Judge Algenon Marbley denied Kroger’s motions to exclude both of the plaintiffs’ expert witnesses — a forensic economist who built the class-wide damages model and a PBM practices expert — and granted class certification for four classes spanning Ohio and Texas. The damages model was simple: subtract the Savings Club price from the amount the insured customer paid. The overcharge was the difference. Kroger argued that simplicity was a methodological flaw. The court held it was a strength.
For damages experts and class action counsel, Kirkbride is a case about the virtue of methodological transparency — and the limits of the Daubert challenge when the expert’s approach is straightforward, grounded in the defendant’s own data, and replicable by anyone with a spreadsheet.
The U&C pricing claim has a specific mechanical structure that makes it unusually well-suited to formulaic damages modeling. Here’s how the pharmacy pricing system works:
When a customer fills an insured prescription, the pharmacy submits a claim to the PBM. The claim includes the pharmacy’s “usual and customary” price for the drug. The PBM uses the U&C price as one of several benchmarks to determine the customer’s copay. If the reported U&C price is higher, the copay is higher. If the reported U&C price reflects the actual cash price most customers pay, the copay should be lower.
Kroger’s Rx Savings Club launched in 2018 and offered deeply discounted generics — $3, $6, or $9 — to members who paid a $36 annual fee and skipped insurance. The plaintiffs alleged that most Kroger pharmacy customers enrolled in the Savings Club, making the Savings Club price the actual “usual and customary” price. But Kroger reported the higher non-Savings-Club retail price to PBMs when processing insured claims. The difference between the two prices — the amount by which the insured customer’s copay was inflated — is the overcharge.
That structure produces a damages model that is almost uniquely simple: for each insured transaction, the overcharge equals the amount paid minus what the customer would have paid at the Savings Club price. The inputs are the insured transaction data (from Kroger’s own records) and the Savings Club prices (also from Kroger’s records). Both datasets exist. Both are in the defendant’s possession. The calculation is arithmetic.
Colin Weir, a forensic economist, built the class-wide damages model. His approach was deliberately simple. Using a dataset of over 158 million pharmacy transactions drawn from Kroger’s own records, Weir identified every instance where an insured consumer paid more than what a Savings Club member would have paid for the same drug. For each transaction, he calculated the overcharge: the amount paid minus the Savings Club price. He then aggregated the overcharges across the proposed class to produce the total damages figure.
The simplicity was intentional and strategic. A formulaic damages model that can be applied uniformly across every class member’s transactions is the gold standard for class certification under Rule 23(b)(3). The predominance requirement demands that common questions — including common damages methodology — predominate over individual questions. A model that says “every insured customer’s overcharge equals X minus Y, calculated from the defendant’s own transaction data” satisfies predominance because the calculation is identical for every class member. No individualized inquiry is required.
Weir identified his data sources, explained the formula, and presented the results. He treated the Savings Club price as the appropriate U&C benchmark — an assumption that tracked the plaintiffs’ legal theory. The methodology was transparent enough that anyone with access to the same data could reproduce the calculation.
Kroger moved to exclude Weir under Rule 702, raising three challenges. Each is a common Daubert attack against damages experts in class actions, and each failed:
Attack one: the expert lacks industry-specific expertise. Kroger argued that Weir lacked specialized knowledge of pharmacy pricing, PBM contracts, and Medicare/Medicaid reimbursement. He had never worked on a case involving U&C pricing or pharmacy reimbursement systems. The court rejected this argument, holding that experts need not have firsthand experience in the specific industry at issue. Weir’s expertise was in forensic economics and data analysis — the skills required to build a class-wide damages model from transactional data. The court distinguished between understanding the data (which Weir had) and understanding the industry’s operational details (which the PBM practices expert, not Weir, addressed).
This holding is consistent with Yador v. Mowatt, where the court held that an expert’s selection of comparable companies was a judgment call that went to weight, not admissibility. In both cases, the court declined to use Daubert as a tool to require narrow subject-matter expertise. The expert’s qualification is to perform the analytical task — building the model, running the calculations, presenting the methodology — not to have worked in the defendant’s industry.
Attack two: the expert relied on counsel’s assumptions. Kroger argued that Weir’s damages model was based on legal assumptions provided by plaintiffs’ counsel — specifically, the assumption that the Savings Club price was the correct U&C benchmark. The court rejected this as a basis for exclusion. Damages experts routinely accept legal assumptions from counsel and build their analyses on those assumptions. The assumption that the Savings Club price is the appropriate U&C benchmark is a legal question for the jury to resolve, not a methodological flaw the expert must independently validate. If the jury accepts the legal theory, the damages model applies. If it doesn’t, the model is moot. The expert’s job is to calculate the damages if liability is established, not to independently prove liability.
Attack three: the methodology is too simple. Kroger argued that Weir’s subtraction formula was not a reliable methodology because it was too basic — that a more sophisticated approach was required to account for the complexities of pharmacy reimbursement. The court found this argument unpersuasive. Simplicity is not a methodological deficiency under Rule 702. A formula that takes Kroger’s own transaction data, applies a straightforward calculation, and produces a replicable result is the definition of reliable methodology. The court emphasized that Rule 702 does not require complexity. It requires that the expert’s method rest on sufficient facts and data, employ reliable principles, and reliably apply those principles to the facts. Subtraction meets all three criteria.
The Kirkbride court’s rejection of the “too simple” argument connects to a principle that runs through the expert admissibility pipeline: the expert’s methodology is evaluated for reliability, not sophistication. A simple model grounded in the defendant’s own data is more reliable than a complex model built on unsupported assumptions.
Compare Kirkbride to the cases where expert methodology failed:
In Matter of Weber, the expert used Microsoft Copilot — a more “sophisticated” tool than subtraction — but the tool produced three different answers on three different computers and the expert couldn’t explain how it worked. Sophistication without transparency failed.
In BBK v. Central Coast Agriculture, the plaintiff’s expert offered a disgorgement analysis with broad assumptions about what portion of the defendant’s revenue was attributable to the infringing use. The assumptions sounded rigorous but lacked factual grounding. Complexity without evidence failed.
In Ramcell v. Alltel, the petitioner’s expert built projections from scratch for a company that had no ordinary-course forecasts. The projections were methodologically detailed but litigation-built, and the court weighted them at only 30%. Detail without an ordinary-course anchor was discounted.
Weir’s model in Kirkbride was the opposite: simple methodology, massive evidentiary foundation (158 million transactions from Kroger’s own records), transparent formula, replicable result. The model doesn’t depend on the expert’s subjective judgment. It depends on data the defendant produced. That combination — simplicity plus evidentiary grounding — is what Rule 702 rewards.
In a class action, the damages expert’s survival under Daubert is often dispositive of class certification. Under Rule 23(b)(3), the plaintiffs must demonstrate that common questions predominate over individual questions. A class-wide damages model that can be applied formulaically to every class member’s transactions satisfies the predominance requirement. Without such a model, each class member’s damages would need to be calculated individually, and the case falls apart as a class action.
That’s why Kroger’s Daubert challenge was really a class certification challenge. If Weir had been excluded, the plaintiffs would have lost their damages model, which would have undermined the predominance showing, which would have defeated class certification. The Daubert motion and the class certification opposition were two fronts of the same fight. The court recognized this and addressed both in the same opinion, denying the exclusion motions and granting class certification for four classes.
For class action counsel: your damages expert’s methodology needs to be designed for Daubert survival from the outset, because the Daubert motion is the defendant’s primary tool for defeating class certification on the predominance requirement. An expert whose methodology is formulaic, data-driven, and grounded in the defendant’s own records is the expert who survives both challenges simultaneously.
Kroger also moved to exclude the plaintiffs’ second expert, who testified about pharmacy benefit manager practices and how PBMs use U&C prices to set copays. The court denied this motion as well, holding that the expert’s opinions were based on sufficient factual foundations and that any alleged weaknesses went to weight, not admissibility.
The second expert’s testimony served a different function than Weir’s: Weir calculated the damages, while the PBM expert explained the mechanism that made the damages possible. Together, the two experts told a complete story: the PBM expert established how Kroger’s U&C reporting translated into inflated copays, and Weir quantified the inflation across the class. Neither expert alone was sufficient. Both together survived because each addressed a distinct element of the damages case with a grounded, defensible methodology.
For litigation counsel assembling an expert team in a complex class action: think about the testimony as a narrative with complementary roles. The mechanism expert explains how the defendant’s conduct caused the harm. The damages expert quantifies how much the harm is worth. Each expert should be prepared to survive Daubert independently — because the defendant will challenge both — but the two together should tell a story neither could tell alone.
Kirkbride doesn’t hold that every simple damages model survives Daubert. The model worked here because the factual structure of the case aligned with a formulaic approach: every transaction had an identifiable overcharge, the overcharge was calculable from two data points (amount paid and Savings Club price), and both data points were in the defendant’s records. Not every case has that structure.
A simple model fails when the damages are not uniform across the class (individualized inquiry is required), when the data sources are unreliable or incomplete (the model’s inputs can’t be verified), when the formula depends on an assumption the expert can’t support with evidence (the assumption is contested and the expert has no factual foundation for it), or when the simplicity masks a methodological gap (the model omits a variable that materially affects the calculation — like the BBK expert’s but-for expense methodology that lacked adequate support).
The test is not whether the model is simple or complex. The test is whether it reliably applies reliable principles to sufficient facts. In Kirkbride, 158 million transactions from the defendant’s own records and a transparent subtraction formula met that standard. In other cases, the same level of simplicity without the same evidentiary foundation would not.
Kirkbride v. Kroger confirms what the expert admissibility pipeline has been building toward: Rule 702 rewards transparency, evidentiary grounding, and replicability. It does not require sophistication. A forensic economist who had never worked a pharmacy pricing case built a damages model from the defendant’s own transaction data using a subtraction formula that anyone could verify. Kroger argued he lacked industry expertise, relied on counsel’s assumptions, and used a methodology that was too simple. All three arguments failed because the model was grounded in 158 million data points the defendant itself had produced.
For damages experts: don’t make the model more complex than the case requires. A transparent formula applied to the defendant’s own data is harder to exclude than an opaque model built on the expert’s assumptions. For class action counsel: design the damages model to serve double duty — Daubert survival and predominance under Rule 23(b)(3). A formulaic, data-driven model accomplishes both. And for defendants’ counsel: if the expert’s model is grounded in your client’s own transaction data and the formula is replicable, the Daubert challenge is unlikely to succeed. The challenge to the assumptions — whether the Savings Club price really is the “usual and customary” price — is a trial issue. The court will let the jury hear the model and decide.
If you’re building a class-wide damages model and need to make sure the methodology survives both a Daubert challenge and the class certification predominance requirement, happy to discuss the approach. The model’s architecture — formulaic, data-driven, transparent — is the design decision that determines whether it carries the case or collapses under scrutiny.
Chris Walton, JD, is President & CEO of Eton Venture Services. He can be reached at [email protected].
Schedule a free consultation meeting to discuss your valuation needs.
