A drug manufacturer’s research pipeline is many things: a bet on science, a bet on regulators, a bet on patents, and a very expensive bet against failure. What it has not traditionally been—at least until now—is a standing invitation for tort plaintiffs to argue, years later, that the company should have bet differently.
That is the question now before the California Supreme Court, which heard oral argument yesterday in Gilead Tenofovir Cases (S283862). The case asks whether California law recognizes a previously unheard-of “duty to commercialize a safer alternative drug.” Under the First Appellate District Court of Appeal’s rule, that duty arises whenever a pharmaceutical manufacturer allegedly “knows” that another formulation in its pipeline is at least as effective, and safer, than the product currently on the market.
The International Center for Law & Economics (ICLE) filed an amicus brief in the case urging reversal. The brief argues that the appellate court’s rule effectively eliminates the longstanding product-defect requirement, rests on a misreading of Brown v. Superior Court, and—most interestingly from a law & economics perspective—would invite courts and juries to second-guess R&D decisions that the patent system and U.S. Food and Drug Administration already heavily structure.
Oral argument, of course, is no oracle of case disposition. Even so, two lines of questioning stood out to me, and both, I think, vindicated concerns we raised in the brief.
The Duty That Couldn’t Be Defined
The most revealing stretch of argument came when the justices repeatedly asked plaintiffs to explain, with any precision, what their proposed duty would actually require pharmaceutical manufacturers to do. The court never got an answer it seemed able to work with.
Plaintiffs argued that manufacturers owe “a duty to act reasonably in making commercialization decisions about an allegedly safer and at least equally effective alternative drug.” But when the justices pressed them on what that meant in practice, plaintiffs largely circled back to the same formulation: manufacturers must “act reasonably.”
The bench kept returning to the obvious follow-up: reasonable according to what standard?
At one point, the court tested the theory with a hypothetical. Could a jury find a pharmaceutical company acted unreasonably by pursuing one research path over another—perhaps backing a less promising candidate that ultimately failed—based on how it allocated scarce R&D resources? Plaintiffs said yes.
That answer immediately raised another concern: hindsight bias. Once a lawsuit is filed years later, after more data emerges and one research path succeeds while another does not, what exactly is the legal rule supposed to tell drug developers ex ante? What is the operating instruction?
Notably, defense counsel faced no parallel interrogation. The court’s questioning there focused mostly on how Gilead’s position fit within existing California law.
The exchange tracked almost perfectly with what our brief predicted. Plaintiffs’ theory is, as we argued, “a recharacterization of a traditional products liability claim, attempting to achieve the same result while avoiding the required showing of defect.”
California has long required plaintiffs alleging injury from a product to prove that the product was defective for a reason. A free-floating “duty to act reasonably,” untethered from any product defect, gives juries little meaningful guidance and leaves manufacturers with no workable way to structure their conduct in advance.
The justices’ discomfort echoed concerns the California Supreme Court itself expressed in Brown v. Superior Court. There, the court warned:
Perhaps a drug might be made safer if it was withheld from the market until scientific skill and knowledge advanced to the point at which additional dangerous side effects would be revealed. But in most cases such a delay … would not serve the public welfare.
The court also raised a closely related question: Why didn’t plaintiffs plead a design-defect claim in the first place?
The answer is straightforward. Plaintiffs have repeatedly conceded—including again at oral argument—that the drug at the center of the case is not defective and should not be removed from the market.
But that concession is the whole problem.
Without a defective product, plaintiffs are left arguing for what our brief describes as “an unprecedented expansion of tort liability” that “would expose manufacturers to potentially unlimited liability for products that are reasonably safe and defect-free.”
The justice who asked the question almost certainly understood that. In many ways, the question itself was the point.
A Jury Is Not a Drug-Development Committee
The second revealing exchange came from a different line of questioning, one aimed squarely at institutional competence. The court asked plaintiffs how their proposed duty fit alongside the patent and regulatory regimes that already govern pharmaceutical development.
Plaintiffs responded largely by criticizing the pharmaceutical industry’s allocation of resources between R&D and marketing. In a sense, that response went directly to one of the core concerns in our amicus brief: a rule like the one adopted in Gilead would effectively give plaintiffs’ attorneys and juries a veto over enormously complex, expensive R&D decisions.
Under the court’s framing, though, the problem ran deeper. The patent system already reflects a deliberate legislative bargain about innovation incentives. Congress decided how long manufacturers receive exclusivity for new inventions. Overlaying that framework with a tort duty requiring companies to commercialize drugs on some judicially determined timetable would replace an explicit statutory scheme with ad hoc, case-by-case litigation judgments.
The court appeared deeply uneasy with that prospect. At one point, the justices described the proposition, in so many words, as “uncomfortable.”
As our amicus brief explained, plaintiffs’ theory “would set an unrealistic and dangerous standard of perfection for drug development.” More fundamentally, though, the theory largely ignores the patent system that already shapes the timing of pharmaceutical commercialization.
That matters because patent terms run from the filing date, not from the date a drug reaches the market. In practical terms, sitting indefinitely on a superior product is not some masterstroke of profit maximization; it is usually economically irrational. Every year spent delaying commercialization is a year of patent exclusivity burned off the clock.
In other words, plaintiffs’ suppression theory depends on a view of the patent system that does not fit how pharmaceutical markets actually work.
As we noted in the brief, if Gilead truly believed Tenofovir alafenamide (TAF) was both safer and equally effective, the company generally would have had every incentive to bring it to market sooner, not later.
A Tough Day for an Expansive Tort Theory
Of course, oral argument is famously unreliable tea-leaf reading. The justice grilling you at the lectern may ultimately write the opinion in your favor; the justice tossing softballs may end up ruling against you.
Still, the substance of the questioning mattered. Plaintiffs repeatedly struggled to explain what “reasonable” commercialization behavior would actually require in practice. At the same time, the court repeatedly drifted toward concerns about institutional competence and the role of legislatures, regulators, and the patent system in structuring pharmaceutical innovation.
That combination feels significant.
If I had to guess, reversal now seems more likely than not. The most plausible paths are the two the defense laid out at argument: either the court reaffirms the defect requirement as the exclusive framework for product-based negligence claims, or it recognizes some broader duty in principle before concluding—under the familiar Rowland framework discussed extensively at argument—that this particular theory goes too far.
Either way, the argument suggested a court searching less for a way to expand tort law than for a limiting principle to stop it from expanding indefinitely.
And that may tell us all we need to know.
