The Supremes Shut Down the Plaintiff’s Bar Crusade Against Monsanto’s Roundup

This past week, the United States Supreme Court held in Monsanto v. Durnell that the Federal Insecticide, Fungicide and Rodenticide Act (FIFRA) “expressly preempts plaintiff [Durnell]’s state-law failure-to-warn claim because the claim would require Monsanto to add a cancer warning to its Roundup products’ label.” Writing for the majority, Justice Kavanaugh delivered what could well be a well-deserved death knell for the barrage of tort claims that plaintiffs have long brought against chemical companies for alleged injuries caused by exposure to commonly used weedkillers and other substances. The Court reversed a $1.25 million award from a Missouri jury and upset the historical trend in its seven-two decision—by complying with federal warning label requirements, these companies cannot be found liable for failure to warn.

The ground for the reversal was the doctrine of federal preemption, under which no state court decision can stand if inconsistent with a federal statute. FIFRA’s preemption clause, entitled “Uniformity,” provides that a “State shall not impose or continue in effect any requirements for labeling or packaging in addition to or different from those required under this subchapter.” The requirements under the subchapter include a standardized warning. By requiring a standardized warning and prohibiting additional or different warnings, Congress adopted an across-the-board solution about what product warning suffices, thereby preventing different juries or judges from imposing inconsistent requirements.

That approach is fully appropriate because nothing turns on a patient’s medical history, which is relevant, if at all, to questions of assumption of risk or contributory negligence. Because these failure-to-warn cases are the only product liability cases where risk cannot be reduced by any kind of improved practices of design or inspection, today’s state tort law claims almost always revolve around the adequacy of warnings. Congress realized that a “uniformity” provision is essential from a planning perspective. If there is no safe harbor like FIFRA, then in every case the plaintiff can, and will, urge the jury to find insufficient whatever warning was supplied, and to presume that had a sufficient warning been provided to the plaintiff, it would have been heeded. No company can expect to avoid an endless procession of lengthy, repetitive, and unpredictable trials over the issue of medical causation in the absence of a safe-harbor position that the preemption statute supplies.

This issue of warning is distinct from and obscures the related issue of medical causation. What does it matter what the warning said or whether a different warning would have been heeded if the product isn’t what caused the harm in the first place? Products like Roundup are sold through hundreds of outlets to tens of thousands of individual users. Large numbers of individuals get non-Hodgkin’s lymphoma from other causes, just as individuals who never smoke can die of lung cancer. The facts of this case illustrate the legal problem well.

John Durnell sued Monsanto (now a subsidiary of the German industrial giant Bayer) claiming he developed non-Hodgkin’s lymphoma from twenty years of exposure to Bayer’s Roundup weedkiller and that, had a cancer warning been on the Roundup label, he would not have used Roundup and therefore would not have developed non-Hodgkin’s lymphoma. As evidence, Durnell merely explained that Roundup contains the common chemical glyphosate, which, he claimed, based on dubious World Health Organization 2015 data, was carcinogenic in humans. But as Justice Brett Kavanaugh stressed repeatedly, glyphosate is among the most studied chemicals in the world, and “regulatory bodies around the world have likewise concluded that glyphosate is not carcinogenic.” Nevertheless, Durnell prevailed before a Missouri jury, securing over $1.25 million. After the Supreme Court reversed, Bayer (the parent of Monsanto) explained that “[g]lyphosate remains the most studied crop protection tool in the world and this decision affirms that the EPA’s safety determination is the law of the land, ensuring that companies cannot be punished under a patchwork of state tort laws for complying with federal labeling requirements.”

There is thus a profound irony because, as products become ever safer, the now dominant warning cases have displaced traditional construction and design cases. Historically, the famous early case of Escola v. Coca-Cola Bottling Co. (1944) arose when a Coke bottle was alleged to have broken in the hands of a waitress as she was removing it from the case in which it was delivered. The implicit assumption in the opinion was that the bottle was defective in its original condition while still in the hands of the defendant, so that the only legal dispute before the California court was how to establish whether the defendant was negligent in handling that bottle’s condition. The California Supreme Court eased the plaintiff’s path by holding that the jury could invoke the doctrine of res ipsa loquitur (the thing speaks for itself) to create a rebuttable presumption of negligence. In a stirring concurrence, Justice Roger Traynor held that a sound legal rule should remove from the plaintiff the burden to prove negligence by holding the firm strictly liable.

In fact, that famous dispute was a tempest in a teapot because the two theories converged 99+ percent of the time. In fact, the real factual issue in Escola was whether a properly made bottle exploded because of its use by the plaintiff or her coworkers was not “normal and proper”: because the bottle had been left too long in the sun or because it had been tossed around in the shop

Due to this focus on upstream conduct, a fraud ring emerged at the time, coaching witnesses on what to say to win these cases. However, insurance premiums increased only slightly, and the combination of improved production and inspection techniques made such cases a rarity. Downstream conduct is key in all product cases because a manufacturer cannot control the activities of parties who take possession of its product after the initial sale, where their use may deviate from standard procedures.

Such deviation from standard norms is not far-fetched. In Wyeth v. Levine (2009), the Vermont Supreme Court let the jury find that the plaintiff, a musician, could recover from Wyeth, which marketed its generic anti-nausea drug Phenergan, even though the treating physician’s assistant injected an excessive dose of the drug into her artery, when a smaller dose was only supposed to be injected into a vein. Thus, that court let the doctor off the hook to soak the rich defendant for huge damages, a double distortion that the preemption rule adopted in Durnell could have prevented. The same was true of the simple design cases of the time. No one would claim that a car was safe if it exploded when the ignition was turned on because gas fumes had accumulated in the carburetor. But it was a very different matter if a driver sped into a brick wall without wearing a seatbelt and then incurred substantial damage because of a misinformed claim that the product was defective in design. Again, preemption is needed, but in the Supreme Court, Justice John Paul Stevens shrugged and did nothing.

The risk that upstream conduct—whether a warning was sufficient—will be the sole focus of inquiry, ignoring core downstream questions, is today far greater with warning cases like Durnell on its two key downstream factual issues: first, whether the product did cause the disease condition alleged, and second, whether an adequate warning would have been sufficient to tell a plaintiff like Durnell to either stop using Roundup or take additional precautions that would have been sufficient to prevent the injury.

In her dissenting opinion, Justice Jackson noted that huge numbers of state and federal judgments let the jury decide both questions. She argued this was not inconsistent with FIFRA:

Durnell’s failure-to-warn claim is not “in addition to or different from” FIFRA’s mandates; it is equivalent to FIFRA’s key labeling requirement—the misbranding prohibition. And Durnell’s claim does not conflict with any other FIFRA “requirement” for §136v(b) purposes because the EPA’s registration of a pesticide and approval of its label does not create a labeling requirement under FIFRA.

Justice Jackson suggested her position was the sounder result on policy grounds. Judicial acceptance of the FIFRA preemption argument, she suggested, “ultimately leaves Durnell without a remedy for the significant harms that he has suffered.” But Justice Jackson is wholly incorrect both that state failure-to-warn actions are compatible with FIFRA’s uniformity provision and that policy militates in favor of this result.

In assuming that the common law action was “equivalent” to the statutory warning, she in effect assumes that, so long as the object of the common law rule was at the highest level of abstraction the same as the statutory rule—patient safety, it matters neither that the federal law and state laws apply wholly different techniques to achieve that common end, nor that liability under her so-called “equivalent” rule is millions of times larger than that which could be imposed under the safe-harbor created by the statute. There is no question that the uniform rule of FIFRA was designed to protect against differing state laws. FIFRA’s no-liability rule is wholly inconsistent with a regime that imposes huge liabilities without limits determined by juries.

On the issue of policy, the classic argument against the preemption regime is that it will cause the FDA to be far too risk-averse in granting approval for standard drugs. The consequences of a drug may vary by the condition of the patient and by drug interactions with other products, and no drug is ever entirely safe. Keeping new drugs off the market means the FDA cannot be held responsible or face severe criticism for losses resulting from their release. In keeping a product off the market, the FDA may well deny or delay the application of valuable products, preventing far more benefit than withholding the product would have prevented harm. Therefore, under a regime in which the FDA is not directly responsible for the sufficiency of the warning and is instead left to state juries, the FDA may be more likely to allow valuable products onto the market.

But this argument is mistaken, because there are many post-release avenues for the FDA to take corrective action without even having to pull the product from the market. The presumption should be that the product remains on the market, where further review can be conducted by both the manufacturer and independent drug testing institutions. Those post-release devices are fully available for Roundup or any other product marketed under FIFRA, such that it is possible to update warnings or instructions without taking the product off the market. But even the option of cancellation remains viable in principle, if necessary.

But it is the FDA that is best positioned to make this determination, not the cumulative effect of individual state juries. For example, before the Court’s ruling, Roundup could easily have been driven off the market by the billions of dollars in tort cases. Justice Jackson ignores that possibility, just as she distorts the record when she assumes that Durnell got his illness from Roundup, thus ignoring alternative causes. Her decision thus fails to prevent rogue juries from finding liability when none exists. Nor is it a reply to say that it is better to be safe than sorry by including a cancer warning on the label. But Monsanto was also right to resist that argument. If the product is safe, then it is dangerous to pretend it is dangerous, if such conduct could lead people not to use a safe product or to refuse to pay for crops from fields protected by Roundup. A false warning is also a misrepresentation, which runs counter to market efficiency and should never be used solely to ward off liability that should never be imposed in the first place.

Richard A. Epstein is a senior research fellow at the Civitas Institute. He is also the inaugural Laurence A. Tisch Professor of Law at NYU School of Law, where he serves as a Director of the Classical Liberal Institute, which he helped found in 2013.