California Confronts the High Liability Costs of Scientific Uncertainty

Much has been said about the eye-popping verdict and the post-trial rulings in the Roundup case tried in San Francisco earlier this year. Johnson v. Monsanto Co., 2018 WL 5246323 (S.F. Super. Ct. Oct. 22, 2018). The court sustained the jury’s award of ~$39 million in compensatory damages, including $37 million in non-economic damages, and its finding that Monsanto was liable for punitive damages. The court reduced the punitive award on due process grounds to a one-to-one ratio, slashing it from $250 million to approximately $39 million. Monsanto recently filed its notice of appeal, and as we await the briefing and argument, a few issues and takeaways merit further discussion, particularly several disturbing issues surrounding the award of punitive damages. We will save for another day (or post) other significant liability and damages issues.

Roundup’s active ingredient, glyphosate, has been the subject of voluminous scientific study over decades, producing, at best, an ongoing scientific debate over whether it can cause Non-Hodgkin’s Lymphoma (NHL), a tragically common blood cancer. The verdict rests on general and specific causation testimony of rather questionable reliability. And because this is the first verdict in a young mass tort litigation, including numerous state court cases and a federal MDL, Monsanto faces enormous potential exposure. There is a real possibility that a series of different juries will reach varying conclusions as to the link between Roundup and NHL, as well as the reasonableness of Monsanto’s conduct.

Given these multiple levels of uncertainty about Monsanto’s liability and the level of potential financial exposure, the punitive damages award raises troubling questions about the appropriate basis for an award of exemplary damages under California law. California reserves punitive damages for cases where the evidence demonstrates clearly and convincingly that the defendant acted with deliberate and conscious disregard for human safety in causing plaintiff’s injury. So the Roundup appeal raises the question – how much uncertainty is (or should be) too much uncertainty and opacity to allow an award of tens of millions of dollars in punitive damages, especially when added to an equally large compensatory award of noneconomic damages?

The Uncertain Science: The first level of uncertainty as to Monsanto’s purported conscious disregard for safety lies in the scientific basis for general causation, the conclusion that glyphosate actually causes NHL. By and large, the stronger the inference of general causation, the greater the basis to require a warning; the clearer the need for a warning, the greater the basis to conclude that the failure to warn connotes a conscious disregard for safety. And vice versa. The same is true in evaluating the reprehensibility of the conduct, which ratchets up/down the appropriate potential award.

Glyphosate-based herbicides have been used ubiquitously for decades, and their risks have been well-studied by the scientific community. Many regulatory bodies and public health organizations have synthesized the body of scientific evidence and evaluated the risks. The consistent consensus has been that glyphosate is not a proven cause of cancer. For example, the EPA and the ECA (European Chemicals Agency) have specifically and repeatedly declined to characterize glyphosate as carcinogenic or probably so. The contrary classification relied on by plaintiffs and their experts is an outlier: the 2015 monograph by the International Agency for Research on Cancer (IARC) classifying glyphosate as a “probable human carcinogen,” published after plaintiff Johnson was diagnosed with NHL. Importantly, in IARC-speak “probable” means “possible.” Accordingly, even IARC considers the link between glyphosate and cancer to be no more than speculative.

To put this in further perspective, courts have largely agreed that an agency’s finding of a causal link between a substance and a disease is inadmissible to prove general causation.  Regulatory and public health bodies ordinarily apply relatively lax causation standards, normally taking a precautionary approach to the scientific evidence, akin to a better-safe-than-sorry standard. To a court focused on finding objective scientific truth to resolve a private dispute, the agency’s conclusion carries little legitimate evidentiary weight but great potential to mislead the jury.

Given this precautionary standard, it would be logical to give an agency’s inability to find a causal link under even this liberal lens heightened exculpatory significance. Regardless, the outlier IARC finding provides no substantial support for any claim that Monsanto consciously disregarded an undeniable risk when it failed to warn users that Roundup causes cancer.

The Unreliable Experts: In light of California’s evidentiary requirement of clear and convincing evidence of conscious disregard, it seems obvious that to support an award of punitive damages a jury’s finding must rest on robust, reliable expert opinion testimony of Roundup’s carcinogenicity.

But that is not the case here. Pretrial, the trial court declined to exclude plaintiff’s experts’ causation testimony because they cited some support in the scientific literature and adhered, at a fairly high level of generality, to standards applicable in their field. The court was satisfied because there was some evidence supporting a statistical association and some evidence that the link was biologically plausible. Hardly a solid foundation for a general causation opinion, and hardly a clear and convincing basis to find that Monsanto deliberately disregarded a compelling body of evidence demonstrating that its product causes cancer. In particular, plaintiff’s experts relied in part on an unreliable “pooled data” study, failed to consider the dose-response relationship, and failed to consider whether the science supported causation given the unusually short latency period the experts attributed to the plaintiff’s injury.

Soon thereafter, a federal court examined the admissibility of the same general causation testimony in the federal MDL. In re Roundup Prods. Liab. Litig., 2018 WL 3368534 (N.D. Cal. July 10, 2018). The MDL court cast serious doubt on the significance of the IARC finding, characterized the evidence of causality as “rather weak” and “too equivocal” to support a firm causation conclusion, and excluded two experts entirely and one partially. Under the notoriously relaxed gatekeeping standards mandated by Ninth Circuit precedent, the court concluded that three of the causation experts were “shaky but admissible.” Ultimately finding it a “close question,” the MDL court denied summary judgment at the general causation stage, finding the experts’ disputes over the evidence to be “reasonable” and capable of generating a triable issue.

Meanwhile, in state court, the San Francisco jury received the testimony of plaintiff’s experts and was allowed to consider punitive damages, returning its rather generous verdict. The trial court concluded that in resolving post-trial motions it was precluded from revisiting its earlier admissibility decisions, and indeed was required to view the expert testimony in the light most favorable to the verdict. Accordingly, any weaknesses or evidentiary errors undermining plaintiff’s expert case had to be disregarded in reviewing the verdict at the trial court level. Even under the deferential substantial evidence rule, the court of appeal will have somewhat more flexibility.

The upcoming appeal is certainly “one worth watching.” Monsanto has been tagged with $37 million in non-economic damages and $39 million in punitive damages, based on its failure to warn of a risk no one in the extremely active regulatory community has classified as “probable” in the quantitative sense (the sense that tort law requires) even with a public health thumb on the evidentiary scale, based largely on the testimony of experts ranging from too unreliable to be admitted to “shaky” at best, and based on risk evidence which Monsanto has “reasonably disputed.” Whether the California court of appeal lets this result stand under these circumstances will say a great deal about whether there are any meaningful safeguards under California law against the imposition of overwhelming liabilities for unproven, uncertain and reasonably debatable risks supported by shaky (at best) expert testimony.