A jury in the Western District of Louisiana made headlines last spring when it awarded a stunning $9 billion in punitive damages to a plaintiff who contended that the diabetes drug Actos caused his bladder cancer.  Last week, the district court cut the award by 99.6 % to approximately $37 million. Despite the impressive scale of the reduction, in our view the remitted award remains unconstitutionally excessive.  Furthermore, the district court’s lengthy opinion reveals significant errors of reasoning that we hope the Fifth Circuit will correct on appeal.  We address three of them here.

First, a new trial rather than remittitur is the appropriate remedy because the aberrational verdict shows that the jury was inflamed and failed to follow the court’s instructions.  The court instructed the jury that punitive damages should be “proportionate” to actual damages and cautioned it not to punish the defendants for alleged injuries to non-parties.  The jury nevertheless set the punishment at 6,000 times the compensatory award—an amount so far in excess of the norm as to defy rational explanation.  The district court surmised that the jury wished to “deter such conduct in the future from corporations whose values are measured in the billions.”

As one of us explained in a recent post addressing the role of wealth in setting punitive damages, however, there is no valid deterrence-based rationale for levying huge exactions against corporate defendants that happen to have a high net worth. The Supreme Court cautioned in State Farm v. Campbell, moreover, that when jurors are urged to increase a corporate defendant’s punishment based on its balance sheet, they may “use their verdicts to express biases against big business, particularly those without strong local presences.”  Here, the verdict so vastly exceeds the scale of permissible punishment that its sheer size compels the inference that the jury misunderstood its task or was swayed by improper factors.  A new trial is the proper remedy for this sort of error.

Second, the district court acknowledged but seriously misapplied Philip Morris v. Williams.  In Williams, the Supreme Court held that the “Due Process Clause forbids a State to use a punitive damages award to punish a defendant for injury that it inflicts upon . . . those who are, essentially, strangers to the litigation.”  Building on the concern expressed in State Farm about the danger of repeated, duplicative punishments, the Williams Court explained that such a punishment would be arbitrary (because the jury would be speculating about how many non-parties had been injured) and would deprive the defendant of its right to defend against the non-parties’ claims.  Thus, the Court ruled, although “evidence of actual harm to nonparties can help to show that the conduct that harmed the plaintiff also posed a substantial risk of harm to the general public, and so was particularly reprehensible,” the jury may not “use a punitive damages verdict to punish a defendant directly on account of harms it is alleged to have visited on nonparties.”  Those other alleged victims can bring their own claims, and juries in those cases may impose punitive damages if they deem that remedy to be warranted.

The district court did not correctly apply these principles. It acknowledged the Supreme Court’s admonition that “one injured party should not be allowed to inflict the punishment for all who might be harmed by the Defendants’ act,” and it noted that 8,000 cases claiming injury from Actos are already pending in federal and state court.  But the district court nevertheless focused on the deterrent effect of the case before it, expressing concern that a punishment of even ten times the seven-figure compensatory award would be too “easily absorbed” by the large defendants to effect sufficient deterrence.  In so reasoning, it failed to recognize that inherent in Williams limitation of punishment to the harms inflicted on the party or parties to the particular case is the principle that no single case or subset of cases should alone be assessed in terms of its efficacy in producing sufficient punishment to deter the company’s entire course of conduct.  Rather, it is the collective result of all foreseeable cases challenging that conduct that should produce the desired effect.

Accordingly, faithful application of the principle underlying Williams required the district court to ask whether the total punishment would be excessive if all 8,000 plaintiffs were awarded punitive damages equal to the amount it would decide was allowable for Mr. Allen.  Had it asked that question, it would have readily seen that $37 million for the harms caused to Mr. Allen would lead to punitive damages of $296 billion for all the harms allegedly caused to the universe of plaintiffs, an amount that would substantially exceed the sum of the net worth of both defendants and the total net sales of Actos over the product’s entire history.  Indeed, if every other plaintiff were awarded compensatory damages equal to the $1,475,000 awarded to the Allens, the compensatory liability alone would approach $12 billion and probably fully satisfy any need for deterrence without adding a penny of punitive damages.

But, you might say, suppose many of these other lawsuits fail and result in no punitive damages?  In fact, the district court expressed precisely that concern, noting that “other Plaintiffs might or might not be able to establish harm to themselves.”  The prospect that the defendant may be exonerated in future cases does not justify imposing disproportionate punitive damages in cases of winning plaintiffs, however.  On the contrary, mixed results in litigation may show that the evidence supporting the kind of aggravated misconduct necessary to justify punitive damages, or even to support any liability, is equivocal, that the product infrequently causes the claimed adverse effects, or that the legal standards are uncertain.

As Judge Posner observed in In re Rhone-Poulenc Rorer, Inc., because different juries inevitably will interpret the same evidence differently, a “decentralized process of multiple trials, involving different juries, and different standards of liability, in different jurisdictions,” is more likely to be fair than a system in which defendants “stake their companies on the outcome of a single jury trial.”  Although Judge Posner was explaining why individual actions are preferable to a class action for cases like this one, that reasoning also explains why it would be a mistake to allow a single jury to punish for an entire course or disproportionate share of conduct alleged to have affected  many potential plaintiffs.

In fact, plaintiffs in the Actos cases have met with mixed success at trial.   The defendants prevailed in the three state cases tried before Allen—once winning a favorable jury verdict and twice persuading the trial courts to overturn verdicts against them.  A few weeks after the Allen trial, juries in Illinois and Nevada exonerated the defendants and found against three plaintiffs.  Last month, a Philadelphia jury awarded a plaintiff who contracted bladder cancer after taking Actos compensatory damages but rejected the claim for punitive damages.   The effect of the district court’s notion that the full measure of deterrence must be accomplished in this case is to deprive the defendants of the benefit of these victories.  The court, in essence, imposed on the defendants a one-way class action.  Their victories have no binding effect in future cases; but even a single loss can be used to punish them in an amount that is wholly disproportionate to the injury in that case.  That result is irreconcilable with Williams and the notions of due process that undergird it.

Finally, the district court misapplied the requirement that there be a reasonable ratio between the compensatory and punitive damages—a rule that it deemed “dissonant” with the deterrence rationale for punitive damages.  Reviewing the Supreme Court’s cases on this subject,  the court found them to contain only “limited guidance” for circumstances involving a “high degree of reprehensibility” and “the need to adequately deter such conduct in the future.”

In fact, however, the Supreme Court has given very clear guidance about the permissible ratio of punitive to compensatory damages, and its decisions refute any suggestion that a 25:1 ratio is permissible when the compensatory damages exceed $1 million, even if the conduct is exceptionally reprehensible.  The Court has clearly stated that “few awards exceeding a single-digit ratio between punitive and compensatory damages, to a significant degree, will satisfy due process” and has held that “[w]hen compensatory damages are substantial, . . . a lesser ratio, perhaps only equal to compensatory damages, can reach the outermost limit of the due process guarantee.”  In addition, the Court has repeatedly expressed concern about “outlier cases [that] subject defendants to punitive damages that dwarf the corresponding compensatories,” observing that the median ratio in all cases is less than 1:1 and holding that 1:1 is the maximum permissible ratio in maritime case—however reprehensible the defendants’ conduct.   The district court’s opinion—largely driven by its concerns about the need to deter large and wealthy companies and its fundamental misconception that it is empowered despite Williams to attempt to accomplish that in a single-plaintiff case—identifies no basis to depart from these clearly established norms.