Application of the Supreme Court’s excessiveness guideposts to cases involving multiple defendants is one of the more confounding problems that arises in punitive damages jurisprudence. The Supreme Court of Texas got the issue right in Horizon Health Corp. v. Acadia Healthcare Co., a case in which several defendants were jointly liable for compensatory damages but individually liable for separate punitive awards.

The plaintiff in Horizon alleged that a competitor conspired to hire away the plaintiff’s employees, causing them to breach their employment contracts and take the plaintiff’s confidential information with them when they left. The defendants were the plaintiff’s former employees and upper management at the competitor.

Initially, the court recognized that it must conduct a separate excessiveness review for each defendant because “[c]onstitutional rights are personal in nature, and, therefore, inquiries into the constitutional excessiveness of an exemplary damages award must be defendant-specific.” “This approach is also consistent with the underlying purpose and focus of exemplary damages—to punish the wrongdoing rather than to compensate the claimant.” In particular, because “defendants act with varying degrees of reprehensibility” and “are assessed by the jury on a per-defendant basis according to each defendant’s culpability, the alleged excessiveness of those damages must also be evaluated on a per-defendant basis.”

The decision to analyze punitive awards separately can be critical when applying the reprehensibility guidepost because it prevents courts from adding up the allegedly wrongful actions of the various defendants and holding each of them liable for the cumulative wrongdoing of the group. Absent proof of collusion between the defendants to engage in the conduct that harmed the plaintiff, each defendant should be held accountable only for the wrongfulness of that defendant’s own actions.

It is worth noting that the court avoided two common pitfalls when applying the reprehensibility factors to the individual conduct of each defendant.

  • The court rejected the contention that the first reprehensibility factor “does not carry weight” in cases involving economic torts. As the court held, the purpose of this factor is to recognize that conduct is less reprehensible when it causes only economic harm and not physical injury. Because Horizon involved economic torts—theft of proprietary information and violation of non-compete obligations—this factor “weigh[ed] against a finding of reprehensibility.”
  • The court correctly held that “Campbell’s fourth reprehensibility factor is about recidivism, not the course of conduct giving rise to the plaintiff’s ultimate harm or the wrongdoer’s attempts to conceal his wrongdoing.” Even though “all of the individual defendants were found to have committed multiple torts,” the court recognized that “these findings do not show recidivism.” Because there was no evidence that the defendants had engaged in similar misconduct in the past, the fourth reprehensibility factor was not present in the case.

Returning to “[t]he more interesting and complicated question” of “how the constitutional excessiveness of an exemplary damages award should be assessed when multiple defendants face joint and several liability for compensatory damages but are individually liable for exemplary damages,” the court’s decision to analyze each punitive award individually raised the question of what number should go in the denominator of the ratio analysis.

The plaintiff in Horizon argued that, because each defendant was jointly liable for the compensatory damages, the entire compensatory award should be used as the denominator in each ratio. The court properly rejected that notion, concluding that, under the ratio guidepost, “the punishment should fit the misconduct rather than a state’s joint and several liability regime.”

Accordingly, “in determining the basis for a constitutionally permissible amount of exemplary damages, courts must consider the harm each defendant actually caused and assess the punishment based on that harm because this approach most closely matches the punishment to each defendant’s misconduct.” Of course, to conduct such an analysis, there must be an apportionment of liability among the defendants—or some other way of assigning a share of the plaintiff’s harm to each defendant. Accordingly, it is generally advisable for defendants in multi-defendant cases to request a verdict form that apportions fault among the defendants even if liability is joint and several.

Apportioning the compensatory damages between the individual defendants in Horizon pursuant to their individual liability, the court then calculated the ratio for each defendant’s punitive award. Instead of the uniform 4:1 ratio employed by the trial court (using the entire compensatory award as the denominator for each punitive award), the Supreme Court of Texas recognized that the actual ratio between the harm caused by a defendant and the punishment being imposed on that defendant ranged from 11:1 to 29:1. This, the court held, was facially excessive punishment.

Indeed, the court held that even a ratio of 4:1 would be excessive in a case that involved only one of the five reprehensibility factors. In reaching that conclusion, the court rejected the contention that the compensatory award of $55,049 justified a higher ratio because it was “small” or “nominal.”

Importantly, the court considered the entire compensatory award—rather than each defendants’ share of the award—for this purpose. That is consistent with the United States Supreme Court’s language establishing this exception, which focuses on the harm experienced by the plaintiff and recognizes that higher ratios might be appropriate to accomplish punishment and deterrence when the defendant’s conduct is reprehensible but the plaintiff’s actual injury is nominal or very small. As the Supreme Court of Texas recognized, “[w]hen the award of exemplary damages is based on ‘substantial’ economic harm suffered, a plaintiff may not argue that the actual damage award is nominal and thus justifies a higher ratio.”