A couple of months ago, the Kentucky Court of Appeals in Grant Thornton LLP v. Yung cut a trial court’s award of punitive damages from $80 million to $20 million—reducing the punitive/compensatory ratio to 1:1.

The case arose from Grant Thornton’s provision to plaintiffs of a tax avoidance strategy that the IRS rejected; and it is unusual in that the excessive award was imposed by the trial judge after a bench trial, not by a jury.

The decision is not yet final: The plaintiffs have filed a motion for discretionary review in the Kentucky Supreme Court in which they challenge the reduction. While we wait to hear whether the high court will weigh in, we will comment on certain aspects of the decision that we found interesting.

First, the Court of Appeals held that a punitive award equal to the compensatory damages (i.e., a 1:1 ratio) is the constitutional maximum “where the harm caused was entirely economic, the plaintiffs were sophisticated business entities who were not financially vulnerable, and the underlying award of compensatory damages was substantial.”

It strikes us as exceedingly unlikely that the Kentucky Supreme Court would want to review this holding. The U.S. Supreme Court has explained that when compensatory damages are “substantial,” a 1:1 ratio between punitive and compensatory damages generally marks the “outermost limit of the due process guarantee.” And many courts—including the Sixth Circuit in several decisions cited by the Kentucky court—have applied this limit.

If anything, the Court of Appeals did not go far enough in reducing the award. As it observed, “not all acts which cause economic harm are sufficiently reprehensible to justify a significant sanction in addition to compensatory damages.” The U.S. Supreme Court instructed in BMW that a multi-million-dollar punitive award—which is “tantamount to a severe criminal penalty”—”cannot be justified on the ground that it was necessary to deter future misconduct without considering whether less drastic remedies could be expected to achieve that goal.”

Here, the Court of Appeals determined that a punishment equal to the compensatory damages “would adequately punish Grant Thornton for its misconduct,” but it did not sufficiently consider whether a “less drastic remedy” also would have sufficed.

In particular, the court gave no weight to the fact that the compensatory damages alone far exceeded Grant Thornton’s “gain” from the alleged misconduct and thus would have a strong deterrent effect in their own right. Similarly, the court failed to appreciate that the amount of money that Grant Thornton spent in defending the plaintiffs in the IRS tax audit represented a dead-weight loss to Grant Thornton that likewise should have a strong deterrent effect.

Second, the Court of Appeals deemed Grant Thornton’s conduct to be more reprehensible because there was evidence that the firm marketed the same or similar tax structures to about 35 other customers. In our view, however, that fact does not justify a higher award.

The Supreme Court held in BMW that “repeated misconduct is more reprehensible than an individual instance of malfeasance,” but it cautioned in State Farm that, before treating the defendant as a recidivist, “courts must ensure [that] the conduct in question replicates the prior transgressions.” The mere fact that Grant Thornton marketed the services in question to other customers does not mean that it defrauded or breached its fiduciary duty to any other customer, and it therefore was irrelevant to assessing reprehensibility.

Third, the Court of Appeals commented that the trial court may have “weigh[ed] punitive damages more harshly” because of its frustration resulting from “certain discovery issues involving Grant Thornton.” To do so, the court explained, would be improper. It held that “any discovery violations must be remedied separately from the underlying conduct.”

This is a useful holding. Plaintiffs often interject allegations about litigation misconduct into trials for the purposes of inflaming the jury and extracting a large punitive award. The Kentucky court’s decision shows that such evidence should not be admitted at trial, much less considered in connection with punitive damages.

In sum, the Court of Appeals made some useful holdings in the course of deeming the punitive award to be unconstitutionally excessive, but should have gone further by recognizing that the compensatory damages and other costs borne by Grant Thornton fully served Kentucky’s interests in deterrence and retribution and should have justified an even greater remittitur, if not outright elimination of the punitive award.