This is a fun paper on a not-so-fun topic — punitive damage awards. In civil cases, juries may award both compensatory and punitive damages against defendants. Punitive damages are not meant to reimburse claimants for their harm, rather discipline defendants for their socially undesired behavior. Under the tax code both types of damages can be considered deducible ordinary and necessary business expenses. The tax-deductibility of punitive damages has been highly criticized. Treasury has continuously expressed its objection to allowing taxpayers to deduct punitive damages arguing it provides a tax break for bad behavior. Both the Obama and the Clinton administrations proposed but failed to make punitive damages non-deductible. What is the correlation of tax deductibility to jury’s decision to award punitive damages?
Scholars such as Polsky and Markel argued that from a practical perspective, jury members are oblivious to the deductibility of punitive damage awards. In fact, the idea that punitive damages are tax-deductible often comes as a surprise even to defendants. This ignorance takes the sting out of the awards and under-punishes defendants. To mitigate this problem, Polsky and Markel recommended to include information on the tax deductibility of the award in jury instructions. They contended that jurors that have such information will have a better sense of how to adjust their punitive damages award. When jurors are explicitly told that punitive damages are tax deductible, it is reasonable to expect they will increase the awards viewing tax-deductibility as a tax break for engaging in the unwanted behavior at subject. Other scholars such as Mogin and Zelenak criticized this proposal arguing correspondingly it will tilt the playing field against defendants and disregards the deterrence, rather than punishment, goal of punitive damages. This study empirically examines such claims.
The authors in this Article argue that if jurors choose to adjust their punitive damages awards for the impact of tax-deductibility, they need to know something about the defendants’ tax rate. Without this knowledge, jurors have virtually no basis to adjust their punitive damages awards even if they want to do so. Accordingly, the authors conduct an experiment in which they give jurors information about the defendant’s historical effective tax rates (“ETR”) to provide context for their decision and allow them to adjust their punitive damage awards. The authors predict that when jurors are faced with a defendant who has committed a socially undesired act, and not only is able to deduct punitive damages but also has managed to pay low taxes for several years (low ETR), they will gross-up punitive damage awards considerably more when a defendant’s ETR is low than when it is high.
The author’s experiment used 113 M-Turk participants that took on the role of jurors and were told that local residents (the plaintiffs) had sued Smith Industries (the defendant), a mining company, for spilling toxic chemicals into a nearby river causing widespread damage to water quality and substantial economic harm to residents and local businesses. Participants were further informed that the defendant had already been ordered to pay $2 million in compensatory damages. The jurors’ task was to determine whether punitive damages were warranted and determine the amount of such damages. Participants were assigned to one of four treatment conditions resulting from manipulating tax-deductibility of punitive damages at two levels (deductible vs. non-deductible). Participants in the deductible condition were told that the defendant’s effective after-tax punishment will be lower than the punitive damages assessed by the jury. Participants in the non-deductible condition were told that the effective after-tax punishment for the defendant will be the same as the damages assessed by the jury. The second manipulation the authors created was the defendant’s ETR (low vs. high). Jurors in the low-ETR condition were told that, over the past ten years, the defendant’s tax rate was 5% per year. Participants in the high-ETR condition were told that the defendant’s tax rate was 30% per year during the previous ten years. After providing their assessment of punitive damages, participants indicated the extent to which they were bothered by the fact that punitive damages were tax-deductible (or not tax-deductible) for Smith Industries. Participants answered a similar question about the extent to which they were bothered by Smith Industries’ ETR. The punitive damages (in dollars) assessed by jurors represented the primary dependent variable and the authors compare it across the four treatment conditions to test whether tax-deductibility and the defendant’s ETR jointly influenced jurors’ punitive damage awards.
The authors’ findings are fascinating. They discover that jurors’ punitive damages awards are higher when they are instructed that damages are tax-deductible compared to when they are instructed that damages are non-deductible. Moreover, jurors’ punitive damages awards are higher for defendants with a low ETR compared to defendants with a high ETR (although not statistically significant). In other words, this study proves that jurors adjust their punitive damages upwards to account for the impact of tax-deductibility considerably more when the defendant’s ETR is low than when the defendant’s ETR is high. This suggests that jurors do not view the question of tax-deductibility independent of the defendant’s ETR while assessing punitive damages. Jurors are considerably more outraged by tax-deductibility when the defendant’s ETR is low than when it is high and this outrage, in turn, drives their punitive damages awards upward.
This Article provides direct evidence to Polsky and Markel’s assertion that instructing jurors about the tax deductibility of punitive damages will affect the level of their award. It also affirms that concerns for under-punishment are not universal (as suggested later by Polsky and Markel) but occur mainly when the defendant’s ETR is low. These findings are economically counter-intuitive as a defendant with a higher historical ETR is likely to benefit more, not less, from tax deductibility of punitive damage awards. In this experiment jurors, on average, awarded damages of $6.54 million when the defendant had a lower ETR and $5.95 million when the defendant had a higher ETR. Accordingly, it seems that Jurors’ are using the defendant’s ETR for reasons other than adjusting their awards for tax-deductibility such as expressing their resentment over firms’ underpayment of tax. This last finding points to an unintended consequence associated with providing additional information to jurors such as tax-deductibility and defendant’s ETR.