A financial economic theory of punitive damages.

• Author: Rhee, Robert J.
• Position: II. Law and Economics Theory and Critique through Conclusion, with footnotes, p.52-87

2. LAW AND ECONOMICS THEORY AND CRITIQUE

   A. Standard Law and Economics Theory

   The standard law and economics theory of punitive damages advances deterrence as the normative goal. The law should optimally deter defendants by causing them to internalize the full cost of all wrongful conduct. (126) This basic idea originates from Judge Learned Hand's famous exposition of the negligence standard in United States v. Carroll Towing Co. (127) The economic theory of punitive damages is based on the idea of minimizing social cost through optimal deterrence. (128)

   The leading proponents of this theory are Mitchell Polinsky and Steven Shavell. They argue that punitive damages should only be awarded if a defendant has a chance to escape liability for the harm it causes. (129) When the tort system operates perfectly in imposing liability for all wrongful conduct, there is no reason for punitive damages since defendants fully internalize the cost of wrongful conduct. Only when a defendant can escape liability should punitive damages be awarded in an amount equal to the escaped liability as a means of preventing the underdeterrence of wrongful conduct. From this perspective, when a defendant has a chance of escaping liability for wrongful conduct, the total damages should be the harm caused multiplied by the reciprocal of the probability of being found liable. (130) This concept can be expressed in a mathematical formula: D = H x [1/P] where D is the total dame ages award, H is the magnitude of the harm, and P is the probability of being found liable. (131) The key point here is that a defendant should pay a liability amount in excess of the compensatory damages equal to the amount of the harm the defendant caused for which he would not otherwise pay. (132)

   The standard law and economics theory is seen at work in Mathias v. Accor Economy Lodging, Inc. (133) There, the plaintiffs suffered bedbug bites while staying at the defendant's motel. The jury awarded each plaintiff $5,000 in compensatory damages and $186,000 in punitive damages, a multiple of 37.2x. (134) Judge Posner rejected the defendant's argument that the punitive damages violated Gore and Campbell:

   The hotel's attempt to pass off the bedbugs as ticks, which some guests might ignorantly have thought less unhealthful, may have postponed the instituting of litigation to rectify the hotel's misconduct. The award of punitive damages in this case thus serves the additional purpose of limiting the defendant's ability to profit from its fraud by escaping detection and (private) prosecution. If a tortfeasor is "caught" only half the time he commits torts, then when he is caught he should be punished twice as heavily in order to make up for the times he gets away. (135)

   The court affirmed the award of punitive damages on the basis that the defendant had escaped liability in many other transactions with motel customers, which justified an upward deviation from Campbell. Thus, Mathias adopts the standard law and economics reasoning that escaped culpable conduct must be "recaptured" through the imposition of punitive damages such that the cost of a defendant's activity is fully internalized.

   B. Critique of the Standard Theory

   Despite the mathematical elegance of the standard law and economics theory, the model is incomplete from the standpoint of how punitive damages should work. Escaped liability is an important factor to consider from a deterrence perspective. It is well known that many legitimate claims go undiscovered or are not pursued for various reasons, (136) and thus defendants "internalize far less than the full cost of the losses they inflict." (137) But the standard law and economics model fails to account for other important variables in the tort and litigation systems.

   In the empirical world, litigation involves risk, and the parties are the risk bearers. Kip Viscusi provides the following critique of law and economics analysis:

   The standard economic theory of punitive damages and incentives casts the analysis within a highly stylized situation of certainty. Does the economic value of the definite harm resulting from an action exceed the benefits? The task then becomes setting a damage amount so that the company recognizes the full costs of its actions and exercises an appropriate degree of care in its activities. These formulations are overly simplistic, however, because they neglect potential risks that affect corporate actions and the linkage of the company's behavior to the societal effects .... To understand fully what legal sanctions may be appropriate when corporate decisions generate social harms, it is necessary to explicitly account for the tradeoffs inherent in corporate risk decisionmaking. (138)

   Like the Hand Formula, the standard law and economics model of punitive damages is built on the edifice of probability and expected value. But the model fails to account for the role of risk in the decisionmaking of parties.

   Probability and variance are entirely different concepts. (139) To illustrate, assume two coin flips: in coin flip A, one gets $20 if the coin is heads and $0 if it is tails; in coin flip B, one gets $40 if the coin is heads and one must pay $20 if it is tails. For both coin flips A and B, the expected value based on the probabilities is $10. However, it is obvious that the variances of the two coin flips are different. In coin flip A, the outcomes are [$20, $0] for a spread of $20, and in coin flip B, the outcomes are [$40, $20] for a spread of $60. The greater the spread of potential outcomes, the greater the risk. Here, coin flip B has more variance, and is thus more risky. A theory of punitive damages must account for the quality of riskiness of punitive damages.

   Another empirical fact is that the litigation system is not composed of homogenous parties. Heterogeneous parties act under a veil of uncertainty and react differently to both similar and dissimilar circumstances. The most important difference among parties is that defendants who are exposed to the largest punitive damage awards are typically corporations, whereas most plaintiffs are individuals. Corporate defendants and individual plaintiffs view and manage risks differently. The former are repeat players in the litigation system. The outcomes of individual lawsuits do not affect the portfolio of corporations for the most part. On the other hand, the outcome of the lawsuit for an individual plaintiff is significant. By virtue of different circumstances, the degrees of variance experienced by the different classes of plaintiffs and defendants are different. Yet the standard law and economics model of punitive damages does not account for this difference. (140)

   If most defendants are corporations, punitive damages and the threat of outlier verdicts must affect corporate finance and risk management, and these factors may affect litigation behavior. Punitive damages can threaten the solvency of firms, require firms to recapitalize lost capital, and otherwise affect the firm's operations just as any large liability or incurred business cost can. These financial effects must be modeled into an economic theory of punitive damages.

   Lastly, in the litigation system, most cases settle. (141) Dispute resolution scholars claim that settlement occurs "in the shadow of the law." (142) Settlement is the large unseen part of the tort system because most disputes are privately ordered. (143) Settlement is irrelevant to the efficiency aspiration of the law only if the aggregate outcomes in settlement reflect the probabilistic value of trial outcomes. (144) Settlements may distort the efficiency aspiration of tort law, thus undermining efficiency and imposing social cost. (145) Tort theory exists in the shadow of settlement, and "[i]ts efficiency aspiration can be achieved only within a system that settles most disputes." (146) Many scholars have recognized the issue--what is the effect of settlement practices on punitive damages, and vice versa? (147) Yet the standard law and economics model does not account for any potential difference between settlement and trial, and the potential effects of punitive damages on either settlement rates or amounts.

   In summary, variance is an important empirical fact of the litigation system. Even a normative model cannot disengage from reality and must account for how the litigation system really works and parties really behave. An economic theory of punitive damages should incorporate the realities of risk, risk management, and corporate finance into a model of how punitive damages actually work and how they should work.

   C. Toward a Complete Theory of Optimal Liability

   Optimal deterrence is achieved through nothing less than full cost internalization. (148) The standard law and economics model concerns escaped liability, but it does not consider other more important factors in the pricing of legal disputes in the settlement process. The threat of punitive damages plays an important role in regulating the value of tort actions, and it can produce inefficiencies through over- or underdeterrence.

   Punitive damages may impose too much legal liability or economic cost on defendants arising from the litigation. In addition to the cash value of legal liability, the threat and the uncertainty of punitive damages create financial costs that function very much like economic penalties for wrongful conduct. Firms can go insolvent. Their cost of equity can increase, resulting in reduced stock value. If a firm must be recapitalized after a large judgment, it could bear the large transaction costs of raising capital. Factoring in information about the adverse litigation outcome, the new capital may cost more. Collectively, these economic costs must be factored into the calculus of cost internalization and deterrence.

   With respect to the cost of legal liability, the method of calculating the amount of "cost" imposed on the tortfeasor must be explored. The...