A financial economic theory of punitive damages.
A. From Constitutional Reformation to Legal Uncertainty
Although punitive damages are well established in American tort law, (38) the Supreme Court has undertaken a major reformation of punitive damages jurisprudence over the past twenty years, which has resulted in significant constitutional restrictions on punitive damages awards. I provide a brief review of the case law to focus specifically on the multipliers in the cases and to provide a sense of the positions of individual Justices.
In Browning-Ferris Industries of Vermont, Inc. v. Kelco Disposal, Inc., the Court addressed the issue of whether the Eighth Amendment applies to punitive damages. (39) The case arose out of the defendant's attempt to drive the plaintiff out of a local disposal business through illegal price-fixing and tortious interference with contracts. (40) The jury awarded compensatory damages of $51,146 and punitive damages of $6 million. (41) The Court, per Justice Blackmun, held that the Eighth Amendment does not apply to civil actions when the government is not a party. (42)
Justice O'Connor, concurring in part and dissenting in part, argued that corporations should be given Eighth Amendment protection against the threat of "overbearing and oppressive monetary sanctions." (43) She introduced the germ of the idea of a quantitative standard. The multiple of punitive to compensatory damages was 117x, far exceeding the highest reported award of punitive damages in the state, and she implied that a multiple of 30x should be considered suspect. (44) She also suggested that the imposition of punitive damages could violate due process, a position that three other Justices shared. (45)
In Pacific Mutual Life Insurance Co. v. Haslip, the Court, per Justice Blackmun, considered whether the Fourteenth Amendment applies to punitive damages. (46) The defendant insurance company was held liable for punitive damages under the theory of respondeat superior when its employee-agent embezzled the plaintiff's premium money and allowed the policy to lapse. The plaintiff was awarded $1,040,000, of which $200,000 was apparently compensatory, including $4,000 in out-of-pocket expenses. (47) The Court approved the punitive damages award, a multiple of 4x. (48) It declined to "draw a mathematical bright line" and instead suggested that the appropriate constitutional calculus is one of "general concerns of reasonableness." (49)
Justice Scalia concurred to reject the idea of substantive due process review for punitive damages. (50) Justice Kennedy concurred to opine that a lack of uniformity in punitive damages cannot be equated to constitutional infirmity since variance in outcomes is the nature of civil litigation in general, and that the usual protections given by states must suffice until common law judges or legislatures initiate systemic changes. (51) Justice O'Connor was the sole dissenter. Describing punitive damages as a "powerful weapon," (52) she argued for a reevaluation of punitive damages, contending that there had been an explosion in the frequency and size of punitive awards and that juries were not constrained in their discretion. (53)
In TXO Production Corp. v. Alliance Resources Corp., the Court addressed whether a punitive damages award of $10 million on top of compensatory damages of $19,000, a multiple of 526x, violated due process. (54) The Court upheld the judgment but was split as to reasoning. Relying on Haslip, the plurality Court, per Justice Stevens, declined to draw mathematical bright lines. (55) The high multiple was not controlling in this case in light of the large amount of money at stake, the defendant's bad faith, a larger pattern of fraud and deceit, and the defendant's wealth). (56) The Court also noted that the "shock dissipates when one considers the potential loss" to the plaintiff of $1 million, a punitive damages multiple of only 10x. (57) Justices Scalia concurred to opine that the Court should "shut the door" on the idea of a substantive due process right to be free of excessive punitive damages. (58) He predicted the following: "The plurality's decision is valuable, then, in that the great majority of due process challenges to punitive damages awards can henceforth be disposed of simply with the observation that 'this is no worse than TXO.'" (59) This prophecy was ultimately proven wrong as the Court continued to delve into the jurisprudence of punitive damages.
In the next case, the landmark decision of BMW of North America, Inc. v. Gore, the Court for the first time invalidated a punitive damages award under the Due Process Clause. (60) The facts are well known. A national car dealership sold a car as new when it had been repaired; at trial, the actual damages were $4,000 and the jury awarded punitive damages of $4 million (later reduced on appeal to $2 million). (61) The issue was whether the 500x multiple comported with due process. The Court, per Justice Stevens, set forth a legal standard composing of three guideposts: (1) the degree of reprehensibility, (2) the disparity between the harm and the punitive damages, and (3) the difference between the punitive damages and the civil penalties imposed or authorized in other cases. (62) Applying these factors, the Court determined that the conduct was not so reprehensible, considering that the harm was "purely economic in nature" and involved an omission of fact that may have had a good-faith basis. (63) The Court considered the 4x multiple in Haslip and the 10x multiple to the potential harm in TXO and determined that "a breathtaking 500 to 1" ratio is constitutionally infirm. (64)
Gore was a close case. Justices Scalia and Thomas dissented on the now-familiar objection to substantive due process review and dismissed the three guideposts as "a road to nowhere." (65) Justices Ginsburg and Rehnquist also dissented but on the grounds that states have an interest in punishment and deterrence, that the procedure satisfied due process, and that the Court was unwisely venturing into the realm of state law. (66)
In State Farm Mutual Automobile Insurance Co. v. Campbell, the Court further strengthened the constraints on punitive damages. (67) In an action for insurance bad faith arising out of the insurer's failure to settle a third-party liability claim, the jury awarded $1 million in compensatory damages and $145 million in punitive damages. (68) The Court, per Justice Kennedy, applied Gore's three guideposts. (69) Additionally, the Court famously laid down a quantitative marker: "Our jurisprudence and the principles it has now established demonstrate, however, that, in practice, few awards exceeding a single-digit ratio between punitive and compensatory damages, to a significant degree, will satisfy due process." (70) Based largely on the fact that the multiple at issue was 145x, the Court struck down the award, (71)
In Philip Morris USA v. Williams, the plaintiff argued that the defendant cigarette manufacturer knowingly and falsely led him to believe that smoking was safe. (72) The jury awarded compensatory damages of $821,000 and punitive damages of $79.5 million, a multiple of 96.8x. Notably, Williams is the only case in the line from Browning-Ferris to Campbell that involved personal injuries. (73) The Court chose not to consider whether this award was grossly excessive. (74) Instead, it focused on whether punitive damages can be based in part on a jury's desire to punish the defendant for harming nonparty victims. This issue--punitive damages based in part on broader harm to others arising from similar activities--was also present in Gore and Campbell. (75) The Williams Court, per Justice Breyer, directly addressed the issue and held that a jury may not punish a defendant for harms caused to others but that it may consider such conduct in determining reprehensibility. (76)
The constitutional journey into the jurisprudence of punitive damages took a detour into federal maritime law with Exxon Shipping Co. v. Baker. (77) The case arose out of the 1989 oil spill in the Prince William Sound where Exxon's ship ran aground. On claims brought by commercial fishermen for lost profit, (78) the jury awarded $287 million in compensatory damages and punitive damages of $5 billion (later remitted to $2.5 billion). The issue was whether federal maritime law constrains punitive damages. The Court, per Justice Souter, was skeptical that verbal formulations could work to promote "systemic consistency." (79) Looking to the median multiple of punitive damages (cited as 0.65x based on empirical scholarly studies), the Court set a 1:1 ratio cap on punitive damages in maritime cases. (80)
Three prominent themes can be seen in the line of cases from Browning-Ferris to Baker. The first theme is obvious but worth mentioning. The Supreme Court increasingly lost confidence in state process and juries. (81) The earlier cases exhibit a reluctance to actively review lower court decisions. Justice O'Connor speculated that the Court's early reticence stemmed from not wanting to open the door to wholesale review of state tort law practices. (82) But that reticence was overcome by the time the Court decided Gore, a case in which the Court substituted its judgment on the degree of egregiousness--ordinarily the fact finder's prerogative--to strike down the award. (83) Gore, Campbell, and Baker exhibit an activist approach of reviewing and superseding state judgments on matters of evidentiary weight and implementation of state policy. (84)
The second theme is the most notable--a move away from a qualitative, multifactor standard to a rule-based numerical cap. (85) In Haslip, the Court declined to "draw a mathematical bright line." (86) This principle was later repeated in TXO, Gore, and Campbell, (87) but the professed rejection of hard-and-fast numeric caps in Campbell seems inconsistent when in the same paragraph the Court wrote "[s]ingle-digit multipliers are more likely to comport with due process." (88) Justice Kennedy, who authored Campbell, seems to have been the bellwether for this shift in the Court's thinking as he was the only Justice to have reversed his earlier position on numerical caps. (89) The analytic framework of a numerical cap is seen in Baker, which rejected a qualitative standard in favor of a bright-line lx multiple limit. (90) Although Baker is a maritime case, the influence of its support for numerical caps and its policy rationale extend beyond its subject-matter jurisdiction. (91)
The third theme is a link among punitive damages, corporate interests, and economic policy. Through all of the punitive damages cases, the Court advanced a pro-business economic agenda by protecting corporations from excessive liability. All of the cases from Browning-Ferris to Baker involved large national corporations. (92) That corporations were the principal defendants raises two economic issues. The Court feared that punitive damages could give juries the power to improperly redistribute the wealth stored in corporations. (93) And, in a line of thinking developed in the later cases, the Court feared that the principle of federalism could be undermined when state tort law conflicts with interstate commerce and the national economy. (94)
The line of cases from Browning-Ferris to Baker spanned nineteen years and involved fifteen Justices. Several Justices have been influential beyond their single vote. Justice O'Connor has been the most vocal critic of punitive damages and advocate of strict constraints, and her forceful dissents in Browning-Ferris, Haslip, and TXO may have influenced her colleagues and the Court's later thinking. Justice Stevens authored the plurality opinion in TXO and the majority opinion in Gore. Justices Kennedy, Breyer, and Souter authored respectively the majority opinions in Campbell, Williams, and Baker. Justices Kennedy and Breyer still sit on the Court today, and both have voted in favor of constitutional constraints. Chief Justice Roberts voted with the majority in Williams and Baker, and Justice Alito did the same in Williams. Justices Scalia and Thomas have opposed the idea of substantive due process limitations on punitive damages. Justice Ginsburg dissented in Gore, Campbell, Williams, and Baker, and she opposes the Court's application, if not the theory, of constitutional constraint.
We do not know whether the law has been shaped into its final form by Gore, Campbell, Williams, and Baker, or whether it will continue to evolve into greater restrictions or even reverse course. Five members of the current Court--Justices Kennedy, Breyer, Scalia, Ginsburg, and Thomas--have staked out well-articulated positions on the issue. Scalia, Thomas, and Ginsburg disfavor constitutional scrutiny, perhaps for different reasons. Four Justices are relative newcomers to the debate: the Chief Justice and Justices Alito, Sotomayor, and Kagan. Chief Justice Roberts and Justice Alito likely favor constraints on punitive damages per their political and pro-business inclinations, (95) an educated guess supported by their respective votes in Williams and Baker. If we assume that Justices Scalia and Thomas continue to believe that the application of substantive due process to punitive damages is wrong and that they will vote based on their core beliefs despite whatever pro-business leanings they may have, the tally is 4 to 3 in favor to continuing the current trajectory of the law. Justices Sotomayor and Kagan, who joined the Court after Baker, are the unknown factors, and they will be the swing voters in future cases.
In addition to changing judicial composition, a theoretical deficit creates uncertainty about the future direction of the law. The Court has not articulated a theory of punitive damages. There is little justification for why it is "unfair" or "arbitrary" or "excessive" to impose punitive damages beyond the assertion that a single-digit multiplier is the limit. Absent a theory with analytic rigor, these characterizations are simply labels attached to a vote. The Court recognizes the common formulation that punitive damages serve the goals of deterrence and retribution. (96) But it has not developed a theory of how retribution and deterrence are served by judicially capping punitive damages. Several Justices have been dismissive of economic analysis, (97) which is odd since that body of scholarship has developed the most sophisticated analysis of deterrence theory.
The current Court composition and the theoretical deficit in the judicial literature have created uncertainty as to the future direction of punitive damages jurisprudence. Justice Kennedy is frequently considered the "swing vote," but in this area Justices Sotomayor and Kagan will play a crucial role in the development of the law. The Court "is in the midst of a major reexamination of punitive damages--the final chapter of which has not yet been written." (98) At the same time, there is a crisis in the theory of punitive damages. (99)
B. Bursting the Myth and Identifying the Central Problem
The Supreme Court case law on punitive damages is commonly seen as being driven by a concern regarding "grossly excessive" punitive damages in the tort system. However, the perception that punitive damages have run amok is empirically false and has been disavowed. In Baker, the Court examined for the first time the well-developed empirical scholarship on punitive damages and concluded that there is no systemic problem with respect to aggregate liability amount. The Court's opinion not only disavowed a major assertion of the problem, but also much of the rhetoric that accompanied it. Two sweeping paragraphs of the opinion are important:
American punitive damages have been the target of audible criticism in recent decades, but the most recent studies tend to undercut much of it. A survey of the literature reveals that discretion to award punitive damages has not mass-produced runaway awards, and although some studies show the dollar amounts of punitive-damages awards growing over time, even in real terms, by most accounts the median ratio of punitive to compensatory awards has remained less than 1:1. Nor do the data substantiate a marked increase in the percentage of cases with punitive awards over the past several decades. The figures thus show an overall restraint and suggest that in many instances a high ratio of punitive to compensatory damages is substantially greater than necessary to punish or deter. The real problem, it seems, is the stark unpredictabiliy of punitive awards. Courts of law are concerned with fairness as consistency, and evidence that the median ratio of punitive to compensatory awards falls within a reasonable zone, or that punitive awards are infrequent, fails to tell us whether the spread between high and low individual awards is acceptable. The available data suggest it is not. A recent comprehensive study of punitive damages awarded by juries in state civil trials found a median ratio of punitive to compensatory awards of just 0.62:1, but a mean ratio of 2.90:1 and a standard deviation of 13.81. Even to those of us unsophisticated in statistics, the thrust of these figures is clear: the spread is great, and the outlier cases subject defendants to punitive damages that dwarf the corresponding compensatories. (100)
In Baker, the Court dealt "a decisive blow" to the claim that punitive damages are out of control. (101) Upon reviewing the empirical scholarship for the first time, the Court silenced much of the noise in the debate over punitive damages. (102)
The leading empirical scholars in this field, Thomas Eisenberg and his coauthors chief among them, have shown that: (1) punitive damages are infrequently awarded, (103) (2) the amount of punitive damages is highly correlated to the amount of compensatory damages, (104) (3) the median ratio of punitive to compensatory damages is less than 1.0, (105) and (4) punitive damages are most likely to be awarded for intentional torts and economic wrongs. (106) The empirical evidence shows that punitive damages are infrequent, stable, and predictable, (107) and that the myth of out-of-control punitive damages is "groundless." (108)
If the aggregate amount of punitive damages is not the problem, then what is? The Baker Court identified the real problem as the variance of awards, which translates into the "stark unpredictability of punitive awards." (109) Finding no empirical or theoretical support for the proposition that variance is a good thing, the Court disapproved of variability between awards for the same or similar conduct and the resulting unpredictability. (110)
C. The Problem of Low Frequency, High Severity Awards
The main thrust of the criticism of punitive damages relates to the predictability of punitive damages at the individual level. (111) Unpredictability suggests arbitrariness and incoherence in the litigation system. Defendants cannot gauge ex ante the range of liability arising from their conduct. Individual judgments, the argument goes, are incoherent due to a lack of case-by-case predictability. Cass Sunstein, Daniel Kahneman, and David Schkade have prominently advanced this argument. (112) They argue that while jurors consistently share moral judgments about a defendant's conduct, they have difficulty in mapping such judgments onto an unbounded scale of dollars. (113) The problem of monetizing shared moral sentiments can lead to erratic, unpredictable, and arbitrary awards. (114)
The problem of uncertainty and unpredictability is a legitimate policy concern. But the statement of the problem must be refined further. First of all, empirical evidence undercuts much of the unpredictability claim. Individual cases are predictable based on statistical inference. If one were to predict in any given case that, first, punitive damages would not be awarded, and second, if punitive damages were awarded at all, the amount awarded would be correlated to the compensatory damages, that prediction would be fairly reasonable as a matter of statistics and probability. Punitive damages are predictable, but obviously predictions are not perfect. Deviations from expectation will occur.
No serious person argues that uncertainty can be eliminated entirely from the litigation system. (115) We live in an uncertain world, and some of the major achievements in economics have dealt with the study of risk and uncertainty. (116) Uncertainty and randomness are inherent in the legal system. Nor should there be any entitlement in the civil litigation system to certainty of outcomes. Certainty in litigation is achieved only with fiat rules like the quantitative caps seen in Campbell and Baker. (117) Otherwise, the legal process is inherently risky as to outcomes on liability and damages. A meritorious case is always subject to uncertainty of outcome. (118) With perfect bilateral omniscience, all cases would settle under the terms dictated by the predicted winner; with perfect single-party omniscience, virtually all cases would go to trial because the omniscient party would not settle for anything less than the winning outcome net of transaction costs. Most cases settle because uncertainty is the governing condition. Uncertainty is a fact of life, and variance of outcomes is a routine feature of the litigation system. (119)
The real policy issues are what is the role of variance with respect to punitive damages, and what degree of variability is acceptable or good? Here, empirical scholarship does not provide the answer. As Eisenberg comments, "How much variation to allow is a question of judgment and policy." (120) A normative theory of variance in punitive damages is needed.
Outlier judgments are unpredictable. (121) Outliers have two attributes. First, the awards are low frequency events. Low frequency is defined here as small probabilities of occurrence based on the class of all cases awarding punitive damages. Since punitive damages are awarded infrequently, (122) a low frequency award is an outlier of the small class of cases (i.e., they are an infrequent occurrence within a set of infrequent events).
Second, outlier awards are low frequency events because they are high severity losses. High severity is defined here as the high dollar value of an award of punitive damages relative to the defendant firm's wealth. Similar to the concept of the marginal utility of money, the same amount of money has different effects depending on a corporate defendant's wealth. High dollar values in absolute terms are not necessarily a high severity award. The $2 million award in Gore is not a large number for a national firm like BMW, but it could be a devastating sum for a smaller firm. Another example: if compensatory damages are $20 billion, say hypothetically for harms related to British Petroleum's oil spill in the Gulf of Mexico, would a $20 billion award of punitive damages be so shocking? Likewise, a high multiple does not necessarily lead to a high severity award. Suppose the compensatory damage award is minimal, say $100, and punitive damages are awarded at a 90x multiple ($9,000). Would such a case have merited the attention of the Supreme Court? Thus, neither a high multiple nor a large absolute value of punitive damages nor both is necessarily problematic.
The rare cases yield a low frequency, high severity award, and thus they are unpredictable. A high severity award exceeds the boundary of ordinary expectation because it imposes financial distress on firms. (123) Financial distress, such as insolvency, substantially disrupts a firm's ordinary operations. It has significant economic consequences and imposes significant economic costs, including the liability amount and the costs associated with disruptions to the firm's operations, transaction costs associated with recapitalization, and insolvency. Moreover, actual awards are not the only economic detriment. The risk of being subject to a low frequency, high severity award has implications on settlement. Implied, but not stated, in Baker is that outlier judgments may coerce settlements, force overpayment, and thus overdeter defendants. (124)
A summary of the essential problem is helpful. The supposed problem of unpredictability is too general a proposition. Unpredictability is tied to the problem of variance. While most awards of punitive damages are predictable within a range of expectation, a small class of awards is highly unpredictable and economically harmful because the awards exceed the range of ordinary expectation as to frequency and severity. Problematic awards occur infrequently. Infrequent events are harmful because they pose severe economic consequences. When an award meets these conditions, it is defined as a low frequency, high severity award. Such extraordinary awards can be like a lightning strike, infrequent as to occurrence and severe as to outcome. (125) The risk associated with these cases affects the whole of the tort and dispute resolution systems.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||I. The Central Problem of Punitive Damages through C. The Problem of Low Frequency, High Severity Awards, p.40-51|
|Author:||Rhee, Robert J.|
|Publication:||Michigan Law Review|
|Date:||Oct 1, 2012|
|Previous Article:||A financial economic theory of punitive damages.|
|Next Article:||A financial economic theory of punitive damages.|