Many believe that punitive damage awards have spiraled out of control. In 2002, a California jury awarded $28 billion in punitive damages to a 64-year-old woman with lung cancer. (1) In 2000, a Florida jury awarded $145 billion in punitive damages to a class of Florida smokers. (2) These are not

isolated decisions; they represent a pattern of extraordinarily high punitive damage awards handed down by juries. (3) States have responded to these excessive awards in three ways. Some have barred punitive damages altogether, (4) others have adopted a cap on such awards, (5) and still others have implemented split-recovery statutes. (6) Of these three responses, the split-recovery system is functionally and constitutionally unique. (7) Often coupled with more searching judicial review of multiple punitive damage awards, (8) this framework shifts a portion of the punitive damage award to society.

Last Term, in Philip Morris USA v. Williams, (9) the Supreme Court held that the Due Process Clause of the Fourteenth Amendment prohibits state juries from punishing a defendant for harm caused to in-state nonparties (potential plaintiffs not before the court), even as the Court reaffirmed a punitive damage framework requiring those same juries to take these harms into account when determining the conduct's reprehensibility. (10) By firmly closing the door to recovery for harms to nonparties--a door left ajar in State Farm Mutual Automobile Insurance Company v. Campell (11)--the Court finally articulated a comprehensive and coherent approach to third-party harms in punitive damage cases. In doing so, however, the Court called into question the primary justification for the split-recovery, multiple punitive-damage review system that states began implementing in the 1980s in response to inflated awards.

In 1997, Mayola Williams filed suit against Philip Morris in Oregon state court. (12) She alleged negligence and deceit against Philip Morris in the death of her husband, Jesse Williams, a heavy cigarette smoker who died of lung cancer at the age of sixty-seven. (13) At trial the plaintiff's attorney asked the jury to consider how many other similarly harmed persons there had been in the State of Oregon over the past forty years. (14) Finding that Mr. Williams's death had indeed been caused by smoking; that he smoked because he believed it was safe to do so; and that Philip Morris "knowingly and falsely" encouraged this belief, the jury returned a compensatory damage award of $821,000 and a punitive damage award of $79.5 million in favor of the plaintiff (a ratio of roughly ninety-seven to one). (15)

After exhausting its state appellate remedies, (16) Philip Morris, claiming that the punitive damage award in favor of Mrs. Williams was constitutionally excessive and procedurally unsound, petitioned the Supreme Court for review. (17) The Court remanded the case for reconsideration in light of State Farm, (18) a then-recent and significant development in the Supreme Court's punitive damages doctrine. (19) The Oregon Supreme Court reviewed and affirmed the punitive damages award, despite the Court's language in State Farm. (20) Specifically, the court found that State Farm, although prohibiting punishment for dissimilar harms, as well as harms to out-of-state parties, did not prohibit a jury from punishing a defendant for similar harms to in-state parties not before the court, and that the extreme reprehensibility and near criminality of Philip Morris's actions justified the high ratio of punitive to compensatory damages. (21) On petition by Philip Morris, the United States Supreme Court granted certiorari.

The Supreme Court vacated and remanded. Writing for the Court, Justice Breyer (22) began by reaffirming both the state's legitimate interest in using punitive damages to punish and deter unlawful conduct and the need to balance these interests against the potentially unfair or arbitrary nature of a discretionary punitive damages award. (23) To balance these competing concerns, the Due Process Clause mandates both procedural requirements and excessiveness review. (24) Deferring the question of excessiveness review, the Court focused instead on the Constitution's procedural requirements, finding that the Due Process Clause "forbids a State to use a punitive damages award to punish a defendant for injury that it inflicts upon nonparties--strangers to the litigation." (25) Not only would such an award eliminate the defendant's ability to "present every available defense," but it would also introduce a "standardless dimension to the punitive damages equation," leaving the jury to speculate about both the number of nonparties and the degree of harm to those parties. (26)

Cognizant of the conflict between this position and the Court's established reprehensibility calculus, which requires a jury to consider evidence of harm to nonparties, the Court suggested that it would not require perfection. (27) The Due Process Clause, the Court held, simply requires that state courts "provide assurance that the jury will ask the right question[s]"--questions that track the thin line between reprehensibility and punishment--and "avoid procedure that unnecessarily deprives juries of proper legal guidance." (28) Simply put, courts cannot sanction procedures that "create an unreasonable and unnecessary risk of ... confusion" between reprehensibility analysis and punishment. (29) Because the Constitution simply requires "some form of protection in appropriate cases," states "have some flexibility to determine what kind of procedures they will implement." (30)

Justice Stevens dissented, adopting, in large part, the reasoning of the Supreme Court of Oregon. (31) He argued that States do have an interest in using punitive damages to punish wrongdoers "for harming persons who are not before the court," so long as the award remains non-compensatory. (32) Punitive damages, unlike compensatory damages, are designed to account for "the public harm [of] the defendant's conduct." (33) Furthermore, Justice Stevens contended that the majority's distinction between direct and indirect use of third-party harm (34) was illusory. (35)

Justice Thomas dissented. Although he joined Justice Ginsburg's dissenting opinion, Justice Thomas wrote separately to reiterate his position in State Farm that "the Constitution does not constrain the size of punitive damages awards." (36)

Justice Ginsburg also dissented. (37) She began by stating that the actions taken by the Oregon court were fully consistent with the approach articulated by the majority; that is, nothing suggested that the trial court failed to provide the jury with proper legal guidance. (38) Justice Ginsburg continued by arguing that the issue presented to the Court was the propriety of the defendant's jury instruction. Finding the instruction confusing, and thus the Oregon court's rejection of it appropriate, she reasoned that the Court should not have gone beyond the issue presented. (39)

The outcome in Philip Morris was widely predicted. Although allowing the jury to punish defendants for similar in-state harms would not have been inconsistent with earlier decisions, it would have complicated an already difficult remedial framework. (40) The Court's decision, however, has larger consequences for the social-interest theory of punitive damages, and is therefore likely to have ramifications for statutes that justify split recovery by focusing on the dual harm to society and the individual.

Academic commentators have already suggested two competing paradigms by which to conceptualize punitive damage awards. (41) The first, traditional, approach is the individual-interest paradigm. (42) Under this theory, punitive awards are by nature retributive and designed to punish egregious or offensive conduct. (43) The second approach, referenced above, is the social-interest paradigm. Under this theory, punitive damage awards are designed to advance social agendas, which are likely to be focused on deterrence. (44) They may, however, be motivated by other justifications, including retribution for antisocial behavior (45) and, theoretically, social compensation. (46)

Over the past 15 years, starting with BMW of North America, Inc. v. Gore (47) and proceeding through State Farm, the Court has consistently struck down attempts to divorce punitive damage awards from the harm before the court. In Gore, the Court made clear that punitive damages could not be designed to account for conduct that may be lawful in other jurisdictions. (48) In State Farm, the Court went further, finding that dissimilar harms, lawful or not, and all out-of-state harms were off-limits for juries, and reminding courts that they "must ensure that the measure of punishment is both reasonable and proportionate to the amount of harm to the plaintiff and [the] damages recovered." (49) In Philip Morris, the Court has finally come full circle, reaffirming its previous position, while clarifying that a punitive award cannot be used to account for similar harm to in-state nonparties. (50)

By prohibiting consideration of third-party harm, the Court has, under the guise of the Due Process Clause, mandated an individual-oriented punitive framework. This renders two goals of the social-interest paradigm constitutionally illegitimate: states may no longer use punitive damages to encourage socially optimal deterrence as traditionally conceived--where the focus is on creating full internalization of harm in a world of partial enforcement--or to advance a social compensation goal. (51)

This effect has passed largely unnoticed. (52) Although modern split-recovery statutes are not necessarily tied to the social-interest paradigm, (53) they have, in practice, been wedded to the social-interest paradigm as state courts "search[ed] [for] a justifying theory." (54) Moreover, the assertion that punitive damages address social rather than individual interests has intuitive appeal in the split-recovery context. With a clear interest in the judgment, society has a better claim to the "windfall" than the individuals to whom the judgment was awarded. Whereas most courts have been guarded in their treatment of exactly how society's interests are vindicated by punitive damages awards, (55) the Court's decision in Philip Morris, by invalidating much of the social-interest paradigm, will nevertheless force a general reassessment of these statutes.

This does not mean that states with split-recovery statutes need to abandon them altogether. As suggested above, it remains clear that states can constitutionally impose limits upon punitive damage recovery. (56) Although Philip Morris outlawed social-interest theories based around social-harm, it does not affect social-interest theories based around individual-harm. What split-recovery states require is a compelling theory of why the state, rather than the individual, is the proper place for the damages to be apportioned, even when the harm the punitive damage award is designed to address is focused upon the individual before the court.

States could articulate a revenue justification for their split recovery frameworks--that the state's interest in realizing revenue justifies imposing what amounts to a large tax on non-compensatory awards. Alternatively, splitting recovery could be used as a method for discouraging certain types of disfavored lawsuits. (57) Both would be consistent with Philip Morris, but neither is especially persuasive given how small and unpredictable a source of revenue punitive damage awards would be and given other, more effective methods for filtering out non-meritorious claims. Ultimately, these statutes can and must be more convincingly reconceptualized along a social-retribution paradigm that conceives of punitive awards as a way to vindicate broader outrage against socially impermissible conduct.

In the meantime, the Court's stronger procedural protections should remedy the most common concerns associated with "blockbuster" punitive damage awards--redundant awards and major windfalls. Without recourse to harm beyond the instant plaintiff, a recurrence of the sizable awards noted above seems unlikely, whether or not juries can cabin their tendency toward excess. This should gut the case for complicated frameworks of state intervention, and may result in a relaxation of punitive damage controls more generally. States may simply find that, under an individual-harm framework caps, rather than split-recovery statutes, are better at resolving the concerns associated with punitive awards that remain after Philip Morris. (58)

Stripped of their most popular foundation, the justification for split-recovery statutes is more uncertain after Philip Morris than at any time since their emergence. A major reassessment of these statutes is likely, and may ultimately result in either the abandonment or a reconceptualization of punitive damages as an expression of social outrage.

(1.) See Bullock v. Philip Morris, Inc., No. BC249171, 2002 WL 31833905 (Cal. App. Dep't Super. Ct. Dec. 18, 2002) (remitting the award to $28 million). On appeal, this case was one of the first to be affected by the United States Supreme Court's decision in Philip Morris. See Bullock v. Philip Morris USA, Inc., No. B164398, 2008 WL 240989 (Cal. Ct. App. Jan. 30, 2008) (vacating judgment as to damages and ordering a retrial in a manner consistent with Philip Morris).

(2.) See Engle v. RJ Reynolds Tobacco, No. 94-08273 CA-22, 2000 WL 33534572 (Fla. Cir. Ct. Nov. 6, 2000), rev'd Liggett Group Inc. v. Engle, 853 So. 2d 434 (Fla. Dist. Ct. App. 2003).

(3.) See Joni Hersch & W. Kip Viscusi, Punitive Damages: How Judges and Juries Perform, 33 J. LEGAL STUD. 1, 6-8 (2004) (identifying more than sixty-three $100 million-plus punitive damage awards between 1985 and 2003, twelve of which were over $1 billion).

(4.) See, e.g., N.H. REV. STAT. ANN. [section] 507:16 (1986). Nebraska is unique in that its constitution bars punitive damage awards. See Distinctive Printing and Packaging Co. v. Cox, 443 N.W.2d 566, 574 (Neb. 1989) (finding punitive damages barred by NEB. CONST. art. VIL [section] 5).

(5.) See, e.g., ALA. CODE [section] 6-11-21 (2005); IND. CODE ANN. [section] 34-51-3-4 (West 1999). Some set the cap at a specific number, others use a ratio to compensatory damages, and still others use a combination of both.

(6.) Nine states currently provide for split-recovery: ALASKA STAT. [section] 09.17.020(j) (2006) (50% to the state); GA. CODE ANN. [section] 51-12-5.1(e)(2) (1998) (75% to the state); 735 ILL. COMP. STAT. 5/2-1207 (2006) (discretionary); IND. CODE ANN. [section] 34-51-3-6 (West 1999) (75% to the state); IOWA CODE ANN. [section] 668A.1(2)(b) (West 1998) (75% to the state); MO. ANN. STAT. [section] 537.675(2) (West 2000) (50% to the state); OR. REV. STAT. [section] 31.735 (2005) (60% to the state); UTAH CODE ANN. [section] 78-18-1(3) (2002) (50% to the state); Dardinger v. Anthem Blue Cross & Blue Shield, 781 N.E.2d 121 (Ohio 2002) (recognizing Ohio courts' common law authority to apportion the award as appropriate). Perhaps in response to Dardinger, other states have expressly prohibited splitting punitive damage awards. See, e.g., ALA. CODE [section] 6-11-21(1) (2005).

(7.) See Victor E. Schwartz et al., I'll Take That: Legal and Public Policy Problems Raised By Statutes That Require Punitive Damages Awards To Be Shared With The State, 68 MO. L. REV. 525, 548-56 (2003) (noting the special due process problems posed by a split-recovery system).

(8.) See, e.g., FLA. STAT. ANN. [section] 768.73(2)(a) (West 2005) ("[P]unitive damages may not be awarded against a defendant in a civil action if that defendant establishes, before trial, that punitive damages have previously been awarded against that defendant ... from the same act or single course of conduct...."); GA. CODE ANN. [section] 51-12-5.1(e)(1) (1998) (limiting recovery to one punitive damages award for product liability actions).

(9.) 127 S. Ct. 1057 (2007).

(10.) See id. at 1064. The Court sidestepped what was certainly the more anticipated of the two questions for which the Court granted certiorari--whether the constitutional ceiling on the punitive damages ratio, originally suggested in Pacific Mutual Life Insurance Company v. Haslip, 499 U.S. 1, 23-24 (1991), could be overcome by a finding that conduct was "highly reprehensible and analogous to a crime." Petition for Writ of Certiorari, Philip Morris USA v. Williams, 127 S. Ct. 1057 (2007) (No. 05-1256).

(11.) State Farm Mut. Auto. Ins. Co. v. Campbell, 538 U.S. 408, 422-23 (2003). State Farm involved claims of bad faith failure to settle within policy limits, fraud, and intentional infliction of emotional distress.

(12.) Ms. Williams filed on behalf of her husband's estate.

(13.) See Philip Morris, 127 S. Ct. at 1060.

(14.) See id. at 1061.

(15.) Id.

(16.) The Oregon State Supreme Court denied review after the court of appeals affirmed.

(17.) These were the same claims ultimately before the Court in 2006. See Philip Morris, 127 S. Ct. at 1062.

(18.) State Farm Mut. Auto. Ins. Co. v. Campbell, 538 U.S. 408, 422-25 (2003) (finding that "[a] defendant's dissimilar acts ... may not serve as the basis for punitive damages," that "each State alone can determine what measure of punishment ... to impose on [one acting] within its jurisdiction," and that "in practice, few awards exceeding a single-digit ratio between punitive and compensatory damages, to a significant degree, will satisfy due process").

(19.) See Philip Morris USA Inc. v. Williams, 540 U.S. 801 (2003).

(20.) See Williams v. Philip Morris Inc., 127 P.3d 1165 (Or. 2006).

(21.) See id. at 1176, 1181.

(22.) Justice Breyer was joined by Chief Justice Roberts and Justices Kennedy, Souter, and Alito.

(23.) See Philip Morris, 127 S. Ct. at 1062.

(24.) See id. To date, the Court has been much clearer on excessiveness review, which takes the form of a three-part test. See BMW of N. Am., Inc. v. Gore, 517 U.S. 559, 574-85 (1996). Procedural requirements have come piecemeal. See, e.g., Honda Motor Co. v. Oberg, 512 U.S. 415 (1994) (requiring judicial review of the size of punitive awards); Cooper Indus. v. Leatherman Tool Group, Inc., 532 U.S. 424 (2001) (requiring de novo review of punitive awards).

(25.) Philip Morris, 127 S. Ct. at 1063.

(26.) Id.

(27.) See id at 1065. As has been widely recognized, the Court's decision puts jurors In an almost impossible situation. North Dakota's Pattern Jury Instruction Commission has attempted to remedy this problem with the following jury instruction, intended to be used in cases where there has been evidence or argument concerning the defendant's harmful conduct toward nonparties:

   In considering an award of exemplary or punitive damages, you may,
   in determining the reprehensibility of the Defendant's conduct,
   consider the harm the Defendant's conduct has caused to others. You
   may not, however, punish the Defendant for harm caused to others
   whose cases are not before you. You may punish the Defendant only
   for harm done to the Plaintiff.

State Bar of North Dakota, Pattern Jury Instruction C-72.07, available at http://www.sband.org/Pattern_Jury_Instructions/viewarticle.asp?ID=645&parent_ category=civil.

(28.) Philip Morris, 127 S. Ct. at 1064. Perhaps the best analog to this split is the one the Court itself suggests. In Witte v. United States, 515 U.S. 389 (1995), the Court held that recidivism statutes do not impose an "additional penalty for the earlier crimes but instead ... a stiffened penalty for the latest crime, which is considered to be an aggravated offense because a repetitive one." Id. at 400 (internal quotation marks omitted). Punitive awards should operate in the same way, with evidence of similar harm to nonparties justifying a higher award only by making the particular harm before the court more offensive.

(29.) Philip Morris, 127 S. Ct. at 1065.

(30.) Id. at 1065. Although the Court did not advance a particular method for meeting this new requirement, it did look favorably upon the jury instruction requested by Philip Morris. Id. at 1064. That instruction went, in part, as follows:

   The size of any punishment should bear a reasonable relationship to
   the harm caused to Jesse Williams by the defendant's punishable
   misconduct. Although you may consider the extent of harm suffered
   by others in determining what that reasonable relationship is, you
   are not to punish the defendant for the impact of its alleged
   misconduct on other persons, who may bring lawsuits of their own in
   which other juries can resolve their claims and award punitive
   damages for those harms, as such other juries see fit.

Williams v. Philip Morris Inc., 127 P.3d 1165, 1175 (Or. 2006).

(31.) See Philip Morris, 127 S. Ct. at 1065-67 (Stevens, J., dissenting).

(32.) Id. at 1066 (internal quotation marks omitted). According to Justice Stevens, awarding "compensatory damages to remedy ... third-party harm might well [deny] due process." Id.

(33.) Id.

(34.) Id. at 1064. The distinction was between punishment for harms to third parties and punishment in light of harms to third parties, the reprehensibility of which increases the punishment for the instant injury accordingly.

(35.) Id. at 1067.

(36.) Id. (Thomas, J., dissenting) (quoting State Farm Mut. Auto. Ins. Co. v. Campbell, 538 U.S. 408, 429-30 (2003) (Thomas, J., dissenting) (quotation marks omitted)).

(37.) Justice Ginsburg was joined by Justices Scalia and Thomas.

(38.) See Philip Morris, 127 S. Ct. at 1068 (Ginsburg, J., dissenting).

(39.) Id. at 1069.

(40.) These complications would have been exacerbated by forcing juries to make similarity determinations about hypothetical claims.

(41.) For a general discussion, see Catherine M. Sharkey, Punitive Damages as Societal Damages, 113 YALE L.J. 347, 356-70 (2003). Both paradigms are consistent with the punitive- and deterrence-based rationale articulated by the Court in Cooper Industries v. Leatherman Tool Group, Inc., 532 U.S. 424, 432 (2001).

(42.) See Sharkey, supra note 41, at 359 (noting that "individually oriented, retributive punishment" remains the prevailing justification for punitive damages). Historically this was the primary, if not exclusive, use of punitive damages. See Thomas B. Colby, Beyond the Multiple Punishment Problem: Punitive Damages as Punishment for Individual, Private Wrongs, 87 MINN. L. REV. 583, 628-29 (2003).

(43.) See Colby, supra note 42, at 628 ("[P]unitive damages ... were consciously limited to the amount necessary to punish the defendant for the wrong done ... to the individual plaintiff only.").

(44.) The social deterrence approach is the most intuitively appealing justification for a split-recovery statute. More generally, it is also the most popular approach to punitive damages among law and economics theorists. See A. Mitchell Polinsky & Steven Shavell, Punitive Damages: An Economic Analysis, 111 HARV. L. REV. 869, 874 (1998) (stating that the primary use of punitive damages should be to facilitate optimal deterrence where there is less than a 100% detection rate).

(45.) See Michael Finch, Giving Full Faith and Credit to Punitive Damages Awards: Will Florida Rule the Nation?, 86 MINN. L. REV. 497 (2002) (noting that split-recovery statutes describe the purpose of punitive awards as vindicating public wrongs).

(46.) See Sharkey, supra note 41, at 389 (suggesting that punitive damages could account for harms that "reach far beyond the individual plaintiff before the court").

(47.) 517 U.S. 559 (1996).

(48.) See id. at 572. In Gore, the plaintiff initiated suit after discovering that his car dealer had failed to disclose that his car had been damaged and repainted before his purchase. As noted above, Gore is best known for establishing the Court's current approach to excessiveness review.

(49.) State Farm Mut. Auto. Ins. Co. v. Campbell, 538 U.S. 408, 426 (2003).

(50.) See Philip Morris USA v. Williams, 127 S. Ct. 1057, 1063 (2007).

(51.) In other words, award-multiplication approaches of the type championed by Professors Polinsky and Shavell, Polinsky & Shavell, supra note 44, are unconstitutional, as are the absent- or quasi-plaintiff and diffuse-harm recovery theories suggested by Professor Sharkey, Sharkey, supra note 41, at 392-401.

(52.) The social compensation goal was a relatively novel position on punitive damages and the social-deterrence goal had been handicapped by State Farm. See 538 U.S. at 427 (stating that the probability of detection (on a national level) has "little to do with the actual harm sustained" and so no bearing on the reasonableness of the size of the award).

(53.) As described below, arguments for splitting the award regardless of its origin do exist.

(54.) Sharkey, supra note 41, at 372, 375; see, e.g., DeMendoza v. Huffman, 51 P.3d 1232, 1238 (Or. 2002) (finding that punitive damages in Oregon are designed simply to "vindicat[e] society's interests," not those of an individual plaintiff); Dardinger v. Anthem Blue Cross & Blue Shield, 781 N.E.2d 121, 145 (Ohio 2002) (finding that "[t]he plaintiff remains a party, but the de facto party is our society").

(55.) Perhaps this is because those state courts that have been the most outspoken in adopting a particular approach--Utah and Oregon--have been consistently overturned by the Supreme Court.

(56.) Courts have allowed states nearly unlimited flexibility in adjusting the size and availability of punitive damage recovery, as evidenced by the proliferation of statutory schemes discussed in this Comment. But see Kirk v. Denver Publ'g Co., 818 P.2d 262, 270-72 (Colo. 1991) (declaring Colorado's split-recovery statute unconstitutional). In fact, the continuing constitutionality of punitive damages generally may require placing them into a more involved statutory framework. See Malcolm E. Wheeler, The Constitutional Case for Reforming Punitive Damages Procedures, 69 VA. L. REV. 269 (1983).

(57.) The method of discouragement may be legitimate in cases where it is likely that the punitive damages would outpace the offense suffered and so overincentivize suit.

(58.) For example, caps solve continuing problems with jury irrationality while split-recovery statutes do not.