« The Legacies of U.S. News's Robert Morse, AALS's Susan Prager, and Us | Main | Best Fiction for Conferences »

Friday, July 11, 2008

Revisiting Cost Internalization and Punitive Damages after Philip Morris

More than a year ago, the Supreme Court handed down its decision in the Philip Morris case. I blogged about it at least a couple times. It's on my mind again, since I'm spending this month, among other things, revising my  Retributive Damages article (coming soon to a Cornell L. Rev. near you), and trying to finish drafting the successor article in the trilogy, Implementing Retributive Damages.

What I want to provoke here is some discussion about whether what Tom Colby (GW) and others call "total-harm" damages  extra-compensatory damages are permissible after Philip Morris if they are characterized simply as augmented damages for the purposes of cost-internalization or optimal deterrence (let's run those two notions together here, shall we?).  In Tom's latest work on punitive damages, which is forthcoming in Yale LJ, and is available in draft here (SSRN version of June 27, 2008), he argues that cost-internalization damages would and should be an available option (constitutionally speaking) so long as state legislatures or courts said they were not engaged in awarding "punitive damages," which is to say, so long as language of condemnation or language of punishment is not explicitly used.

My own view is that Colby's reading of the implications of Philip Morris for augmented or "total-harm" damages could logically be correct. But it seems a weird and ultimately, to my mind, wrong way to read what the litigants and the majority of the Court thought they were up to in the Philip Morris dispute and in this opinion. My extended thoughts are after the jump, and basically excerpted from my rough IRD draft. I invite your thoughts about the future of optimal deterrence/cost-internalization in the comments. (Btw, I will probably have more to say on Tom's article in other posts or perhaps in a response I might draft.)

Update: I made the mistake of conflating Tom's conception of "total-harm" damages with cost-internalization. As Tom kindly reminded me, his notion of "total-harm" damages, as used in his 2003 article, included a retributive component in addition to a deterrence signal.

To my mind, the proponents of the cost-internalization approach now face substantial difficulties on account of the Supreme Court’s recent decision in Philip Morris USA v. Williams. Implicitly repudiating the language of its TXO decision, which permitted punitive damages awards to consider the harm or potential harm from the conduct “as a whole,”[i] the Court reversed course. Writing for a 5-4 majority, Justice Breyer held that the Due Process Clause forbids punishing a defendant for harms to nonparties to the instant litigation because the defendant would not have the ability to “defend against the charge,” depriving the defendant of notice and imposing a substantial degree of arbitrariness and uncertainty in punishment.[ii] Thus, in assessing the putative excessiveness of punitive damages in a given case, a court must scrutinize whether a jury is punishing a defendant an amount of punitive damages that considers the harms actually suffered by strangers to the litigation.

By restricting the permissible scope of harm and potential harm, the Philip Morris decision now raises questions about whether full cost internalization is forbidden when establishing the amount of punitive damages. The ambiguity is subtle and has been brought to my attention by Professor Tom Colby’s forthcoming article on the subject.[iii] Professor Colby argues that states could constitutionally pursue “total harm” damages meant only to achieve cost-internalization so long as the state said there was nothing “punitive” to this cost-internalization approach. That would leave punitive damages, in his view, to pursue only the goals associated with what I have been calling "victim vindication" or "aggravated" damages.[iv]

The problem with Colby's narrow reading of Williams is that it reads the Court’s holding into, effectively, a nullity, giving with one hand a right that defendants would want and have fought repeatedly for and then stripping away the value of that right with the other hand. It’s hard to believe that the Court and litigants would be divided so sharply on an issue with such apparent lack of consequence--by letting total harm damages erupt without any procedural or substantive safeguards so long as the courts or legislatures say they're not punishing, they're just regulating. And unlike United States v. Booker -- which did recognize a defendant’s right only before demolishing its significance -- there aren’t separate opinions in Philip Morris regarding the “merits” and the “remedy.” Philip Morris is just one opinion and therefore susceptible to less plausible charges of schizophrenic reasoning.

To my mind, the better reading of the case undermines the feasibility of a cost-internalization agenda, but for two reasons, it does not destroy the agenda in one fell swoop.

First, the goal of achieving cost internalization remains feasible to the extent that jurisdictions make available class action or other aggregative litigation strategies that protect the rights of defendants; once a class is certified, at that point, nonparties become parties to the litigation. That said, more class actions will only permit the consideration of actual litigants, not the “quasi-parties” on whose behalf Professor Sharkey also advocated.

Second, the prospect of obtaining augmented damages outside the class action also continues to survive Philip Morris to the extent the defendant caused potential harm to the plaintiff(s). In other words, under my proposed reading of the case, the operative question (for purposes of cost-internalization) is now: what harm or potential harm did the defendant cause this particular plaintiff (rather than what harm or potential harm of this sort did the defendant cause)? The potential harm to the plaintiff, which the Court still permits as a consideration, plausibly encompasses the likelihood that the defendant would escape compensating this particular plaintiff. And to those that worry that Philip Morris eviscerates optimal deterrence, consider cases like those presented in the Supreme Court’s TXO decision, where an augmented damages award could be justified based on potential harm to the particular plaintiff on account of the tortfeasor’s concealment or deceit. If we assume a situation where a plaintiff faced a 1/3 chance of discovering that it was this defendant who caused the harm to the plaintiff, then that should, in theory, allow a jury pursuing cost internalization to impose augmented damages that are double the compensatory damages. (Per Polinsky and Shavell, “the total damages imposed on an injurer should equal the harm multiplied by the reciprocal of the probability that the injurer will be found liable when he ought to be.”[v] Thus, if a harm of 10 is caused and there is 1/3 chance the D will be held liable, then augmented damages should be 20 in addition to the 10 in compensatory damages, for a total of 30, or 3 times the harm.)

In sum, under my reading of Philip Morris, if a multiplier were based solely on the likelihood that the harm or potential harm to the plaintiff would be detected and compensated, it would pass constitutional muster, as suggested in Justice Breyer’s concurring opinion in BMW v. Gore.[vi] But Professor Colby is correct to emphasize that augmented damages need no finding of reprehensibility to warrant their award, and in such a situation, they should not logically be thought of as part of the punitive damages constitutional analysis.[vii] Instead, they involve a fact-finding determination, much like the amount of compensatory damages. And because augmented damages are essentially a regulatory tool to facilitate cost-internalization, states should be permitted to pursue different regulatory strategies, with tort law as one of several instruments to pursue that goal. A jury verdict form that reflects this ambition simply has to limit the jury’s consideration appropriately. The Appendix to [the IRD and RD articles] makes such an effort though it does so based on my reading of Philip Morris’s implications for cost-internalization, not on what I think would be logically required on a blank slate.

I should add that I don’t particularly have a dog in the fight; if it turns out Colby’s reading of cost-internalization after Philip Morris is regarded as correct, then I’d be happy to have the jury instructions I offer in the Appendix to my article -- the ones dealing with cost-internalization -- adjusted accordingly.

Standard of Review for Augmented Damages


What’s important about this decoupling of purposes (separating out the retributive purpose from the cost-internalization perspective) is that it logically calls for a different standard of review on appeal. In Cooper Industries, the Supreme Court announced that in federal cases the amount of punitive damages should be reviewed de novo, and that decision was predicated in part on the claim that determinations of punitive damages involve a moral component that has to be weighed against a defendant’s due process rights. That holding is only true for the retributive (and arguably the complete, not optimal, deterrence) part of extra-compensatory damages; by their nature,  augmented damages for the sake of cost-internalization involve only matters of empirical estimation (ie, what’s the likelihood the defendant would escape having to compensate this plaintiff), and thus are no different than compensatory damages (which ask, among other things, what kind of pain and suffering did the defendant’s action cause?), and thus determinations of augmented [total-harm] damages deserve greater deference when reviewed on appeal.

[i] Gore, 517 U.S. at 584 (indicating punitive damages award may reflect harm caused to all in-state purchasers);

[ii] Philip Morris, 127 S.Ct. at 1063 (slip op. at 6).

[iii] Thomas B. Colby, Clearing the Smoke from Philip Morris v. Williams: The Past, Present, and Future of Punitive Damages, 118 Yale L.J. (forthcoming 2009; manuscript on SSRN).

[iv] Professor Colby thinks, based on his understanding of the constitutional issues, that a public interest in retributive justice cannot be pursued outside the criminal law because of the absence of procedural safeguards. In this respect, he seems to agree with the view articulated by Redish, Emory L.J. But this perspective, to my mind, requires a finding that an intermediate sanction cannot be imposed even if it is accompanied by intermediate levels of procedural safeguards. My view is that with some basic safeguards, we can tolerate and embrace the use of intermediate sanctions. And given what the Supreme Court has repeatedly said in the last thirty years about the “quasi-criminal” purposes of punitive damages, there’s no reason to suspect that a space for an intermediate sanction cannot be carved out.

[v] See Polinsky & Shavell, supra note, at 889.

[vi] BMW of N. Am. Inc. v. Gore, 517


559, 562 (1996) (Breyer, J., concurring).


While the Court has in the past referred to deterrence as a goal of punitive damages, cost-internalization is about optimal or efficient deterrence, whereas what’s probably being adverted to by the Court is the goal of complete deterrence. The goal of complete deterrence signals that the conduct is prohibited whereas the goal of cost-internalization suggests that the defendant’s conduct is “priced.” Because cost-internalization refers to a different kind of deterrence (optimal) than the one (of complete deterrence) probably subsumed under the Court’s pronouncements in the constitutional analysis, Colby’s conclusion is highly plausible as a matter of theory and logic; but for the reasons I mentioned in the text, I doubt this conclusion is what the Court meant in Philip Morris.


Posted by Dan Markel on July 11, 2008 at 08:49 AM in Article Spotlight, Criminal Law, Current Affairs, Dan Markel, Retributive Damages, Torts | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference Revisiting Cost Internalization and Punitive Damages after Philip Morris:


Feed You can follow this conversation by subscribing to the comment feed for this post.

The news of this decision is that civil defendants explicitly have nearly all procedural due process rights, and that torts are not only for making the plaintiff whole, but for punishment. The hyprocrisy is less.

The remaining, unmentioned but implied procedural due process right is one to a fair hearing. The overwhelming majority of all claims, even those without exemplary damages, are pretextual, weak, and filled with a host of cognitive biases. Once that gets explicitly mentioned, then the biases inherent in all tort claims will violate the Constitution. The biggest tort of all with be a Constitutional tort caused by the filing of weak, biased, claims, and the forbearance of the pro-litigation biased courts.

Posted by: Supremacy Claus | Jul 11, 2008 12:28:05 PM

Post a comment