Monday, July 17, 2017

Quantum, Proportionality, and Uncertainty in Punitive Damages

Airbus Helicopters SAS v Bell Helicopter Textron Canada Limitée 2017 FC 170 Martineau J [FC Damages]
            2,207,787 / helicopter landing gear

As I noted in my post on Airbus’ entitlement to punitive damages, most jurisdictions do not award enhanced or punitive damages for patent infringement, or for civil actions more generally, because, as explained by Lord Reid in Cassell & Co Ltd v Broome [1972] UKHL 3 [1972] 1 All ER 801, it is undesirable to expose a party to a quasi-criminal sanction when both entitlement and quantum are so vaguely defined. With respect to quantum, he noted “[t]here is no limit to the punishment except that it must not be unreasonable.”

In US patent law uncertainty regarding quantum is addressed by limiting the total award to a maximum of three times compensatory damages assessed: 35 USC § 284. This means that the punitive damages are limited to twice the compensatory damages. So, in this case, the award of $1,500,000, comprised of $500,000 in compensatory damages and $1,000,000 in punitive damages, would be the maximum allowable under US law.

In Canadian law there is no such hard limit, either with respect to patents or more generally, though this may be because punitive awards have tended to be more modest and less of often granted. In Whiten v. Pilot Insurance Co 2002 SCC 18, [70], the SCC recognized the problem of uncertainty relating to quantum:

[70 [T]he incantation of the time-honoured pejoratives (“high-handed”, “oppressive”, “vindictive”, etc.) provides insufficient guidance (or discipline) to the judge or jury setting the amount. . . . A more principled and less exhortatory approach is desirable.

See similarly [39]. Thus the purpose of the principled approach is to provide guidance in assessing quantum. Does it?

The Court in Whiten stressed that the “governing rule for quantum is proportionality,” with relation to “the objectives for which the punitive damages are awarded,” namely “retribution, deterrence and denunciation” [74]. Proportionality requires that the award is neither too large nor too small to achieve these objectives [111]. What do these terms mean? Deterrence looks “to deter the defendant and others from similar misconduct in the future [94]. It is forward looking, and it looks not to the particular defendant, but to future parties in a similar position. The idea is that those future parties will look at what happened to the defendant, and decide that the wrongful act isn’t worth it. In contrast, retribution “reflect[s] the moral blameworthiness of that particular offender,” while its “sibling” denunciation “communicate[s] society's condemnation of that particular offender's conduct,” with “a symbolic, collective statement that the offender's conduct . . . encroach[es] on our society's basic code of values” R. v. M. (C.A.) [1996] 1 SCR 500 [81]. It seems, then, that retribution and denunciation both look to the particular defendant, and so are “siblings,” but retribution is directed at the defendant themselves, while denunciation speaks to the community at large. Deterrence in contrast looks to future parties who have not yet done anything wrongful.

The difficulty with this principled scheme, as I see it, is that it actually provides very little guidance in assessing quantum. Quantum must in the end be expressed as a number. Deterrence, as discussed below, naturally lends itself to quantification, but the siblings of denunciation and deterrence communicate moral values and judgments, which are by the nature almost impossible to quantify.

So, in addition to describing the overarching objectives, the SCC in Whiten provided a list of more specific “dimensions” [111] to be considered in assessing the quantum (listed by Martineau J at [384]). The first is that the award must be “(i) Proportionate to the Blameworthiness of the Defendant’s Conduct” [112]. This apparently goes to the siblings, as denunciation was expressly mentioned by the SCC in that discussion, while R v M(CA) tell uses that moral blameworthiness goes directly to retribution. The SCC begins its discussion of blameworthiness by saying “[t]he more reprehensible the conduct,” this higher the limits to the potential award [112]. This implies that the award must be proportionate to the degree of reprehensibility. What does that mean? The SCC held that punitive damages should be awarded only if there has been “high-handed, malicious, arbitrary or highly reprehensible misconduct that departs to a marked degree from ordinary standards of decent behaviour.” This suggests that that “reprehensible” is more or less synonymous with “high-handed, malicious, arbitrary.” Putting this together, the principled approach tells us that the quantum must be proportionate to the degree to which the behaviour is high-handed, malicious, arbitrary or reprehensible. It looks to me that this does little more than replicate the “time-honoured pejoratives,” which the Court disparaged, with the addition of the word “proportionate.” I don’t see that adding a qualifier “proportionate” reduces uncertainty when applied to concepts which are inherently unquantifiable.

The SCC in Hill stated that punitive damages is “the means by which the jury or judge expresses its outrage at the egregious conduct of the defendant” [196] (quoted in Whiten [68]). (Similarly, the Court noted in Whiten than compensatory damages “may be disproportionately small to the level of outrage the jury wishes to express” [123]). The guidance regarding the quantum of punitive damages is that they must be “proportionate” to that level of outrage. But how can “outrage” be quantified?

The problem with the Whiten approach is evident in this case. Martineau J went through all the sub-factors set out by the SCC as bearing on moral blameworthiness, but so far as quantification is concerned, I see no point in going through them in detail, because even if I agree with this assessment on each one, that would say nothing as to whether the final quantum of $1m in punitive damages was proportionate to the degree of outrage, since neither “blameworthiness,” nor “outrage,” nor “proportionate,” have any numbers attached. Martineau J was outraged by Bell’s conduct, and he felt $1m was proportionate to Bell’s blameworthiness [401], and I don’t see any basis for second guessing that, since I don’t know, numerically, how outraged Martineau J was, or what proportionality factor was applied to that unknown degree of outrage. In my earlier posts I have tried to explain why I am not as outraged by Bell’s conduct as was Martineau J, but even so, for the same reason I can’t say that the award is proportionate, neither can I say it is not. I can say that I’m less outraged than Martineau J, but I don’t know how to put a number on my outrage, and I certainly don’t know how to put a number on his.

No doubt reviewing the various factors and objectives is more principled that saying merely that the quantum should not be unreasonable, but the putative point of providing “a more principled approach” was that “the incantation of the time-honoured pejoratives. . . provides insufficient guidance” [70]. It is not apparent to me that the approach set out in Whiten provides any better guidance than the time-honoured pejoratives.

Deterrence is a different matter. It does lend itself to quantification, as we can ask “How much of penalty would be necessary to dissuade future parties from the wrongful act?” As Martineau J stated “one of the most fundamental aspect of punitive damages is to ensure that the defendant does not treat compensatory damages merely as a license to get its way irrespective of the legal or other rights of the plaintiff (Whiten at para 124). . . . As stated by the Supreme Court of Canada, it is perfectly acceptable to use punitive damages to relieve a wrongdoer of its profits where compensatory damages would amount to nothing more than an expense paid to earn greater profits while flouting the law” [395]. This is an excellent observation, which begs the question, when, exactly, would compensatory damages be inadequate to ensure compliance with the law?

Punitive damages are by no means necessary for effective deterrence. Compensatory damages may be intended to compensate the victim of the wrong, but because they involve a payment from the wrongdoer, they will necessarily also have a deterrent effect. If a party is rational, then it will not infringe if it expects that the cost of infringement, including expected liability, will exceed the benefit. If a rational party is 100% sure that it will get caught, then it will not infringe, even if punitive damages are never awarded. If an accounting of profits is awarded, it will lose the entire benefit, and it will have to pay its own legal costs as well as part of the costs of the patentee, as well as suffering the other losses such as the cost of executive time diverted from productive projects. The same is normally true if damages are awarded, as the patentee’s damages are typically as large or larger than the infringer’s profits.

This simple argument leads to the rationale for punitive damages that is the most widely accepted in the academic law and economics literature, namely the problem of underdetection: see Polinsky & Shavell, “Punitive Damages: an Economic Analysis,” 111 Harvard Law Review 869 (1998). If the likelihood of being detected is very low, say 10%, then, on average, the benefit in the 90% of cases in which the infringer escapes detection may well outweigh the compensatory liability imposed in the 1 case in 10 in which it is caught. If punitive damages are imposed proportionately to the likelihood of detection – ten times compensatory damages would be appropriate in this example – this would restore the deterrent effect. This explains why punitive damages may properly exceed compensatory damages.

The main difficulty with this underdetection analysis is that it will often be very difficult, or impossible, to accurately calibrate the quantum of punitive damages. But there are few perfect tools in law, and in my view, this analysis largely rings true, both as a matter of policy and as an explanation for the cases. Whiten, for example, is a paradigmatic example. After the Whiten’s house burned down, Pilot Insurance intentionally breached its contract of insurance, asserting arson when all the evidence, including that of its own investigators, was to the contrary, and cutting off the family with no money in the middle of winter, in the expectation that impoverished homeowner in desperate circumstances would not be able to afford to litigate, and with the intent of forcing the Whitens to settle for much less than the claim was worth. In the circumstances, it was entirely rational for Whiten to expect that it would not get caught, and punitive damages were entirely appropriate, purely from a deterrence perspective.

The courts don’t usually discuss punitive damages in terms of the probability of underdetection, (though an entitlement to an accounting has been discussed in very much these terms). But the reference to “vulnerability” of the plaintiff (Whiten esp [114]-[116]) captures an important class of cases in which punitive damages are appropriate because of underdetection. When the plaintiff is vulnerable financially, as in Whiten, or otherwise, as in Norberg v. Wynrib, [1992] 2 SCR 226, (Whiten ibid), or in Cinar 2013 SCC 73, the wrongdoer can often exploit that vulnerability to commit a wrong without fear of being caught. In this case Bell argued that “a defining characteristic of all the cases with awards of $1 million or more is the vulnerability of the plaintiff who has suffered as a result of the actions and violations carried out by a powerful defendant” [390]. Martineau J rejected this as an attempt to limit punitive damages to certain categories of claims [393], referring to Chanel 2016 FCA 111, apparently as a counter-example [394]. In my view, Martineau J was right to say that punitive damages should not be restricted on a categorical basis, but at the same time, Bell was right to argue that vulnerability is an important indicia. This is because cases in which the plaintiff is vulnerable, in the sense of being dependent on the defendant, or financially weaker than the defendant, is just one subset of cases in which the problem of underdetection arises. In particular, pirates, whose strategy is to avoid detection, hide their assets, and generally evade the force of the law, are a prime example of a party exploiting the problem of underdetection, even though the IP owner such as Chanel cannot be said to be vulnerable to the pirate in terms of financial or emotional dependency. A farmer is more powerful than a mouse, but the problem of underdetection is very real. Thus, the problem of underdetection captures vulnerability of the plaintiff as an important sub-class, but it is a broader concept.

It might be said that this argument fails unless the defendant is perfectly rational, and able to make the detailed cost-benefit assessment. Not so. It is true that the argument fails if the defendant is completely irrational, and makes its decisions without any regard to the consequences. But in that case, the entire deterrence objective fails as a justification for punitive damages, as no prospective penalty can deter someone who acts without regard to consequences. More generally, and more realistically, most parties are probably imperfectly rational – they have regard to consequences, but do not necessarily make explicit cost-benefit calculations, or if they do, those calculations are imperfect. Imperfect rationality does not undermine the underdetection argument; on the contrary, underdetection explains the cases even better if we assume imperfect rationality. I have noted that the courts don’t have the information necessary to tailor punitive damages exactly to the probability of underdetection, but if defendants are not perfectly rational, perfect calibration of the award would be pointless. For punitive damages to have a deterrent effect, it is enough that they alter the defendant’s calculus. With a low probability of detection and compensatory damages, an imperfectly rational defendant will think, “What are the chances that I’ll get caught, and if I do, I’ll only get a slap on the wrist, so I’ll risk it.” With a low probability of detection and punitive damages, an imperfectly rational defendant will think, “I probably won’t get caught, but if I do, it’s sure going to hurt, so maybe it’s not worth it.” With an imperfectly rational defendant, it is enough to have a looser proportionality – make sure the penalty stings, relative to the profits gained by the defendant.

On the facts of this case, the probability of underdetection must have been essentially zero. There were only a couple of major players in the industry, the landing gear is a distinctive and highly visible part of the final product, which was, moreover, being actively advertised by Bell, and the patentee was a large and sophisticated party with ample resources to pursue litigation. Consequently, in my view, the deterrence objective cannot justify punitive damages in this case. If Bell believed the patent was probably valid, the inevitable prospect of compensatory damages, or being required to account for their own profits, plus bearing their own legal costs plus a portion of Eurocopter’s costs, would provide an ample deterrent. And to the extent that it believed that the patent was invalid, then it was doing nothing wrong. On the facts, it seems more likely that Bell wasn’t engaging in a rational cost-benefit calculation in any event. As Martineau J said, “In pursuing the project, Bell acted in a foolhardy manner” [FC Liability [431], quoted at [400]]. It seems likely, that Bell simply didn’t think about the consequences, and if it had, it would have realized that its course of action was foolish, and, as Martineau J pointed out, contrary to its own policy manuals [ibid]. If so, then punitive damages are not warranted from a deterrence perspective. Deterrence, recall, is forward looking. The question is whether similarly placed parties in the future would be adequately deterred by compensatory damages. Occasional mistakes are inevitable, since parties are not perfectly rational, and if Bell had thought about it, the prospect of compensatory damages and costs would have been adequate deterrence. To return to Martineau J’s observation that “[a]s stated by the Supreme Court of Canada, it is perfectly acceptable to use punitive damages to relieve a wrongdoer of its profits where compensatory damages would amount to nothing more than an expense paid to earn greater profits while flouting the law” [395], on the facts of this case I can’t see why compensatory damages would be an expense that would allow Bell to earn greater profits from its infringement.

None of this is to criticize Martineau J. The SCC in Whiten clearly endorsed a calculus of outrage in assessing punitive damages. I can't help but feel that even a proper application of the Whiten principles has done little to mitigate the problem it identified, that "Substantial awards are occasionally assessed at figures seemingly plucked out of the air" [39].

Finally, in discussion deterrence Martineau J stated that “Bell’s outrageous conduct caused irremediable damages that simply cannot be corrected by an award of damages or an account of profits” [430]. Taken in the context of the deterrence discussion, this seems to be saying, in effect, that compensatory damages are not an adequate deterrent because they are not fully compensatory. I’m a bit puzzled by this statement, because Martineau J in this very decision had just quantified Airbus’ loss, and he didn’t point to any specific heads of damages that were unquantifiable, either in the course of assessing compensatory damages, or in this paragraph. If he means that damages are generally not fully compensatory, then this statement would be contrary to the Federal Courts’ steadfast refusal to grant interlocutory injunctions, on the view that damages are almost always adequate to make the patentee whole.

No comments:

Post a Comment