Friday, October 27, 2017

Territoriality and Remedies for Transnational Infringement

AstraZeneca Canada Inc v Apotex Inc 2017 FC 726 Barnes J [Omeprazole Accounting]
            1,292,693 / omeprazole formulation / LOSEC

Omeprazole Accounting indirectly raises an interesting and difficult issue respecting the principle of territoriality as applied to monetary remedies in transnational litigation. As I understand the facts,* Apotex manufactured its omeprazole product in Canada and exported some of it for sale in the US. Litigation ensued in both jurisdictions, resulting in determinations that the manufacture in Canada infringed the ‘693 patent, and the sale in the US infringed US Patent 4,786,505 (the US equivalent of the ‘693 patent). In the US litigation, AstraZeneca was awarded reasonable royalty damages in the amount of $76m: AstraZeneca AB v Apotex Corp, 985 F Supp 2d 452 (2013). The question was how to allow for those US damages in assessing the Canadian accounting of Apotex’s profits from the US export sales.

It appears that (unsurprisingly) Apotex’ profits on the US sales exceeded the reasonable royalty damages payable under the US judgment. While the parties agreed that double recovery of the US damages was not allowed, AstraZeneca argued that Apotex should be required to disgorge all of the profits on the US sales, less the US damages. This was evidently on the logic that AstraZeneca had two separate causes of action, and should be allowed to recover on both, so long it is did not get double recovery for the same loss. Apotex, on the other hand, argued that it should not be required to disgorge any US export profits [243]-[235]. This argument was based on res judicata, and more specifically cause of action estoppel [241]: “by opting to claim a recovery in the United States AstraZeneca must now accept that award as full satisfaction of its entitlement from the infringement of the 693 Patent for Apotex’s sales into the United States” [240].

Barnes J rejected Apotex’s argument, saying “the causes of action in the two proceedings arose under different patents, involved distinct acts of infringement and were tried in jurisdictions where different substantive legal principles applied” [244]. He noted in particular that an accounting of profits is not available in the US, and that the temporal scope of the infringing acts is different, as the US patent expired in 2005, before the Canadian patent. “It cannot be the case that, by proceeding first in the United States, AstraZeneca should be taken to have abandoned its claim for ongoing Canadian infringement post-dating the expiry of the United States patent” [245]. I am not an expert on res judicata, but on my understanding of the law, Barnes J’s conclusion seems to me to be correct. Cause of action estoppel requires that the material facts giving rise to both actions are the same: Danyluk 2001 SCC 44, [54]. A key material fact in the US litigation is the sale of omeprazole in the US, which is not relevant to infringement by manufacture in Canada, while a key fact in the Canadian litigation was the manufacture in Canada, which is not relevant to infringement by sale in the US.

With res judicata disposed of, Barnes J’s holding follows from the principle that the infringer should account for the profits caused by the infringement. However, while there can be no recovery without causation, causation is not the only limit on recovery. The other potentially relevant limit is the principle of territoriality. Barnes J did not address this principle, presumably because it was not raised by the parties, but the US Federal Circuit has addressed territoriality in several cases, including Power Integrations, Inc. v. Fairchild Semiconductor Int'l, Inc., 711 F.3d 1348 (Fed.Cir.2013), Carnegie Mellon University v. Marvell Technology Group, Ltd. 807 F.3d 1283 (Fed Cir 2015), and WesternGeco LLC v. ION Geophysical Corp., 791 F.3d 1340 (Fed.Cir.2015). (See here and here for Professor Cotter’s remarks on these cases.)

In Power Integrations, “the patentee, a chip supplier, lost contracts to supply a prospective customer with computer chips in the United States and abroad because the accused infringer became a competitor for such contracts as a result of the U.S. infringing sales. If the accused infringer had been precluded from U.S. infringement, the patentee alleged that the accused infringer could not have competed for the contracts which necessarily involved supplying chips both in the United States and abroad.” (WesternGeco 1350). Invoking the principle of territoriality, the Federal Circuit held that the patentee was not entitled to lost profit damages on the foreign sales it lost, notwithstanding that “those foreign sales were the direct, foreseeable result of Fairchild's domestic infringement” Power Integrations 1371. In Carnegie Mellon the facts are summarized quite briefly, but if I understand correctly, the patented invention was incorporated in a simulator used by Marvell in the design of chips, which was carried out in the US. The infringement resulted in a “design win” leading to both US and foreign sales: 1291, 1309-10. Thus the infringement in the US caused the foreign sales. The Federal Circuit nonetheless refused to allow reasonable royalty damages in respect of chips made and sold abroad, in light of “the general bar on extraterritorial application of our patent laws” 1305. The Federal Circuit summarized in Carnegie Mellon:

In the lost-profits context, this court indicated in Power Integrations that, where the direct measure of damages was foreign activity (i.e., making, using, selling outside § 271(a)), it was not enough, given the required strength of the presumption against extraterritoriality, that the damages-measuring foreign activity have been factually caused, in the ordinary sense, by domestic activity constituting infringement under § 271(a). 711 F.3d at 1371–72. We think that the presumption against extraterritoriality, to be given its due, requires something similar in the present royalty setting. Although all of Marvell's sales are strongly enough tied to its domestic infringement as a causation matter to have been part of the hypothetical-negotiation agreement, that conclusion is not enough to use the sales as a direct measure of the royalty except as to sales that are domestic (where there is no domestic making or using and no importing).

It is quite clear, therefore, that as a matter of US law, the bar on extraterritorial application of the patent laws may bar recovery of a foreign loss even though that loss was directly caused by domestic infringement. It is not enough that the foreign loss was caused by the US infringement; the loss itself has to be sufficiently linked to the US territory.

What exactly, in addition to causation, constitutes a sufficient link to the US is a different question, and not a simple one. I must admit that I do not find the Federal Circuit’s reasoning in these cases to be entirely transparent, so it is not easy to apply the analysis to different facts. With that said, consider a hypothetical in which Apotex manufactured an omeprazole formulation in the US that infringed AstraZeneca’s 505 patent, and sold it in a foreign country in competition with AstraZeneca. In the foreign country, AstraZeneca’s patent was held invalid, and so Apotex did not infringe. Under US law, continued manufacture in the US would be enjoined, but would damages be recoverable in respect of the foreign sales? This hypothetical is different from the cases noted above, in that the product sold abroad was directly manufactured in the US where the “making” was itself an infringement. Judge Wallach, dissenting in WesternGeco, 1354ff, cited a number of cases, which, as the majority put it at 1352, “suggest that profits for foreign sales of the patented items themselves are recoverable when the items in question were manufactured in the United States and sold to foreign buyers by the U.S. manufacturer.” The majority dismissed these cases as not “remotely similar” to the facts at hand. Thus it would undoubtedly be open to the Federal Circuit to distinguish this hypothetical on that basis, and allow recovery in respect of the foreign sales. However, the Court did not say that recovery would be allowed in such cases. I must say that I don’t see the difference in principle (and neither did Judge Wallach); in the hypothetical, the making itself does not give rise to the loss, which is caused by the foreign sale, just as in Power Integrations and Carnegie Mellon. I conclude that it is an open question as to whether recovery would be allowed in my hypothetical, though my best guess is that it would be refused. (I haven’t done an exhaustive review of US cases, and it may be that the answer in US law is clearer than it appears to me.)

Backing away from the details of US law, what is clear is that the US cases illustrate a very different approach to the problem of coordinating monetary remedies when a single loss is caused by separate acts in two different jurisdictions. The difficulty is to ensure full recovery, without double recovery, and while respecting the laws of each jurisdiction. For the Federal Circuit, the loss itself is allocated to one of the two jurisdictions. Full recovery is allowed in that jurisdiction, without allowance for any recovery in the other jurisdiction, and recovery is prohibited in the other. (In the US cases discussed above, recovery is prohibited in the US, leaving the patentee to its remedy in the foreign jurisdiction.) This obviously prevents double recovery, but it also allows full recovery, at least when the act infringes in both jurisdictions. The approach used in Omeprazole Accounting, on the other hand, ties the loss to both jurisdictions, but splits the recovery between the jurisdictions. This obviously allows full recovery, and double recovery is avoided by each court explicitly taking into account any award in the foreign jurisdiction.

Nothing much would turn on the difference if the patent laws were the same in all jurisdictions, but when the law is different, either substantive or remedially, the approaches diverge. It may appear that the US approach is more insular, in that it does not consider the foreign remedy explicitly, but I would suggest that in fact the US approach is premised on prioritizing respect for the laws of the foreign jurisdiction, while the approach of splitting recovery prioritizes respect for the laws of the domestic jurisdiction. For example, suppose the foreign act does not infringe in the foreign jurisdiction because there is no valid patent in effect. If the loss is tied to both jurisdictions, then under the split recovery approach, the patentee should be allowed recovery for the loss arising from the foreign act, on the same principle as Barnes J allowed recovery for post-expiry US foreign sales, namely that the patentee should be entitled to recover for all losses caused in fact by the Canadian infringement. The US courts, on the other hand, would say that the fact that the patentee is uncompensated under the US approach stems from its failure to seek a foreign patent:

Our patent system makes no claim to extraterritorial effect; `these acts of Congress do not, and were not intended to, operate beyond the limits of the United States,' and we correspondingly reject the claims of others to such control over our markets. To the degree that the inventor needs protection in markets other than those of this country, the wording of 35 U.S.C. ss 154 and 271 reveals a congressional intent to have him seek it abroad through patents secured in countries where his goods are being used.
                        Deepsouth Packing Co. v. Laitram Corp., 406 US 518, 531 (1972)

As this passage indicates, the advantage in principle of the US approach is that it better respects the laws of foreign jurisdictions. See also Professor Brian Love’s brief in support of Marvell, on appeal to the Fed Cir, arguing that the district court rule which allowed recovery “effectively transforms every U.S. patent into a worldwide patent,” and similarly Professor Bernard Chao’s article, Patent Imperialism, arguing against worldwide causation.

In this case, Barnes J reasoned that “[i]t cannot be the case that, by proceeding first in the United States, AstraZeneca should be taken to have abandoned its claim for ongoing Canadian infringement post-dating the expiry of the United States patent” [245]. While this is true as a matter of res judicata, territoriality poses a distinct limit on recovery. The US courts would likely say that AstraZeneca should not have any claim to a remedy for the post-expiry US sales, not because it abandoned that claim by suing in the US, but because it should never have had a claim based on US sales in the first place. To allow recovery in Canada has the effect of improperly extending the term of protection for sales taking place in US territory. Under US law the product is no longer protected by a patent, and therefore free competition should be allowed in respect of US sales, unimpeded by the happenstance of where the products are made. The same problem arises even with respect to sales during the term of the US patent. Allowing AstraZeneca to obtain an accounting of Apotex’ profits in respect of those sales, deducting the US damages award, effectively converts the sanction for the US sales from a reasonable royalty into a disgorgement of profits. Rightly or wrongly, the US legislature has made a deliberate policy decision to eliminate an accounting of profits as a remedy for patent infringement. The Canadian recovery thwarts that policy.

Examples can be multiplied. Suppose that in this case Apotex had been liable in the US for treble damages for willful infringement, but no punitive damages had been awarded in Canada, where such awards are much less common. Would the entire US award have been deducted from the Canadian accounting? If so, that would undermine the policy respecting willful infringement which the US saw fit to enact in respect of an infringement occurring in its own territory. If not, then the effect of the Canadian award would be to subject Apotex to punitive damages, contrary to Canadian policy. Or suppose the infringing sales had been subject to a positive defence under US law, but the infringing act in Canada was not. To allow full recovery in Canada for the foreign sales would effectively negate that US defence; to deny recovery entirely in Canada would under-compensate under Canadian law for a Canadian infringement.

The counter-argument to all this is that disgorgement of Apotex’s profits on the US sales is necessary to satisfy the policy underpinning the Canadian patent laws. The US response would no doubt be that it is not possible to respect the distinct policies of the different patent laws simultaneously. One set of laws must take priority, and it should be the law of the jurisdiction most closely connected to the loss itself. The US sales are more closely tied to the US, where they actually took place, and so US law should govern.

I have emphasized the strengths of the US position, not because I believe it should necessarily prevail, but because I think that it has not been sufficiently acknowledged in Canada. While prioritizing Canadian law over US law was the effect of the method used by Barnes J, I must emphasize that it was not the basis for his decision, as the issue of territoriality was not argued before him. His holding followed simply from the principle of causation, and his (correct, in my view), rejection of res judicata. Similarly, while awards for loss of foreign sales have been made in other Canadian cases, this does not imply a rejection of the US approach to territoriality, as I do not believe the point has been argued.

My basic point is that when a single loss is caused by separate acts in two different jurisdictions, and the law is different in those jurisdictions, either substantively or remedially, it does not appear to me to be possible to respect the legal principles of both jurisdictions simultaneously. The problem is inherently difficult. Is the answer to prioritize the jurisdiction most closely connected to the loss, even if it is the foreign jurisdiction? Is it to prioritize domestic law? Is some third way possible? I have no answer to any of those questions.

* The parties had settled most issues, and asked the Court for guidance on specific points, so the facts were more briefly described than is usual.

UPDATE: In the course of reviewing for the Competition Law course, I am teaching this semester, I realized that in Pro-Sys 2013 SCC 57  [40] and Sun-Rype 2013 SCC 58 [21], the SCC addressed a similar argument, to the effect that indirect purchaser actions should not be allowed because of the risk of double recovery under s 36 of the Competition Act if direct purchasers in a foreign jurisdiction, such as the US, and indirect purchasers in Canada, could both bring claims in respect of the same breach of the Act. The SCC dismissed this concern, saying “the judge may deny the claim or modify the damage award in accordance with an award sought or granted in the other jurisdiction in order to prevent overlapping recovery.” This clearly endorses what I have called the “split recovery” approach, which was used by Barnes J in this case. The patent context might potentially be distinguished on the basis that the principle of territoriality plays out differently in the patent context, though it must be noted that the SCC in Sun-Rype also rejected a challenge to application of the Competition Act to actions by foreign defendants [44]-[47]. On the whole, Pro-Sys and Sun-Rype are not determinative in the patent context, but they do support the view that the split recovery approach is appropriate.

No comments:

Post a Comment