Nova v Dow: Doctrinal Implications

Nova Chemicals Corp v Dow Chemical Co 2022 SCC 43 Rowe J: Wagner CJ, Moldaver, Karakatsanis, Brown, Martin, Kasirer and Jamal JJ concurring; Côté J dissenting affg Nova Chemicals Corporation v Dow Chemical Company 2020 FCA 141 Stratas JA: Near, Woods JJA affg Dow Chemical Co v Nova Chemicals Corp 2017 FC 350, 2017 FC 637 Fothergill J

2,160,705 / film-grade polymers / ELITE SURPASS

The Intuition / The Legal Background / Causation as a Matter of Fact / The Concession / What Role for “But For” Causation in Identifying the NIO? / Summary of the Summary / Causation Concept in the Absence of an NIO / What is the NIO? / The Value of the Invention / Rivett on the Facts / Three Policy Arguments / The Source of the Chilling Effect / Miscellaneous Policy Issues

In Nova v Dow, Rowe J, writing for an 8-1 majority, addressed the proper method of calculating an accounting of profits in the patent context. I have a two-part article on the decision forthcoming in the IPJ. Part I of that article analyzes Nova v Dow at a doctrinal level. What did Rowe J mean by “cause”? What did he mean by “the non-infringing option”? What did he mean by “the value of the invention”? I summarized Part I in a series of blog posts, listed above. Part II addresses the policy implications of Nova v Dow. In the first post summarizing Part II, I argued that the policy arguments advanced by Rowe J in support of his approach are either entirely misguided, or support a “but for” approach to an accounting. Further, while Rowe J asserted that his approach would not chill legitimate non-infringing competition because his approach is not intended to be punitive, the chilling effect does not depend on the label we attach to a remedy. In the next post I discussed the source of the chilling effect and showed that Rowe J’s departure from “but for” causation does indeed risk a chilling effect. That post left aside the question of the extent of that chilling effect and the appropriate policy implication, which will be discussed in the post following this one. In this, the fourth post on Part II, I will summarize my thoughts on the likely doctrinal implications of Nova v Dow. A subsequent post will consider the policy implications and the appropriate response to the decision. (The third post on Part II dismisses two minor policy arguments made by Rowe J.)

The accounting context

The most basic question is how Nova v Dow will impact the accounting of profits remedy in the patent context. This question is remarkably difficult to answer given that it was the focus of the entire decision. The problem is that Rowe J did not provide any guidance on the key issues. While he implicitly rejected “but for” causation, he did not specify any alternative. On the central issue of what constitutes an appropriate a non-infringing option, the key holding is that the NIO is “any product that helps courts isolate the profits causally attributable to the invention” [58]. In the absence of any clear causation concept, saying that the NIO is anything that is helpful is, well, unhelpful. Moreover, he held that both causation and the nature of the NIO are matters of fact. Since Rowe J did not articulate any guiding principle it is difficult to predict how the law will evolve. Ultimately, my best guess is that the courts will develop a non-exhaustive list of factors, including market substitutability, technical similarity, and “but for” causation, which are all considered and balanced in light of the facts of the case.

Many doctrines involve balancing a list of factors. This does not raise any problems in itself—so long as the factors are all relevant to some principled underlying inquiry. For example, a variety of factors may be relevant to reconstructing the hypothetical negotiation in assessing reasonable royalty damages, with more or less weight given to any particular factor on the facts of a case. That does not necessarily result in unpredictability, because the various factors all address the same principled question, namely what are the terms of the licence the parties would have entered into had they bargained ex ante. The weight to be given to each factor turns on the degree to which, on the facts, that factor is relevant to the ultimate inquiry. The difficulty in with Nova v Dow is that there is no underlying principle. This means that even if the courts develop a list of “relevant” factors, those factors will not be relevant to any underlying principle, because there is no underlying principle. This means that there will be no way to predict whether one factor rather than another will have more weight in any given case; that will likely turn on arbitrary and idiosyncratic factors such as the mannerisms and persuasiveness of the expert witnesses.

Consequently, if I suspect that Nova v Dow will result in long-term uncertainty, where there is really no way of predicting whether an NIO is appropriate other than going to trial.

In her dissent, Côté J stated that “[t]here must be some legal requirements or standards [for a non-infringing option] that guide courts and the parties to the litigation” [193]. The “must” in that sentence is aspirational rather than descriptive. There should be some legal requirements or standards to provide guidance—but, since Nova v Dow, there aren’t. No doubt the FCA would like to provide guidance, but only time will tell whether that will be possible in the face of Rowe J’s holding that it is purely a matter of fact as to whether a particular alternative is an NIO.

There will also be uncertainty as to the applicability of Rowe J’s approach to different heads of profit. Rowe J explicitly endorsed “but for” causation in the context of springboard damages, while he used a different causation concept in the context of the non-infringing option. So “but for” causation is used sometimes, but not always, when assessing the amount to be disgorged. There is an open question as to whether “but for” causation will be used in assessing products from convoyed goods (unpatented goods normally sold with the patented product), fixed costs (which were addressed in the FCA but not the SCC), and the cost of capital. The most likely scenario is that these heads of profits will be addressed on an ad hoc basis, in which “but for” causation may or may not be applied in each case. However, we can expect clarity to emerge as courts specifically address these heads of profit. 

In particular, in the Nova v Dow FCA decision Stratas JA held that "the full costs approach [to deduction of fixed costs] is principled and sound" [154]. This means that the infringer will always be able to deduct some portion of its fixed costs. This issue was not addressed by the SCC, but if the SCC had addressed some coherent causation principle, then Stratas JA's approach to fixed costs might have been subject to revision. As it is, I expect it will stand, even though it is not based either on "but for" causation or on whatever causation concept was being used by Rowe J. While we now have doctrinal clarity regarding the deductibility of fixed costs, I doubt this will lead to predictability. While Stratas J A clearly held that some proportion of fixed costs are always deductible, he did not provide any principle to determine what that proportion might be — except that whatever the principle is, it is not "but for" causation. In my view, "but for" causation is the only principled approach and consequently, Stratas JA's full costs approach is likely to also lead to long-term unpredictability and arbitrary results. I am coming around to the view that this may even more important in practice than the question of the appropriate NIO. An infringer will always be able to point to very substantial fixed costs and if any significant proportion of those costs is regularly deducted, this may significantly reduce the quantum of the accounting.

Further, in the article, I point out that there is a real puzzle in allowing for the cost of capital. It is possible that deductibility of the cost of capital will turn on whether the infringing product is funded with retained capital or external capital, even though this clearly has no bearing on whether the profits are caused by the invention.

The patent damages context

The next question is whether Rowe J’s approach to the NIO will extend to patent damages. The straightforward argument for extending it to the damages context is that prior to Nova v. Dow, patent damages and an accounting were symmetrical. As the Court of Appeal stated in Merck 2015 FCA 171 [60] in the course of holding that the Schmeiser differential profits approach extends to damages, “the significance of [Schmeiser] is that if a court may consider a defendant’s resort to a non-infringing alternative when calculating the infringer’s profit, there is no reason in principle to ignore such conduct when calculating the patentee’s lost sales.” If the Schmeiser approach applies to damages as much as an accounting, then it would seem to follow that the Nova v Dow approach does also. On the other hand, Rowe J focused throughout his decision on the accounting remedy, and there is not the slightest hint of a suggestion that the principles would extend to damages as well. Further, Stratas JA, whose judgment in the Court of Appeal was affirmed by the Supreme Court, expressly held that his approach to an accounting did not apply in the damages context: Nova v. Dow FCA 2020 FCA 141 [45], [67], [76]. Consequently, it would be entirely fair to read the SCC decision as being confined to the accounting context. I suspect that whether the Court of Appeal decides to extend Rowe J’s approach to the NIO to the damages context depends on the view it takes of the soundness of his decision. A similar question arises with respect to s 8 damages under the PM(NOC) regime. I expect that will go the same way as damages more generally.

Beyond patent law

An accounting or disgorgement of the wrongdoer’s profits is a remedy that is not limited to patent law, but I very much to doubt that Nova v Dow will have any influence beyond patent law. It is difficult to identify any principle which might be extended to other cases, and Rowe J’s focus on the profits causally attributable to “the invention” is a straightforward basis for holding that Nova v Dow does not apply more broadly.