Nova v Dow: What Role for “But For” Causation in Identifying the NIO?

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

In Nova v Dow, Rowe J held that when an accounting is granted, the infringer must disgorge all profits caused by the invention. While the argument turned on whether “but for” causation was the applicable causation concept, Rowe J never addressed this question; he did not explain what it means for profits to be “caused” by the invention, except to say that causation is a matter of fact. In my last post, I noted that Rowe J implicitly rejected the position that the differential profits rule from Schmeiser reflected the direct application of “but for” causation; it is necessary to apply the three-step test which Rowe J set out in Nova v Dow in which causation is addressed at Step 2. However, Rowe J did not address the nature of the causation requirement at that step. This post considers what role, if any, there is for the application of “but for” causation at Step 2 of the three-step test.

Step 2 requires the court to “[d]etermine whether there is a non-infringing option [NIO] that can help isolate the profits causally attributable to the invention” [15]. In my last post, I noted that we cannot infer anything about the causation concept from the holding on the facts, as it turned entirely on a concession made by Nova at trial.

It is nonetheless clear that “but for” causation is not the exclusive causation concept in identifying the NIO. Rowe J expressly rejected Nova’s submission that “an infringer’s ‘best non-infringing option’ is synonymous with the infringer’s ‘most profitable’ alternative option” [41]. Rowe J also endorsed the use of conventional canola as the appropriate non-infringing option in Rivett, while emphasizing that conventional canola was in fact unavailable to the infringer: [64]. Rowe J also rejected a hypothetical proposed by the intervener CGPA Canadian Generic Pharmaceutical Association which clearly applied “but for” causation: [41]. Further, the structure of the three-step test implicitly rejects “but for” causation as the sole causation concept, because Step 2 asks “whether” there is an NIO that can help isolate the profits causally attributable to the invention; if “but for” causation were the causation concept being used, then there is always a non-infringing option, namely what the infringer would in fact have done but for the infringement.

This makes it clear that the NIO is not determined solely by finding what the infringer would in fact have done but for the infringement. However, none of the foregoing excludes the possibility that “but for” causation might be helpful in identifying the NIO in some circumstances. Other aspects of the decision imply that “but for” causation may indeed play a role in identifying the NIO at Step 2. As discussed here, Rowe J suggested that even on the facts in Nova v Dow, commodity grade plastic might have been an appropriate comparator if Nova had not made its concession to the contrary at trial [72]. This is consistent with his definition of the NIO as any product that “helps” identify the profits caused by the invention [58]. Further, Rowe J also relied on “but for” causation in holding that springboard profits are recoverable [81]. This shows that “but for” causation is not entirely excluded from the accounting analysis.

Putting all this together, it is clear that the causation concept at issue at Step 2 is not “but for” causation as such; that is, the NIO is not to be identified simply by asking what the infringer would have done but for the infringement. At the same time, “but for” causation is not entirely irrelevant either; what would have happened but for the infringement might be used at Step 2 if the trial judge finds it helpful in identifying the profits causally attributable to the invention.

For example, suppose there is a case in which documentary evidence is introduced showing that the infringer had narrowed down its expansion plans to two options, Option A, which involved novel technology (which turned out to be infringing) and Option B, which is within the tried and tested technical capability and expertise of the company, but which is otherwise unrelated to Option A. A PowerPoint presentation from the meeting at which the decision was made is introduced into evidence. It shows that the infringer’s expected profit is $30m under Option A and $20m under Option B; and that even Option B is more profitable than another other course of action. A summary slide says:

Choice narrowed to Option A or Option B.

In favour of Option A

The value to us of the novel technology is $10m.

In favour of Option B

Tried and tested technology

If Option A turned out to be infringing, it seems reasonably clear that it would be open to a trial judge, applying Nova v Dow, to find this evidence helpful in assessing the value of Option A to the infringer. Certainly that would be true if Option B was similar to Option A, either technically or from a consumer market perspective (as discussed in more detail in the forthcoming paper), but Rowe J’s reference to the “pail and crate” market implies that it would also be open to the court to find such evidence helpful even if Option B is very different from Option A.

Of course, I have constructed this to be a particularly compelling example where “but for” causation seems helpful in establishing the value of the non-infringing option; but that is enough to establish the principle.

So, I would suggest that evidence as to what the infringer would have done but for the infringement should normally be admissible and relevant as one factor in determining what constitutes the NIO, even though it will not be determinative. I will consider other potentially relevant factors in my full article, which is forthcoming in the IPJ, and in subsequent posts.