The case is Dow Chemical Co. v. Nova Chems. Corp., 2017 FC 350 (Apr. 19, 2017), decision by Mr. Justice Fothergill. In an earlier proceeding, Mr. Justice O'Keefe found the patent in suit, which relates to film-grade polymers, valid and infringed. The more recent proceeding focused on the monetary award, and addressed several interesting issues. Norman Siebrasse has published detailed posts on these issues on Sufficient Description (see here, here, here, and here), and I'll just summarize the important issues here.
1. First, Dow sought an award of "reasonable compensation" under section 55(2) of the Canadian Patent Act for Nova's pre-grant use of what later issued as the '705 Patent. Section 55(2) reads:
A person is liable to pay reasonable compensation to a patentee and to all persons claiming under the patentee for any damage sustained by the patentee or by any of those persons by reason of any act on the part of that person, after the application for the patent became open to public inspection under section 10 and before the grant of the patent, that would have constituted an infringement of the patent if the patent had been granted on the day the application became open to public inspection under that section.
As I discuss in my book, many countries award some form of compensation for the unauthorized use of what later issues as a patented invention, for the period of time between publication and grant (though to my knowledge there aren't too many published decisions addressing the question of how such awards should be calculated). The counterpart in the U.S., for example, is Patent Act § 154(b), which (unlike the Canadian statute) expressly conditions such an award on the defendant having had "actual notice of the published patent application." (According to the Federal Circuit, however, "actual notice" effectively means "actual knowledge," and doesn't necessarily require an affirmative act on the part of the patent owner. For my critique of the decision reaching this conclusion, see here.) Anyway, the court in Dow concluded that "reasonable compensation" means a reasonable royalty, which in turn implies use of a hypothetical bargaining construct. The problem in the present case, however, was that in the but-for world in which the defendant had sought permission to use the invention, no bargain would have been reached, because the minimum amount Dow would have accepted exceeded the maximum amount Nova would have agreed to pay--a not uncommon circumstance, given that often the profit margin of a practicing patent owner with an effective monopoly will exceed the profit margin of the would-be user under duopoly. (In the but-for world, in other words, the plaintiff would have excluded the defendant and denied permission to use, rather than granting a license.) A lost profits award therefore would make more economic sense than a reasonable royalty, but according to Professor Siebrasse there is other Canadian precedent holding (rightly or wrongly) that courts can't award lost profits as such for pre-infringement use, only a royalty. The Dow court gets around this dilemma by holding that the royalty should be Dow's minimum willingness to accept, in effect granting a lost profit without calling it that. This would appear to me to be the correct economic result under the circumstances.
2. For the period covering the issuance of the patent through the end of the year 2015, Dow opted for an accounting of the defendant's profits. One interesting issue here was whether the award of profits could include a "springboard" award, that is, an award reflecting in part the additional the profit the defendant was able to earn post patent expiration, because the infringement provided it with a springboard to market entry (earlier market entry than would have been possible absent the infringement). The court held that this was permissible under Canadian law, stating that "springboard damages are nothing more than a type of loss to be proven with evidence, and I see no reason why this principle should operate differently to a plaintiff’s gains in the context of an accounting of profits" (para. 124). The second difficult issue was whether the defendant should be able to deduct any portion of its fixed costs, which would result in a reduction of the profit it would owe to Dow. Courts throughout the world have reached differing conclusions on this issue (see, e.g., my book pp. 206-08, 271-73, 325-26), and while there are reasonable economic arguments in support of both sides of the issue I'm inclined to agree with Stephen Margolis that the better rule is to allow such a deduction at least where the defendant would have employed that overhead to other purposes absent the infringement. In any event, in the present case the court held that it would be appropriate under the circumstances to allow Nova to deduct its apportionable overhead.
3. A third set of issues related to prejudgment interest. With regard to the pre-infringement royalty, the court used Dow's internal rate of return as the benchmark, rather than the otherwise applicable bank rate, which seems correct if the goal is to restore the plaintiff to the position it would have occupied but for the infringement. In keeping with Mr. Justice O'Keefe's ruling in the earlier proceeding, however, the court did not award compound interest (which of course is not consistent with the goal of compensation, though Canada is hardly alone in its reluctance to award compound interest). With regard to the award of profits, the court used the defendant's rate of return and did award compound interest (which seems more correct as a matter of economics).
4. Finally, the court had to decide what date to use to convert the award from U.S. to Canadian dollars, and it opted for the date of judgment. I think this generally makes sense for reasons of administrative convenience, rather than converting the award on a year-by-year (or even more granular) basis in cases in which the infringement spans a lengthy period of time.