Nova Chemicals Corp. v. Dow Chemical Co., 2022 SCC 43

SUPREME COURT OF CANADA
Citation: Nova Chemicals Corp. v. Dow Chemical Co., 2022 SCC 43			Appeal Heard: April 20, 2022 Judgment Rendered: November 18, 2022 Docket: 39439
Between:   Nova Chemicals Corporation Appellant   and   The Dow Chemical Company, Dow Global Technologies Inc. and Dow Chemical Canada ULC Respondents   - and -   Bell Canada, Rogers Communications Canada Inc., TELUS Communications Inc., Vidéotron ltée and Canadian Generic Pharmaceutical Association Interveners       Coram: Wagner C.J. and Moldaver, Karakatsanis, Côté, Brown, Rowe, Martin, Kasirer and Jamal JJ.
Reasons for Judgment: (paras. 1 to 87)	Rowe J. (Wagner C.J. and Moldaver, Karakatsanis, Brown, Martin, Kasirer and Jamal JJ. concurring)
Dissenting Reasons: (paras. 88 to 227)	Côté J.

 

 

 

Note: This document is subject to editorial revision before its reproduction in final form in the Canada Supreme Court Reports.

 

 

Nova Chemicals Corporation                                                                        Appellant

v.

The Dow Chemical Company,

Dow Global Technologies Inc. and

Dow Chemical Canada ULC                                                                    Respondents

and

Bell Canada,

Rogers Communications Canada Inc.,

TELUS Communications Inc.,

Vidéotron ltée and

Canadian Generic Pharmaceutical Association                                        Interveners

Indexed as: Nova Chemicals Corp. v. Dow Chemical Co.

2022 SCC 43

File No.: 39439.

2022: April 20; 2022: November 18.

Present: Wagner C.J. and Moldaver, Karakatsanis, Côté, Brown, Rowe, Martin, Kasirer and Jamal JJ.

on appeal from the federal court of appeal

                    Intellectual property — Patents — Infringement — Remedies — Accounting of profits — Non‑infringing option — Springboard profits — Patentee permitted to seek accounting of profits following successful infringement claim against infringer — Reference judge awarding patentee sum equal to infringer’s actual revenue selling patented product minus its actual full costs and also awarding springboard profits — Court of Appeal upholding reference judge’s award — Whether lower courts erred in calculating infringer’s profits under accounting of profits — Whether theoretical profits that infringer could have earned by selling unrelated product can be considered as non‑infringing option in calculation of profits that infringer must disgorge — Whether patentee entitled to springboard profits.

                    Nova made and sold products covered by Dow’s patent for metallocene linear low‑density polyethylenes, which are thin but strong plastics. The Federal Court permitted Dow to seek an accounting of profits to be assessed by reference. The reference judge awarded Dow a sum equal to Nova’s actual revenue from selling the patented plastics minus its actual full costs associated with producing the patented plastics. Accordingly, Nova was only permitted to subtract its actual cost of producing ethylene, the main ingredient in the patented plastics, not the higher market price of ethylene. The reference judge also concluded that Dow was entitled to “springboard profits” — profits that arise post‑patent‑expiry but that are causally attributable to infringement of the invention during the period of patent protection. The Federal Court of Appeal upheld the entirety of the award to Dow.

                    Held (Côté J. dissenting): The appeal should be dismissed.

                    Per Wagner C.J. and Moldaver, Karakatsanis, Brown, Rowe, Martin, Kasirer and Jamal JJ.: When a patentee chooses an accounting of profits as a remedy for patent infringement, the infringer must disgorge all the profits they gained that are causally attributable to the invention. This may involve consideration of the hypothetical profits that an infringer could have earned by selling a non‑infringing option. Springboard profits may also be available.

                    An accounting of profits requires that the infringer disgorge all profits causally attributable to infringement of the invention after the grant of the patent. An accounting of profits is sometimes presented as a choice between the (1) differential costs, (2) full costs, and (3) differential profits approaches. Under the differential costs and full costs approaches, the infringer is required to disgorge the difference between actual revenues earned by selling the infringing product and the actual costs associated with producing the infringing product; under the differential profits approach, the infringer is required to disgorge the difference between the actual profits earned by selling the infringing product and the profits it could have earned had it sold the best non‑infringing option. But these approaches are not completely distinct as all three start with determining the infringing product’s actual revenues and costs. It is therefore more appropriate to conceptualize an accounting of profits as a three‑step test. At step 1, the court should calculate the actual profits earned by selling the infringing product — i.e., revenue minus (full or differential) costs. At step 2, the court should determine whether there is a non‑infringing option that can help isolate the profits causally attributable to the invention from the portion of the infringer’s profits not causally attributable to the invention — i.e., differential profits. It is at this step that judges should apply the principles of causation. At step 3, if there is a non‑infringing option, the court should subtract the profits the infringer could have made had it used the non‑infringing option from its actual profits, to determine the amount to be disgorged.

                    When calculating the infringer’s profits in Step 1, courts should only consider actual revenues and costs. If an infringer is an inefficient manufacturer, and, as a result, makes less profit than theoretically possible, the patentee cannot claim profits that the infringer “should have” made, but if the infringer is an efficient manufacturer, the patentee is entitled to all the profits actually made, even if the patentee could not have achieved similar profit levels.

                    At Step 2, a “non‑infringing option” is any product that helps courts isolate the profits causally attributable to the invention from the profits not causally attributable to the invention. It is not an infringer’s “most profitable” alternative sales product that it “would have” and “could have” sold had it not infringed. The latter approach is flawed for two reasons. First, it is grounded on the incorrect premise that an accounting of profits is designed to ensure that the infringer ends up no worse off than had it never infringed. Instead, the goal of an accounting of profits is to ensure that all profits causally attributable to the invention are disgorged to the patentee. Second, this reading of non‑infringing options would distort the purpose of an accounting of profits and, in turn, undermine the patent bargain underlying the Patent Act . If an infringer is allowed to use any prior profitable business venture as a non‑infringing option, an infringer would always be incentivized to switch its business capacity to a more profitable infringing product, creating a form of business insurance for infringers: an infringer could always use their previous product lines as a non‑infringing option and protect those profits in the event their new product infringes a patent.

                    Whether there is a non‑infringing option is a question of fact. There are no strict rules around this factual exercise, and the non‑infringing option need not be a strict market substitute for the patented product. The onus is on the infringer to adduce sufficient evidence to satisfy the court that the profits from its infringing product arose by virtue of features other than the patentee’s invention and that there is a non‑infringing option that can help the courts isolate this value.

                    Springboard profits are legally permissible. An accounting of profits requires that the infringer disgorge all profits causally attributable to infringement of the invention, regardless of when the profits materialize, including profits an infringer earns because of an accelerated entry into the market. A patent provides a time‑limited monopoly to make, use, and sell an invention during the life of the patent. This monopoly gives the patentee the right to build sales capacity and market share without any market competition. The patentee can then use this market advantage against competitors after the patent expires. An infringer that begins selling a patented invention before the patent expires can build sales capacity and market share for their own version of the patented product. Then, after the patent expires, the infringer can use this sales capacity and market share to earn profits that it would not have earned but for the infringing activity that occurred during the life of the patent. A portion of such post‑expiry profits may be causally attributable to infringement of the invention. Failing to disgorge those profits would leave gains that are causally attributable to infringement of the invention in the hands of the infringer and would be unfair to third parties that waited for patent expiry to compete with the patentee.

                    In the instant case, the reference judge did not err in refusing to deduct the market price of ethylene as a cost that Nova incurred by making the patented plastics. Nor did the reference judge err when he concluded that all the profits Nova earned by selling the patented plastics were causally attributable to Dow’s invention. His conclusion is supported by two facts. First, he found that customers only purchased Nova’s infringing plastics because they contained the features captured by Dow’s patent. Second, Nova did not establish that there were relevant non‑infringing options that would help the court isolate the profits causally attributable to Dow’s invention from profits attributable to non‑inventive features of the infringing product. Nova bore the evidentiary onus to establish this fact. Instead, Nova conceded before the reference judge and again before the Federal Court of Appeal that there were no non‑infringing options available for the purpose of applying the differential profits approach. As the reference judge did not misapply the law, there is no basis for sending the matter back for redetermination.

                    Nor is there any basis to interfere with the decision to award springboard profits to Dow. By infringing Dow’s patent, Nova entered the market created by the invention early, built market share, and used that market advantage to earn profits post‑patent‑expiry that were causally attributable to infringement of the invention, during the period of patent protection. The reference judge found no support for Nova’s argument that its springboard profits were already accounted for by its payment of pre‑grant reasonable royalties, and Nova conceded that there were no non‑infringing options that would reduce the springboard profits award.

                    Per Côté J. (dissenting): The appeal should be allowed. The focus of an accounting of profits in the patent context must be on the infringer’s profits that are causally attributable to its infringement. This is determined using “but for” causation. This approach is consistent with the gain‑based nature of the remedy and ensures that the analysis remains unwaveringly focused on the infringer and the profits it earned — whether during the life of the patent or after its expiry — as a result of its infringement.

                    An accounting of profits, which focuses on the wrongdoer’s gains, seeks to restore the wrongdoer to the position it would have been in but for its wrongdoing. “But for” causation is therefore appropriate, as the focus is on restoring the wrongdoer to the position it would have occupied but for its wrongdoing. The remedy of an accounting of profits may serve either or both of two equitable purposes: a restorative purpose and a prophylactic deterrence purpose. While an accounting of profits can have a deterrent purpose in the patent context, as it can deter intentional and efficient infringements, the primary purpose of the remedy in this context is restorative. As in other contexts, a causal relationship between the infringement and the profits is required, but, in the patent context, there is an acute focus on causation: the patentee is entitled only to the profits causally attributable to the infringement of the patent. The differential profit approach is the preferred approach in this context because it best approximates the causal contribution of the infringement. It isolates the value of the patent in the hands of the infringer by comparing the infringer’s actual profits with the profits that the infringer could have and would have earned but for the infringement. Once the profits causally attributable to the infringement have been isolated, no other apportionment would appear to be necessary to restore the infringer to the position it would have occupied, but for the infringement.

                    Reading the case law as showing that the purpose of an accounting of profits is to isolate the value of the patent in the hands of the infringer is preferable for four reasons. First, the statement of the law in the leading cases underscored that the proper focus is on the infringement. Second, the application of the differential profit approach confirms that focus. Third, a narrow interpretation that focuses exclusively on the value of the invention is inconsistent with the proposition that the profits to be disgorged are those that, on a common sense view of causation, were caused by the breach. The phrase “a common sense view of causation” clarifies that “but for” causation applies. Fourth, a focus on the infringement is necessary to capture all of the potential profits that the infringer may derive from its breach of the patentee’s statutory monopoly. The advantage of the patented technology over the non‑infringing options does not turn solely on whatever value we might decide is attributable to the inventive quality of the technology; it turns on the actual difference in profitability.

                    Under the differential profits approach, a non‑infringing option does not have to be a true a consumer substitute for the patented product. There are two principled reasons for this. First, a true consumer substitute requirement is legally irrelevant in an accounting of profits given the different purposes and focus of this remedy. An accounting of profits is not compensatory, as the profits to be disgorged are calculated exclusively by reference to the defendant’s wrongful gain, irrespective of whether it corresponds to damage suffered by the plaintiff. It is therefore legally irrelevant whether consumers would view the product that the infringer could have and would have made in the hypothetical world as a substitute for the patented product. What matters is the profits the infringer derived from its breach of the patentee’s statutory monopoly. Second, limiting non‑infringing options to true consumer substitutes distorts the causation analysis. By focusing on the value of the patent in the abstract, the analysis ignores the value that the infringement brought to the infringer. It needlessly disadvantages infringers that could have and would have produced a different product that would not have been a consumer substitute for the patented product or that would have competed in a different consumer market. Conversely, it risks incentivizing efficient infringement by benefiting infringers that could not or would not otherwise have produced a consumer substitute for the patented product. Without any principled justification, it prevents the application of the differential profit approach, forcing an infringer to disgorge gains that are not causally attributable to its infringement.

                    Courts are to determine the appropriate hypothetical “best non‑infringing option” to be used based on what an infringer could have and would have done but for the infringement. An examination of what the infringer could have done involves an objective assessment of what was feasible in the circumstances, while an examination of what the infringer would have done involves a more subjective assessment of its alternative course of conduct. In an accounting of profits case, the infringer’s best non‑infringing option can be, and likely often will be, a consumer substitute for the patented product, but the focus should be on what the infringer could have and would have done but for the infringement. The real world informs both the “could have” and the “would have” branches of the hypothetical analysis and the burden of establishing what the infringer could have and would have done lies on the infringer, although both sides can adduce evidence on this point.

                    It is of no moment that certain profits — known as springboard profits — causally attributable to infringement do not accrue until after the patent has expired. So long as the profits are causally attributable to the infringement, they remain subject to disgorgement.

                    In the instant case, upholding the conclusion that there was no non‑infringing option is problematic for three reasons. First, it is based on two factual findings which clearly show that the distinct consumer market remained central and legally relevant in the reference judge’s analysis. Second, it does not take into account the reference judge’s conclusion that the differential profit approach was legally unavailable because the non‑infringing option had to be a true substitute. Third, the majority is equivocal and unclear about the legal requirements for a non‑infringing option. Whether such an option exists cannot be a question of fact alone. It must be a question of mixed fact and law.

                    Even though the reference judge made specific findings about what Nova could have and would have done but for the infringement, he applied case law which required a true consumer substitute, thereby preventing the use of the differential profit approach. This distorted the causation analysis by preventing a comparison between Nova’s actual profits and its hypothetical profits had it not infringed, and forces Nova to disgorge millions of dollars in profits that have not been shown to be causally attributable to its infringement of Dow’s patent.

                    Nor should Nova’s concession that there were no direct non‑infringing alternatives to its infringing products deny it the relief it seeks. This concession was based on the understanding that a non‑infringing alternative had to be a true consumer substitute in order for the differential profit approach to apply. As the reference judge erred in law by holding that a non‑infringing option needed to be a true consumer substitute for the purposes of the differential profit approach, Nova’s concession is not determinative of the outcome.

                    Finally, Dow should be entitled to an accounting of all profits that Nova earned as a result of its infringement, including springboard profits. The causation analysis for any post‑expiry period should be based on the same hypothetical “but for” reasoning embodied in the differential profit approach. However, on the record available, the Court cannot determine how to apply the differential profit approach to the facts, and the springboard profits were not properly calculated. The application of the differential profit approach and the calculation of springboard profits should be remitted to the Federal Court.

Cases Cited

By Rowe J.

                    Applied: Monsanto Canada Inc. v. Schmeiser, 2004 SCC 34, [2004] 1 S.C.R. 902; considered: Monsanto Canada Inc. v. Rivett, 2009 FC 317, [2010] 2 F.C.R. 93, rev’d in part 2010 FCA 207, [2012] 1 F.C.R. 473; Monsanto Canada Inc. v. Janssens, 2009 FC 318, 343 F.T.R. 234, rev’d in part 2010 FCA 207, [2012] 1 F.C.R. 473; Lubrizol Corp. v. Imperial Oil Ltd., [1997] 2 F.C. 3; referred to: AlliedSignal Inc. v. Du Pont Canada Inc. (1998), 78 C.P.R. (3d) 129; Unilever PLC v. Procter & Gamble Inc. (1993), 47 C.P.R. (3d) 479; Apotex Inc. v. ADIR, 2020 FCA 60, 172 C.P.R. (4th) 1; AlliedSignal Inc. v. Du Pont Canada Inc. (1995), 61 C.P.R. (3d) 417; Bayer Inc. v. Cobalt Pharmaceuticals Co., 2016 FC 1192, 142 C.P.R. (4th) 374; Seedlings Life Science Ventures, LLC v. Pfizer Canada ULC, 2021 FCA 154; Beloit Canada Ltée v. Valmet Oy (1994), 55 C.P.R. (3d) 433; Merck & Co., Inc. v. Apotex Inc., 2015 FCA 171, [2016] 2 F.C.R. 202, aff’g 2013 FC 751, [2015] 1 F.C.R. 405; Snell v. Farrell, [1990] 2 S.C.R. 311; Apotex Inc. v. Sanofi‑Synthelabo Canada Inc., 2008 SCC 61, [2008] 3 S.C.R. 265; Teva Canada Ltd. v. Pfizer Canada Inc., 2012 SCC 60, [2012] 3 S.C.R. 625; Apotex Inc. v. Wellcome Foundation Ltd., 2002 SCC 77, [2002] 4 S.C.R. 153; Atlantic Lottery Corp. Inc. v. Babstock, 2020 SCC 19; Bayer Aktiengesellschaft v. Apotex Inc. (2001), 10 C.P.R. (4th) 151; Laboratoires Servier v. Apotex Inc., 2008 FC 825, 67 C.P.R. (4th) 241, aff’d 2009 FCA 222, 75 C.P.R. (4th) 443; Cinar Corporation v. Robinson, 2013 SCC 73, [2013] 3 S.C.R. 1168; AFD Petroleum Ltd. v. Frac Shack Inc., 2018 FCA 140, 157 C.P.R. (4th) 195; AstraZeneca Canada Inc. v. Apotex Inc., 2015 FC 671; Teva Canada Limited v. Janssen Inc., 2018 FCA 33, 420 D.L.R. (4th) 493; Apotex Inc. v. Eli Lilly and Company, 2018 FCA 217, 161 C.P.R. (4th) 411.

By Côté J. (dissenting)

                    Warman International Ltd. v. Dwyer (1995), 182 C.L.R. 544; Monsanto Canada Inc. v. Schmeiser, 2004 SCC 34, [2004] 1 S.C.R. 902; Apotex Inc. v. Wellcome Foundation Ltd., 2002 SCC 77, [2002] 4 S.C.R. 153; Teva Canada Ltd. v. Pfizer Canada Inc., 2012 SCC 60, [2012] 3 S.C.R. 625; Free World Trust v. Électro Santé Inc., 2000 SCC 66, [2000] 2 S.C.R. 1024; Bristol‑Myers Squibb Co. v. Canada (Attorney General), 2005 SCC 26, [2005] 1 S.C.R. 533; Merck & Co., Inc. v. Apotex Inc., 2015 FCA 171, [2016] 2 F.C.R. 202; Lubrizol Corp. v. Imperial Oil Ltd., [1997] 2 F.C. 3; Atlantic Lottery Corp. Inc. v. Babstock, 2020 SCC 19; Southwind v. Canada, 2021 SCC 28; Strother v. 3464920 Canada Inc., 2007 SCC 24, [2007] 2 S.C.R. 177; Beloit Canada Ltd. v. Valmet‑Dominion Inc., [1997] 3 F.C. 497; Bayer Aktiengesellschaft v. Apotex Inc. (2002), 16 C.P.R. (4th) 417; Bayer Inc. v. Cobalt Pharmaceuticals Co., 2016 FC 1192, 142 C.P.R. (4th) 374, aff’d 2018 FCA 32, [2018] 4 F.C.R. 58; AlliedSignal Inc. v. Du Pont Canada Inc. (1995), 61 C.P.R. (3d) 417; Pfizer Canada Inc. v. Teva Canada Ltd., 2016 FCA 161, 400 D.L.R. (4th) 723; AFD Petroleum Ltd. v. Frac Shack Inc., 2018 FCA 140, 157 C.P.R. (4th) 195; Apotex Inc. v. Eli Lilly and Company, 2018 FCA 217, 161 C.P.R. (4th) 411; Apotex Inc. v. ADIR, 2017 FCA 23, 406 D.L.R. (4th) 572; Apotex Inc. v. ADIR, 2020 FCA 60, 172 C.P.R. (4th) 1; Pettkus v. Becker, [1980] 2 S.C.R. 834; Philip Morris Products S.A. v. Marlboro Canada Ltd., 2015 FC 364, [2015] F.C.J. No. 1564 (QL); Hodgkinson v. Simms, [1994] 3 S.C.R. 377; Canadian Aero Service Ltd. v. O’Malley, [1974] S.C.R. 592; Lac Minerals Ltd. v. International Corona Resources Ltd., [1989] 2 S.C.R. 574; Canson Enterprises Ltd. v. Boughton & Co., [1991] 3 S.C.R. 534; Cadbury Schweppes Inc. v. FBI Foods Ltd., [1999] 1 S.C.R. 142; ADIR v. Apotex Inc., 2015 FC 721, 482 F.T.R. 276; Reading & Bates Construction Co. v. Baker Energy Resources Corp., [1995] 1 F.C. 483; Constellation Brands US Operations Inc. v. Société de vin internationale ltée, 2021 QCCA 1664; Wellcome Foundation Ltd. v. Apotex Inc., [2001] 2 F.C. 618; Celanese International Corp. v. BP Chemicals Ltd, [1999] R.P.C. 203; Monsanto Canada Inc. v. Schmeiser, 2002 FCA 309, [2003] 2 F.C. 165; Housen v. Nikolaisen, 2002 SCC 33, [2002] 2 S.C.R. 235; Cinar Corporation v. Robinson, 2013 SCC 73, [2013] 3 S.C.R. 1168; Monsanto Canada Inc. v. Rivett, 2009 FC 317, [2010] 2 F.C.R. 93, rev’d in part 2010 FCA 207, [2012] 1 F.C.R. 473; Monsanto Canada Inc. v. Schmeiser, 2001 FCT 256, 12 C.P.R. (4th) 204.

Statutes and Regulations Cited

Patent Act , R.S.C. 1985, c. P‑4, ss. 42 , 55(1) , (2) , 57 .

Authors Cited

Andrews, Kurtis, and Jeremy de Beer. “Accounting of Profits to Remedy Biotechnology Patent Infringement” (2009), 47 Osgoode Hall L.J. 619.

Bently, L., et al. Intellectual Property Law, 5th ed. New York: Oxford University Press, 2018.

Charles, Michael E. “Monetary Remedies in Intellectual Property Litigation”, [2007] J. Bus. Valuation 159.

Cotter, Thomas F. Comparative Patent Remedies: A Legal and Economic Analysis. New York: Oxford University Press, 2013.

Devonshire, Peter. Account of Profits. Wellington, N.Z.: Brookers, 2013.

Devonshire, Peter. “Account of Profits for Breach of Fiduciary Duty” (2010), 32 Sydney L. Rev. 389.

Gergen, Mark P. “Causation in Disgorgement” (2012), 92 B.U.L. Rev. 827.

Hughes, Roger T., and Dino P. Clarizio. Hughes & Woodley on Patents, 2nd ed. Toronto: LexisNexis, 2005 (loose‑leaf updated March 2022, release 87).

Perry, Stephen J., and T. Andrew Currier. Canadian Patent Law, 4th ed. Toronto: LexisNexis, 2021.

Seaman, Christopher B., et al. “Lost Profits and Disgorgement”, in C. Bradford Biddle et al., eds., Patent Remedies and Complex Products: Toward a Global Consensus. New York: Cambridge University Press, 2019, 50.

Siebrasse, Norman. “A Remedial Benefit‑Based Approach to the Innocent‑User Problem in the Patenting of Higher Life Forms” (2004), 20 C.I.P.R. 79.

Siebrasse, Norman. Non‑Infringing Baseline as an Alternative to “But For” Causation, October 19, 2020 (online: http://www.sufficientdescription.com/2020/10/non-infringing-baseline-as-alternative.html; archived version: https://www.scc-csc.ca/cso-dce/2022SCC-CSC43_1_eng.pdf).

Siebrasse, Norman V., Alexander J. Stack and the Cole & Partners IP Litigation Support Group. “Accounting of Profits in Intellectual Property Cases in Canada (2007)” (2008), 24 C.I.P.R. 83.

Smith, Lionel, and Jeff Berryman. “Disgorgement of Profits in Canada”, in Ewoud Hondius and André Janssen, eds., Disgorgement of Profits: Gain‑Based Remedies throughout the World. New York: Springer, 2015, 281.

Snell’s Equity, 34th ed. by John McGhee and Steven Elliott. London: Thomson Reuters, 2020.

                    APPEAL from a judgment of the Federal Court of Appeal (Stratas, Near and Woods JJ.A.), 2020 FCA 141, [2021] 1 F.C.R. 551, 452 D.L.R. (4th) 318, 177 C.P.R. (4th) 145, [2020] F.C.J. No. 928 (QL), 2020 CarswellNat 3799 (WL), affirming a decision of Fothergill J., 2017 FC 350, [2018] 2 F.C.R. 154, [2017] F.C.J. No. 441 (QL), 2017 CarswellNat 1891 (WL). Appeal dismissed, Côté J. dissenting.

                    Andrew Bernstein, Sheila Block, Nicole Mantini and Jonathan Silver, for the appellant.

                    Steve Garland, Jeremy Want, Daniel Davies and Matthew Burt, for the respondents.

                    Audrey Boctor and Danielle Marcovitz, for the interveners Bell Canada, Rogers Communications Canada Inc., TELUS Communications Inc. and Vidéotron ltée.

                    Andrew Brodkin, Harry Radomski and Jordan Scopa, for the intervener the Canadian Generic Pharmaceutical Association.

The judgment of Wagner C.J. and Moldaver, Karakatsanis, Brown, Rowe, Martin, Kasirer and Jamal JJ. was delivered by

 

                   Rowe J. —

I.               Introduction

[1]                              This appeal concerns proper calculation of the quantum payable by the appellant (“Nova”) to the respondents (“Dow”) as a remedy for patent infringement. Central to this appeal is the patent law remedy labelled an “accounting of profits”. An accounting of profits requires that a party who infringes a patent (the “infringer”) disgorge all the profits they gained that are causally attributable to the invention.

[2]                              Nova asks this Court to deduct theoretical profits that it could have earned on an entirely unrelated product had it not infringed, from the profits that the company earned by selling products that infringed Dow’s patent. The difference between these values, Nova submits, should be the quantum payable to Dow.

[3]                              I disagree. While in certain circumstances the hypothetical profits that an infringer could have earned by selling a non-infringing option are relevant to a calculation of an accounting of profits, this is not such a case. A “non-infringing option” is any product that helps courts isolate the profits causally attributable to the invention from the profits which arose at the same time the infringing product was used or sold, but which are not causally attributable to the invention. Whether there is a relevant non-infringing option that can assist the court in this calculation is a question of fact. The infringer has the onus of establishing that there is a relevant non-infringing option. On the basis of the record before the reference judge, there were no relevant non-infringing options to consider. He therefore fairly determined that all of Nova’s profits were causally attributable to Dow’s invention. I see no reason to disturb this factual finding.

[4]                              Additionally, for the first time in Canadian law, the reference judge awarded “springboard profits” to Dow. Springboard profits are profits that arise post-patent-expiry but that are causally attributable to infringement of the invention during the period of patent protection. Nova appeals this award, which was upheld at the Federal Court of Appeal. On this point, too, Nova’s appeal must fail. Springboard profits are an extension of the fundamental principle that, in calculating an accounting of profits, the infringer must disgorge all profits causally attributable to infringement of the invention. It is irrelevant when the profits arise, provided they are causally connected to infringement. Like non-infringing options, whether profits that arise post-patent-expiry are causally attributable to infringement of the invention, during the period of patent protection, is a question of fact. On the basis of the record before him, the reference judge determined that some of Nova’s post-patent-expiry profits were causally attributable to infringement of the invention: by infringing Dow’s patent, Nova had entered the market created by the invention early, built market share, and used that market advantage to earn profits post-patent-expiry that were causally attributable to infringement of the invention during the period of patent protection. I see no reason to disturb his factual findings on this point.

[5]                              It follows that I would dismiss Nova’s appeal.

II.            Introduction to Patent Law Remedies

[6]                              To understand this decision, an introduction to three concepts is needed: (1) remedies in patent law; (2) the calculation methods for the remedy at issue in this appeal (an “accounting of profits”); and (3) this Court’s leading authority on calculating an accounting of profits (Monsanto Canada Inc. v. Schmeiser, 2004 SCC 34, [2004] 1 S.C.R. 902). I explain each below. After introducing these topics, I will (4) clarify the methods of calculating an accounting of profits.

A.           Remedies in Patent Law

[7]                              Below I define three patent law remedies and explain how they relate to each other. The three statutory remedies I define are (1) reasonable compensation, (2) damages, and (3) an accounting of profits.

Term	Definition
Reasonable Compensation	Reasonable compensation can be granted for any loss caused by the infringer’s use of the invention between the patent’s publication and the grant of the patent.   This remedy is authorized by s. 55(2)  of the Patent Act , R.S.C. 1985, c. P-4 .   This remedy typically entitles a patentee to a “reasonable royalty” (S. J. Perry and T. A. Currier, Canadian Patent Law (4th ed. 2021), at §§17.93-17.94).   A reasonable royalty is “that which the infringer would have had to pay if, instead of infringing the Patent, [the infringer] had come to be licensed under the Patent” (AlliedSignal Inc. v. Du Pont Canada Inc. (1998), 78 C.P.R. (3d) 129 (F.C. (T.D.)), at para. 199, quoting Unilever PLC v. Procter & Gamble Inc. (1993), 47 C.P.R. (3d) 479 (F.C. (T.D.)), at p. 571 (text in brackets in original)). “The test is what rate would result from negotiations between a willing licensor and a willing licensee” (para. 199).
Damages	Damages compensate the patentee for all pecuniary losses causally attributable to infringement after the grant of the patent.   This remedy is authorized by s. 55(1)  of the Patent Act .   Damages can include lost profits on sales or due to depression of prices, and lost income from licensing opportunities, among others (Perry and Currier, at §17.9).
Accounting of Profits	An accounting of profits requires that the infringer disgorge all profits causally attributable to infringement of the invention after the grant of the patent.   This remedy is authorized by s. 57(1) (b) of the Patent Act .   This remedy is an alternative to an award of damages (Apotex Inc. v. ADIR, 2020 FCA 60, 172 C.P.R. (4th) 1, at para. 35). It is an equitable, discretionary remedy (AlliedSignal Inc. v. Du Pont Canada Inc. (1995), 61 C.P.R. (3d) 417 (F.C.A.), at pp. 444-46). Judges may consider practical consequences, including expediency, misbehaviour by litigants, and whether the patentee practices the invention itself when exercising this discretion (K. Andrews and J. de Beer, “Accounting of Profits to Remedy Biotechnology Patent Infringement” (2009), 47 Osgoode Hall L.J. 619, at p. 641; Bayer Inc. v. Cobalt Pharmaceuticals Co., 2016 FC 1192, 142 C.P.R. (4th) 374, at paras. 6 and 10; Seedlings Life Science Ventures, LLC v.  Pfizer Canada ULC, 2021 FCA 154, at paras. 76 and 79-81 (CanLII)).

 

B.            Calculating an Accounting of Profits

[8]                              As previously stated, this appeal addresses the correct method of calculating an accounting of profits. Courts have calculated an accounting of profits in three ways: (1) differential costs, (2) full costs, and (3) differential profits. I explain each below.

Term	Definition
Differential Costs	The differential costs approach requires the infringer to disgorge the difference between actual revenues earned by selling the infringing product and the actual costs associated with producing the infringing product.   Infringers can deduct any expense incurred because of the infringing activity. Any category of expense, so long as it is directly attributable to the infringing activity, is deductible. These can be thought of as “direct” expenses: “. . . all expenses — variable, current, increased, fixed, or capital — that are directly attributable to the infringement, are deductible” (Andrews and de Beer, at p. 646).   Infringers cannot deduct expenses they would have incurred in the absence of infringement (Monsanto Canada Inc. v. Rivett, 2009 FC 317, [2010] 2 F.C.R. 93, at para. 30; Andrews and de Beer, at pp. 643-44).
Full Costs	The full costs approach also requires the infringer to disgorge the difference between actual revenues earned by selling the infringing product and the actual costs associated with producing the infringing product.   But, unlike the differential costs approach, the full costs approach allows infringers to deduct all direct costs and a portion of the indirect expenses incurred to make the infringing product.   Indirect expenses are costs which did not arise solely because the infringer manufactured the infringing goods (i.e., expenses that would have been incurred in the absence of infringement) (Rivett, at para. 32; Andrews and de Beer, at pp. 643 and 646).   Simply put, infringers can deduct more costs from their revenue (and thus reduce the profits payable to the patentee) under the full costs approach as opposed to the differential costs approach.
Differential Profits	The differential profits approach requires an infringer to disgorge the difference between the actual profits (revenue minus costs) earned by selling the infringing product and the profits it could have earned had it sold the best “non-infringing option” (Schmeiser, at para. 102).   As I will explain, the main issue in this appeal is the appropriate scope of a non-infringing option. I will expand on this concept in the next section.   The onus is on the infringer to prove that there is a non-infringing option (see Beloit Canada Ltée v. Valmet Oy (1994), 55 C.P.R. (3d) 433 (F.C. (T.D.)), at p. 456).

C.            Schmeiser

[9]                              Schmeiser is this Court’s leading authority on calculating an accounting of profits. The decision addressed an alleged patent infringement and the appropriate remedy flowing from that infringement. Most relevant to this decision was the Court’s acceptance of the differential profits approach and its use of a non-infringing option.

[10]                          By way of background, Monsanto brought a patent infringement claim against Mr. Schmeiser. Monsanto held a patent on a genetic modification to canola seeds, marketed as Roundup Ready Canola. The genetic modification to the canola “dramatically” increased the plant’s tolerance to herbicides containing glyphosate (para. 8). Farmers could spray the crop with a glyphosate herbicide after the plants had emerged from the soil without killing the crop.

[11]                          This Court determined that Monsanto’s patent was valid and that Mr. Schmeiser infringed that patent by planting Roundup Ready Canola without a licence. Monsanto requested and was granted an accounting of profits as a remedy. The company argued Mr. Schmeiser was required to disgorge profits he earned from selling the harvested canola seeds.

[12]                          This Court disagreed. It stressed it was “settled law” that, under an accounting of profits, a patentee “is only entitled to that portion of the infringer’s profit which is causally attributable to the invention” (para. 101). The preferred means of determining this was the “value-based or ‘differential profit’ approach” (para. 102). To determine which portion of Mr. Schmeiser’s profits were causally attributable to Monsanto’s invention, this Court compared the profits Mr. Schmeiser earned from the sale of Roundup Ready canola with the profits he could have earned had he used the best non-infringing option: regular, non-genetically modified canola seeds. The difference between these two values was zero: Mr. Schmeiser did not spray the crop with a glyphosate herbicide and, therefore, did not benefit from the increased yields that Monsanto’s patent offered. The entirety of Mr. Schmeiser’s profits were attributable to non-patented features of the product — the canola seed itself. None of the profits he earned were causally attributable to the invention.

[13]                          From Schmeiser, it can be understood that a non-infringing option is any product that helps courts isolate the profits causally attributable to the invention from the profits which arose at the same time the infringing product was used or sold, but which are not causally attributable to the invention. As stated by the Court, a patentee is only entitled to profits “causally attributable to the invention” (para. 101).

D.           Simplified Method of Calculating an Accounting of Profits

[14]                          An accounting of profits is sometimes presented as a choice between the (1) differential costs, (2) full costs, and (3) differential profits approaches (see Rivett, at para. 28). But these approaches are not completely distinct. The starting point of analysis for all three methods is determining the infringing product’s actual revenues and costs. Even in the differential profits approach, courts must first calculate the infringer’s actual profits earned by selling the infringing product (i.e., deduct the actual cost of producing the infringing product from actual revenue). This calculation must precede the comparison between actual profits earned on the infringing product and the profits an infringer could have earned had they sold the best non-infringing option.

[15]                          It is therefore more appropriate to conceptualize an accounting of profits as a three-step test:

Step 1: Calculate the actual profits earned by selling the infringing product — i.e., revenue minus (full or differential) costs.

Step 2: Determine whether there is a non-infringing option that can help isolate the profits causally attributable to the invention from the portion of the infringer’s profits not causally attributable to the invention — i.e., differential profits. It is at this step that judges should apply the principles of causation. Causation “need not be determined by scientific precision: it is ‘essentially a practical question of fact which can best be answered by ordinary common sense’” (Merck & Co., Inc. v. Apotex Inc., 2015 FCA 171, [2016] 2 F.C.R. 202, at para. 44, quoting Snell v. Farrell, [1990] 2 S.C.R. 311, at para. 328).

Step 3: If there is a non-infringing option, subtract the profits the infringer could have made had it used the non-infringing option from its actual profits, to determine the amount to be disgorged.

[16]                          Step 2 is the principal issue in this appeal. As I will set out, we are called on to clarify what is a non-infringing option. We are not called on to resolve whether differential costs or full costs is the preferred method for calculating an infringer’s costs in Step 1. While this was dealt with in the decision under appeal, we did not receive submissions on this issue (see 2020 FCA 141, [2021] 1 F.C.R. 551, at paras. 143-64). Accordingly, I do not deal further with this issue.

III.         Facts

[17]                          Nova and Dow are competitors in the plastics industry. In 1994, Dow filed a patent for new plastics (“patented plastics”). The patented plastics were metallocene linear low-density polyethylenes that had superior strength and processability characteristics as compared to conventional linear low-density polyethylenes. In plain language, the patented plastics are thin but strong plastics, used in items like garbage bags and food wrappings. The patented plastics serviced a market demand for items with these characteristics. Dow’s patent was issued in 2006 and expired in 2014.

[18]                          Nova made and sold plastics covered by Dow’s patent. Dow brought an infringement claim against Nova. In the infringement proceedings, Nova challenged the validity of Dow’s patent on a variety of grounds. The Federal Court rejected these arguments, holding that the patent was valid and that Nova had infringed the patent (2014 FC 844, 129 C.P.R. (4th) 199, aff’d 2016 FCA 216, 142 C.P.R. (4th) 339). As a remedy, Dow was permitted to seek an accounting of profits, to be assessed by a subsequent reference.

[19]                          The infringement decision is not at issue. This appeal relates only to the remedy.

IV.         Decisions Below

A.           Federal Court, 2017 FC 350, [2018] 2 F.C.R. 154

[20]                          The main issue before the reference judge was the proper approach to assessing the quantum of profit that Nova had to disgorge as a result of infringement. Nova argued it should be permitted to subtract the market cost of ethylene from the revenues it earned selling the patented plastics.

[21]                          Ethylene is the main ingredient in the patented plastics. Nova produces its own ethylene at a cost that is lower than market price. Consequently, Nova’s cost to manufacture the patented plastics was significantly less than if the company had purchased the ethylene at market price.

[22]                          The reference judge rejected Nova’s argument. He held that only the actual costs incurred to produce the patented plastics were appropriately deductible in Step 1. The market price of ethylene was not a “cost” that Nova incurred (paras. 137-40).

[23]                          Notably, Nova did not argue that selling ethylene at market prices was a non-infringing option for the purposes of the differential profits approach. Nova conceded that there were no direct non-infringing options to the patented plastics (para. 146).

[24]                          Instead, in the alternative, Nova argued that the reference judge should deduct the “full cost” it incurred to manufacture the patented plastics (para. 134). The reference judge accepted this argument. The reference judge awarded Dow a sum equal to Nova’s actual revenue from selling the patented plastics minus its actual full costs (as described in para. 8 above) associated with producing the patented plastics.

[25]                          Also relevant to this appeal is the reference judge’s decision to award springboard profits. Dow argued that some of the profits Nova earned by selling the patented plastics after patent expiry were causally attributable to Nova’s patent infringement. Had Nova not infringed the patent, it would have taken time to reach the same level of sales that it enjoyed after the patent’s expiry. As a result, Dow argued, these profits were causally attributable to Nova’s infringement and, accordingly, had to be disgorged.

[26]                          Nova advanced two arguments in response. First, it argued that springboard profits could not be awarded under Canadian law. Second, it argued, in the alternative, that, if springboard profits were permissible, then an award of reasonable compensation under s. 55(2) (as described in para. 7 above) was sufficient to account for the company’s post-patent-expiry sales.

[27]                          The reference judge concluded that Dow was entitled to springboard profits. Some of Nova’s post-patent-expiry profits were causally attributable to Nova’s infringement of the invention during the period of patent protection and they were not covered by the compensation award under s. 55(2) (paras. 123 and 130). Those profits had to be disgorged. Regardless of when profits materialize, the law demands that infringers disgorge all profits causally attributable to infringement of the invention.

B.            Federal Court of Appeal, 2020 FCA 141, [2021] 1 F.C.R. 551

[28]                          On appeal, Nova again argued that the court should use the market price of ethylene, rather than the actual costs it incurred to manufacture ethylene, when calculating the profits payable to Dow. However, the company’s submissions as to why this was the correct calculation differed from those at trial. 

[29]                          At trial, Nova asked to deduct the market price of ethylene as a “cost”. On appeal, Nova advanced two new arguments. First, Nova argued that, had it not used the ethylene to make the patented plastics, it could have earned a profit by selling that ethylene to third parties. It was therefore entitled to deduct the profits it could have earned by selling ethylene to third parties from the profits payable to Dow (paras. 90˗92). Second, and in the alternative, Nova argued that a portion of the profits earned by selling the patented plastics were due to its own manufacturing efficiencies, rather than Dow’s patent. The court should thus apportion those profits out of the amount awarded to Dow (para. 98).

[30]                          Finally, Nova argued that springboard profits were inappropriate.

[31]                          A majority of the Federal Court of Appeal rejected Nova’s arguments and upheld the entirety of the award to Dow.

[32]                          The dissenting judge disagreed on the issue of whether Nova could reduce the profits payable to Dow because it manufactures ethylene at below-market cost. In her view, Nova did not need to disgorge the profits attributable to its below-market-cost manufacturing of ethylene because those profits were not caused by making and selling the patented plastics. The profits earned from ethylene therefore needed to be “apportioned” from the profits earned by selling the patented plastics.

V.           Issues

[33]                          This appeal raises two issues:

(1)                   Did the lower courts err in calculating the profits payable to Dow under an accounting of profits?

(2)                  Is Dow entitled to springboard profits?

VI.         Analysis

A.           Issue 1: Did the Lower Courts Err in Calculating the Profits Payable to Dow Under an Accounting of Profits?

[34]                          Before this Court, Nova again changed its argument as to why the quantum awarded to Dow was inappropriately high. While previously the company focused its arguments on “Step 1” of the accounting of profits calculation, before this Court it focused on “Step 2”.

[35]                          My analysis of this issue proceeds as follows. First, I explain why the reference judge did not err in refusing to deduct the market price of ethylene as if it were a cost that Nova had incurred. Second, I provide a brief summary of Nova’s new arguments before this Court. In the third and fourth sections, I set out the conceptual underpinnings of an accounting of profits and non-infringing options, respectively. An understanding of these topics is necessary to assess Nova’s new arguments. Fifth, I apply these principles to the facts of this case and explain why Nova’s arguments must fail.

(1)          The Market Price of Ethylene Was Not a Deductible Cost

[36]                          Nova effectively conceded before this Court that the reference judge did not err in refusing to deduct the market price of ethylene as a true cost associated with making the patented plastics. This was sensible. When calculating the infringer’s profits in Step 1, courts should consider only actual revenues and costs.

[37]                          This conclusion flows from the principle that a patentee must take the infringer as they find them. If an infringer is an inefficient manufacturer, and, as a result, makes less profit than theoretically possible, the patentee cannot claim profits that the infringer “should have” made (N. V. Siebrasse, A. J. Stack et al., “Accounting of Profits in Intellectual Property Cases in Canada (2007)” (2008), 24 C.I.P.R. 83, at p. 87). The opposite is also true. If an infringer is an efficient manufacturer, the patentee is entitled to all the profits actually made, even if the patentee could not have achieved similar profit levels. Accordingly, the reference judge did not err by refusing to deduct the market price of ethylene as a “cost” in Step 1 (F.C. reasons, at paras. 139-40).

(2)          Nova’s New Argument: Pail and Crate Plastics Were Valid Non-infringing Options

[38]                          Nova advances new arguments before this Court, which focus on Step 2. Nova argues that had it not manufactured the patented plastics, it would have used its manufacturing capacity to make and sell entirely different plastics: high˗density polyethylenes, which are inflexible plastics used to make items like pails, crates, and buckets (“pail and crate plastics”). The sales of pail and crate plastics would have generated profit that Nova argues is properly deductible from the profits earned by infringing.

[39]                          The pail and crate plastics markets and patented plastics markets do not overlap. A potential buyer interested in the patented plastics would not purchase pail and crate plastics as an alternative (F.C. reasons, at paras. 146-49; F.C.A. reasons, at para. 97). Nevertheless, Nova says the only way to determine which of its profits are causally attributable to Dow’s invention is by subtracting the profits it would have made had it sold pail and crate plastics from the profits it made from the patented plastics. Nova submits that this approach follows from Schmeiser.

[40]                          As discussed, Schmeiser held that the differential profits approach is the “preferred” means of determining what profits are causally attributable to the invention. Under differential profits, courts subtract the profits the infringer could have made using the “best non-infringing option” from the “profit[s] attributable to the invention” (Schmeiser, at para. 102). The patentee is entitled to the difference between these two sums.

[41]                          Nova submits that an infringer’s “best non-infringing option” is synonymous with the infringer’s “most profitable” alternative option. Under their reading of Schmeiser, courts would determine the profits payable to the patentee by subtracting (a) the profits the infringer could have made had they pursued any other alternative course of conduct from (b) the infringing product profits (i.e., profits payable = (b) minus (a)). My colleague endorses Nova’s approach, as do several interveners. The Canadian Generic Pharmaceutical Association, for example, suggested that if a drug manufacturer earned $1 million a year selling pain medication, but switched some pain medication production capacity to make a patented cancer drug that generated $2 million in yearly profits, the infringer would only need to disgorge $1 million per year under the differential profits approach. This flows from the fact that the company would have made $1 million on their pain medication had they never infringed. If an infringer proves on a balance of probabilities that they would have made other profits (had they never infringed), they can deduct that sum from the profit gained by infringing, even if the products are entirely unrelated.

[42]                          I disagree. To explain why, it is necessary to understand the purpose of (1) an accounting of profits and (2) non-infringing options.

(3)          Purpose of an Accounting of Profits

[43]                          The Patent Act  is designed to encourage research and development (Apotex Inc. v. Sanofi-Synthelabo Canada Inc., 2008 SCC 61, [2008] 3 S.C.R. 265, at para. 64). It does this through the “patent bargain”: an inventor discloses their useful invention to the public in exchange for a time-limited market monopoly on that invention (Teva Canada Ltd. v. Pfizer Canada Inc., 2012 SCC 60, [2012] 3 S.C.R. 625, at para. 32; Apotex Inc. v. Wellcome Foundation Ltd., 2002 SCC 77, [2002] 4 S.C.R. 153, at para. 37; Perry and Currier, at §3.1; R. T. Hughes and D. P. Clarizio, Hughes & Woodley on Patents (2nd ed. (loose-leaf)), at §1). This bargain mutually benefits the public and the inventor. The public benefits by receiving innovations in science and technology. The inventor benefits because they receive a time-limited market monopoly. The inventor can use the monopoly to generate profits and compensate themselves for the time, effort, and risk associated with making the invention.

[44]                          An accounting of profits is a remedial tool designed to protect the patent bargain. It does this by (1) disgorging the profits earned from patent infringement to the patentee and (2) ensuring that infringers are deterred but not punished for infringement (ADIR, at para. 37; Atlantic Lottery Corp. Inc. v. Babstock, 2020 SCC 19, at para. 24).

[45]                          Disgorgement is necessary because allowing infringers to appropriate the benefits of the patent monopoly for themselves “discourages research and development, and the disclosure of useful inventions” (Merck, at para. 42). If infringers could keep the profits earned from patent infringement, they could appropriate the time, effort, and risk associated with making the invention for their own benefit. This would make disclosure less likely, and the public would receive fewer innovative products.

[46]                          To fulfill the purpose underlying an accounting of profits, the infringer must disgorge to the patentee the “portion of the infringer’s profits which is causally attributable to the invention” (Schmeiser, at para. 101 (emphasis added)). Courts isolate “the value that the invention has brought to the product” (Rivett, at para. 61). They then ensure that the defendant will “disgorge any profits improperly received by the defendant as a result of its wrongful use of the plaintiff’s property. Such profits, having been earned through the use of the plaintiff’s property, rightly belong to the plaintiff” (Bayer Aktiengesellschaft v. Apotex Inc. (2001), 10 C.P.R. (4th) 151 (Ont. S.C.J.), at para. 12; see also Laboratoires Servier v. Apotex Inc., 2008 FC 825, 67 C.P.R. (4th) 241, at para. 504, aff’d 2009 FCA 222, 75 C.P.R. (4th) 443).

[47]                          Deterrence flows from disgorgement. The incentive to infringe is minimized if an infringer has to disgorge all profits causally attributable to the invention (Andrews and de Beer, at p. 640). To be clear, an accounting of profits is not the only remedial tool available to deter infringement and protect the patent bargain. Alternative remedies, including damages, elevated costs, injunctive relief, and punitive damages, may also be available to counterbalance any incentive to infringe. An accounting of profits, alongside these other remedial tools, protects the patent bargain. For example, it discourages efficient infringement: when an infringer’s profits exceed the damages suffered by the patentee.

[48]                          But deterrence should not be conflated with punishment. An infringer can be liable for patent infringement even if they had no knowledge of the patent or genuinely believed that the patent was invalid (Schmeiser, at para. 49). An accounting of profits should therefore discourage infringement but do no more. This requires disgorging only the profits causally attributable to the invention. Requiring infringers to disgorge anything more would constitute punishment and risk chilling public innovation and competition. Disgorging anything less would reduce the incentive to invent (Merck, at para. 42; ADIR, at para. 39).

[49]                          I would pause to note that Justice Côté and I agree on many points, including that an accounting of profits is an equitable remedy (although I would underscore that, in these circumstances, the remedy must serve the purposes of the Patent Act ); that “[t]he remedy should not punish potential competitors for competing in areas they reasonably believe are not properly covered by a patent”, as I indicate in my preceding paragraph; and that differential profits should isolate profits attributable to “unauthorized use of the patentee’s invention” (paras. 148 and 157). Our major difference relates to the methodology by which to isolate such profits. While we agree that this requires having regard to a non-infringing option, if there is one, where we disagree is how to give practical effect to that term. That is our fundamental point of difference. I will seek to explain the way in which the jurisprudence, read in the context of the scheme and purposes of the Patent Act , calls for that term to be given effect.

(4)          Purpose of Non-infringing Option

[50]                          In my view, Schmeiser’s use of the term “non-infringing option” should be interpreted with the goals of an accounting of profits in mind: (1) to disgorge the “portion of the infringer’s profit which is causally attributable to the invention” and (2) to deter, but not to punish, infringers (Schmeiser, at para. 101).

[51]                          Understood in this light, a non-infringing option helps courts isolate the profits causally attributable to the invention from the profits which arose at the same time the infringing product was used or sold, but which are not causally attributable to the invention. For example, when a company sells a product with a component that infringes a patent, courts need to determine the profit generated by (1) the patented invention and (2) the rest of the non-patented product. To generate these sums, courts should compare the profits generated from the sale of the infringing product with the patented feature — i.e., the actual profits — with the profits the infringer would have made on the sale of a similar product without the patented feature — i.e., the non-infringing option (Perry and Currier, at §17.50). Non-infringing options are “generally used in cases where an infringement allows the infringer to commercialize a good in a more profitable manner than [they] could have without the infringement” (Cinar Corporation v. Robinson, 2013 SCC 73, [2013] 3 S.C.R. 1168, at para. 80). The concept helps courts recognize the limited nature of a patent, as “a patent does not confer a complete monopoly if a defendant could make or sell a non-infringing version of the patented invention” (ADIR, at para. 40).

[52]                          Schmeiser, this Court’s leading authority on accounting of profits, affirms that this is the purpose of non-infringing options. As discussed, in Schmeiser, Monsanto’s patent related to a particular canola gene that permitted a farmer to spray the crop with a glyphosate˗based herbicide, which promised to increase harvesting yields. To determine which portion of Mr. Schmeiser’s profits were causally attributable to this invention, this Court compared the profits Mr. Schmeiser earned from the sale of the patented Roundup Ready canola with the profits he could have earned had he used the best non-infringing option: regular, non-genetically modified canola seeds. Since Mr. Schmeiser never sprayed his seeds with herbicide, none of his profits were causally attributable to the invention. All the profits earned were causally attributable to non-patented features of the sold product — the canola seed itself. The use of a non-infringing option therefore helped this Court determine that, despite infringing the patent, none of Mr. Schmeiser’s profits were causally attributable to the invention.

[53]                          Non-infringing options can also help courts determine when some, but not all, of the infringer’s profits are causally attributable to the invention. For example, in Rivett and the related judgment in Monsanto Canada Inc. v. Janssens, 2009 FC 318, 343 F.T.R. 234, the Federal Court required the defendants to disgorge a portion of their soybean profits because some of their profits were causally attributable to Monsanto’s invention. The defendants infringed Monsanto’s patent by planting genetically modified soybean seeds without licence. Like Schmeiser, the genetically modified crop allowed producers to spray a particular herbicide on soybean plants without killing them. However, the defendants, unlike Mr. Schmeiser, profited from the invention by spraying their crop with herbicide and increasing harvesting yields (Rivett, at para. 95; Janssens, at para. 27).

[54]                          As instructed by Schmeiser, the trial judge in both Rivett and Janssens used a non-infringing option — non-genetically modified soybean seeds — to determine how much of the defendants’ profits were causally attributable to Monsanto’s invention. The Federal Court of Appeal, which heard the appeals together, agreed and relied on evidence demonstrating that farmers who planted Monsanto’s genetically modified soybean and sprayed the crop with herbicide earned 18 percent higher profits than farmers using traditional soybean seeds (2010 FCA 207, [2012] 1 F.C.R. 473). The defendants therefore had to disgorge 18 percent of their profits.

[55]                          Schmeiser, Rivett, and Janssens reveal the utility of non-infringing options: the concept helps courts isolate profits causally attributable to the invention from profits causally attributable to non-inventive aspects of the infringing product. Since Mr. Schmeiser did not spray his canola with herbicide, none of his profits were causally attributable to Monsanto’s invention. All of Mr. Schmeiser’s profits were generated by the sale of canola seeds. Monsanto did not invent or hold a patent on canola seeds (Schmeiser, at paras. 16-17 and 21). Similarly, in Rivett and Janssens, a non-infringing option helped the court isolate the profits causally attributable to Monsanto’s invention from the profits attributable to non-inventive features of the infringing product, e.g. the profits attributable to the extra yield caused by the herbicide˗resistant gene versus the profits attributable to the soybean seed itself (Andrews and de Beer, at p. 625). Said differently, considering a non-infringing option ensured that the court did not erroneously extend Monsanto’s patent by entitling it to profits generated by product features it did not invent. Demanding that infringers disgorge profit that was not causally attributable to the invention would undermine the patent bargain by granting the patentee a windfall and punishing the infringer.

[56]                          Courts recognized the need to consider non-infringing options in this manner before Schmeiser. In Lubrizol Corp. v. Imperial Oil Ltd., [1997] 2 F.C. 3 (C.A.), for example, Imperial Oil sold motor oil containing a patented dispersant additive. The Federal Court of Appeal concluded that the infringer might not need to disgorge all of its profits derived from the sale of the infringing product because some of the profits may be causally attributable to the motor oil itself, which the patentee did not invent:

It may be possible for Imperial to show that some part of the profits made on the infringing sales are not profits “arising from” the infringement in that they are not caused by but simply made on the occasion of such infringement. . . .

. . . Form must not be allowed to triumph over substance. While motor oil containing the dispersant additive was properly claimed in the patent (it would seem likely that the dispersant is useless except as an additive to motor oil) and while that claim was properly found to have been infringed, the reality is that Lubrizol did not invent motor oil and that Imperial’s motor oils contain other additives than the one here in issue. The terms of the judgment quoted by the Prothonotary in the above extract from his reasons make it plain that it is the presence of the additive (“carboxylic derivative compositions”) claimed in the Meinhardt patent which caused Imperial’s motor oils to infringe. Thus, it is possible that such oils have achieved their market share and attendant profits for reasons other than the presence of Lubrizol’s patented additive. A finding that Imperial’s motor oils infringed the Lubrizol patent does not necessarily amount to a finding that all the profits from the sales of such motor oils are profits arising from the infringement. That is an issue of fact to be decided on the reference.

. . .

. . . to allow Lubrizol to take profits which Imperial succeeds in showing were solely attributable to some non-infringing feature of its motor oil would be to judicially sanction Lubrizol’s unjust enrichment at Imperial’s expense. [Emphasis added; paras. 9-15.]

[57]                          The Federal Court of Appeal in Imperial Oil did not have the benefit of Schmeiser. However, its reasoning accords with that in Schmeiser. To isolate the profits causally attributable to the invention from the profits attributable to non-inventive features of the infringing product, the court would compare the infringer’s profits from the infringing product with what they would have earned with a non-infringing option: motor oil without the patented additive.

[58]                          In sum, a non-infringing option is any product that helps courts isolate the profits causally attributable to the invention from the profits which arose at the same time the infringing product was used or sold, but which are not causally attributable to the invention.

[59]                          A non-infringing option is not, as Nova and my colleague contend, an infringer’s “most profitable” alternative sales product that it “would have” and “could have” sold had it not infringed. This approach is flawed for two reasons.

[60]                          First, it is contrary to Schmeiser. Nova and my colleague’s argument is grounded on the incorrect premise that an accounting of profits is designed to ensure that the infringer ends up no worse off than had it never infringed (Côté J.’s reasons, at para. 146). But Schmeiser did not hold that the purpose of an accounting of profits was to determine what profits the infringer “could have earned” but for the infringement. Nor did it ever suggest that the focus was on the “value of the invention in the hands of the infringer” (Côté J.’s reasons, at para. 91; see also para. 118). These phrases are never used in Schmeiser. Nor are they present in the application of the law to the facts of Schmeiser¸ as there is no discussion in Schmeiser about what Mr. Schmeiser “could have earned” using non-infringing canola seeds. Instead, Schmeiser affirmed that the goal of an accounting of profits is to ensure that all profits causally attributable to the invention are disgorged to the patentee:

It is settled law that the inventor is only entitled to that portion of the infringer’s profit which is causally attributable to the invention . . . .

. . .

The difficulty with the trial judge’s award is that it does not identify any causal connection between the profits the appellants were found to have earned through growing Roundup Ready Canola and the invention. On the facts found, the appellants made no profits as a result of the invention. [Emphasis in original; paras. 101-3.]

[61]                          My colleague dismisses Schmeiser’s clear instruction that courts must focus on isolating the profits “causally attributable to the invention” (Schmeiser, at para. 101; see also para. 102). She suggests this explicit phrasing is an inadvertent ambiguity (Côté J.’s reasons, at para. 173). I cannot accept this. The language in Schmeiser is clear, unambiguous, and binding on this Court. I do not believe it would be appropriate for this Court to ignore this instruction from Schmeiser and adjust the focus of an accounting of profits to align with new terminology not used in the judgment.

[62]                          Second, this reading of non-infringing options would distort the purpose of an accounting of profits and, in turn, undermine the patent bargain underlying the Patent Act . If an infringer is allowed to use any prior profitable business venture as a non-infringing option, an infringer would always be incentivized to switch its business capacity to a more profitable infringing product. At worst, the infringer would keep all the profits they would have earned selling the non-infringing products that they sold before. At best, the infringer keeps some or all of the extra profits earned from infringement. Reading “non-infringing option” as Nova and my colleague suggest would have the effect of creating a form of business insurance for infringers: an infringer could always use their previous product lines as a non-infringing option and protect those profits in the event their new product infringes a patent.

[63]                          This distortion of the purpose of an accounting of profits gives rise to unacceptable consequences, one being that the quantum of profits to be disgorged would vary with the size of the infringing business and the breadth of its product lines. Nova’s suggested approach disproportionately benefits large corporations (like itself) that have diverse product lines. Such businesses have multiple products that they “could have” and “would have” produced had they not infringed. Nova, for example, produces numerous non-infringing plastics. In circumstances like these, my colleague’s reasons would provide no incentive not to infringe. At best, Nova would retain all profits from infringement. At worst, Nova could keep the profits it would have made on any of its other product lines. Nova’s approach would allow it and other large enterprises to infringe with relative impunity. This undermines the bargain provided for in the Patent Act .

[64]                          By contrast, smaller businesses would be disproportionately disadvantaged by Nova’s approach. The facts of Rivett are instructive. Mr. Rivett was unable to establish that he “could have” planted regular soybean seeds, as they were not available for purchase in his locale (Rivett, at para. 63). Despite this unavailability, under the reading of Schmeiser adopted in these reasons and by the Federal Court in Rivett, it was appropriate to consider regular soybean seeds as a relevant non-infringing option. Considering what profits Mr. Rivett would have earned with regular soybean seeds allowed the courts to isolate the “portion of the infringer’s profit which [was] causally attributable to the invention” (the patented genes that made the plants glyphosate herbicide˗resistant) from the profits attributable to non-patented features of the sold product (the soybean seed itself), as per the instructions in Schmeiser (para. 101).

[65]                          My colleague sees no issue with such consequences. In fact, she implies that Rivett was wrongly decided. In her view, Mr. Rivett should have been forced to disgorge all of his profits to Monsanto, including those not causally attributable to Monsanto’s invention (Côté J.’s reasons, at paras. 198-99). I cannot endorse such an outcome. This approach is detached from the Patent Act . It ties the profits to be disgorged to the size of the infringer’s business, while ignoring the value of a patent.

[66]                          In sum, I see no reason to depart from Schmeiser and adopt the definition of “non-infringing option” proposed by Nova and adopted by my colleague. While the approach advanced by Nova is in its interests, it has the result of favouring the strong and adversely affecting the weak, a consequence that accords with no principle of equity of which I am aware.

[67]                          Whether there is a non-infringing option that can assist courts in isolating the profits causally attributable to the invention is a question of fact. There are no strict rules around this factual exercise. Despite my colleague’s assertions that I hold otherwise, the non-infringing option need not be a strict market substitute for the patented product. The onus is on the infringer to adduce sufficient evidence to satisfy the court that the profits from its infringing product arose by virtue of features other than the patentee’s invention and that there is a non-infringing option that can help the courts isolate this value (Beloit, at p. 457; Bayer Aktiengesellschaft, at paras. 15-16; AFD Petroleum Ltd. v. Frac Shack Inc., 2018 FCA 140, 157 C.P.R. (4th) 195, at paras. 62-63). Typically, non˗infringing options will be most relevant when a patent covers only part of the product sold. In those situations, the profits generated by the sale of the infringing product may be attributable to inventive and non-inventive features of the product. But non-infringing options may also be relevant when the entire product sold is patented. Ultimately, “the question is whether the patent contributes the whole value of the thing that was sold, or merely a part”, “not whether the patent is the whole thing that was sold, or merely a part” (N. Siebrasse, “A Remedial Benefit-Based Approach to the Innocent-User Problem in the Patenting of Higher Life Forms” (2004), 20 C.I.P.R. 79, at p. 109 (emphasis in original)).

(5)          Application 

[68]                          In my view, the reference judge did not err when he concluded that all the profits Nova earned by selling the patented plastics were causally attributable to Dow’s invention. This is supported by two facts.

[69]                          First, the reference judge found that customers only purchased Nova’s infringing plastics because they contained the features captured by Dow’s patent, i.e., Nova sold the unique, patented thin and durable plastics covered by Dow’s patent. Dow’s invention had created a distinct market and Nova could only service that market because it sold infringing plastics. The two companies were the only companies serving the distinct market created by the patented plastics (F.C. reasons, at paras. 70 and 73-76).

[70]                          Second, Nova did not establish that there were relevant non-infringing options that would help the court isolate the profits causally attributable to Dow’s invention from profits attributable to non-inventive features of the infringing product. As explained earlier, the existence of an appropriate non-infringing option is a question of fact. Nova bore the evidentiary onus to establish this fact. It failed to discharge this onus. Before the reference judge, Nova conceded that there were no non-infringing options available for the purpose of applying the differential profits approach (F.C. reasons, at para. 146). This is also confirmed by the trial transcripts:

[Reference judge]: Now, here Nova is not seeking to deduct profits on foregone plastics. It merely wants to satisfy the court that the fixed costs would have been covered and therefore should be deducted.

 

[Nova’s trial counsel]: Right. That’s true.

 

(R.R., at p. 27)

[71]                          Nova reiterated this position at the Federal Court of Appeal. In Nova’s responding factum on cross-appeal, it wrote:

In Schmeiser, the Supreme Court recognized that the “preferred method” for achieving this goal is the differential profits approach — “where profits are allocated according to the value contributed to the defendant’s wares by the patent.” This method allocates only profits that the defendant derived as a result of the invention (i.e., profits that are causally attributable to the invention) by comparing them with profits that would have been earned on a true substitute or direct “non-infringing alternative” product.

All parties agreed that the differential profits method was not the appropriate method in this case. [Footnotes omitted.]

[72]                          Nova’s concession before the reference judge is sufficient to dispose of the appeal on this issue. Nova’s concession at trial is underscored by its continued concession at the Federal Court of Appeal. Permitting Nova to fundamentally alter its argument here would allow the company to treat this Court as one of first instance. It is not our role to make factual findings as to whether hypothetical profits from pail and crate plastics sales could assist with determining what portion of Nova’s profits were causally attributable to Dow’s invention. To evaluate Nova’s argument properly, this Court would need to re-evaluate the whole of the factual record. This is not our role.

[73]                          I also do not accept Nova’s request that we send this matter back to the reference judge for reconsideration. This decision affirms established law on an accounting of profits, as set out in Schmeiser. The reference judge did not misapply Schmeiser. There is no basis for sending the matter back for redetermination in these circumstances.

B.            Issue 2: Is Dow Entitled to Springboard Profits?

[74]                          An accounting of profits requires that the infringer disgorge all profits causally attributable to infringement of the invention, regardless of when the profits materialize. For example, if an infringer earns additional profits after the patent expires because of an accelerated entry into the market, the infringer should disgorge those post-patent-expiry profits. This principle is known as “springboard profits” or “springboard relief”.

[75]                          For the first time in Canadian law, the reference judge awarded springboard profits to Dow. The Federal Court of Appeal upheld this decision, agreeing with the reference judge that “springboard profits ‘are nothing more than a type of [gain] to be proven with evidence’” (F.C.A. reasons, at para. 126 (text in brackets in original), citing F.C. reasons, at para. 124).

[76]                          Nova challenges the decision to award springboard profits on three grounds. The company argues (1) that springboard profits are not legally permissible; (2) that Nova has already compensated Dow for its “ramp-up” into the market through its payment of a reasonable royalty (as described in para. 7 above); and (3) alternatively, that if springboard profits are legally permissible, the springboard profits should be reduced to account for the fact that Nova would have earned profits by selling pail and crate plastics had it not infringed.

[77]                          None of these arguments are persuasive. As I explain below, (1) springboard profits are legally permissible; (2) whether post-patent-expiry profits are causally attributable to patent infringement is a question of fact and Nova has not shown palpable and overriding error in the reference judge’s findings; and (3) for the reasons given above, Nova is not entitled to deduct the profits it could have earned from pail and crate plastics. There is therefore no basis to interfere with this part of the reference judge’s award to Dow.

(1)          Springboard Profits are Legally Permissible

[78]                          Nova submits that Canadian courts have never awarded springboard profits before. That is true. But Canadian courts have recognized and awarded springboard damages (AstraZeneca Canada Inc. v. Apotex Inc., 2015 FC 671, at para. 7 (CanLII); Teva Canada Limited v. Janssen Inc., 2018 FCA 33, 420 D.L.R. (4th) 493, at paras. 107-12; Merck & Co., Inc. v. Apotex Inc., 2013 FC 751, [2015] 1 F.C.R. 405, at para. 183, aff’d 2015 FCA 171, [2016] 2 F.C.R. 202; Apotex Inc. v. Eli Lilly and Company, 2018 FCA 217, 161 C.P.R. (4th) 411, at para. 114; see also Siebrasse, Stack et al., at pp. 96-97).

[79]                          The purpose underlying springboard damages is sound, having regard to the scheme and purposes of the Patent Act . A patent provides a time-limited monopoly to make, use, and sell an invention during the life of the patent (Patent Act , s. 42 ). This monopoly gives the patentee the right to build sales capacity and market share without any market competition. The patentee can then use this market advantage against competitors after the patent expires (M. E. Charles, “Monetary Remedies in Intellectual Property Litigation”, [2007] J. Bus. Valuation 159, at pp. 160 and 167).

[80]                          An infringer that begins selling the patented invention before the patent expires interferes with the patentee’s right to build sales capacity and market share in the absence of competition. This can reduce the patentee’s post-patent-expiry profits. If the patentee can prove that it lost sales post-patent-expiry as a result of infringing activity occurring during the life of the patent, the patentee is entitled to springboard damages to compensate for that loss.

[81]                          A corresponding purpose underlies an accounting of profits. By infringing during the life of the patent, the infringer can also build sales capacity and market share for their own version of the patented product. Then, after the patent expires, the infringer can use this sales capacity and market share to earn profits that it would not have earned but for the infringing activity that occurred during the life of the patent. Thus, a portion of such post-expiry profits may be causally attributable to infringement of the invention. Failing to disgorge those profits would leave gains that are causally attributable to infringement of the invention in the hands of the infringer. It would also be unfair to third parties that waited for patent expiry to compete with the patentee.

[82]                          I agree with the reference judge and the Federal Court of Appeal that springboard profits are “nothing more than a type of [gain] to be proven with evidence”. The Patent Act  does not bar disgorgement of profits that materialize after patent expiry. To the contrary, artificially limiting an accounting of profits in the manner suggested by Nova would leave gains that are causally attributable to infringement of the invention in the hands of the infringer and thus undercut the bargain underlying the Patent Act . Judges should determine whether any profits earned post-patent-expiry are causally attributable to infringement of the invention during the period of patent protection. If they are, such profits should be disgorged.

(2)          The Reasonable Royalty Award Did Not Cover Nova’s Post-expiry Profits

[83]                          The reference judge found no support for Nova’s argument that its springboard profits were already accounted for by its payment of pre-grant reasonable royalties (paras. 122-24). The Federal Court of Appeal agreed. It held that royalties paid to cover the post-filing, pre-grant time period were unrelated to the quantum of profits Nova earned as a result of infringement of the invention, during the period of patent protection (paras. 137-40).

[84]                          Before this Court, Nova again argued that it had already compensated Dow for its “ramp-up” into the market by virtue of its payment of a reasonable royalty (as described in para. 7 above; see A.F., at para. 126). I see no merit in this argument. Disgorgement of springboard profits is a means to prevent an infringer from gaining from its infringement that is different from, but (in the circumstances of this case) complementary to, compensating a patentee by way of a reasonable royalty. The need to disgorge springboard profits is a factual question: are there any profits earned post-patent-expiry that are causally attributable to infringement of the invention, during the period of patent protection? Nova has not shown any palpable and overriding error in the reference judge’s findings in this regard.

(3)          Dow is Entitled to All Springboard Profits Causally Attributable to Infringement of the Invention

[85]                          In the alternative, Nova argues that the springboard profits should be reduced to account for the fact that, had it not infringed Dow’s patent, it would have earned profits by selling pail and crate plastics. For the reasons given above, I reject this argument. Like profits earned during the life of the patent, a patentee is entitled to all the infringer’s profits causally attributable to infringement of the invention, even if they materialize after the patent expires. Nova conceded that pail and crate plastics were not a non-infringing option, so no proper basis has been shown to reduce the reference judge’s springboard profits award in the circumstances.

VII.      Disposition

[86]                          For the foregoing reasons, I would dismiss the appeal.

[87]                          Dow is entitled to its costs throughout.

The following are the reasons delivered by

 

                   Côté J. —

TABLE OF CONTENTS
Paragraph
I.      Overview	88
II.    Background and Context	95
A.    Patent Bargain and Remedies Available for Patent Infringement	97
(1)      Damages	101
(2)      Accounting of Profits	105
(3)      Election Between Damages and an Accounting of Profits	109
B.    Post‑Schmeiser Case Law	111
C.    Facts	119
(1)      Nova’s Infringement of Dow’s Patent	119
(2)      Nova’s Production of SURPASS	121
D.    Judicial History	124
(1)      Reference Judge’s Decision: Federal Court, 2017 FC 350, [2018] 2 F.C.R. 154	124
(2)       Federal Court of Appeal Decision, 2020 FCA 141, [2021] 1 F.C.R. 551	128
III.   Issues	130
IV.   Analysis	132
A.    Principles Governing an Accounting of Profits	133
(1)      Equitable Purposes That May Be Served by an Accounting of Profits	133
(2)      Equitable Purposes Served by an Accounting of Profits in the Patent Context	140
(3)      Role of Causation	152
B.    The Differential Profit Approach Should Have Been Available	186
(1)      An Infringer’s Best Non‑Infringing Option Does Not Need to Be a True Consumer Substitute	187
(2)      How Are Courts to Determine the Reasonable Hypothetical to Be Used in an Accounting of Profits?	203
(3)      Availability of the Differential Profit Approach in the Present Case	214
C.    Springboard Profits Are Available	221
V.    Disposition	226

I.               Overview

[88]                         This appeal concerns the principles that govern the remedy of an accounting of profits. Specifically, this appeal is about how an accounting of profits operates as a remedy for patent infringement. The appellant, Nova Chemicals Corporation, infringed a patent owned by the Dow Chemical Company and licensed to Dow Global Technologies Inc. and Dow Chemical Canada ULC (collectively “Dow”). The issues in the courts below related to the proper approach for an accounting of profits, an equitable remedy that has been aptly described as “ancient and notoriously difficult in practice” (Warman International Ltd. v. Dwyer (1995), 182 C.L.R. 544 (H.C.A.), at p. 556 (footnote omitted)).

[89]                         This appeal therefore provides our Court with an opportunity to elucidate how an accounting of profits operates as a remedy for patent infringement in a way that aligns with the purposes of the Patent Act , R.S.C. 1985, c. P‑4 . An accounting of profits is a flexible, equitable remedy that applies differently depending on the context. The context of an infringement of intellectual property is quite different than other contexts in which an accounting of profits is available, such as that of a breach of a fiduciary duty. This difference has profound implications for how the remedy operates. Specifically, in the patent context, a more rigorous analysis of causation is both appropriate and necessary. This more rigorous analysis — epitomized in the differential profit approach — helps to properly isolate the profits to be disgorged. It ensures that the infringer disgorges only those profits that are causally attributable to the infringement of the patent.

[90]                         The differential profit approach was affirmed by this Court in Monsanto Canada Inc. v. Schmeiser, 2004 SCC 34, [2004] 1 S.C.R. 902, as being the “preferred means of calculating an accounting of profits” for patent infringement (para. 102). This approach compares the actual profits earned by the infringer as a result of the infringement with the profits that it could have and would have earned had it not infringed the patent. In other words, it compares the infringer’s actual profits with its hypothetical profits from using its best non‑infringing option.

[91]                         As I will explain, the differential profit approach should have been available in this case. There is no principled reason why this approach should be available only where the hypothetical non‑infringing option is a true consumer substitute for the embodiment of the patent, which, in these reasons, I will refer to as the “patented product”. Rather, any hypothetical non‑infringing option that can be established in evidence — including one involving a product that services a different consumer market — can be used. However, the patent infringer must establish on a balance of probabilities that, but for the infringement, it could have and would have pursued this alternative course of action. The goal is not to isolate the value of the invention in the abstract. Rather, as is apparent from the application of Schmeiser, the goal is to isolate the profits causally attributable to the infringement. This determines the value of the invention in the hands of the infringer.

[92]                         Despite its claim to the contrary, I am of the respectful view that the majority — relying on a line of post‑Schmeiser damages cases — adopts a simplified method of calculating an accounting of profits in which the differential profit approach can be applied only when an infringer’s non‑infringing option would compete in the same consumer market as the patented product. At the very least, the majority claims that it is legally relevant whether the patented product and the potential non‑infringing option would compete in the same consumer market. The majority’s approach fails to give sufficient consideration to the equitable nature of the remedy and to its appropriate use in the patent context. The majority endorses an unclear approach to causation, upholding an award of profits in which the reference judge failed to sufficiently isolate the profits causally attributable to Nova’s infringement of Dow’s patent.

[93]                         The consequences of such an approach to causation are significant, both for the parties in this case and because of the precedent it establishes. For the parties, it leads to an inequitable outcome that, according to Nova, inflates the award of profits by over $300 million. This approach also distorts the goals of the Patent Act . Moreover, its purported simplicity is belied by its inability to isolate the profits causally attributable to the infringement in many plausible scenarios, including the case at hand. The majority’s analysis is also inconsistent. Indeed, when dealing with the profits the infringer earned during the life of the patent, the majority emphasizes that the causation analysis isolates the profits causally attributable to the invention. However, when dealing with profits earned after the expiry of the patent (i.e. springboard profits), the majority shifts the focus, emphasizing that the causation analysis isolates the profits causally attributable to the infringement. With respect, the majority’s method of calculating an accounting of profits is conceptually impoverished and doctrinally flawed.

[94]                         For the following reasons, I would allow the appeal and remit the matter to the Federal Court.

II.            Background and Context

[95]                         My colleague Rowe J. provides some general background relating to remedies in patent law, the calculation of an accounting of profits, and this Court’s decision in Schmeiser. However, and I say this with respect, I believe he has not sufficiently canvassed all of the concepts needed to fully understand the issues in this appeal. Similarly, while I agree with much of my colleague’s recitation of the facts and judicial history, there are significant contextual factors that are not discussed in his reasons.

[96]                         Below, I discuss important background relating to (1) the patent bargain and the remedies available for patent infringement and (2) the post‑Schmeiser case law regarding non‑infringing options that has developed in the Federal Courts. I then discuss important context pertaining to (3) the facts relating to Nova’s infringement of Dow’s patent and (4) the judicial history and the manner in which the case was argued in the courts below.

A.           Patent Bargain and Remedies Available for Patent Infringement

[97]                          The grant of a patent is not “an accolade or civic award for ingenuity” (Apotex Inc. v. Wellcome Foundation Ltd., 2002 SCC 77, [2002] 4 S.C.R. 153, at para. 37). Rather, the patent system is based on a bargain — that is, a quid pro quo — between the inventor and the public: “. . . the inventor is granted exclusive rights in a new and useful invention for a limited period in exchange for disclosure of the invention so that society can benefit from this knowledge” (Teva Canada Ltd. v. Pfizer Canada Inc., 2012 SCC 60, [2012] 3 S.C.R. 625, at para. 32; see also Free World Trust v. Électro Santé Inc., 2000 SCC 66, [2000] 2 S.C.R. 1024, at para. 13; Wellcome, at para. 37). The Patent Act  therefore confers a time‑limited statutory monopoly. In exchange for disclosure of the invention, the patentee is granted “the exclusive right, privilege and liberty of making, constructing and using the invention and selling it to others to be used” (Patent Act , s. 42 ). The patent bargain is the fundamental policy rationale underlying the Patent Act  (Teva, at para. 32).

[98]                          The infringement of the patentee’s statutory monopoly has been described as a statutory tort (S. J. Perry and T. A. Currier, Canadian Patent Law (4th ed. 2021), at §17.1; Bristol‑Myers Squibb Co. v. Canada (Attorney General), 2005 SCC 26, [2005] 1 S.C.R. 533, at para. 134, per Bastarache J., dissenting, but not on this issue). As this Court stated in Schmeiser, “the governing principle is whether the defendant, by his actions, activities or conduct, appropriated the patented invention, thus depriving the inventor, in whole or part, directly or indirectly, of the full enjoyment of the monopoly the patent grants” (para. 49). Liability thus arises from any breach of the patentee’s monopoly and is absolute. The infringer’s intention, or its knowledge about its use of the patent, is irrelevant (Schmeiser, at paras. 49 and 86).

[99]                          Once infringement has been proven, a number of different remedies are available to a successful plaintiff. Often, as in the case at bar, the issue of liability for patent infringement is bifurcated from the issue of calculating the proper monetary award.

[100]                      My colleague highlights three different remedies available under the Patent Act : reasonable compensation, damages, and an accounting of profits. With respect, a more comprehensive summary of damages and an accounting of profits is required. I provide such a summary below. However, before I elaborate in greater detail on these two remedies, it is important to emphasize that the remedies set out above are not the only ones available for patent infringement. Indeed, the Patent Act  offers a number of other remedies to a successful plaintiff. Injunctive relief is available under s. 57(1) (a) of the Patent Act . Delivery up or destruction, in which the infringing product is ordered to be delivered up to the patentee or destroyed, are also available (Perry and Currier, at §§17.117‑17.118). Other remedies — such as legal costs, interest on an award of damages or profits, and punitive damages — are also available to a successful plaintiff (Perry and Currier, at §17.5).

(1)          Damages

[101]                      Damages may be awarded by a court for any monetary loss suffered by the patentee as a result of the patent infringement. The infringer is liable to the patentee for “all damage sustained . . . after the grant of the patent, by reason of the infringement” (Patent Act , s. 55(1) ). Damages are compensatory and can include the following: lost profits on direct sales of the patented product; lost profits on sales of related goods or services; lost profits due to the depression of prices; lost profits due to increased costs; and lost profits on royalties from licensing agreements (Perry and Currier, at §§17.9 and 17.22‑17.35).

[102]                      The purpose of damages in the context of a patent infringement is the same as it is for other torts: to compensate the wronged party for the wrongdoing. Damages are therefore a loss‑based remedy aimed at making up the plaintiff’s loss. This remedy places the patentee in the position it would have been in but for the infringement (Perry and Currier, at §17.15).

[103]                     When calculating damages, courts strive to achieve perfect compensation: they want neither to overcompensate nor to undercompensate the plaintiff. It is noteworthy that this concern is particularly acute in the patent context. Undercompensation may discourage innovation. However, in Merck & Co., Inc. v. Apotex Inc., 2015 FCA 171, [2016] 2 F.C.R. 202 (“Lovastatin”), the Federal Court of Appeal also discussed the equal problem of overcompensation:

At the heart of this bargain with the inventor, and at the heart of the Act, is the concept of balance between the benefit conferred on the public through the disclosure of a new and useful invention, and the benefit conferred on the inventor through the grant of a monopoly. Thus, in the event of infringement, under‑compensation of an inventor discourages research and development, and the disclosure of useful inventions. Equally, over‑compensation of an inventor chills potential competition to the extent that a potential infringer is uncertain about the scope and validity of a patent. The balance at the heart of the Act requires perfect compensation. [Emphasis added; para. 42.]

[104]                      Ultimately, the availability of damages as a remedy for patent infringement helps to preserve the patent bargain by ensuring that patentees are fully compensated for any damage they suffer as a result of the infringement.

(2)          Accounting of Profits

[105]                      An accounting of profits — sometimes referred to as an account of profits, the disgorgement of profits, or disgorgement — is an equitable remedy. It operates to strip the defendant of unlawful gains and is non‑punitive (Schmeiser, at para. 101; Lubrizol Corp. v. Imperial Oil Ltd., [1997] 2 F.C. 3 (C.A.), at para. 15). As a gain‑based remedy, it is “calculated exclusively by reference to the defendant’s wrongful gain” (Atlantic Lottery Corp. Inc. v. Babstock, 2020 SCC 19, at para. 23). Gain‑based remedies stand in contrast to loss‑based remedies, such as compensatory damages or equitable compensation (Southwind v. Canada, 2021 SCC 28, at para. 67).

[106]                      The statutory basis for awarding an accounting of the infringer’s profits is s. 57  of the Patent Act , which states:

(1) In any action for infringement of a patent, the court, or any judge thereof, may, on the application of the plaintiff or defendant, make such order as the court or judge sees fit,

 

(a) restraining or enjoining the opposite party from further use, manufacture or sale of the subject-matter of the patent, and for his punishment in the event of disobedience of that order, or

 

(b) for and respecting inspection or account,

 

and generally, respecting the proceedings in the action.

[107]                      The statutory availability of an accounting of profits helps to uphold the patent bargain by ensuring that the infringer will not be able to retain any profits causally attributable to its infringement, even if the patentee suffers no compensable damage from the violation of its statutory monopoly. The remedy deters competitors from infringing a patentee’s statutory monopoly even where the infringement can be described as “efficient”. An efficient infringement — which is similar in nature to the concept of an efficient breach in contract law — is one in which an infringer’s profits arising from its infringement exceed the damages payable to the patentee. An accounting of profits therefore acts as a remedy to prevent an infringer from retaining any profits causally attributable to its infringement. This ensures that the patent monopoly is upheld and respected: only the patentee can exploit the invention for gain.

[108]                      Although an accounting of profits has been made statutorily available as a potential remedy, it has not been shorn from its equitable origins; it remains an equitable remedy. Lest it be forgotten, such remedies are always subject to the discretion of the court (Strother v. 3464920 Canada Inc., 2007 SCC 24, [2007] 2 S.C.R. 177, at para. 74), and equitable considerations apply when assessing whether a remedy is fit in the circumstances (Beloit Canada Ltd. v. Valmet‑Dominion Inc., [1997] 3 F.C. 497 (C.A.), at paras. 106‑25; Bayer Aktiengesellschaft v. Apotex Inc. (2002), 16 C.P.R. (4th) 417 (Ont. C.A.), at paras. 10‑15).

(3)          Election Between Damages and an Accounting of Profits

[109]                      Given that an accounting of profits is an equitable and discretionary remedy, a plaintiff in a patent infringement case does not have an absolute right to make an election between damages and an accounting of profits. The decision to award an accounting of profits, upon request, remains a discretionary one for the court. However, “[i]t is common practice in cases of patent infringement to allow a plaintiff to elect between damages and an accounting of profits” (Bayer Inc. v. Cobalt Pharmaceuticals Co., 2016 FC 1192, 142 C.P.R. (4th) 374, at para. 6, aff’d 2018 FCA 32, [2018] 4 F.C.R. 58; AlliedSignal Inc. v. Du Pont Canada Inc. (1995), 61 C.P.R. (3d) 417 (F.C.A.), at pp. 444‑46; Perry and Currier, at §17.80).

[110]                      In the case at hand, the issue of liability was bifurcated from the determination of the appropriate quantum of an award. After Dow successfully proved that Nova was liable for infringement of its patent, Dow was permitted “to elect after due inquiry and full discovery, either an accounting of profits of the defendant or all damages sustained by reason of infringement” (2014 FC 844, 129 C.P.R. (4th) 199, at para. 283).

B.            Post‑Schmeiser Case Law

[111]                     Since this Court’s decision in Schmeiser, case law has developed in the Federal Court and the Federal Court of Appeal regarding how courts should determine the appropriate non‑infringing option, both in relation to claims for damages and in relation to claims for an accounting of profits. It is important to take account of these post‑Schmeiser decisions, as they ultimately played a significant role in the judicial history of this case.

[112]                     One line of cases that has developed in the Federal Court of Appeal pertains to the legal relevance of non‑infringing alternatives when calculating an award of damages (see, e.g., Lovastatin; Pfizer Canada Inc. v. Teva Canada Ltd., 2016 FCA 161, 400 D.L.R. (4th) 723 (“Pfizer (2016)”); AFD Petroleum Ltd. v. Frac Shack Inc., 2018 FCA 140, 157 C.P.R. (4th) 195; Apotex Inc. v. Eli Lilly and Company, 2018 FCA 217, 161 C.P.R. (4th) 411 (“Cefaclor”)). I note that this case law largely refers to a “non‑infringing alternative” rather than using the language from Schmeiser, in which the Court referred to a “non‑infringing option”.

[113]                     In Lovastatin, the Federal Court of Appeal recognized the legal relevance of a non‑infringing alternative when calculating damages. As it explained, where a defendant can make and sell a non‑infringing alternative to the embodiment of the patent, the patent does not give the patentee a complete monopoly on the market — the patentee has a monopoly on the invention, but consumers may also choose to purchase the non‑infringing alternative product (para. 48). Given that the patent does not confer a monopoly on the entire market for the product, the availability of a non‑infringing alternative becomes legally relevant when assessing a claim for damages. If “the infringer could and would have made and sold a non‑infringing alternative, these sales may well reduce the patent owner’s sales” (para. 49). Therefore, if the infringer can establish that it could have and would have produced a non‑infringing alternative product, the court can consider this when calculating the damages payable to the patentee. The infringer’s hypothetical sales of the non‑infringing alternative, which would have competed in the same market as the patentee’s product, reduce the number of sales that the patentee actually lost due to the infringement.

[114]                     However, as the Federal Court of Appeal noted in Lovastatin, “[w]hen considering the effect of legitimate competition from a defendant marketing a non‑infringing alternative, a court is required to consider [whether] the alleged non‑infringing alternative [is] a true substitute and thus a real alternative” (para. 73 (emphasis added)). In Cefaclor, the Federal Court of Appeal emphasized that consideration of whether the non‑infringing alternative is a true substitute is particularly important in non‑pharmaceutical patent infringement cases: this “very important question . . . usually turns on whether the product at issue would be considered a true substitute by the consumer” (para. 54 (emphasis added)).

[115]                     In the context of damages, there is some potential analytical value to the requirement that the infringer show that it could have and would have made an alternative product that would have competed in the same market as the patented product. It is the infringer’s hypothetical sales of the non‑infringing alternative that reduce the damages payable to the patentee because those sales reduce the patentee’s lost sales (and therefore lost profits). If the non‑infringing alternative were not a true alternative for the consumer such that it could not displace sales of the patented product, then it would not be legally relevant when calculating damages.

[116]                     A brief example is illustrative. A patentee has a patent on a specific product, which I will refer to as a patented widget. An infringer breaches the patentee’s statutory monopoly by selling 100 patented widgets. The patentee elects damages and argues that, but for the infringement, it would have sold 100 patented widgets, so the infringer’s unauthorized sales caused it (the patentee) to lose 100 sales. The infringer, however, can argue that it could have and would have sold a non‑infringing alternative to the patented widgets had it not infringed the patent. These unpatented widgets, while not identical to the patented ones, would have competed in the same market. If the infringer can show that it could have and would have sold, for example, 25 unpatented widgets, then the patentee’s lost sales would be reduced to 75.

[117]                     In summary, in the damages context, a non‑infringing alternative is relevant to the extent that the infringer’s hypothetical sales of the non‑infringing alternative reduce the patentee’s lost sales. The patentee’s lost sales — and its corresponding lost profits from those sales — are a head of damages. If the infringer can show that it could have and would have sold a non‑infringing alternative product that would have competed in the same market, then the patentee’s damages based on lost sales may be reduced.

[118]                     The damages case law dealing with non‑infringing alternatives has been cited approvingly in accounting of profits cases (see Apotex Inc. v. ADIR, 2017 FCA 23, 406 D.L.R. (4th) 572 (“ADIR (2017)”), at para. 34; Apotex Inc. v. ADIR, 2020 FCA 60, 172 C.P.R. (4th) 1 (“ADIR (2020)”), at paras. 48‑49). Indeed, there has been a convergence in the law with respect to damages and accounting of profits post‑Schmeiser. Since being decided in 2015, Lovastatin — including its requirement that a non‑infringing alternative be a “true substitute” for or “real alternative” to the patented product — has been applied in both damages cases and accounting of profits cases.

C.            Facts

(1)          Nova’s Infringement of Dow’s Patent

[119]                     In April 1994, Dow filed an application for a patent for “Fabricated Articles Made From Ethylene Polymer Blends”, which became Canadian Patent No. 2,160,705 (“705 Patent”). The patent was published in November 1994 but not issued until August 2006. Dow’s 705 Patent covers certain grades of polyethylene used to make plastic film products, like plastic garbage bags and food wrappings. Dow sold products made from its patented grades of polyethylene in Canada under the name “ELITE”.

[120]                     In December 2010, Dow sued Nova, claiming that the polyethylene marketed by Nova under the name “SURPASS” infringed the 705 Patent. Nova defended the claim, asserting that the 705 Patent was invalid. In September 2014, the Federal Court found that Dow’s patent was valid and that it had been infringed by Nova (2014 FC 844, 129 C.P.R. (4th) 199). This was affirmed on appeal (2016 FCA 216, 142 C.P.R. (4th) 339).

(2)          Nova’s Production of SURPASS

[121]                     Ethylene is a raw material used to make polyethylene. Ethylene was used by Nova in the production of both its infringing polyethylene, SURPASS, as well as other non‑infringing grades of polyethylene.

[122]                     Nova operated two primary business lines in Canada: (1) the Olefins unit, which produced ethylene; and (2) the Polyolefins unit, which produced polyethylene. Nova kept separate business records for the two units. These records indicate that Nova did not pay the market price for the ethylene used to manufacture SURPASS. Rather, as the reference judge found, it was able to produce ethylene in the Olefins unit “at a significant discount compared to the market price” (2017 FC 350, [2018] 2 F.C.R. 154, at para. 137). Nova paid an internal “transfer price” for that ethylene.

[123]                     During the period of patent infringement, Nova used a small quantity of its total ethylene production to make infringing grades of polyethylene. The rest of Nova’s ethylene was either sold to customers at market prices or used to make other non‑infringing grades of polyethylene, including “pail and crate” polyethylene.

D.           Judicial History

(1)          Reference Judge’s Decision: Federal Court, 2017 FC 350, [2018] 2 F.C.R. 154

[124]                     At trial, Nova conceded that there were “no ‘direct non‑infringing alternatives’” to the infringing SURPASS products (para. 146). Accordingly, it conceded that, based on its reading of post‑Schmeiser case law, the differential profit approach was unavailable. Nova then framed its argument about deducting ethylene costs, rather than profits, in the calculation of an accounting of profits. It argued that it should be permitted to deduct the market price of ethylene (i.e. the average third‑party selling price) rather than its actual incurred costs (i.e. the lower transfer price it paid). In the alternative, Nova argued that it should be permitted to deduct the full costs — which included variable, fixed, and capital costs — associated with its manufacture of ethylene (para. 134).

[125]                     The reference judge held that the differential profit approach was not available. When reciting the law on the availability of that approach, the judge cited Lovastatin, a damages case, holding that the infringer’s best non‑infringing option “must be a true substitute or real alternative” (para. 143). He also rejected Nova’s primary argument that it should be permitted to deduct the market price of ethylene rather than its actual incurred costs, as an “accounting of profits should be based on actual revenues and costs” (para. 139). However, he agreed with Nova’s alternative argument and permitted it to deduct its full costs, including both variable and fixed costs (para. 140).

[126]                     When calculating Nova’s deductible fixed costs associated with its manufacture of ethylene, the reference judge made the following important findings of fact:

I am satisfied that if Nova had not manufactured the infringing products, it would have worked assiduously to fill out the PE2 plant with other products: pail and crate or other resins that form part of Nova’s “product wheel”. I am further satisfied that Nova would have sold these other products within North America or to Asian markets. [para. 158]

[127]                     Ultimately, Nova was ordered to disgorge approximately $644 million to Dow (2017 FC 637).

(2)          Federal Court of Appeal Decision, 2020 FCA 141, [2021] 1 F.C.R. 551

[128]                     In the Court of Appeal, Nova recast its position, advancing two principal apportionment arguments. First, Nova argued that “had it not infringed Dow’s patent, it would have produced ethylene anyway and would have made lawful profits” from selling this ethylene at the market price (para. 90). Nova submitted that it should be credited for these hypothetical profits because it “would have earned . . . ethylene profits even without infringing” (para. 90, citing Nova’s revised memorandum of fact and law, at paras. 24, 55 and 56). Second, Nova argued that a portion of its profits was attributable to its unique ability to produce ethylene at a significant discount. These profits were not causally attributable to the patent but were instead attributable to Nova’s “Alberta Advantage”, and so they should be apportioned.

[129]                     The majority of the Federal Court of Appeal ultimately rejected both of these arguments, upholding the reference judge’s award of profits. It held that Nova’s first argument was purely hypothetical and therefore legally irrelevant to an accounting of profits. The majority further held that the reference judge had declined to make the requisite finding of fact with respect to that argument. The dissenting judge agreed with the majority’s recitation of the law. However, she concluded that the reference judge had erred by not considering the issue of causation, which pertained directly to the legal test of apportionment.

III.         Issues

[130]                     There are two main issues in this appeal. First, did the Federal Court of Appeal err in law by failing to apportion Nova’s ethylene profits in calculating the award to Dow? This turns on whether, as a matter of law, the differential profit approach can be applied only when an infringer’s best non‑infringing option is a true consumer substitute for the patented product.

[131]                     Second, did the Federal Court of Appeal err in law in upholding the award to Dow of springboard profits earned on products made and sold after the patent expired?

IV.         Analysis

[132]                     Given the complexity and difficulty associated with the accounting of profits remedy, a return to first principles is warranted. My analysis proceeds in three sections. First, I outline the principles governing an accounting of profits. Second, I assess whether the courts below erred in failing to apportion Nova’s ethylene profits by declining to apply the differential profit approach. This assessment entails consideration of whether, as a matter of law, an infringer’s best non‑infringing option needs to be a true consumer substitute for the patented product. Third, I assess whether the courts below erred in awarding springboard profits.

A.           Principles Governing an Accounting of Profits

(1)          Equitable Purposes That May Be Served by an Accounting of Profits

[133]                      While there are certain fundamental principles inherent in the operation of this remedy, the flexible and context‑sensitive nature of equitable remedies ensures that they are appropriately calibrated to address the wrong they are intended to redress. Indeed, as Dickson J. stated in Pettkus v. Becker, [1980] 2 S.C.R. 834, at pp. 847‑48, “[t]he great advantage of ancient principles of equity is their flexibility: the judiciary is thus able to shape these malleable principles so as to accommodate the changing needs and mores of society, in order to achieve justice.”

[134]                      For a proper understanding of how the remedy of an accounting of profits works in the patent context, it is important to begin by identifying the equitable purposes that may be served by this remedy. These purposes were examined in Strother, this Court’s leading case on the remedy. Strother involved a lawyer who had failed to advise a client about a favourable tax ruling from which the client could have benefited. In breach of his fiduciary duty, the lawyer had subsequently taken advantage of a business opportunity with another client related to that ruling, making millions in the process. The lawyer was ordered to account for the profits derived from the breach.

[135]                      Writing for the majority, Binnie J. stated that the remedy of “‘disgorgement’ of profit . . . may be directed to either or both of two equitable purposes”: (1) a prophylactic purpose aimed at deterring the wrongful conduct; and (2) a restitutionary purpose aimed at restoring profit that properly belongs to the plaintiff but has been wrongly appropriated by the defendant (paras. 75‑76 (emphasis added)). The restitutionary purpose is applicable, for example, to the “wrongful exploitation by the defendant of the plaintiff’s intellectual property” (para. 76).

[136]                      Before I elaborate in greater detail on how these two equitable purposes affect the operation of the remedy, a preliminary point of clarification is in order regarding the use of the term “restitution”. As this Court recently explained in Atlantic Lottery, restitution and disgorgement are two distinct gain‑based remedies: “. . . disgorgement requires only that the defendant gained a benefit (with no proof of deprivation to the plaintiff required), while restitution is awarded in response to the causative event of unjust enrichment . . ., where there is correspondence between the defendant’s gain and the plaintiff’s deprivation . . .” (para. 24 (emphasis omitted)).

[137]                      There is an apparent incongruity between the definition of “disgorgement” in Atlantic Lottery and the potential purpose of disgorgement as explained in Strother. In Atlantic Lottery, the Court stated that disgorgement and restitution are distinct gain‑based remedies, whereas, in Strother, it stated that disgorgement may have a restitutionary purpose. In my view, the description, in Strother, of disgorgement as having a restitutionary purpose is an example of the “inconsistent nomenclature” that has abounded in the jurisprudence (Atlantic Lottery, at para. 23). Properly understood, the restitutionary purpose referred to in Strother is more appropriately described as a restorative purpose. It functions to restore the wrongdoer’s improper gain to the plaintiff, as the gain properly belongs to the plaintiff.

[138]                      In the interests of “greater precision” and in order to avoid the “ambiguity inherent in such terminology” (Atlantic Lottery, at para. 23), I will refer to a restorative purpose rather than a restitutionary one. In my view, this more precise terminology accurately captures what was meant in Strother while avoiding the unnecessary confusion discussed in Atlantic Lottery. The purpose of the remedy is to restore the wrongdoer, as “the wrong disgorgement rectifies is the wrongdoer’s use of the claimant’s entitlement without consent. [The wrong] is not using the entitlement in combination with the wrongdoer’s own resources to produce wealth” (M. P. Gergen, “Causation in Disgorgement” (2012), 92 B.U.L. Rev. 827, at p. 830). “But for” causation is therefore appropriate, as the focus is on restoring the wrongdoer to the position it would have occupied but for its wrongdoing. Damages, which are a loss‑based remedy, restore the plaintiff to the position it would have been in but for the infringement. It is only logical that an accounting of profits, which focuses on the wrongdoer’s gains, seeks to restore the wrongdoer to the position it would have been in but for its wrongdoing (see Philip Morris Products S.A. v. Marlboro Canada Ltd., 2015 FC 364, [2015] F.C.J. No. 1564 (QL), at para. 17).

[139]                      Although he does not refer to Strother, my colleague describes an accounting of profits as a remedial tool that protects the patent bargain by (1) requiring infringers to disgorge the profits earned from patent infringement and (2) ensuring that infringers are deterred but not punished (Rowe J.’s reasons, at para. 44). It is unclear to me whether my colleague is affirming what was said in Strother, altering the terminology used in Strother (as I am doing), or departing from the discussion of the equitable purposes of an accounting of profits from Strother altogether.

(2)          Equitable Purposes Served by an Accounting of Profits in the Patent Context

[140]                      In order to highlight the equitable purposes served by an accounting of profits in the patent context, it is helpful, in my view, to contrast how the remedy operates in a fiduciary context. Indeed, as I said above, the focus of an accounting of profits in the patent context is quite different than it is in the context of a breach of a fiduciary duty. Specifically, in the patent context, the focus is on restoration — that is, restoring to the patentee the profits that the infringer gained as a result of its exploitation of the patented invention. Deterrence is not an overriding purpose, although it does play a limited role in helping to preserve the patent monopoly.

[141]                      My colleague, in his approach, fails to sufficiently identify and explain the equitable purposes served by an accounting of profits in the patent context. Since this Court’s decision in Strother, much of the accounting of profits case law has simply stated that this remedy has two equitable purposes: restitution and deterrence. At para. 44 of his reasons, my colleague cites ADIR (2020) for this proposition. With respect, that decision does not accurately describe what was said in Strother. This Court stated in Strother that the accounting of profits “remedy may be directed to either or both of two equitable purposes” (para. 75). In my view, it is important to understand conceptually how the remedy operates in the patent context and how its operation differs when there has been a breach of a fiduciary duty. I do not deny that an accounting of profits can have a deterrent purpose in the patent context, as it can deter intentional and efficient infringements. However, the primary purpose of the remedy in this context is restorative. Describing the deterrent purpose inaccurately risks distorting the causation analysis.

(a)       Purpose of an Accounting of Profits in the Fiduciary Context

[142]                      When there has been a breach of a fiduciary duty, the importance of the prophylactic deterrence purpose of the accounting of profits remedy cannot be understated. Equity abhors a breach of a fiduciary duty. This Court has long recognized the unique policy goals underlying fiduciary relationships, which equity seeks to uphold. As La Forest J. stated in Hodgkinson v. Simms, [1994] 3 S.C.R. 377, at p. 453,

[t]he law of fiduciary duties has always contained within it an element of deterrence. . . . In this way the law is able to monitor a given relationship society views as socially useful while avoiding the necessity of formal regulation that may tend to hamper its social utility.

 

(See also Canadian Aero Service Ltd. v. O’Malley, [1974] S.C.R. 592, at pp. 607 and 610; Lac Minerals Ltd. v. International Corona Resources Ltd., [1989] 2 S.C.R. 574, at pp. 672‑73, per La Forest J.; Canson Enterprises Ltd. v. Boughton & Co., [1991] 3 S.C.R. 534, at pp. 571‑74, per La Forest J., and pp. 546‑47, per McLachlin J.)

[143]                      Equitable remedies are flexible and context-sensitive. As La Forest J., writing for the majority, stated in Canson,

[w]here a situation requires different policy objectives, then the remedy may be found in the system that appears more appropriate. This will often be equity. Its flexible remedies such as constructive trusts, account, tracing and compensation must continue to be moulded to meet the requirements of fairness and justice in specific situations. . . .

 

[E]quity cannot be rigidly applied. Its doctrines must be attuned to different circumstances. Quite obviously not all fiduciary obligations are the same. It would be wholly inappropriate to interpret equitable doctrines so technically as to displace common law rules that achieve substantial justice in areas of common concern, thereby leading to harsh and inequitable results. [p. 588]

[144]                      Given equity’s acute sensitivity to the need to preserve a fiduciary relationship and deter the faithless fiduciary, the absence of a fiduciary relationship may significantly attenuate the deterrence purpose of an equitable remedy. Indeed, in Cadbury Schweppes Inc. v. FBI Foods Ltd., [1999] 1 S.C.R. 142, which involved a breach of confidence in a non‑fiduciary relationship, Binnie J. emphasized that “[t]he overriding deterrence objective applicable to situations of particular vulnerability to the exercise of a discretionary power . . . does not operate here” (para. 30 (citations omitted)). This played a considerable role in weighing the equities of the case and fashioning the appropriate remedy.

(b)      Purposes of an Accounting of Profits in the Patent Context

[145]                      An accounting of profits may serve either or both of two equitable purposes: a restorative purpose and a prophylactic deterrence purpose. It does not need to serve both purposes. As Professor Devonshire explains, the function of an accounting of profits for an infringement of a patent is very different than it is for a breach of a fiduciary duty. When used to remedy a breach of a fiduciary duty, an accounting of profits “gives effect to concepts such as prophylaxis and the vindication of trust and confidence. In intellectual property cases the remedy reflects the more limited objective of disgorging gains attributable to infringement of the plaintiff’s interests” (P. Devonshire, Account of Profits (2013), at p. 131 (footnotes omitted)).

(i)            Restorative Purpose

[146]                      As noted, the restorative purpose of an accounting of profits in the patent context is to place the infringer in the position it would have been in but for the infringement. In the case of loss‑based remedies, such as damages or equitable compensation, the aim is to compensate the injured plaintiff, restoring it to the position it would have been in but for the infringement. However, an accounting of profits is a gain‑based remedy, so the focus is on the wrongdoer’s gains and the purpose is to restore the wrongdoer to the position it would otherwise have occupied by forcing it to disgorge the profits attributable to its wrongdoing.

(ii)         Prophylactic Deterrence Purpose

[147]                      With respect to the prophylactic deterrence purpose of an accounting of profits, in my view, there is no overriding deterrence objective in the patent context. As in Cadbury Schweppes, the absence of a fiduciary relationship means that deterrence is not an overriding objective. Moreover, there are countervailing policy concerns about overdetermined gains (i.e. windfalls) and over‑deterrence that are especially acute with respect to the disgorgement of profits, particularly in the patent context. This Court discussed concerns about over‑deterrence in this context in Free World Trust. In my view, Binnie J.’s comments at para. 42 remain apposite:

The patent system is designed to advance research and development and to encourage broader economic activity. Achievement of these objectives is undermined however if competitors fear to tread in the vicinity of the patent because its scope lacks a reasonable measure of precision and certainty. A patent of uncertain scope becomes “a public nuisance” (R.C.A. Photophone, Ld. v. Gaumont‑British Picture Corp. (1936), 53 R.P.C. 167 (Eng. C.A.), at p. 195). Potential competitors are deterred from working in areas that are not in fact covered by the patent even though costly and protracted litigation (which in the case of patent disputes can be very costly and protracted indeed) might confirm that what the competitors propose to do is entirely lawful. Potential investment is lost or otherwise directed. Competition is “chilled”. The patent owner is getting more of a monopoly than the public bargained for. There is a high economic cost attached to uncertainty and it is the proper policy of patent law to keep it to a minimum. [Emphasis added.]

[148]                      I recognize that Binnie J. was discussing the need to ensure that the boundaries of a patent are both fair and reasonably predictable. In my view, the same logic applies to the disgorgement remedy. The remedy should not punish potential competitors for competing in areas they reasonably believe are not properly covered by a patent: if a patentee were able to obtain a windfall by being awarded the infringer’s profits that were not causally attributable to the infringement, the remedy would become punitive and the scope of the patent monopoly would be unduly expanded. The patent bargain would be distorted and competition would be chilled.

[149]                      Concerns about over‑deterrence have also been common in the academic literature on the accounting of profits remedy, both generally and in the patent context (see, e.g., Devonshire, at pp. 58‑71 and 131‑33; Gergen, at p. 837; C. B. Seaman et al., “Lost Profits and Disgorgement”, in C. B. Biddle et al., eds., Patent Remedies and Complex Products: Toward a Global Consensus (2019), 50, at pp. 60 and 72‑73; T. F. Cotter, Comparative Patent Remedies: A Legal and Economic Analysis (2013), at p. 69; N. Siebrasse, “A Remedial Benefit‑Based Approach to the Innocent‑User Problem in the Patenting of Higher Life Forms” (2004), 20 C.I.P.R. 79, at pp. 96‑98).

[150]                      I further note that any significant deterrent purpose of an accounting of profits in this context was effectively rejected by the Federal Court of Appeal in ADIR (2017). In that case, the lower court had dismissed the relevance of using a non‑infringing option. According to the Federal Court, if the infringer could establish a non‑infringing option in which it made the same hypothetical profits, it would have “a perfect shelter against the consequences of any future patent infringement in Canada” (ADIR v. Apotex Inc., 2015 FC 721, 482 F.T.R. 276, at para. 121). However, the Federal Court of Appeal appropriately, and in a manner consistent with this Court’s decision in Schmeiser, rejected the concern about a “perfect shelter”. It emphasized that other available remedies — such as elevated costs, injunctive relief and punitive damages — counterbalance any incentive to infringe (Lovastatin, at para. 71).

[151]                      To be clear, I am not saying that an accounting of profits does not have any deterrent purpose in the patent context. However, in this context, the deterrent effect of disgorgement is limited and must be viewed in conjunction with other available remedies. Specifically, the statutory availability of an accounting of profits helps to deter infringement to the extent that it ensures that, even if the patentee suffers no or minimal compensable damage, the infringer will not be able to retain any profits causally attributable to its infringement. The remedy therefore deters competitors from intentionally infringing a patent even where the infringement can be described as “efficient”. This ensures that the patent monopoly is upheld and respected.

(3)          Role of Causation

[152]                      Given the different purposes of an accounting of profits, the question becomes how this equitable remedy operates in different circumstances. The answer to this question lies in the approach to causation. For both a breach of a fiduciary duty and a patent infringement, a causal relationship between the breach or infringement and the profits is required (Strother, at para. 79; Schmeiser, at para. 101; Lubrizol).

[153]                      However, there are important differences in how causation is approached. A comparison between the causation analysis in Strother and that in Schmeiser is illustrative. Differences in context justify the subtle, yet significant, differences in the approach to causation. This is entirely consistent with the flexible nature of equitable remedies (see, e.g., Hodgkinson, at p. 443). My intention here is to illuminate the differences in the causation analysis generally. I should not be understood as providing comprehensive guidance on how an accounting of profits operates for a breach of a fiduciary duty. That is not at issue in this appeal.

(a)          Apportionment Is Inappropriate When Deterrence Is the Overriding Purpose

[154]                      In Strother, this Court ordered the faithless fiduciary to “account for profit earned from the personal financial opportunity he pursued in breach of his fiduciary duty” (para. 85). The Court emphasized that Mr. Strother “must disgorge profits related to his breaches of loyalty” and that “[i]f the prophylactic purpose of the equitable remedy is to be achieved, [Mr.] Strother cannot be permitted to pocket the money thus derived from a personal interest in conflict with his fiduciary duty” (paras. 86‑87).

[155]                      Before this Court, Mr. Strother argued that he should be permitted to apportion his profits depending on the extent to which they were attributable to the breach. Binnie J. rejected the apportionment argument, emphasizing the prophylactic deterrence purpose of the remedy in this context:

In my view, this is not a case for apportionment. The Court’s purpose here is prophylactic rather than restitutionary. We are not therefore engaged in allocating an amount of profit amongst different contributing sources (or “profit drivers”). Strother acquired a personal financial interest in one client that conflicted with his duty to provide full and candid advice to another concurrent client. He should not be permitted to profit from that conflicting interest even though it is justly said that his own skill and experience were major contributors to those profits. Apportionment in these circumstances would reward the breach and undermine achievement of the prophylactic purpose. [Emphasis added; emphasis in original deleted; para. 96.]

[156]                      The gravity of the wrongdoing and equity’s concern for preserving the integrity of fiduciary relationships may therefore influence the strictness of limiting principles such as causation (P. Devonshire, “Account of Profits for Breach of Fiduciary Duty” (2010), 32 Sydney L. Rev. 389, at pp. 394‑404). The prophylactic deterrence purpose of an accounting of profits “advances the policy of equity, even at the expense of a windfall to the wronged [party]” (Strother, at para. 77). In these circumstances, apportionment — that is, the allocation of profits among different sources — may be inappropriate. As Professor Gergen notes, “the usual test of but‑for causation gives way to an expanded test to treat conduct as a cause of an event despite an imagined (and sometimes a non‑imagined) competing cause if reasons of policy or fairness justify disregarding the competing cause” (p. 835).

[157]                      In contrast, the absence of a comparable and overriding deterrent purpose in a patent infringement case means that there is an intense focus on the profits causally attributable to the infringement. This is epitomized by the differential profit approach, which seeks to isolate the various “profit drivers”, to use the term from Strother, that are attributable to the infringer’s unauthorized use of the patentee’s invention.

(b)          Differential Profit Approach

[158]                      In the patent context, there is an acute focus on causation: it is settled law that the patentee is entitled only to the profits causally attributable to the infringement of the patent (Schmeiser, at para. 101; Lubrizol). However, isolating these profits has proven notoriously difficult, which is evident from both the jurisprudence and the academic commentary.

[159]                      In Reading & Bates Construction Co. v. Baker Energy Resources Corp., [1995] 1 F.C. 483 (C.A.), at p. 493, Létourneau J.A. noted that “the exercise of this remedy has been associated with a number of practical difficulties” (see also AlliedSignal, at pp. 444‑46; Beloit, at para. 113; Constellation Brands US Operations Inc. v. Société de vin internationale ltée, 2021 QCCA 1664, at paras. 38‑40 (CanLII)).

[160]                      Academics have similarly noted the challenges posed by the apportionment of profits in the patent infringement context. For example, Professor Siebrasse describes the challenges faced by courts when attempting to apportion profits among different profit drivers:

The goal of an accounting of profits is to force the wrongdoer to disgorge all profits made by the defendant as a result of his infringement. This can quickly lead to difficult questions, because profits may spring from the use of many complementary non‑infringing assets. The traditional solution was to award the plaintiff only that portion of the defendant’s overall profits which was attributable to the infringement. Determining when and how to conduct this “apportionment” of profits has been among the most difficult legal questions . . . . [Emphasis added; endnote omitted.]

 

(N. V. Siebrasse, A. J. Stack et al., “Accounting of Profits in Intellectual Property Cases in Canada (2007)” (2008), 24 C.I.P.R. 83, at p. 112; see also L. Bently et al., Intellectual Property Law (5th ed. 2018), at pp. 1348‑49; L. Smith and J. Berryman, “Disgorgement of Profits in Canada”, in E. Hondius and A. Janssen, eds., Disgorgement of Profits: Gain‑Based Remedies throughout the World (2015), 281, at pp. 292‑93.)

[161]                      The difficulties inherent in the apportionment exercise were significantly reduced by this Court’s decision in Schmeiser, in which the differential profit approach was affirmed as being the “preferred means of calculating an accounting of profits” in relation to infringements of intellectual property rights (para. 102). Indeed, there is a principled reason why the differential profit approach is the preferred approach in this context: it best approximates the causal contribution of the infringement.

[162]                      The differential profit approach isolates the profits earned by the infringer as a result of the infringement. This approach therefore isolates the value of the patent in the hands of the infringer. It does so by comparing the infringer’s actual profits with its hypothetical profits had it used the best non‑infringing option (Schmeiser, at para. 102). In other words, this approach compares actual profits with the profits that could have and would have been earned but for the infringement.

[163]                      This is why the differential profit approach has also been called the “value‑based” approach. Through a comparison between actual profits and hypothetical profits based on the best non‑infringing option, the differential profit approach subsumes the apportionment analysis. Once the profits causally attributable to the infringement have been isolated, “[n]o other apportionment would appear to be necessary to restore the [infringer] to the position it would have occupied, but for the infringement” (Cotter, at p. 205).

[164]                      In my view, there is no doubt that this is what the Court intended to lay down as the governing test in Schmeiser. A brief examination of the history of the remedy of an accounting of profits leading up to that case will demonstrate this.

[165]                     Prior to Schmeiser, the differential profit approach had been rejected by the Federal Court of Appeal in Reading & Bates. The language used by the court is telling: “. . . one has to look at the profits that the appellant actually made through the infringing acts, not the profit that he could have made had he used a non‑infringing method” (p. 496; see also Wellcome Foundation Ltd. v. Apotex Inc., [2001] 2 F.C. 618 (C.A.), at paras. 19‑21).

[166]                     While the Court in Schmeiser did not rely on the Federal Court of Appeal’s jurisprudence, it did refer to Celanese International Corp. v. BP Chemicals Ltd, [1999] R.P.C. 203. In that case, the English High Court engaged in an extensive discussion on the point at issue here. Laddie J. described two approaches to an accounting of profits: the “incremental” or differential profit approach and the “apportionment” approach. The first was explained as follows by counsel for BP Chemicals Ltd.: “The correct approach in assessing the profits derived from infringement is to assess the value of the benefits arising from using the guard bed, compared to the alternative which BP were most likely to have used otherwise, absent infringement” (para. 16). In other words, the differential profit approach relies on “but for” causation to put the defendant in the position it would have been in but for the infringement. The second approach, which Laddie J. adopted, was described by him as follows: “Apportionment looks at the profits actually made on the whole process or article and, where appropriate, splits them between those parts which infringe and those which do not” (para. 34). He added that it “should not matter that similar profits could have been made in another, non‑infringing way” (para. 39).

[167]                     Thus, Laddie J. expressly rejected the differential profit approach in favour of apportionment. While Celanese was cited by this Court in Schmeiser, at para. 101, for the proposition that causal attribution of profits is necessary, it is plain that the apportionment approach was rejected in favour of the differential profit approach using “but for” causation in the very next paragraph. It can hardly be suggested that this Court had not read Laddie J.’s judgment when it decided Schmeiser. Yet that is what is suggested by my colleague’s reasons, insofar as he reads Schmeiser as not requiring the application of the “but for” test and instead requiring courts to focus exclusively on the abstract value of the invention.

[168]                     The facts of Schmeiser confirm this analysis. Mr. Schmeiser had been ordered by the Court of Appeal to account for all of the profit obtained from his crops, “not the difference between that profit and the profit from the sale of an alternative crop that was not grown” (Monsanto Canada Inc. v. Schmeiser, 2002 FCA 309, [2003] 2 F.C. 165, at para. 79). But that difference is precisely what this Court ordered, i.e., $0. Mr. Schmeiser was therefore restored to the position he would have been in had he not infringed the patent.

[169]                     Further support for my proposition is evidenced by the fact that, in Schmeiser, this Court cited Siebrasse (2004). A single excerpt from that article suffices to show that Professor Siebrasse — and, by extension, the majority in Schmeiser — preferred the “but for” test which my colleague now rejects:

The argument in favour of the differential-profit approach is that it is simply the application of “but for” causation to an accounting of profits.

 

. . . the causation requirement leads directly to the conclusion that the plaintiff is not entitled to all her loss, but only to her “differential loss” — namely, the difference between the “injured position,” which is the plaintiff’s position in fact, and the “original position,” which the plaintiff would have been in but for the wrong.

 

From this we see that the differential‑profit approach to an accounting is simply a more specialized statement of this general principle. It says that the defendant’s profit caused by the infringement is the difference between the profit that the defendant in fact made and the profit that the defendant would have made but for the infringement, on the supposition that but for the infringement the defendant would have used the next best non‑infringing method. [Emphasis added; p. 91.]

Thus, it was indeed held in Schmeiser that the “purpose of an accounting of profits was to determine what profits the infringer ‘could have earned’ but for the infringement”, at least insofar as doing so is necessary to restore the infringer to the position it would have been in had it not infringed the patent (Rowe J.’s reasons, at para. 60).

[170]                     As I note further below, a few years before it decided in this case, the Federal Court of Appeal itself held that “but for” reasoning was essential to the calculation of an accounting of profits. In ADIR (2017), at para. 40, it stated that “the [c]ourt is to consider a hypothetical world where the infringing conduct did not take place”. To put the point beyond any doubt, the court said the following:

For clarity, the issue I would remit to the Federal Court is whether Apotex would have and could have obtained quantities of non‑infringing perindopril . . . and, if so, whether Apotex would have and could have used non‑infringing perindopril for sales to its affiliates in the United Kingdom and Australia. [Emphasis added; para. 65.]

And mere months prior to rendering judgment in this case, the Federal Court of Appeal had resort to the differential profit approach as I apply it (ADIR (2020)).

[171]                     In sum, and with respect, contrary to what Stratas J.A. said in the decision under appeal — and to what my colleague now states — consideration of what the defendant could have and would have done is of central importance to the calculation of profits on an accounting, as it helps to restore the defendant to the position it would have been in but for the infringement. I would add that it is unclear to me on what basis the Federal Court of Appeal departed from its own precedents.

[172]                      My colleague criticizes me for stating that the focus of an accounting of profits is on the “value of the invention in the hands of the infringer”, as this phrase was “never used in Schmeiser” and it would be inappropriate for this Court to “adjust the focus of an accounting of profits to align with new terminology not used in the judgment” (paras. 60‑61). It is unclear what, specifically, my colleague finds objectionable about this phrase. In my view, it is simply another way of stating that the infringer must disgorge the profits causally attributable to the infringement — this is precisely why the court examines and isolates the value derived from the infringer’s use of the invention.

[173]                      With respect, I am of the view that the phrase “causally attributable to the invention”, which was used in Schmeiser, at para. 101, is ambiguous and creates confusion. It seems to imply that an accounting of profits is intended to value the patent in the abstract rather than to remedy a specific infringement by a specific defendant at a specific point in time. To the extent that my colleague is suggesting that the purpose of an accounting of profits is to value the patent in the abstract, I disagree and I believe that this is a misreading of Schmeiser. The phrase “causally attributable to the invention” should not be understood in that manner. Rather, the purpose is to isolate the value of the patent in the hands of the infringer. My interpretation of Schmeiser, and my consistent use of the phrase “causally attributable to the infringement”, are preferable for four reasons.

[174]                      First, although it is not included in my colleague’s block quotation at para. 60 of his reasons, the phrase “causally attributable to the invention” used in Schmeiser was cited to Lubrizol and Celanese, at para. 37. Both of these cases clearly indicate that the focus is on the wrongdoing — that is, the infringement of the patent — and not on the purported abstract value of the patent. In Lubrizol, for example, the Federal Court of Appeal stated: “It may be possible for Imperial to show that some part of the profits made on the infringing sales are not profits ‘arising from’ the infringement in that they are not caused by but simply made on the occasion of such infringement” (para. 9 (emphasis added)). In Celanese, the court similarly underscored that the proper focus is on the infringement:

Although an account may give rise to a very different figure to that on an inquiry as to damages, they both proceed on a common principle of legal causation. On an inquiry the court is trying to determine what damage has been caused, in a legal sense, by the defendant’s wrongful acts. It has to decide whether the breach was the cause of the loss or merely the occasion of it . . . . In an account the court is trying to determine what profits have been caused, in a legal sense, by those acts. [para. 37]

[175]                      Second, the application of the differential profit approach in Schmeiser also confirms that the focus is on the infringement. In Schmeiser, Monsanto’s patent pertained to genetically modified genes and cells that could be inserted into plants to increase their tolerance to glyphosate herbicides such as Roundup. Canola seed containing the patented genes and cells was marketed under the name “Roundup Ready Canola”. It was not the abstract value of the patented invention that mattered, but rather the value of the patented invention in Mr. Schmeiser’s hands. Because Mr. Schmeiser had not sprayed his fields with Roundup, he had gained no “agricultural advantage” and his “profits were precisely what they would have been” but for the infringement (para. 104 (emphasis added)).

[176]                      Third, a narrow interpretation that focuses exclusively on the value of the invention, and not on the value derived from the infringement, is inconsistent with the proposition that the profits to be disgorged are those that, “on a common sense view of causation, were caused by the breach” (Schmeiser, at para. 101, quoting Canson, at p. 556 (emphasis added)). The phrase “a common sense view of causation” clarifies that “but for” causation applies, even though other common law requirements for legal causation, such as foreseeability and remoteness, do not readily apply in equity (Southwind, at para. 75). The application of common sense “but for” causation is therefore appropriate when isolating the profits causally attributable to the infringement.

[177]                     Fourth, a focus on the infringement, rather than on the purported value of the invention, is necessary to capture all of the potential profits that the infringer may derive from its breach of the patentee’s statutory monopoly. There is a difference between looking, in the abstract, at the value that the patent brings to the product and looking at the value that resulted from the infringement. Professor Siebrasse articulates this difference clearly:

No one knows the value contributed by the inventive quality of the patented technology and it does not matter. The rationale for the patent system is that it provides an incentive to invent. That incentive is provided by the private return to the patentee from exploiting the invention. In a world without any infringement, that private return is determined by the advantage that the patentee is able to extract in real‑world competition with various rivals using non‑infringing technology. The advantage of the patented technology over the non‑infringing options does not turn solely on whatever value we might decide is attributable to the inventive quality of the technology; it turns on the actual difference in profitability, which in turn does depend on things like distribution advantages, including the practical availability or otherwise of the putative alternatives, as well as all the other elements that go into making a real product and bringing it to market.

 

. . .

 

. . . The benefit as a result of the infringement is typically not the value attributable to the inventive quality of the patented technology; it is the difference between what the infringer in fact made and what it would have made in the “but for” world in which it used the best non‑infringing option.

[Emphasis in original.]

 

(N. Siebrasse, Non‑Infringing Baseline as an Alternative to “But For” Causation, October 19, 2020 (online))

[178]                      With respect, I am of the view that my colleague is imprecise and inconsistent in describing the focus of an accounting of profits. In most of his reasons, he refers to this focus as being the disgorgement of profits “causally attributable to the invention” (see, e.g., paras. 1, 3, 12‑13, 15, 40, 46‑48, 50‑53, 55, 57‑58, 60‑61, 64, 67 and 71 (emphasis added)). However, when discussing springboard profits in particular, my colleague describes the focus of this remedy as being the disgorgement of profits “causally attributable to infringement of the invention” (e.g., paras. 4, 27, 74, 81‑82 and 84‑85 (emphasis added)).

[179]                     My colleague needs this shift in focus to justify his conclusion on springboard profits. Without this shift, he would not be able to justify an award of springboard profits, as such profits have absolutely nothing to do with the purported “value of the invention”. They are the profits the infringer earns after the expiry of the patent that can be attributed to the infringement of the patentee’s statutory monopoly during the life of the patent. By infringing the patent during the life of the patent, the infringer builds sales and market share that increase its profits after the patent expires. In short, springboard profits are profits that an infringer derives from its infringement, rather than profits derived from the purported value of the invention itself.

[180]                      I am of the respectful view that the phrase “causally attributable to infringement of the invention” is inaccurate and incompatible with the nature of patent infringement. An infringer that uses a patented invention without authorization does not infringe the invention; rather, it infringes the patent itself, which is the patentee’s statutory monopoly over the use and exploitation of the invention.

[181]                      Moreover, I would note that this phrase is novel, as it is conspicuously absent from Schmeiser. If, as my colleague states, the exact language from Schmeiser is “clear, unambiguous, and binding on this Court” such that it would be inappropriate to “adjust the focus of an accounting of profits to align with new terminology not used in the judgment”, then he must justify an award of springboard profits in light of exactly what was stated in Schmeiser: “. . . the inventor is only entitled to that portion of the infringer’s profit which is causally attributable to the invention . . .” (Rowe J.’s reasons, at para. 61; Schmeiser, at para. 101 (emphasis added)). With respect, my colleague cannot have it both ways. He cannot criticize my approach, which focuses on the value derived from the infringement, while also adopting it when calculating springboard profits. Profits earned during the life of the patent cannot and should not be calculated in a different manner than profits earned after its expiry.

[182]                     Relatedly, my colleague’s reasons largely avoid any use of the words “but for” or “hypothetical”. His reasons refer to “hypothetical” profits only twice (at paras. 3 and 72), and “but for” is used only when discussing springboard profits (para. 81). This merely adds to the confusion about the proper operation of the differential profit approach. It is especially problematic given the obvious “tension between two Federal Court of Appeal panels on the use of a ‘but‑for’ analysis in cases involving accounting of profits” (Perry and Currier, at §17.60). The decisions to which the authors refer are the one at issue in this appeal and ADIR (2020). While my colleague cites the latter case at various points in his reasons, it is unclear to me which approach he endorses. Is hypothetical “but for” reasoning prohibited in an accounting of profits, or is it permissible? Or is such reasoning permissible only when calculating springboard profits? My colleague does not address these issues.

[183]                     Clearly, in my view, hypothetical “but for” reasoning is both appropriate and necessary when applying the differential profit approach. This approach is inherently premised on the creation of a hypothetical “but for” world. This Court’s decision in Schmeiser and subsequent jurisprudence from the Federal Court of Appeal clearly demonstrate the relevance of such reasoning (see, e.g., Schmeiser, at para. 104; ADIR (2017), at paras. 40‑41, 61, 71 and 79; ADIR (2020), at paras. 47‑51).

[184]                     The academic literature on an accounting of profits in patent cases has also discussed the relevance of “but for” causation and the creation of reasonable hypotheticals (see, e.g., Cotter, at pp. 197 and 205; Siebrasse, Stack et al., at pp. 89, 112‑13 and 121‑22; Siebrasse (2004), at pp. 91‑93; Seaman et al., at p. 50; Gergen, at p. 832; Devonshire (2013), at p. 133; Perry and Currier, at §§17.59‑17.63; Smith and Berryman, at p. 292).

[185]                      To summarize, the focus of an accounting of profits in the patent context must be on the infringer’s profits that are causally attributable to its infringement. In other words, what profits did the infringer earn as a result of its infringement of the patentee’s statutory monopoly and its exploitation of the invention? This is determined using “but for” causation. This approach is consistent with the gain‑based nature of the remedy and ensures that the analysis remains unwaveringly focused on the infringer and the profits it earned — whether during the life of the patent or after its expiry — as a result of its infringement.

B.            The Differential Profit Approach Should Have Been Available

[186]                     In my view, this appeal directly raises the following legal questions. First, does a non‑infringing option have to be a true substitute (i.e. a consumer substitute) for the patented product? This is a key issue raised by Nova both in its factum and in oral argument. Dow also addressed this issue in its factum as well as in oral argument. Second, how are courts to determine the appropriate hypothetical “best non‑infringing option” to be used? I will answer these questions in turn.

(1)          An Infringer’s Best Non‑Infringing Option Does Not Need to Be a True Consumer Substitute

[187]                     For the reasons that follow, I conclude that, under the differential profit approach, a non‑infringing option does not need to be a true consumer substitute for the patented product. There are two principled reasons for this conclusion: (1) a true consumer substitute requirement is legally irrelevant in an accounting of profits given the different purposes and focus of this remedy when compared to damages; and (2) limiting non‑infringing options to true consumer substitutes distorts the causation analysis.

(a)           A True Consumer Substitute Requirement Is Legally Irrelevant Given the Different Purposes and Focus of an Accounting of Profits

[188]                     An accounting of profits and damages are two remedies that ultimately serve different purposes and therefore have a different focus (Snell’s Equity (34th ed. 2020), by J. McGhee and S. Elliott, at para. 20‑037). A true consumer substitute requirement is legally irrelevant in an accounting of profits in light of the different purposes and focus of this remedy.

[189]                     To reiterate, an award of damages is compensatory. An infringer’s hypothetical sales of a non‑infringing product are legally relevant to the extent that they reduce the patentee’s lost sales of the patented product. This can be the case only if the two products would have competed in the same market. In contrast, an accounting of profits is not compensatory, as the profits to be disgorged “are calculated exclusively by reference to the defendant’s wrongful gain, irrespective of whether it corresponds to damage suffered by the plaintiff” (Atlantic Lottery, at para. 23). It is therefore legally irrelevant whether consumers would view the product that the infringer could have and would have made in the hypothetical world as a substitute for the patented product. What matters is the profits the infringer derived from its breach of the patentee’s statutory monopoly. This is most accurately determined through the differential profit approach, which uses “but for” causation to identify the infringer’s best non‑infringing option.

[190]                     My colleague states that “[t]here are no strict rules” concerning the “factual exercise” of determining whether there is a non‑infringing option and that such an option “need not be a strict market substitute for the patented product” (para. 67). In my view, the elimination of a “strict” requirement that the non‑infringing option be a true consumer substitute is a positive and welcome development. However, and with respect, my colleague’s analysis, and his upholding of the reference judge’s legal conclusion, are problematic for three reasons.

[191]                     First, the reference judge’s two factual findings on which my colleague relies clearly show that the “distinct [consumer] market” remained central and legally relevant in the judge’s analysis. According to my colleague, the reference judge did not err in holding that there were no non‑infringing options because, for one thing, customers that purchased Nova’s SURPASS did so due to the features captured by Dow’s patent, as Dow’s invention had “created a distinct market and Nova could only service that market because it sold infringing plastics” (Rowe J.’s reasons, at para. 69 (emphasis added)). Clearly, the “distinct [consumer] market” remains legally relevant under my colleague’s approach. My colleague also concludes that the reference judge made no error in holding there were no non‑infringing options because Nova did not establish a relevant non‑infringing option and even conceded that there were no such options (Rowe J.’s reasons, at para. 70). However, it is noteworthy that this concession was ultimately based on Nova’s understanding — shared by the reference judge and the Court of Appeal — that a non‑infringing option was required by law to be a true consumer substitute.

[192]                     Second, my colleague does not take into account the reference judge’s conclusion that the differential profit approach was legally unavailable. Citing Lovastatin, a damages case, the judge held that the non‑infringing option “must be a true substitute or real alternative” (para. 143). On my colleague’s own approach, therefore, the judge erred in law. This ought to answer any concerns about the legitimacy of this appeal in light of Nova’s concessions, a point upon which I expand below.

[193]                     Third, my colleague is equivocal and unclear about the legal requirements for a non‑infringing option. Whether such an option exists cannot be, as my colleague states, “a question of fact” (para. 67). There must be some legal requirements or standards that guide courts and the parties to the litigation. It is, and must be, a question of mixed fact and law. I do not deny that the determination of whether a non‑infringing option exists is a factually suffused task that, absent an error of law or an error in principle, is reviewable on a standard of palpable and overriding error (Housen v. Nikolaisen, 2002 SCC 33, [2002] 2 S.C.R. 235, at paras. 26‑29; Cinar Corporation v. Robinson, 2013 SCC 73, [2013] 3 S.C.R. 1168, at para. 78). However, there must be some legal standards that guide the analysis and, in my view, this appeal requires our Court to set out those standards clearly. The approach I endorse regarding how courts are to determine whether a non‑infringing option exists involves consideration of what the infringer could have and would have done but for its infringement.

(b)          A True Consumer Substitute Requirement Distorts the Causation Analysis

[194]                     Limiting non‑infringing options to consumer substitutes (or at least maintaining that the consumer market for the patented product is relevant to determining whether a non‑infringing option exists) distorts the causation analysis. By focusing on the value of the patent in the abstract, the analysis ignores the value that the infringement brought to the infringer. The problem with focusing on the value of the patent in the abstract is that it needlessly disadvantages infringers that could have and would have produced a different product that would not have been a consumer substitute for the patented product. Conversely, it risks incentivizing efficient infringement by benefiting infringers that could not or would not otherwise have produced a consumer substitute for the patented product. I explain this below.

[195]                     A true consumer substitute requirement works to the disadvantage of an infringer that could have and would have produced a non‑infringing product that would have competed in a different consumer market. Without any principled justification, it prevents the application of the differential profit approach. This is a distortion of the causation analysis, as it forces an infringer to disgorge gains that are not causally attributable to its infringement.

[196]                     This distortion can be seen in the present case. The reference judge made specific findings about what Nova could have and would have done but for the infringement. Indeed, with respect to Nova’s PE2 plant where the infringing grades of polyethylene were produced, the judge found as a fact that “Nova could use the PE2 plant flexibly to replace all infringing products with [non‑infringing alternatives]” (para. 86 (emphasis added)). He also found that Nova “would have worked assiduously to fill out the PE2 plant with other products:” and that it “would have sold these other products within North America or to Asian markets” (para. 158 (emphasis added)). Nevertheless, the application of the damages case law, which required a true consumer substitute, prevented the use of the differential profit approach. This distorted the causation analysis by preventing a comparison between Nova’s actual profits and its hypothetical profits had it not infringed.

[197]                     In this case, my colleague’s approach forces Nova to disgorge millions of dollars in profits that have not been shown to be causally attributable to its infringement of Dow’s patent. As the dissenting judge in the Federal Court of Appeal noted, not all of Nova’s profits appear to be causally attributable to that infringement: “To put the issue in context, the total amount ordered to be paid to Dow was approximately $644 million, and according to NOVA the failure to apply an apportionment of profits resulted in inflating the award by more than $300 million” (para. 187). The differential profit approach should have been available, and its proper application would have ensured that Nova disgorged only the profits that were causally attributable to its infringement.

[198]                     While the consequences for Nova in this case are particularly inequitable and harsh, the approach my colleague has endorsed will have the opposite effect for other patent infringers. Indeed, a related causation problem arises with respect to an infringer that could not or would not have produced a non‑infringing option. The theoretical existence of an alternative product that services the same consumers as the patented product allows the infringer to retain profits derived from the breach of the patentee’s statutory monopoly. If anything, this creates the very form of “business insurance for infringers” discussed by my colleague (para. 62). So long as there is a non‑infringing option that is a true consumer substitute, then a would‑be infringer has an incentive to infringe. If caught, it will have to disgorge only the difference between its actual profits and the baseline profits that theoretically could have been earned. This is so even if these theoretical profits are not based on a reasonable hypothetical (i.e. a hypothetical that considers what the infringer could have and would have done but for the infringement). Such an approach to causation permits a would‑be infringer to “efficiently breach” a patentee’s statutory monopoly, thereby rewarding intentional and brazen infringements. This undermines the patent bargain.

[199]                     This distortion in the causation analysis is evident in Federal Court’s decision in Monsanto Canada Inc. v. Rivett, 2009 FC 317, [2010] 2 F.C.R. 93, rev’d in part 2010 FCA 207, [2012] 1 F.C.R. 473. In that case, Mr. Rivett infringed Monsanto’s patent — the same patent at issue in Schmeiser — by planting genetically modified soybean seeds. Mr. Rivett’s evidence was that he had “no choice but to plant the ROUNDUP READY soybean seed as there were no conventional soybeans available at the local co‑op or from the one local farmer whom he asked” (para. 60). The reference judge, whose decision was affirmed on this point on appeal, held that “the fact that conventional soybeans are a non‑infringing alternative to ROUNDUP READY soybeans is sufficient to permit the Court to use that crop as a comparator, whether or not such conventional seed was in fact available to the defendant in 2004” (para. 63). However, if an infringer could not or would not have used a non‑infringing option that would have competed in the same consumer market, why should it benefit from its theoretical existence? While I would not speculate on what the appropriate hypothetical should have been in Rivett, suffice it to say that the unavailability of conventional soybeans means that their use as a comparator was not a reasonable hypothetical.

[200]                     My colleague criticizes my approach, stating that it gives rise to “unacceptable consequences, one being that the quantum of profits to be disgorged would vary with the size of the infringing business and the breadth of its product lines” (para. 63). According to my colleague, this “benefits large corporations”, while “smaller businesses would be disproportionately disadvantaged” (paras. 63‑64). With respect, this is simply not the case. Indeed, it is my colleague’s approach that creates a windfall for Dow, a large corporation, in the case at hand.

[201]                     In any event, the size of the infringer is completely irrelevant. Nowhere in the Patent Act  is it indicated that an accounting of profits must ensure that all infringers disgorge the same amount for their respective infringements of a patent. What my approach does is emphasize that, rather than focusing on the value of the patent in the abstract, the court should focus on the individual infringer and the benefits it derives from its infringement of the patent. This is entirely consistent with the gain‑based nature of the remedy. With respect, my colleague’s view about the effect on the remedy on small infringers and large infringers finds no support either in the Patent Act  or in the conceptual underpinnings of the accounting of profits remedy. I could not say it better than Stratas J.A.: “. . . some judges start invoking ‘equity’ as a reason to award whatever seems, to them, appropriate and fair. Such an approach is antithetical to a legal system governed by the rule of law that prizes consistent and predictable rulings” (F.C.A. reasons, at para. 9). Although I disagree with Stratas J.A. on the disposition of this appeal, I agree with him on the necessity of applying rules impartially rather than on the basis of irrelevant considerations like the size of the companies involved in the litigation.

[202]                     Ultimately, I am of the view that a rigorous application of “but for” causation is critical to the proper application of the differential profit approach. With respect, my colleague’s approach distorts the causation analysis. By doing so, it both under‑deters and over‑deters. It under‑deters to the extent that it permits an efficient infringement. It over‑deters to the extent that, when there is no consumer substitute for the patented product, it fails to isolate the value attributable to the infringement and instead assumes that all profits are attributable to it.

(2)          How Are Courts to Determine the Reasonable Hypothetical to Be Used in an Accounting of Profits?

[203]                     In my view, the reasonable hypothetical to be used in an accounting of profits must be based on what an infringer could have and would have done but for the infringement. An examination of what the infringer could have done involves an objective assessment of what was feasible in the circumstances, while an examination of what the infringer would have done involves a more subjective assessment of its alternative course of conduct. However, as I will explain, this subjective assessment is tempered and influenced by the infringer’s conduct in the real world.

[204]                     I am not suggesting that there was any discussion in Schmeiser of what the infringer could have done. I readily acknowledge that the Court never referred to what Mr. Schmeiser could have done but for his infringement of the patent. However, the Court did create a general framework that compared actual profits with hypothetical profits, without comprehensively detailing how hypothetical profits were to be determined.

[205]                     In my view, an infringer’s hypothetical profits are most easily and accurately determined by examining what the infringer could have and would have done but for the infringement. The case law subsequent to Schmeiser — both from this Court and from the Federal Court of Appeal — has affirmed this approach. For example, in Cinar, this Court clarified that the “‘differential profit’ approach . . . is generally used in cases where an infringement allows the infringer to commercialize a good in a more profitable manner than he could have without the infringement” (para. 80 (emphasis added)). Since Schmeiser and Cinar, the case law relating to how courts should construct the hypothetical world has continued to evolve. Courts should examine what an infringer could have done and what it would have done. This is not inconsistent with Schmeiser; it is simply a development and refinement of the law that has occurred since that case which I would endorse rather than reject, which I believe my colleague does. The “could have” and “would have” branches of the analysis help courts to create a reasonable hypothetical world in which the infringement did not occur.

(a)           “Could Have” and “Would Have” Branches of the Analysis

[206]                     Since Schmeiser, the Federal Court of Appeal has developed a robust jurisprudence on the construction of reasonable hypotheticals. This jurisprudence clarifies the evidence that is relevant to establishing what the infringer could have and would have done (see, e.g., Lovastatin, at paras. 73‑95; Pfizer (2016), at paras. 47‑52; Cefaclor, at paras. 31‑43; ADIR (2020), at paras. 53‑55 and 90‑127). The “could have” and “would have” branches have been considered both when assessing damages (e.g., Lovastatin) and in the accounting of profits context (e.g., ADIR (2020)).

[207]                     In my view, the case law on how courts are to construct the hypothetical world is relevant and applicable in the accounting of profits context. The only caveat is that the hypothetical world is not limited to true consumer substitutes, as it is in the damages context. To be clear, in an accounting of profits case, the infringer’s best non‑infringing option can be, and likely often will be, a consumer substitute for the patented product. However, being a true consumer substitute is not an essential requirement and is not even legally relevant. The focus should be on what the infringer could have and would have done but for the infringement.

[208]                     The damages case law on constructing the hypothetical world is therefore informative. For example, as the Federal Court of Appeal noted in Lovastatin, at para. 90:

•                        The real world informs our construction of the “but for” world.

 

•                        Conduct in the real world is “very important” to what would have happened in the “but for” world.

 

•                         Findings of fact from the liability decision are relevant to constructing the “but for” world.

 

•                         “Brazen” infringement in the real world makes it very difficult to prove that the defendant would have deployed the non‑infringing alternative in the “but for” world.

[209]                     The real world informs both the “could have” and the “would have” branches of the hypothetical analysis. If the infringer’s best non‑infringing option was not available as an option in the real world, then the infringer could not have pursued that option in the hypothetical world. Similarly, if the infringer brazenly infringed the patent, then this too informs what it would have done in the hypothetical world. If the infringer knowingly had the option of complying with the law but brazenly chose not to do so, it cannot then seek to benefit from the hypothetical profits it could have earned had it obeyed the law — this is not a reasonable hypothetical.

[210]                     I understand brazen infringement to be an intentional infringement committed for the purpose of surreptitiously securing a benefit from the patent. I would not consider an infringement to be brazen simply because the infringer intentionally produced or used a product that it honestly and reasonably believed was not covered by the patent. This could be because (1) the infringer reasonably and genuinely believed the patent was invalid; (2) the infringer was an innocent infringer; or (3) the infringer, while perhaps not innocent, did not intentionally attempt to reap the benefits of the patent surreptitiously.

[211]                     This is consistent with Schmeiser. Mr. Schmeiser could not be classified as an innocent infringer, since he had actively cultivated seeds containing Monsanto’s patented invention. However, while he was not necessarily innocent, there was no finding that he had intentionally infringed the patent, as this was held to be irrelevant to liability (see Monsanto Canada Inc. v. Schmeiser, 2001 FCT 256, 12 C.P.R. (4th) 204, at para. 115). As well, and importantly, Mr. Schmeiser had not sprayed Roundup after the crop commenced growing.

(b)          Persuasive Burden

[212]                     In an accounting of profits case, the burden of establishing what the infringer could have and would have done logically lies on the infringer itself (ADIR (2020), at para. 51), although both sides can adduce evidence on this point.

[213]                     To the extent that the infringer makes an implausible argument about what it could have and would have done, it risks not satisfying its burden. The infringer does this at its own peril. If the infringer cannot establish its “best non‑infringing option” on a balance of probabilities, then the differential profit approach cannot be used. In these circumstances, the court may properly conclude (after determining that no further apportionment is necessary) that it is appropriate to apply either the full costs approach or the differential costs approach. This is entirely consistent with the notion that the differential profit approach is the “preferred means of calculating an accounting of profits” (Schmeiser, at para. 102). It also addresses my colleague’s concern about “business insurance”.

(3)          Availability of the Differential Profit Approach in the Present Case

[214]                     Dow raises a procedural argument regarding why Nova should be denied the relief it now seeks. Dow argues that, at trial, Nova abandoned its argument that it could have sold non‑infringing products as substitutes for its infringing product, SURPASS. According to Dow, “Nova cannot, on appeal, walk back unequivocal concessions it made at trial and that the trial judge relied on in rendering his decision” (R.F., at para. 9).

[215]                     I disagree with Dow’s characterization of Nova’s concession. That concession needs to be contextualized in light of the post‑Schmeiser case law. The damages case law and accounting of profits case law dealing with non‑infringing options or alternatives became intertwined in the Federal Courts after Schmeiser. Nova — like the reference judge and the Federal Court of Appeal — understood the damages case law on non‑infringing alternatives to be relevant and binding. Lovastatin had been incorporated into the accounting of profits cases, and its requirement that a non‑infringing alternative be a true consumer substitute was viewed as a requirement for the application of the differential profit approach. Indeed, the reference judge held that the differential profit approach requires a comparison with a non‑infringing alternative, which “must be a true substitute or real alternative” (para. 143).

[216]                     Based on this understanding of the law, Nova abandoned its argument about hypothetical profits, believing that the differential profit approach was unavailable. Instead, it suggested that its hypothetical costs could be deducted. On appeal in the Federal Court of Appeal, Nova “cast its apportionment arguments differently” (para. 84). In this Court, Nova once again recast its arguments, asserting that the differential profit approach should have applied.

[217]                     I agree with the Federal Court of Appeal that, although Nova may have recast its legal arguments, it has not “rais[ed] any new legal issu[e]” but rather has “offer[ed] different legal arguments on that same issue” (para. 86). The primary issue, although characterized differently, has always been one of apportionment and causation: to what extent were all of Nova’s profits causally attributable to its infringement of Dow’s patent?

[218]                     Given my conclusion that a non‑infringing alternative does not need to be a true consumer substitute for the purposes of the differential profit approach, Nova’s concession that there were no consumer substitutes is not determinative of the outcome. With respect, the reference judge erred in law by holding that a non‑infringing option needed to be a true consumer substitute for the patented product.

[219]                     Additionally, the differential profit approach is preferred, at least in part, because of the way in which it subsumes any apportionment analysis. Apportionment is simply the manner by which a court separates the profits made as a result of the infringement from those made as a result of something else. The court looks at the different profit drivers and apportions profits based on whether or not they are causally attributable to the infringement. Even though Nova conceded in the courts below that the differential profit approach could not apply, I see no reason why apportionment would have been inappropriate. The differential profit approach is simply a particularly effective and precise form of apportionment analysis. Accordingly, even if that approach could not be applied based on Nova’s concession, the courts below erred by not excluding the profits not causally attributable to its infringement of Dow’s patent.

[220]                     On the record available, this Court cannot determine how to apply the differential profit approach to the facts. Therefore, I would remit the matter to the Federal Court.

C.            Springboard Profits Are Available

[221]                     As I said above, springboard profits are an infringer’s post-patent expiry profits that result from the infringement of the patentee’s statutory monopoly during the life of the patent. By infringing the patentee’s statutory monopoly, the infringer is able to build sales and market share that increase its profits after the expiry of the patent.

[222]                     Nova asserts that springboard profits are not available because a “patent provides a time‑limited monopoly over the patented product” (A.F., at para. 124). Nova argues that “an accounting must end when the patent expires” and that, therefore, there is no basis for a post‑expiry award of springboard profits (A.F., at para. 125).

[223]                     I would reject this argument. Although the patent provided Dow with a monopoly only for a limited period, Dow should be entitled to an accounting of all profits that Nova earned as a result of its infringement. It is of no moment that certain profits causally attributable to Nova’s infringement did not accrue to it until after the patent expired. So long as the profits were causally attributable to its infringement, they remain subject to disgorgement.

[224]                     That said, as I discussed above, the causation analysis for any post‑expiry period must be based on the same hypothetical “but for” reasoning embodied in the differential profit approach.

[225]                     It follows that I do not agree that the springboard profits were properly calculated in this case. The case should be remitted to the Federal Court to determine the post‑patent expiry profits that Nova could have and would have made but for the infringement. These profits should then be compared with Nova’s actual post‑expiry profits.

V.           Disposition

[226]                     For these reasons, I would allow the appeal. I would remit the matter to the Federal Court for determination. The differential profit approach should be available in this case.

[227]                     With respect to costs, Nova is entitled to its costs in this Court and in the Federal Court of Appeal.

                    Appeal dismissed with costs throughout, Côté J. dissenting.

                    Solicitors for the appellant: Torys, Toronto.

                    Solicitors for the respondents: Smart & Biggar, Ottawa.

                    Solicitors for the interveners Bell Canada, Rogers Communications Canada Inc., TELUS Communications Inc. and Vidéotron ltée: IMK, Montréal.

                    Solicitors for the intervener the Canadian Generic Pharmaceutical Association: Goodmans, Toronto.