Kapischke on Interim Licenses and AILIs

Justus Kapischke has published an article titled Interim Licences and Anti-Interim-Licence Injunctions:  Semi-strict non-interference or rules for the race?, 5/2026 GRUR 275.  Here is the abstract:

This article analyses and provides background information on two recent decisions of the LG Munich I and the LD Mannheim.  In both decisions, the courts react negatively to the English practice of awarding interim licence declarations in FRAND disputes, enjoining implementers from applying for such relief.  The LD Mannheim further offers its opinion on the limits of final FRAND determinations in England.

The decisions at issue are, of course, Judgment of Nov. 26, 2025, LG Munich I, 21 O 12112/25, and InterDigital VC Holdings, Inc. v. Amazon.com, Inc., UPC_CFI_936/2025 (LD Mannheim Dec. 22, 2025), both of which entered anti-interim license injunctions (AILIs) directed against Amazon.  (Both decisions are also excerpted in this same issue of GRUR, at pp. 313-30.  The appeal from the Mannheim LD decision will be heard on May 28.)  The author ably illustrates the incompatibility of the English approach to FRAND disputes (under which the court’s task, ultimately, is to set a rate) and the German/UPC approaches (which to date have focused on conduct), writing that “[a] foreign court making injunctions unavailable by ordering the SEP holder to grant a licence interferes with the incentive structure of injunction focused approaches,” whereas “conduct-based injunctions may force settlement before any court had the chance to establish FRAND terms either by determining them or by confirming the FRANDness of the SEP-holder’s offer, frustrating a rate-setting approach” (p.177).  He questions some of the analysis of the above two decisions, noting that they both seem premised on the coercive effect of an English declaration concerning interim licenses (effectively equating them to antisuit injunctions), even though “it is clear that interim licences do not force an SEP-holder to accept rate-setting proceedings in the implementer’s preferred forum” (p.179).  True, the SEP holder’s refusal to comply with the declaration could result in its being deemed an unwilling licensor, but the author states that the SEP holder’s seeking an AILI is “unlikely to change anything about that” (p.180).  (In addition, it would seem to me, the odds that the SEP holder would obtain an injunction in the U.K. are pretty negligible anyway, aren’t they?)  In addition, he notes both courts’ emphasis on the territorial nature of patents, which however in a strictly jurisdictional sense “appears not to be the current law in Germany or the EU as exemplified by the ECJ’s judgment in BSH” (p.182).  I’m glad to see someone making this connection, which seems quite important to me.  The author argues, however, that strict territoriality is not the Mannheim LD’s position, but rather only what he refers to as “semi-strict” territoriality in the sense that “courts can, on behalf of SEP holders, impose their view of global FRAND licences on implementers by way of injunctions,” but implementers “may not use similarly coercive measures as means of imposing global FRAND determinations on SEP holders, since this would interfere with foreign infringement proceedings” (p.182).  I’m not entirely sure of the author’s view here, but I’m not convinced of the merits of this distinction.  (I plan to explore the question more deeply in a forthcoming essay I will contributing to the Research Handbook on Extraterritoriality and Intellectual Property, which I mentioned here the other day.  Consistent with the author’s observation that “both sides can at least plausibly accuse each other of having taken the first ASI-like measure” (id.), it has long seemed to me that when courts start issuing ASIs, AASIs, AILIs, declarations concerning interim licenses, and so on, it is awfully difficult to say which jurisdiction is the one violating the comity norm.)  The author closes by suggesting that it would be helpful “to obtain authoritative constructions of the relevant (F)RAND undertakings from the French of Swiss courts,” and also if SEP holders “offer[ ] to refrain from seeking injunctive relief if the implementer agrees to be bound by a rate-setting procedure in Germany or at the UPC” (id.).  Might the German courts and/or UPC ever head in this direction?      

Lavie and Shapira on Rivalrous Remedies

Shay Lavie and Roy Shapira have posted an article on ssrn titled Rivalrous Remedies, Vand. L. Rev. (forthcoming).  Here is a link to the paper, and here is the abstract:

Legal scholarship typically conceptualizes enforcement as operating directly on wrongdoers: either by enjoining them ex ante or by imposing monetary sanctions on them ex post. Yet across a wide range of legal fields, courts and lawmakers have long employed a different and largely untheorized instrument. Instead of sanctioning the wrongdoer directly, these doctrines deter misconduct by conferring a legal advantage on the wrongdoer’s rival: a business competitor, a litigation counterparty, or another strategically positioned actor. For example, when a patent holder forces customers to sign aggressive licensing agreements, courts may allow the patent holder’s competitors to freely infringe the patent. And in family law, when one parent alienates a child from the other, courts often respond not by fining the alienator parent but by granting the alienated parent additional visitation rights.

 

This Article is the first to systematically analyze this institutional design, which we term “rivalrous remedies.” In the process, the Article makes three contributions. Conceptually, it identifies the conditions under which regulating behavior indirectly—by empowering rivals—can outperform classic remedies. Rivalrous remedies can leverage the superior information and incentives of rivals. At the same time, these remedies operate effectively only in certain market structures and pose risks of overuse and spillover harms to third parties.

 

Descriptively, the Article demonstrates the prevalence of rivalrous remedies by analyzing nine doctrines across diverse legal fields: from false advertising to defamation law to civil procedure. While each doctrine has been criticized by scholars in their respective fields, viewing them through a common lens reveals a shared institutional logic: they emerged to address chronic underenforcement problems. In particular, these doctrines respond to underenforcement by shifting who enforces (not necessarily the direct victim) and/or how enforcement occurs (through largely self-executed mechanisms requiring minimal judicial involvement).

 

Normatively, the Article offers guidance for courts interpreting existing rivalrous remedies, identifying when they should be expanded, constrained, or combined with traditional remedies. It also proposes the adoption of a new rivalrous remedy in trade secret law to address enforcement failures surrounding consumers’ “right to repair.”

Although I don’t necessarily with everything in the paper, its presents an intriguing analysis of a variety of doctrines—including one that I have written on from time to time (including in my forthcoming book, Wrongful Patent Assertion), namely IP misuse.  On misuse in particular, the authors argue that traditional remedies against, say, overextensive licensing practices are often ineffective, because the direct victim may have difficulty proving (for example) an antitrust violation (or, in the copyright sphere) fair use.  Patent and copyright misuse law can counter this potential underenforcement problem by conferring an advantage (unenforceability of the subject IP) on the patent or copyright owner’s rivals; in principle, this “rivalrous remedy” should serve as a deterrent against the overextensive licensing practice in the first place.  Of course, how effective that deterrent is depends on the rivals being aware of the overreaching, and being confident that they would prevail on the misuse defense if sued for infringement.  (One possible countervailing consideration, it occurs to me, is that under the traditional understanding of misuse the penalty of unenforceability lasts only until the misuse is purged, so a rival might still be hesitant to make any irreversible investments in reliance on its ability to lawfully access the subject IP.)  Moreover, as the authors note, “the patent misuse doctrine is imperfect,” because “[c]ourts sometimes have a hard time distinguishing between practices that do not hurt competition and those that do” (p.22).  All that said, this is an interesting way of thinking about misuse and doctrines in other bodies of law such as false advertising (where the idea of investing competitors with the right to sue has sometimes been thought of as, in effect, deputizing them to vicariously avenge the direct injuries suffered by consumers) and antitrust.  I think one could probably extend the analysis to the tort of interference with prospective business relations as well, insofar as that body of law confers upon a disappointed rival some recourse against a person who has used improper means to discourage third parties from doing business with the rival—though my sense is that, in its current form, this tort is rarely successful.  

China and ASIs

I don’t think I have previously this mentioned on this blog, but Professor Tim Holbrook and I have signed a contract with Oxford University Press to coedit an edited volume titled Research Handbook on Extraterritoriality and Intellectual Property Law.  I will have more to say about this project over the course of the next year or so, but for now will just note that we have assembled a list of about forty contributors, including ourselves, to address the many issues that may arise in evaluating the geographic scope of IP rights.

One group of such issues, of course, relates to standard-essential patents; among the relevant issues are whether courts can or should award global royalties, with or without consent of the parties, and whether they can or should award interim relief such as antisuit injunctions, anti-antisuit injunctions, or declarations concerning interim licenses.  In connection with antisuit injunctions, as readers are by now aware, a WTO arbitration panel last year ruled in favor of the EU's appeal from the original panel decision, in finding that China’s antisuit injunction policy violated TRIPS articles 28.1 and 28.2, read in light of article 1.1 second sentence (see my post here).  It is therefore notable that, as has been reported elsewhere, on April 1 the EU published an announcement that China had announced its withdrawal of the antisuit injunction policy at the WTO Dispute Settlement Body meeting in September.  The minutes of that meeting are available here, and the EU-China dispute is Item 2 (pages 6-9).  China expresses its agreement with the original panel opinion and its disagreement with the arbitration panel, before stating in paragraph 2.4 that the Supreme People’s Court had issued a notice on September 25 “stating that the so-called ‘ASI policy’, to the extent it ever existed, had been withdrawn and had no continuing effect upon how Chinese courts evaluated requests for anti-suit injunctions in the context of SEP litigation.”  The aforementioned EU announcement states that “[a]fter some initial technical issues were resolved, this notice is currently accessible online from outside China too” (if any readers can point me to it, I would appreciate it), but that “[g]iven the unwritten nature of China’s anti-suit injunction policy, the European Commission will continue to closely monitor the situation to ensure China’s full compliance . . .  and take further action, if necessary.”

In light of these developments, it was interesting to read an article by Yao Jianjun (former vice president of the Xi’an Intermediate People’s Court) in an issue of China Patents & Trademarks, No. 4, 2025, that recently came my way.  The article, appearing in English translation at pages 12-22, is titled Application of Anti-Suit Injunctions in SEP Disputes.  It provides an overview of Huawei v. Conversant, which was the first of five cases from 2020 in which a Chinese court (here, the SPC) granted an ASI (here, against the enforcement of an injunction granted against Huawei in Düsseldorf).  The author discusses where ASIs fit within Chinese civil procedure law, as well as the five conditions specified by the SPC for granting an ASI (“the impact of the enforcement of foreign judgments on Chinese litigation,” “the necessity of ASIs as a preservation measure,” the “balance of interests of both parties,” “the impact of ASI on the public interest,” and “international comity”) and how they played out on the facts of Huawei v. Conversant.  The author agrees with the court’s decision, though he also notes some of the limitations of ASIs, and cautions against the sequence of lawsuits being a dispositive factor (insofar as this would encourage races to the courthouse).  Apparently the article was written before the EU-China matter was completed, since the author references its existence but not its conclusion.

One thing I found interesting in the article, and which I plan to discuss in the essay I will be contributing to the edited volume I mentioned at the beginning of this post, is the author’s emphasis on the perceived need to counteract the impact of proceedings initiated in another country which, he writes, may “impair[ ] the legitimate rights and interests of the applicant,” thus requiring “a remedy . . . to the party that has suffered damage to the injunction” (pp. 14-15).   In this regard, he writes, if Conversant had applied to enforce an injunction in Germany, the result would have been that Huawei either would have exited the German market or would have settled with Conversant (at a rate, he says, that would have been 18.3 times higher than the rate determined by the Chinese first instance court); and that "such potential damage . . . may substantially harm Huawei's legitimate rights and interests" (p.18).  This framing of the issues bears some resemblance the "effects" test (as used, for example, in U.S. antitrust litigation), under which courts sometimes justify the exercise of prescriptive jurisdiction directed against extraterritorial conduct when such conduct has effects within the prescribing jurisdiction.  From an economic perspective, the application of the effects test is akin to the concept of internalizing an externality—though in this context, the twist is that the externality is caused by the foreign jurisdiction’s toleration or authorization of conduct which the domestic court would like to forbid.  Relevant to this point, a generation ago Professor Joel Trachtman wrote an intriguing paper titled Externalities and Extraterritoriality: The Law and Economics of Prescriptive Jurisdiction, in Economic Dimensions in International Law 642 (Jagdeep S. Bhandari et al. eds., 1997), in which he posits that, if we think of nations participating in a market for prescriptive jurisdiction, and of (in some cases) even having the ability to engage in some measure of Coasean bargaining, there is a range of possible options for allocating the right to prescribe.  Some allocations may be more efficient than others, under the circumstances at hand.  An unavoidable aspect of this framework, however, is determining exactly what counts or should count as a harmful externality—that is, one that causes cognizable harm, a topic that my colleague Claire Hill also has written about.  In the context of ASIs, of course, the country against which the ASI is directed (in effect though not in form, since the form of the injunction is inter partes) presumably believes that it is doing nothing wrong by entertaining a case that is lawfully before it, or by issuing an injunction against the infringement of domestic patent rights in accordance with domestic law; and indeed, if the WTO arbitration panel approach prevails, such conduct is in general privileged and does not cause cognizable harm under international law.  But one could also imagine an alternative rule under which the right to issue an ASI to prevent the perceived negative domestic effect (suffered by a domestic firm or by a domestic court) of a foreign injunction prevails over the foreign court's right to entertain a case or enter an injunction.  Determining which rule makes more sense from a policy standpoint might perhaps depend on how absolute the prohibition on ASIs is under the first approach, and how substantial the domestic effect must be under the second.  (Thinking of these issues from a slightly different angle, as I mention in a discussion of Professor Christopher Drahozal’s game theoretic approach to ASIs in my forthcoming book Wrongful Patent Assertion, in a given case reasonable minds may differ whether it is the country issuing the ASI, or the country against which the ASI is issued, that is deviating from the "comity" norm under which nations are largely left to determine their own domestic policies.)  At this stage, I'm not sure exactly where my analysis is going to lead, but this seems like an interesting problem for analysis through a law-and-econ lens; and if readers have any thoughts or suggestions, I’d be delighted to hear them.     

Blogging Break This Week

 Apologies, but once again I am swamped with other work.  I hope to resume blogging next week. 

FRAND in Japan

As I note in my forthcoming book Wrongful Patent Assertion, in 2014 a Grand Panel of the IP High Court held, in Samsung Elecs. Co., Ltd. v. Apple Japan Godo Kaisha, Judgment of May 16, 2014, 2013 (Ne) 10043, that the owner of a FRAND-committed SEP may abuse its rights by seeking injunctive relief against a willing licensee; and “[i]n a subsequent decision, Imation Corp. Japan v. One-Blue LLC, Tokyo Dist. Ct., Feb. 18, 2015, Case No. 2013 (Wa) 21383, the court held that the defendant (the manager/operator of a patent pool that included some SEPs) had engaged in unfair competition by sending a warning letter to three retailers threatening injunctive relief, because under the Samsung v. Apple decision, the act of seeking an injunction for the infringement of a FRAND-committed SEP against a willing licensee is an abuse of right.  The [latter] decision therefore appears to read Samsung v. Apple as holding that it is an abuse of right per se to seek injunctive relief against a willing licensee for the infringement of a FRAND-committed SEP.  Cf. Yuzuki Nagakoshi & Katsuya Tama, Japan Without FRANDS?  Recent Developments on Injunctions and FRAND-Encumbered Patents in Japan, 44 AIPLA Q.J. 243 (2016) (critiquing Samsung v. Apple for arguably creating such a per se rule and for ignoring, as part of the abuse of rights calculus, the rightsholder’s subjective intent; and proposing that Japanese courts hearing FRAND cases instead consider whether the rightsholder has satisfied its statutory duty to negotiate in good faith).”  Other than these two cases, however, to my knowledge there hasn’t been much any Japanese case law involving FRAND-committed SEPs, until last year when three new district court decisions—all involving the same plaintiff, Pantech Corp.—came out.  Two of these have recently made available in translation on the IP High Court’s website; the third, not yet.  They all appear to agree, however, that it is an abuse of right for the owner of a FRAND-committed SEP to seek injunctive relief against a willing licensee.

The first of the three, Pantech Corp. v. ASUS Japan Co., 2022 (Wa) 7976 (Tokyo Dist. Ct., Apr. 10, 2025), involves Japanese Patent No. 4982653, titled “Method of transmitting and an apparatus for transmitting ACK/NACK information and a method of transmitting and receiving and an apparatus for receiving ACK/NACK signals.”  The court concludes that ASUS’ products infringe and that the claims in suit are valid.  On the issue of remedy, however, the court states that “a FRAND-encumbered standard essential patent holder's act to demand an injunction against a standard essential patent implementer based on the standard essential patent is impermissible as an abuse of right, unless there is a special circumstance where the standard essential patent implementer has no willingness to obtain a license under FRAND terms” (p.25).  After a review of the parties’ negotiating history (pp. 25-30), the court concludes that the defendant was a willing licensee, stating:

the reason for which the Plaintiff and the Defendant failed to reach an agreement on a FRAND rate in spite of prior consultations relating to the abovementioned findings and settlement talks in this lawsuit is that the FRAND rates presented by both parties were far too divergent, as explained in detail later (No. 7). According to both parties' assertions mentioned above and the entire import of oral arguments, the cause for this divergence is considered to be as follows: in major countries in the world, the Unwired Planet v. Huawei judgment, as presented by the Plaintiff, and other court precedents on FRAND rate calculation methods have developed internationally in response to changes of the times; on the other hand, in Japan, there have been no court precedents based on the abovementioned international development for about 10 years following the Apple v. Samsung Grand Panel judgment, and also, even taking into account various circumstances that appeared in this case, a FRAND rate calculation method has not necessarily been established in Japanese business practice (p.31).

Citing article 102(3) of the Japanese Patent Act (on reasonable royalties) and the IP High Court’s 2019 judgment in NeoChemir (see my blog post here), the court then sets out principles for calculating the FRAND rate:

In light of the global nature and the enormous number of standard essential patents implemented in the manufacturing of products conforming to a standard, when determining a FRAND rate, it is necessary [i] to take into consideration a royalty rate set in the actual license agreement for the standard essential patent, or if it is indefinite, a global average royalty rate in the industry; [ii] to presume the individual values of all standard essential patents to be the same because it is practically difficult to find the individual values of the enormous number of patents, and to calculate the value of a single standard essential patent by dividing the value of all standard essential patents by the total number of standard essential patents, while ensuring that the total amount of royalty obtained through aggregation of royalty rates of the standard essential patents remains  within a reasonable scope; and [iii] to also take into consideration, in such a case, possible contributions to sales and profits if all standard essential patents are used in the product. Furthermore, [iv] as a FRAND rate should fundamentally be agreed upon globally as soon as possible through good-faith negotiations between a FRAND-encumbered standard essential patent holder and a standard essential patent implementer, under the literally fair, reasonable, and non-discriminatory terms, a reasonable FRAND rate should be determined by comprehensively taking into account the negotiation process between the parties and the standard essential patent implementer's willingness to obtain a license under FRAND terms, and other circumstances appearing in the lawsuit, from the viewpoint of facilitating the agreement (p.35).

Ultimately, the court calculates a royalty for the one patent in suit by multiplying the sales revenue from the defendant’s products (redacted) by the aggregate royalty burden for LTE-standard-related patents (which the court figures at 9% for LTE products, based on an analysis of findings made in Unwired Planet, TCL v. Ericsson, and Huawei v. Samsung, and 8% for 5G products (p.39)), and then dividing by the number of LTE-standard-related patents (which the court estimates to be 1,300, taken into account likely validity, etc., see pp. 40-41)).  The rate itself is redacted.

The second and third decisions are both actions by Pantech against Google Japan G.K.  I understand that both cases involve the same patent, Japan Patent No. 6401224 (“Method for mapping a physical hybrid automatic repeat request indicator channel”), but different accused products (the Google Pixel 7, since discontinued, and the Pixel 7a, respectively).  In the first of the two, 2023 (Wa) 70501 (Tokyo District Court June 23, 2025), a translation of which is available here, the Tokyo District Court agreed that a FRAND-committed SEP owner’s demand for an injunction “is impermissible as an abuse of right, unless there is a special circumstance where the standard essential patent implementer has no willingness to obtain a license under FRAND terms" (p.39).  The court nevertheless held that Pantech was entitled to an injunction because Google’s conduct during the course of a court-supervised settlement negotiation (occurring in July 2024) demonstrated that Google was an unwilling licensee.  In dicta, however, the court found no fault with Google’s pre-settlement negotiations behavior and thus would not have found Google unwilling on the basis of that conduct alone (see pp. 47-50).  In the second one, however—according to the commentary cited below, since a translation has not been uploaded yet—the Osaka District Court on July 10, 2025 concluded that Pantech was not entitled to an injunction, because Google had shown itself to be a willing licensee through the end date of negotiations the court took into account (November 30, 2023).  The commentaries below do not say anything about the Osaka court actually determining the FRAND royalty, and some more recent commentary indicates that by early January (with other cases involving newer models of Google phones pending) Pantech and Google had settled on global terms, following a recommendation by the Tokyo district court.

For commentary that was helpful in drafting this post, see Masachi Chucho, Pantech v. ASUS: A Recent FRAND Judgment from Japan (available here); Takeshi Komatani, Japan’s First SEP Injunction:  Pantech v. Google and the High Bar for Establishing Unwillingness (available on AIPPI’s website here); and Kenji Tosaki, Takahiro Hatori & Yujiro Fukuhara, The Japanese court first judgment to grant an injunction based on a FRAND-committed SEP (available on Chambers and Partners’ website here; original, longer version available here).   

Meanwhile, as discussed elsewhere (see, e.g., here, here, here, and here), in early 2026 the Tokyo District Court also issued two new documents, Guidelines for Patent Infringement Litigation Based on Standard Essential Patents and SEP Mediation Procedures.  I may have more to say about these in due time.   

UPC’s Hamburg L.D. Concludes that Infringing Offer Caused No Proven Damages

The decision is Fives ECL v. REEL GmbH, UPC_CFI_274/2023, issued on February 11, 2026.  The decision is the subject of a recent blog post on ip fray.  Last June, I blogged about an earlier decision (of the UPC Court of Appeal) in this dispute, writing that “the patent owner had obtained from a German national court a judgment of infringement, prior to June 1, 2023, and a declaration that the defendant would be liable for damages; but it thereafter pursued its damages claim before the UPC.  The UPC Court of Appeal held that the UPC was competent to hear the damages claim (but left open the question of whether national or UPC law would apply to that claim).”  The current decision holds that German domestic law applies to the damages claim, but that it wouldn’t matter in any event in view of the Intellectual Property Rights Enforcement Directive (paras. 100-10).  More interesting, however, is the court’s conclusion that, although the defendant was adjudicated to have made an infringing offer, it isn’t liable for any damages.  A copy of the original decision, in German, is linked to above; below, I use a machine translation that I have compared with the original.

The patent in suit is EP  c1 740 740 B1, for a “compact service module which is intended for electrolytic aluminium production plants.”  Plaintiff and defendant compete “in the market for special purpose cranes, which are used worldwide in various countries in aluminum furnaces as part of aluminum production” (para. 5).  In December 2016, the parties submitted competing bids for a project in Bahrain, which was to be built by Bechtel.  The first offers, dated December 2, were for twelve service modules and auxiliary bridges.  Defendant’s offer was higher than plaintiff’s.  Bechtel then requested that the parties provide a price “in the event these two parts were split and continued separately,” which the parties responded to on December 15.  Defendant offered a price reduction, while the plaintiff did not (para. 70).  Plaintiff didn’t get the contract; but then Bechtel decided to reopen the bidding process, and plaintiff submitted a new bid on February 21, 2017, which included a price decrease of €6,500,000.  Apparently plaintiff was then awarded the contract, but it (successfully) sued the defendant for having made an infringing offer, which it then followed up with this claim for lost profits.

The court rejects the claim for lost profits, for failure of proof as to amount and for lack of proof of causation.  From what I gather, the plaintiff didn’t proffer the December or February offers to the court (see paras. 18, 47), but rather sought to rely on evidence of (1) the defendant’s typical profit margin, according to the latter’s publicly available financial data from 2011-16, and (2) “projects implemented by the plaintiff in the past” (para. 51).  The former, however, is not sufficient proof, given that “defendant’s activities span multiple business segments,” and also that the submitted evidence reports the defendant’s gross margin (Bruttomarge), which may not be comparable to the profit margin for this project.  The latter as well is not sufficient, because the projects “predate the tender for the [present project] by eight to thirteen years,” “included contracts with a wide range of volumes,” and were reflective in part of both the greater market power enjoyed by the plaintiff at those earlier dates and a better economic environment generally for aluminum manufacturers and suppliers.  In addition, the court casts doubt on the technical advantage provided by the patented technology in comparison with alternatives (discussed further below), and concludes that in any event there is no evidence that any assumed advantage over the state of the art could justify the plaintiff’s alleged margin.  Moreover, the defendant’s bid included an “erection and installation concept” that Bechtel favored, to the point of requesting the plaintiff to include a similar concept in its proposal when bidding reopened—which “contradicts the assumption that the plaintiff would have prevailed in a hypothetical scenario without the defendant’s patent-infringing bid,” insofar as “the defendant had a competitive advantage over the plaintiff that was independent of the” machinery at issue (paras. 76-77).

Finally, we come to the causation issue.  The court begins this section by stating that it cannot “be ruled out that, even if the defendant had submitted an alternative offer that did not infringe the patent, the plaintiff would have had to reduce its offer of December 2/15, 2016” (para. 78).  Here, the analysis gets a bit confusing, with the court first seeming to indicate that the existence of a noninfringing alternative is irrelevant to the plaintiff’s entitlement to lost profits, and then appearing to walk it back:

79 As a general rule, the claim—in this case, the asserted loss of profits—cannot be countered by the defense of lawful alternative conduct (see BGH, GRUR 2024, 1201, para. 43 et seq. – Verdampfungstrockneranlage). According to the case law of the Federal Court of Justice (BGH), the defense that the damage would have occurred even if lawful conduct had been adopted may be relevant for the attribution of the damage. The relevance of the defense depends on the protective purpose of the respective infringed provision (BGH NJW 2017, 1104, para. 24; BGHZ 194, 194 = GRUR 2012, 1226, para. 35 – Flaschenträger).

 

80 In the case of a patent infringement, the defense that the same economic result could have been achieved through non-infringing acts cannot, in principle, lead to the exclusion of a claim for damages. A patent does not preclude third parties from competing with the right holder by offering non-infringing products. However, the offering and placing on the market of the protected subject matter is reserved to the right holder. A culpable infringement of this exclusive right must result in the infringer having to compensate for the resulting damage even if he could have offered other products. These principles also apply to the patent-infringing offering of a product.

 

81 These principles do not apply in the present case. This is because it is undisputed between the parties that the customer would always have requested a second offer in order to foster competition . . . .

 

82 If one therefore assumes that a non-infringing alternative offer must be included in the assessment, it cannot be definitively established that the plaintiff would certainly have been awarded the contract with its original offer.

The court then goes on to explain why the defendant’s Pavlodar model would have been both technically and economically more attractive than the proposals the plaintiff submitted in December 2016.  As a consequence, as stated above, the plaintiff fails to prove that it “would certainly have been awarded the contract with its original offer” (paras. 83-89).

I’ve noted (what I view as) similar inconsistencies in the German courts’ analyses of noninfringing alternatives before, for example in my June 2024 post on the Verdampungstrockneranlage decision cited above.  Maybe it’s fair to say, however, that under German law, the existence of a noninfringing alternative doesn’t necessarily preclude the plaintiff from recovering damages for infringement, but that the plaintiff still must present evidence as to the amount of those damages; and where, as here, it proceeds instead with an untenable lost profits theory, it gets nothing, though perhaps under other circumstances it would still be entitled a reasonable royalty—as may have been the case in Verdampfungstrockneranlage, where it is conceivable that there was some value to the defendant in making an infringing offer within Germany, as opposed to a noninfringing offer somewhere else, even if the end result would have been the same in that the defendant would have been awarded the contract (for a noninfringing project carried out outside of Germany).  If so, the German position may not be all that different from the U.S. position, under which (as reflected in cases such as Grain Processing Corp. v. Am. Maize-Prods. Co., 189 F.3d 1341 (Fed. Cir. 1999)), the existence of a noninfringing alternative means that the plaintiff can’t recover a lost profit (because it wouldn’t have made the allegedly forgone sales even absent the infringement), but the plaintiff may still recover a reasonable royalty reflecting the cost saving the defendant incurred by having used the patented technology over the noninfringing technology. 

In any event, my most recent effort to compare and contrast the law of noninfringing alternatives can be found at page 144 of my book Remedies in Intellectual Property Law, where I note that, although Canadian and French case law seems more or less consistent with the U.S. approach, the U.K. courts “continue to follow the House of Lords’ 1888 decision in United Horse-Shoe & Nail Co. v. John Stewart & Co., holding that the existence of noninfringing alternatives is irrelevant” to both lost profits and awards of infringer’s profits (though the U.K. courts recognize that noninfringing alternatives are relevant to reasonable royalties); and that “German courts also have held that the existence of noninfringing alternatives does not preclude an award of lost profits, though such evidence can affect the amount awarded,” citing both Verdampfungstrockneranlage  and Flaschenträger.

Federal Circuit Reverses Judgment Awarding Extraterritorial Damages

The case in Trustees of Columbia University v. Gen Digital Inc., precedential opinion by Judge Dyk (joined by Judges Prost and Reyna), published this morning.  This is a very complicated matter, involving among other things an inventorship dispute that devolved into a contempt order against trial counsel for the defense (which contempt order is reversed in a separate appellate decision also handed down today, which I need not address here), as well as two previous appeals on claim construction and validity (which I also will not address).  The current decision involves questions of patent eligibility, claim construction, and damages; and as is my typical practice for purposes of this blog, I will focus only on the last of these.

There are two patents in suit, both of which related “primarily to protecting computer systems from viruses and other malicious activity” (p.2).  The claims at issue consist of one system claim, two method claims, and a computer-readable medium claim, all of which allegedly are infringed by software marketed by the defendant under the Norton brand.  The district court denied a motion to dismiss for lack of patent eligibility.  Then jury then returned a verdict of willful infringement and awarded damages of $185,112, 727; the district court awarded enhanced damages and fees.  On appeal, the Federal Circuit reverses and remands for further proceedings on the issue of patent eligibility--but “[b]ecause other issues may arise on the remand,” the court addresses one remaining issue of claim construction as well as the issues pertaining to damages and fees.  As noted, I will address the damages and fees issues only.  The most interesting of these—at least in my view, since I’ve written a fair amount now about this topic—is whether, on the facts of this case and assuming the patents in suit are valid, the patentee is entitled to damages reflecting foreign sales of Norton software.

As readers may be aware, the general rule that seems to be emerging from Supreme Court, Federal Circuit, and district court case law over the last few years is that, although U.S. patents are territorial rights, if the defendant engages in the unauthorized manufacture, use, or sale of patented products in the United States, and this domestic infringement causes-in-fact and proximately causes either (1) the plaintiff to lose sales that the plaintiff would have made to foreign customers, or (2) the defendant to make sales abroad that the defendant otherwise would not have made, the plaintiff is entitled to recover, respectively, either its own lost profit on its lost foreign sales, or a royalty reflecting some portion of the profit the defendant would have expected to earn from the defendant's foreign sales, as of the date of the hypothetical ex ante bargain.  In view of this precedent, the district judge gave the jury the following instruction:

Columbia is entitled to damages based on sales to customers located outside of the United States if you find that the infringing product sold to those customers was made in or distributed from the United States, even if the infringing product is delivered to a customer and used by the customer outside the United States (p.23; emphasis added by the Federal Circuit).

The jury found that the defendant (referred to throughout the opinion as “Norton”) sold antivirus software abroad, and that “the infringing product” was made in or distributed from the United States.  (The jury did not find that the sales to foreign customers “substantially occurred in the United States,” (p.24 n.7)).  The appellate panel nevertheless agrees with Norton that “no reasonable jury could conclude that any infringing copies of Norton’s software that were sold to customers outside the United States were made in the United States or distributed from the United States” (p.24). 

This seems correct to me under the governing standards for determining what an infringing software product is, principally Microsoft Corp. v. AT&T Corp., 550 U.S. 437 (2006).  As the panel explains:

Microsoft establishes that software in the abstract—that is, software not physically encoded in a “tangible copy” like a CD or hard drive—is akin to a “blueprint” or “a schematic, template, or prototype.” Id. at 449–50. If someone abroad builds an infringing product based upon a blueprint that exists in the United States, for example, then the product was still made abroad. See id. at 442. So too, software is not tangible—or capable of infringing the asserted claims—until tethered in a particular copy of the software encoded in a computer-readable medium (p.25).

Applying this principle:

The system claim, ’322 patent, claim 27, includes a “processor.” Like the apparatus claim at issue in Microsoft, this claim is not infringed until a particular instance of software is installed onto a computer with a processor. See Centillion Data Sys., LLC v. Qwest Commc’ns. Int’l, Inc., 631 F.3d 1279, 1288 (Fed. Cir. 2011). Because the instances of software sold to customers located abroad are not installed on a computer in the United States, those instances were not made in or distributed from the United States.

 

The same conclusion follows as to the other claims asserted here. A method claim is only infringed when the claimed process is performed; it is not infringed by the mere existence of software that, if installed on a computer, could perform the method. See Ericsson, Inc. v. D-Link Sys., 773 F.3d 1201, 1219 (Fed. Cir. 2014). Because the infringing software is only capable of performing either of the claimed methods once installed on a computer, the versions installed abroad also cannot give rise to domestic infringement. In any event, “[t]here is no established recognition in patent law of direct infringement by ‘making’ a ‘method.’” See Brumfield, 97 F.4th at 879. The methods here were not “made” in the United States nor “distributed” from the United States.

 

This leaves only claim 11 of the ’322 patent, the computer-readable medium claim. Columbia argues that this claim must be treated differently, because it does not require that a particular version of software be first installed on a computer with a processor to be infringing. It is true that claim 11 does not require software to be installed on a device with a processor, but claim 11 does still require that the software be encoded in a particular “non-transitory computer-readable medium.” ’322 patent, claim 11. While a non-transitory computer-readable medium may be created on a server in the United States, that medium is not exported abroad. The computer-readable media sold to foreign customers are only created once the foreign computer encodes the software on its hard drive, which occurs outside the United States. These computer-readable media are—like the apparatuses in Microsoft—created outside the United States and therefore cannot be domestically infringing. Under the logic the Court applied in Microsoft, these cannot constitute infringing products that were made in or distributed from the United States (pp. 25-26).

Columbia tries a few additional arguments on appeal, but none of them work.  The one that might have worked, had it been presented at trial, was that “the jury could have found that the domestic infringement involved in creating its master copies, which enabled the foreign sales, were the cause of the foreign sales damages. However, the jury was not instructed, and Columbia did not seek an instruction, that they could grant a reasonable royalty for foreign sales based on this theory. We cannot reform the damages theory actually presented to the jury in favor of an alternative that was not, even if the alternative would have been legally valid. . . . We thus need not reach the question of whether Columbia’s theory of foreign damages was proper under the causation theory of Brumfield.” (p.25).  My initial reaction is that that theory probably wouldn’t have worked either, because the causal connection between the domestic manufacture of the master copies and the foreign sales is too tenuous to satisfy proximate causation, though I would want to know more about the underlying facts to assert that opinion with confidence.  (Alternatively, if the domestic manufacture of the master copies could have been outsourced, then in my opinion outsourcing should count as a noninfringing alternative, and any royalty awarded for the resulting foreign sales should reflect only the cost saving, if any, of domestic over foreign manufacture of the master copies.  Whether the courts would agree with me on this remains to be seen.)  The appellate court also rejects arguments that Norton could be liable as a joint infringer with the foreign customers, or that Norton could be liable under an inducement theory (pp. 27-28).

As for willfulness and enhanced damages, the court affirms the finding of willfulness, primarily on the basis of evidence that Norton was aware in advance of “‘the Columbia professors’ designs and work before the patents issued’ including the provisional application,” and the lack of evidence that, during the relevant time period, Norton was aware of and acted upon its subsequently asserted objectively reasonable defenses (pp. 21-22).  The court nevertheless vacates the district judge’s enhancement of actual damages (2.6 times the actual damages) and the award of attorneys’ fees, in part because the amount awarded and the finding of exceptionality were based on the vacated finding of contempt of court.  In addition, the case was close (on patent eligibility) and Norton’s assertion of allegedly “repetitive” arguments did not amount to litigation misconduct.