Claim preclusion, issue preclusion and the Kessler doctrine: game, set and match!

The famous Kessler Twins

Just a few days ago, while drafting a post on the 'actual controversy' requirement in declaratory judgment actions, I came across an interesting article by Michael G. Munsell. The article briefly mentioned an obscure theory I had never heard of. 'The Kessler doctrine', the author explained, 'bars an infringement suit against a customer or seller who has previously prevailed against the patent owner because of invalidity or noninfringement'. I made a mental note to dig deeper, as I was curious to understand how the Kessler doctrine correlated with res judicata (after all, I thought, which additional arguments could the patent owner raise against the alleged infringer's customer, after losing suit against the former?). Luckily, the Federal Circuit was kind enough to look into the matter in Brain Life LLC v Elekta Inc., an interesting opinion published on Monday. The panel discussed how a final judgment of non-infringement prevents relitigation under three different doctrines (claim preclusion, issue preclusion, and Kessler doctrine), reviewing the characteristics and limitations of each of them.

To understand the Federal Circuit's decision, it is necessary to go back to 1997, when MIDCO sued Elekta, alleging that three products sold by the defendant (GammaKnife, GammaPlan, and SurgiPlan) infringed US Patent No. 5,398,684. The patent at issue contained only two independent claims, respectively disclosing an apparatus and a method for generating a video image from a variety of scanner imaging sources, including a computerized axial tomography image, a nuclear magnetic resonance image, and an X-ray. During discovery, MIDCO focused on the apparatus claims and did not oppose Elekta's motion for dismissal of the method claims prior to trial. Accordingly, the district court dismissed the method claims without prejudice. The jury rendered a verdict in favor of the plaintiff, finding that Elekta's products infringed claim 1 of the '684 patent. On appeal, the infringement finding was reversed, and the case remanded to the district court to enter judgment of non-infringement as a matter of law in favor of Elekta.

A few years after these events, MIDCO licensed its patent to a third party, which, in turn, licensed it to Brain Life. The licensee filed suit against Elekta (and several other defendants), maintaining that four of its products (GammaKnife, GammaPlan, SurgiPlan, and ERGO++) infringed the method claims of the '684 patent. At first instance, the district court granted summary judgment in favor of the defendant. Finding that there was no material difference between the accused products and those adjudicated in the previous proceeding between MIDCO and Elekta, the court held that allowing the plaintiff to assert the method claims would amount to impermissible claim splitting. Thus, it concluded that 'Brain Life cannot revisit the decision MIDCO made many years ago to forego including in its infringement action [the method claim], when it could have been asserted in the first litigation, and revive that allegation now against products that are essentially the same as the accused products that were found not to infringe'.

The Federal Circuit reviewed the district court's decision under the three doctrines mentioned above. First, it discussed the applicability of claim preclusion. Under the law of the Ninth Circuit, claim preclusion applies where the prior suit (i) involved the same claim or cause of action as the later suit, (ii) reached a final judgment on the merits, and (iii) involved the same parties or privies. The presence of a final judgment, which ended a cause of action, bars the parties from relitigating both the claims that were brought and those that could have been brought in the prior suit. In this perspective, the doctrine of claim preclusion effectively prevents claim splitting (as expressly acknowledged, for example, in Shaver v F.W. Woolworth Co. - the Tenth Circuit, in Hartel Springs Ranch of Colorado Inc. v Bluegreen Co., reviewed several decisions that recognized a similar correlation). The court highlighted that, according to the teaching of Nystrom v Trex Co. Inc. and Foster v Hallco Mfg. Co., the party that demands the application of the doctrine of claim preclusion in a patent case must demonstrate that the accused product is 'essentially the same' as that accused in the prior suit.

Brain Life acknowledged the similarity between the products litigated in the MIDCO trial and those accused in the later suit. However, it contended that the final judgment rendered in the prior suit could not bar the assertion of the method claims, which were dismissed without prejudice. The Federal Circuit disagreed, clarifying that '[w]hile the dismissal without prejudice allowed for the possibility that acts of infringement of the method claims could be subject to a future cause of action, that possibility was cut-off for all such acts predating the final MIDCO judgment once that judgment was entered'. The court observed that the doctrine of claim preclusion does not prevent Brain Life from asserting either the apparatus or method claims against Elekta, for acts of infringement occurred after the judgment rendered in the prior suit. The panel recited the recent decision of Aspex Eyewear Inc. v Marchon Eyewear Inc., which clearly explained that 'if the party could not have asserted particular claims - because the tortious conduct in question had not occurred at that time - those claims could not have been asserted and therefore are not barred by res judicata'.

The opinion then looked at the doctrine of issue preclusion, or collateral estoppel, under which 'once an issue is actually and necessarily determined by a court of competent jurisdiction, that determination is conclusive in subsequent suits based on a different cause of action involving a party to the prior litigation' (Montana v United States). Actual litigation and determination of an issue of fact or law in a final judgment is the element that distinguishes issue preclusion from claim preclusion. The court found that the method claims originally asserted by MIDCO against Elekta 'were not fully, fairly, and actually litigated to finality'. Therefore, it held that issue preclusion does not bar a second suit on those claims. Similarly, as the ERGO++ product was not included among the accused products in the prior suit, issue preclusion cannot bar the assertion of any claim of the '684 patent in relation to said product.

Finally, the panel turned to the Kessler doctrine, suggesting that it 'fills the gap between [claim preclusion and issue preclusion], allowing an adjudged non-infringer to avoid repeated harassment for continuing its business as usual post-final judgment in a patent action where circumstances justify that result'. In particular, this doctrine applies to new acts of (alleged) infringement, which are excluded from the applicability of claim preclusion. The Kessler doctrine was created, in 1907, by the Supreme Court in Kessler v Eldred: Eldred, after losing an infringement suit brought against Kessler's electric lighters, had sued a customer of Kessler that was selling the same products accused in the first suit. According to the Federal Circuit, the Kessler decision went beyond the perimeter of the doctrine of claim preclusion, as the Supreme Court 'granted Kessler a limited trade right to continue producing, using, and selling the electric lighters that were the subject of the first suit and to do so without fear of allegations of infringement by Eldred - even when the acts of infringement occurred post-final judgment and even when it was third-parties who allegedly engaged in those acts of infringement'.

The court acknowledged that its past reliance on the Kessler doctrine has been sparse, but recalled the cases of Rubber Tire Wheel Co. v Goodyear Tire and Rubber Co. and MGA Inc. v General Motors Corp. The latter described Kessler as a doctrine that 'bars a patent infringement action against a customer of a seller who has previously prevailed against the patentee because of invalidity or noninfringement of the patent; otherwise, the effect of the prior judgment would be virtually destroyed'. The Federal Circuit suggested that the Kessler doctrine may be less valuable today than at the time of its creation, as the strict doctrine of mutuality that was applied at that time has been replaced by the development of defensively applied issue preclusion. Applying Kessler, the court found that, after the final judgment of non-infringement, Elekta was free to continue engaging in the accused commercial activity as a non-infringer, in relation to all the products adjudicated in the MIDCO litigation. 'Simply', explained the court, 'by virtue of gaining a final judgment of noninfringement in the first suit - where all of the claims were or could have been asserted against Elekta - the accused devices acquired a status as noninfringing devices, and Brain Life is barred from asserting that they infringe the same patent claims a second time'. The panel reiterated, however, that this conclusion does not apply to Elekta's ERGO++ product, which was never included in the prior suit. Thus, it reversed the district court's judgment regarding this product, remanding for further proceedings.

Who bears the burden of proof in declaratory judgments of non-infringement? The patentee, answers the US Supreme Court

On Wednesday, the US Supreme Court published its decision in the case of Medtronic Inc. v Mirowski Family Ventures LLC, ruling that, when a licensee brings an action for declaratory judgment of non-infringement of a licensed patent, the patentee bears the burden of persuasion on the issue of infringement. The Supreme Court unanimously reversed the judgment of the Federal Circuit, as suggested by the briefs submitted by the Solicitor General (here), and by a number of law professors (here).

In 1991, Medtronic and Mirowski concluded a licensing agreement, which allowed the former to implement, in its products, patented technologies held by the latter, in exchange for royalty payments (the facts of the case are described in detail here). A supplementary agreement (Litigation Tolling Agreement), signed in 2006, established that, if Mirowski believed that one of the products manufactured by Medtronic infringed one of its patents, the licensee could either (i) pay the additional royalties due, or (ii) seek a declaratory judgment of non-infringement, continuing to pay royalties into an escrow account (upon conclusion of the declaratory judgment action, these royalties were to be assigned to the prevailing party). These provisions essentially prevented the licensor from filing a counterclaim for infringement, as, even in case of alleged infringement, the agreement would remain in force.

In 2007, Mirowski gave Medtronic notice of infringement, in relation to two patents allegedly violated by several Medtronic products. As established under the agreement, the licensee filed a declaratory judgment action of non-infringement, paying all the relevant royalties into an escrow account. The District Court found that the patentee had failed to prove infringement, noting that the burden of proof of infringement fell upon Mirowski, even if it was the defendant in the action. The Court of Appeals for the Federal Circuit disagreed, acknowledging that the patentee normally bears the burden of proving infringement, but carving an exception for cases in which the patentee (i) is a defendant in a declaratory judgment action, and (ii) cannot file a counterclaim for infringement due to the continued existence of a license. The Federal Circuit essentially considered that, in such exceptional cases, the patentee is merely assuming a passive position, as the licensee is the only party seeking to disturb the status quo ante:

The contract at issue here required MFV to identify products it believed were covered by the contract. After MFV identified those products, Medtronic was required to either pay royalties on them, or sue for declaratory judgment that the products were not covered. Medtronic is unquestionably the party now requesting relief from the court: it already has a license; it cannot be sued for infringement; it is paying money into escrow; and it wants to stop. In contrast [...] MFV seeks nothing more than to be discharged from the suit and be permitted to continue the quiet enjoyment of its contract. In other words, it is Medtronic and not MFV that is asking the court to disturb the status quo ante and to relieve it from a royalty obligation it believes it does not bear. Consistent with the above, for the court to disturb the status quo ante, Medtronic must present evidence showing that it is entitled to such relief. 

The Supreme Court reversed the Federal Circuit's decision (for a discussion of the first part of the court's decision, which dealt with issues of jurisdiction, see here). The court observed that, according to settled case law, the Declaratory Judgment Act has merely limited procedural purpose (Skelly Oil Co. v. Phillips Petroleum Co.), and leaves substantive rights unchanged (Beacon Theatres Inc. v Westover). Therefore, as the burden of proof is a substantive aspect of a claim (Raleigh v Illionois Department of Revenue), the Declaratory Judgment Act cannot alter the general rule valid in patent infringement actions, under which the burden of proving infringement rests upon the patentee (Under Sea Industries Inc. v Dacor Corp.).

The judges noted that 'shift[ing] the burden depending upon the form of the action could create postlitigation uncertainty about the scope of the patent', highlighting that the Restatement (Second) of Judgments expressly acknowledges the possibility of relitigating an issue in a subsequent suit, when the burden of persuasion has shifted. The infringement issue could thus potentially remain unresolved, if the alleged infringer loses the declaratory judgment action, and the patentee similarly loses a subsequent infringement action brought against the former:

Suppose the evidence is inconclusive, and an alleged infringer loses his declaratory judgment action because he failed to prove noninfringement. The alleged infringer, or others, might continue to engage in the same allegedly infringing behavior, leaving it to the patentee to bring an infringement action. If the burden shifts, the patentee might lose that action because, the evidence being inconclusive, he failed to prove infringement. So, both sides might lose as to infringement, leaving the infringement question undecided, creating uncertainty among the parties and others who seek to know just what products and processes they are free to use.

The court also argued that the patentee is 'in a better position than an alleged infringer to know, and to be able to point out, just where, how, and why a product (or process) infringes a claim of that patent'; shifting the burden of persuasion would force the licensee to negate 'every conceivable infringement theory'. Reciting MedImmune Inc. v Genentech Inc., the Supreme Court further noted that imposing the burden on the licensee would create a significant obstacle to the use of declaratory judgments actions in similar cases:

In [MedImmune], we wrote that the "very purpose" of that Act is to "ameliorate" the "dilemma" posed by "putting" one who challenges a patent’s scope "to the choice between abandoning his rights or risking" suit. [...]. In the absence of the declaratory judgment procedure, Medtronic would face the precise dilemma that MedImmune describes. Either Medtronic would have to abandon its right to challenge the scope of Mirowski’s patents, or it would have to stop paying royalties, risk losing an ordinary patent infringement lawsuit, and thereby risk liability for treble damages and attorney’s fees as well as injunctive relief. [...] As in MedImmune, the declaratory judgment action rescues Medtronic from this dilemma.

These reasons were deemed sufficient to justify an exception to the basic rule described in Schaffer v Weast, where the court had ruled that plaintiffs normally bear 'the risk of failing to prove their claims'. The judges also explained that the public interest favors the maintenance of a well-functioning patent system, but requires, at the same time, that 'patent monopolies [be] kept within their legitimate scope' (Precision Instrument Manufacturing Co. v Automotive Maintenance Machinery Co.). The latter objective would be hindered, if the rules on the burden of proof effectively discouraged licensees from filing declaratory judgment actions. Thus, '[t]he general public interest considerations are, at most, in balance', and suggest retaining the general rule imposing the burden of proving infringement upon the patentee.

In reversing the Federal Circuit's decision, the Supreme Court rightfully took into consideration a series of theoretical (e.g. the nature of the declaratory judgment action and its objectives) and practical issues (e.g. the difficulty, for the licensee, to prove non-infringement under any conceivable infringement theory, and the risk of an unresolved dispute in case of relitigation), which undoubtedly support its conclusion. The court did not discuss other controversial aspects of the case, for example the risk of an abusive use of non-infringement declaratory judgment actions by licensees (the judges merely commented that 'litigation can occur only in the presence of a genuine dispute, "of sufficient immediacy and reality", about the patent's validity or its application'), or the possibility, for the patentee, to bring a counterclaim seeking a declaratory judgment of future infringement under T. Lang et al. v Pacific Marine and Supply Co. (in which the Federal Circuit taught that '[i]f the controversy requirement is met by a sufficient allegation of immediacy and reality, we see no reason why a patentee should be unable to seek a declaratory judgment of infringement against a future infringer when a future infringer is able to maintain a declaratory judgment action for noninfringement under the same circumstances').

It is interesting to note, as observed by Professor Cotter here, that the path taken by the Supreme Court mirrors the approach employed by German courts ('the patentee always has the burden of proof with regard to the infringement issue, even in negative declaratory actions' - for a similar principle applied by Italian courts, see this case), but differs from the ruling of the Court of Appeal for England and Wales in the recent case of Generics [UK] Ltd v Yeda Research and Development Co Ltd & Anor ('the burden of establishing non-infringement fell on [the plaintiff that brought the declaratory judgment action'). The latter case, however, should probably be read in conjunction with Baxter Healthcare Corporation et al. v Abbott Laboratories, where the patentee was required to plead its case, even if the burden of proof remained on the plaintiff (see IPKat post here). From this point of view, even jurisdictions which provide for a shifting of the burden of proof are likely to introduce additional requirements, in order to identify the perimeter of the underlying infringement dispute that prompted the licensee to file the declaratory judgment action.