Mystery Science: What Lemley and His Colleagues Get Wrong in Their Push for SCOTUS to Review TCL v. Ericsson

Image Source: Wikipedia

In December 2019, the United States Court of Appeals for the Federal Circuit issued a decision in a standard essential patent (SEP) appeal involving Ericsson and TCL Communication Technology—a closely watched case that many thought would shed light on what constitutes a FRAND (fair, reasonable and non-discriminatory) offer of a licensing royalty rate relative to standard essential patents (SEPs). See TCL Communication Technology Holdings Ltd. V. Telefonaktiebolaget LM Ericsson, 943 F.3d 1360 (Fed. Cir. 2019).

To summarize a detailed history, essentially Ericsson proposed two alternative license offers to TCL: so-called Option A and Option B. Option A proposed a lump-sum payment of $30 million for its first $3 billion in sales with percentage running royalties ranging between. .8% and 2%. Option B proposed only running royalties ranging between .8% and 1.5%, but with a $2.00 floor and a $4.50 cap. Both options also included a “release payment” for TCL’s past unlicensed sales. No license deal was struck, litigation ensued. Ericsson was denied a jury trial.

The Federal Circuit, in a decision authored by Judge Chen (joined by Judges Newman and Hughes) concluded that the release payment was in substance a request for relief for TCL’s past infringement of Ericsson’s patented technologies without a license, which required a jury trial. Unfortunately, the Federal Circuit did not reach the question of whether the district court erred in its FRAND analysis. TCL appealed the decision to the U.S. Supreme Court on May 1 and several amicus briefs have now been filed in support of the petition being granted.

The amici mostly believe the Federal Circuit was incorrect in this ruling, including a small group of law professors, led by Mark Lemley.

Below are excerpts taken from the Summary of the Argument and the introduction to the Argument in the professors’ amicus filing. I’ve taken the liberty of providing my thoughts in the format of comments from the peanut gallery, or perhaps as a patent law equivalent to Mystery Science Theater 3000, which I like to do from time to time. In order to differentiate my thoughts/comments from the professors’ amicus, my comments are italicized, colored, indented and tagged with the IPWatchdog logo.

________________

THE BRIEF: The Federal Circuit’s decision requiring that juries decide what a promise to license patents on “fair, reasonable, and nondiscriminatory” (FRAND) terms means is premised on that court’s conclusion that enforcing a FRAND commitment on a patent owner was the equivalent of and a “substitute” for a patent damages lawsuit of the sort that must be tried to a jury. It is not. It is a contract dispute over the terms of a license, or, as a minimum, a question of equitable estoppel. The Federal Circuit misunderstood the nature of the FRAND commitment, conflating a restitutionary remedy as part of a license dispute brought by the implementer with the very different case of a patent damages lawsuit brought by the patent owner. Properly understood, the nature and scope of the FRAND commitment is not equivalent to a patent infringement action and not something a jury must decide.

MY TAKE: My initial reaction is that this entire paragraph is false, but perhaps it would be more accurate to characterize it as gobbledygook. The professors either don’t understand what FRAND means, or they are conflating FRAND with some kind of mythical, exalted, high-minded contractual principle that is devoid of anything to do with patent infringement. FRAND stands for fair, reasonable and non-discriminatory. It is not a promise to license, or a promise to allow everyone the unquestioned right to use patents that may be standard essential. The FRAND commitment is a promise to negotiate and offer to license fairly and reasonably. But a license is a two-way street, which requires both an offer and an acceptance. In other words, the standard essential patent owner cannot both offer and accept the license themselves, and without a willing implementer there will never be a license. When such an unwilling licensee chooses to adopt standard essential patents, they are choosing not only to enter into license agreement, but they also choose to use technologies that read on the patent claims owned by an SEP holder. This constitutes infringement. It is utterly ridiculous to suggest that FRAND is incompatible with a patent infringement action. If you don’t pay you don’t have a license, and if you don’t have a license you are an infringer, period.

The professors also pretend that they can divine the meaning of FRAND without looking at the SSO (standard setting organization) agreements.  They are simply substituting what they would like to see for what is, and assuming others will not challenge them. There is no “promise to license patents” but only an agreement if certain prerequisites are met, as in the case of ETSI. There is nothing in any SSO agreement that says what rate is fair, and SSOs jealously stay away from making such proclamations.

THE BRIEF: In requiring a jury in all FRAND cases, the Federal Circuit departed from the rule that has been applied in other circuits, including the Ninth Circuit and district courts in the Fifth and Seventh Circuits. It threw into disarray what was becoming a settled process for evaluating FRAND obligations.

MY TAKE: That other courts have adopted procedures that violate the Seventh Amendment to the U.S. Constitution is not a reason to overrule the Federal Circuit. Frankly, the suggestion from a bunch of law professors that a clearly unconstitutional procedure be adopted nationwide because it is becoming “a settled process” is absurd. One has to wonder what other constitutional rights are vulnerable to expediency in the view of the professors?

THE BRIEF: If not corrected, the Federal Circuit’s rule will have a variety of other undesirable procedural implications. Because the Circuit’s decision depends on the conclusion that money paid under a FRAND commitment is compensation for an act of patent infringement, it threatens to turn every patent license dispute not just into a federal case but into a case that must go to the Federal Circuit, not the regional circuits.

MY TAKE: Of course money paid is for an act of infringement. That is SEP / FRAND basics. Ask yourself this: If money paid for past unlicensed activity is not compensation for patent infringement, exactly what is it? The FRAND commitment is triggered only in the event of infringement. If there is no infringement of a standard essential patent any FRAND commitments are a nullity. That is precisely why some innovators are finding it advantageous to litigate patents that are not standard essential. 

If an entity refuses to enter into a license agreement and they engage in activities covered by the exclusive rights granted to a patent owner, that entity is a patent infringer. This is not a new, or radical concept. When a patent owner needs to sue a tortfeasor for an act of patent infringement and one or both parties wish to appeal the decision beyond the district court level, the appeal is to the Federal Circuit. Again, this is not a new or radical concept. The jurisdiction to hear patent appeals is well-established and mandatory. See 28 U.S.C. 1295. The fact that a patent infringement action must go to the Federal Circuit is hardly a reason for the Supreme Court to grant certiorari.

THE BRIEF: That has never been the law. It is inconsistent with this Court’s limited approach to Federal Circuit jurisdiction in Gunn v. Minton, 133 S.Ct. 1059 (2013). But it flows inexorably from the faulty logic of the Federal Circuit’s holding.

MY TAKE: Once again, the professors engage in the very type of conflation and misleading argumentation that they would scold students for making. Gunn v. Minton dealt with malpractice, which is not a federal issue. The specific question was whether a claim of malpractice arising from patent litigation was properly appealed to the Federal Circuit. The Supreme Court ruled that it was not because a malpractice cause of action does not raise a federal issue of patent law. Gunn v. Minton simply does not deal with a direct appeal from a patent infringement litigation and citing to it as some authority here, where Ericsson brought a patent infringement action and TCL raised invalidity defenses, is beyond absurd.

THE BRIEF: Setting an industry standard can create problems when that standard is arguably covered by one or more patents, however.

MY TAKE: Isn’t that what SEPs are all about? One has to wonder whether the professors realize just how naïve this statement sounds. They do understand that an industry standard when set will define technology that will be covered by many thousands of patents owned by many dozens of patent owners, right? That is precisely why BOTH the implementers and innovators are supposed to negotiate in good faith. What creates the problems is when implementers refuse to engage innovators, which is a well-documented and growing problem. Unfortunately, too often in this era of weak patent rights in the U.S., implementers choose to take without paying and refuse to negotiate—something known as efficient infringement, which former Director of the USPTO David Kappos has called “stealing”. See also Explaining Efficient Infringement. This efficient infringement, or stealing, makes it necessary for patent owners to bring infringement lawsuits to seek redress for infringement that occurred prior to entering into a license. That implementers choose to refuse to enter into an arms-length negotiation is entirely unsurprising, but the fact that the patent owner committed to a FRAND obligation does not change the fact that unlicensed activity that is covered by a patent is patent infringement.

THE BRIEF: Patent owners whose technology is used deserve to get paid.

MY TAKE: Finally, the professors are right about something.

THE BRIEF: But patent owners whose technology is adopted as a standard can gain an enormous windfall because they can hold up implementers of that technology, threatening to shut down critical infrastructure that the entire industry has adopted unless they are paid an exorbitant fee. See Thomas F. Cotter, Patent Holdup, Patent Remedies, and Antitrust Responses, 34 J. corP. L. 1151, 1152 (2009); Joseph Farrell et al., Standard Setting, Patents, and Hold-Up, 74 Antitrust L.J. 603, 604–05 (2007); Mark A. Lemley & Carl Shapiro, Patent Holdup and Royalty Stacking, 85 Tex. L. Rev. 1991 (2007).

MY TAKE: The professors are wrong, and their positions have been discredited. First, notice that they cite themselves. Second, notice that the citations are more than a decade old, which is more than a small coincidence, oversight or laziness. This idea that SEPs are responsible for hold-up surfaced over a decade ago in no small part because of the writing of Lemley and Shapiro. Since the theory first gained popularity, however, there has been no actual evidence of hold-up, a difficult and inconvenient reality for those who are so invested in the theory. To the contrary, there has been evidence of the exact opposite; namely that implementers are refusing to enter into licenses. For example, even Google has charged Apple with being a uniquely unwilling licensee. And Google is not alone. Nokia has similarly charged Apple with being an unwilling licensee, and so too has Qualcomm most recently charged Apple with being an unwilling licensee, and the district court recognized that the theory has been the basis for declaratory relief in other cases. Indeed, there is a mountain of evidence that implementers are even refusing to enter into negotiations in the first place or have policies that encourage copying as a means for enabling faster growth. See here. Thus, while there is no actual evidence to support hold-up, there is great evidence supporting hold-out (i.e., refusal to license and in some cases refusal to negotiate).

“[T]hose who advocate using antitrust law to reduce the risk of an alleged ‘hold-up’ have yet to identify any harm to the competitive process,” Assistant Attorney General Makan Delrahim said in a speech on October 21, 2019. “[A]ctual evidence of hold-up remains scant even after a decade has passed since the theory was first introduced.  The gulf between the theory and practice is especially troubling as many advocates ignore the real risk of hold-out by potential licensees of the chosen SEP technology.” Delrahim is, of course, correct. See also Why Incentive for ‘Patent Holdout’ Threaten to Dismantle FRAND (Richard Epstein & Kayvan Noroozi, 2017).

Furthermore, one of the particularly insidious aspects of refusing to negotiate or simply refusing to license is that when patent owners are forced to litigate they are likely to ultimately receive less compensation than FRAND once litigation costs and time is accounted for. Given that so many implementers are engaging in either refusal to negotiate or refusal to license, one has to wonder whether there is some concerted effort. Regardless of whether there is collusion, or just independent recognition of the weakened patent rights owned by SEP owners, forcing those who have agreed to a FRAND commitment to litigate inevitably has the effect of driving down the asking price for future licenses.

A better question for these professors to ask—and one courts will eventually be forced to rule on—is whether innovators are required to maintain FRAND commitments to entities that refuse to negotiate in good faith. It seems rather obvious the answer should be no, and implementers can’t be in any hurry for any court to actually answer that question.

THE BRIEF: Most SSOs balance those competing concerns by requiring participants to license their patents on FRAND terms. Patentees get paid, but in return for having their technology widely adopted as an industry standard they agree to forego the right to hold up implementers and to accept only a reasonable and nondiscriminatory license fee. That FRAND commitment is a license, though courts and commentators have sometimes differed on whether it is a binding contract or one implied through equitable doctrines of estoppel and acquiescence. See Mark A. Lemley, Intellectual Property Rights and Standard- Setting Organizations, 90 Calif. L. Rev. 1889 (2002).

MY TAKE: Once again, Professor Lemley cites himself for a proposition that is clearly wrong. First, as already mentioned, the mythical hold-up that Lemley and others have so carefully crafted as the boogeyman of implementers does not exist. Second, FRAND is not a blanket commitment to license as Lemley would have the reader believe. Third, Lemley’s theory would only make sense if the implementer were required contractually required to license the SEPs, which is simply not the case.

Ask yourself this: What is a patent owner to do when an implementer engages in otherwise infringing activity and they refuse to enter a license on any terms? Lemley would have you believe that provokes a contract case, but that is both a legal and factual impossibility. There is no license, which means there is no contract between the parties. The FRAND commitment is between the innovator and the SSO. The fact that there is no license between the implementer and the innovator (i.e., patent owner) is precisely the problem that requires litigation and because there is no license (i.e., no contract) the dispute cannot possibly be a contract dispute. Instead, the dispute is a patent infringement dispute whereby the patent owner is suing for past infringement.

THE BRIEF: When the parties agree to a FRAND commitment but cannot agree on what payment would be “reasonable,” courts are frequently called upon to supply that missing term in the deal, just as courts supply missing terms in other commercial contracts. But they are doing so to give effect to the commitment the patent owner made to the SSO – taking a fair price in exchange for adoption of their patented technology and foregoing the opportunity for holdup.

MY TAKE: First, what the professors assert as frequent never happens. In order for this to actually be true—namely that parties agree to a FRAND but cannot agree on the payment—they would have to be agreeing to patent validity, what the standard covers, what the patents cover and admitting infringement. This is something that never happens. Even in this case, TCL argued that the patents were invalid and not infringed, so they were not agreeing that they owed any royalty at all, let alone a FRAND. Indeed, I challenge the professors to find a single instance of parties agreeing to validity, essentiality, infringement, and only disagreeing on the amount of payment.

Second, there is no evidence of hold-up no matter how many times the professors say it or cite to themselves. Nevertheless, this and other similar disputes are much more than supplying a missing term to a deal. There is no deal.

Third, the adoption of technologies covered by the patents in question was an act of infringement, which occurred outside the scope of a license, which means the activities constituted patent infringement. Ericsson had every right to seek compensation for past damages for patent infringement. To suggest otherwise would be to incentivize implementers to even further engage in the hold-out that they are already engaging in knowing that all they would eventually have to do, if sued, is pay some royalty that at worst would have been reasonable had they negotiated fairly and not taken steps as a tortfeasor to disrespect the property rights of the SEP holder in the first place.

THE BRIEF: The Federal Circuit decision in this case fundamentally upset that balance by treating the FRAND commitment, not as a license or estoppel dispute, but as a mutant form of patent infringement remedy, one that it held must be set by a jury under the Seventh Amendment. This Court should correct that error.

MY TAKE: I have never seen an SSO document that says it is a license. In fact, if you look at ESTI, it has express language that says if the parties do not agree, the parties must look to the law of the host nation to resolve the issue.

The Federal Circuit gets plenty wrong, but they got this case exactly right. If the Supreme Court does take the case, it will be to affirm the Federal Circuit.