When is a Covenant just a Covenant? Of Meso, Newman and Transactions with Patents

Nov 11, 2015 | Labs Blog

Originally posted on New Private Law blog by Jorge Contreras. BiolawToday.org.

jorge-contreras_wvflnypetition for certiorari was recently filed with the Supreme Court in Meso Scale Diagnostics LLC v. Roche Diagnostics GmbH (Del. 2015). In it, Petitioner poses the following question: “Whether a covenant, promise, or agreement not to sue for the infringement of a federal patent is a license of that patent as a matter of federal law.”

Though phrased simply, the question is an important one for reasons that extend well beyond the immediate dispute between the parties (which itself is too complex to describe here). Among other things, its answer has serious implications for the doctrine of patent exhaustion and the treatment of industry-wide commitments to limit the enforceability of patents in contexts such as standard-setting (commitments I have elsewhere termed “patent pledges”).

In Meso, Petitioner asks whether, and under what circumstances, a covenant not to sue (a “CNS”) for patent infringement should be deemed to convey a property-like license interest in the subject patent. Readers of this blog may be familiar with Prof. Chris Newman’s recent article A License is Not a “Contract Not to Sue”: Disentangling Property and Contract in the Law of Copyright Licenses (98 Iowa L. Rev. 1101 (2013)). Newman convincingly argues that a copyright license embodies more than a mere contractual relationship between a licensor and a licensee. Instead, he claims, such a license may be more properly construed as a form of usufructory property interest akin to a leasehold or servitude on a parcel of land. He concludes, as his title declares, that a license is not merely a “contract not to sue”. (Though Newman’s article deals primarily with licenses of copyrights, his reasoning extends to patents as well.)

The question posed by Petitioner in Meso is the converse of that posed by Newman. Whereas Newman asks (and answers) whether a license is merely a CNS, Petitioner asks whether, and when, a CNS should be construed as a license. The difference between these two questions is subtle but important. Newman takes as his starting point a license agreement under which the patentee desires to grant to the licensee a bundle of rights that have characteristics that we usually associate with a property interest (e.g., the licensee’s rights survive the licensor’s bankruptcy, they continue if the licensor sells the underlying property, etc.). Thus, even if the licensor does not expressly state that it is granting a property interest to the licensee, we should impute one from the surrounding context and language. Hence, a “license” should be understood to grant a property interest, and not merely to be a personal promise by the licensor (a covenant) not to sue the licensee for infringement.

The Petitioner in Meso, however, starts from a different premise. It begins with contractual language that does notoutwardly purport to be a license. Rather, the contractual language with which it is concerned is phrased (taking the facts, arguendo, as Petitioner has represented them) as a “covenant not to sue”. In this case, Petitioner asks whether the CNS is properly construed as a license, or whether it remains a simple contractual commitment. In other words, if the parties enter into what purports to be CNS, without more, must that agreement be interpreted to convey a propertyinterest (a license) in the underlying patent?

The answer to this question is of more than theoretical interest, and has implications far beyond the parties’ dispute in Meso. One such implication pertains to the doctrine of patent exhaustion, reinvigorated by the Court’s decision in Quanta Computer v. LG Electronics, 553 U.S. 617 (2008). Under this venerable doctrine, once a patented article is sold by the patentee or its authorized licensee, the patentee’s rights in that article are exhausted and it may not enforce the patent or collect royalties from subsequent users or sellers of the article. Thus, in the seminal case of Adams v. Burke, 84 U.S. (17 Wall.) 453 (1873), Burke, an undertaker who purchased a patented coffin lid from Lockhard & Seyle, the patentee’s authorized dealer in Cambridge, could not be sued under the same patent by Adams, the patentee’s dealer in Natick.

A critical question in such cases, then, is who has been licensed under the patent. If, in Adams v. Burke, Lockhard & Seyle were not “licensed” by the patentee, then Burke, who purchased its coffin lid from them, would have been answerable to Adams for infringement. Accordingly, the terms of the agreement between the patentee and Lockhard & Seyle are of paramount importance. If the agreement conveys a license, then the patent rights are exhausted. If not, then the rights are not exhausted and downstream buyers are potentially liable for infringement.

With the exhaustion doctrine firmly in mind, patentees have attempted any number of clever contractual maneuvers to avoid exhausting their patent rights. One such tactic, itself nearly a century old, was to grant a CNS rather than a license. By doing so, it was hoped, the patentee made a personal promise not to bring suit against the manufacturer or seller of a patented article, but stopped short of granting it a property-like license under the patent. This tactic was put to the test, and failed, in De Forest Radio Telephone Co. v. United States, 273 U.S. 236 (1927). In De Forest, the Court held that AT&T’s commitment not to “interfere with the immediate manufacture” of vacuum tubes by the United States constituted a license, thus exhausting the patentee’s rights in those tubes.

De Forest has long been cited for the proposition that a license, even when wrapped in the sheep’s clothes of a CNS, does not thereby shed its character as a license of patent rights (see, e.g., TransCore, LP v. Electronic Transaction Consultants Corp., 563 F.3d 1271 (Fed. Cir. 2009)). While the facts in Meso are far more involved than these, if Petitioner’s assertion is correct that, through artful drafting, a patent license can successfully be masked as a mere CNS, the result could have important implications for patent exhaustion.

A second area in which the question posed in Meso (though unrelated to the facts of that case) could have significant implications relates to the burgeoning phenomenon of “patent pledges”. As I discuss in a recent article (Patent Pledges, 47 Ariz. St. L.J. __ (2015, forthcoming)), patent pledges are public commitments made by patent holders to refrain from enforcing their patents to the fullest degree. Such commitments are made for a variety of reasons, frequently in the context of developing and promulgating technical interoperability standards such as Wi-Fi, 4G, USB and Bluetooth. In some of these cases, the patentee may commit to license patents essential to those standards on terms that are “fair, reasonable and non-discriminatory” (FRAND), but in other cases it may commit not to assert the patents at all (i.e., granting a CNS).

While the FRAND licenses granted by patent holders in these cases are indisputably “licenses” that survive transfer of the underlying patents, etc., the status of a general CNS issued to a standards body or to the public at large is less clear. Are such covenants really licenses that convey property-like interests, or are they merely promises binding on the patentee? In another recent article, I assess the legal theories available to enforce these general pledges (contract, antitrust, property), and conclude that the optimal analytical approach is rooted in promissory estoppel. A theory that is grounded in property, which would conceivably convey property interests to an amorphous, undefined set of potential beneficiaries, could present both practical and theoretical problems, particularly when (as is often the case) the set of patents as to which the commitment is made is unspecified.

The question posed in Meso presents the Court with an opportunity to clarify the now-fuzzy boundary between CNS and licenses of intellectual property. While it seems reasonable to disregard CNS language used by clever patentees simply to avoid exhaustion of their rights and to treat such covenants as licenses per De Forest and its progeny, it also seems a mistake to treat every covenant made with respect to a patent as a property-like license of rights. Newman has convincingly argued that a license is more than a covenant. But at what point does a covenant become a license? To paraphrase the famed legal realist, is it true that “sometimes, a covenant is just a covenant”? The answer is not apparent, but could have serious ramifications for industries dependent on intellectual property transactions.