In a previous post, I discussed the Federal Circuit’s recent Princo decision, which limited the affirmative defense of patent misuse. The effect of the decision is to give licensors more room to use certain restrictive license terms. Princo is a good occasion to take stock and reconsider what types of practices may constitute patent misuse. What follows is not breaking news, but I hope it is a useful refresher.
Starting with the easy stuff, Congress amended Section 271 of the Patent Act in 1988 to make clear that certain practices are not patent misuse: (1) enforcing a patent against infringement; (2) refusing to license a patent (to anyone, or to someone in particular); (3) licensing a patent only on condition that the licensee purchase another product or take a license to another patent, unless the licensor has “market power” in the market for the patented product (generally speaking, market power is the power of a company to raise prices or restrict competition in a given product market, usually as a result of considerable market share or other competitive advantages); and (4) using one’s invention or authorizing another to use it in ways that (to simplify a bit) knowingly facilitate infringement of the patent by a third party (“contributory infringement” — in other words, a defendant cannot avoid liability for infringement by claiming misuse on the grounds that the patent holder knew he might infringe and licensed the patent anyway). To this list of statutory exceptions, Princo adds practices that do not directly use the patent to gain leverage and, it seems, those that are not anticompetitive.
Many other license restrictions will not constitute misuse because they do not impermissibly enlarge the scope of a patent. No misuse occurs when a licensor grants less than the full rights in a patent, for example by imposing territorial or field of use limitations, output restrictions or quality requirements. Many license practices are unlikely to cause misuse when they are warranted by legitimate business concerns and not clearly anticompetitive. One example would be a “grant-back” clause where a license is granted with a requirement that the licensee grant back to the licensor nonexclusive rights in any improvements (some case law has found misuse in grant backs under some circumstances, such as those involving unrelated subject matter and exclusive grant backs involving competitors).
By contrast, looking to several (mostly decades old, but still good law) Supreme Court decisions and Princo, we can say that some types of practices can constitute misuse, at least in the right circumstances.
“Tie-in” arrangements constitute the first set of practices. A “tie in” is an antitrust concept that describes agreements (in the patent setting) where a seller conditions the sale of a patented product (or the license of the invention) on the buyer’s willingness to purchase an unpatented product. For example, the Supreme Court in Morton Salt (1942) concluded that the holder of a patent in a machine for adding salt to canned foods would sell the machine only if buyers purchased its unpatented (and utterly fungible) salt tablets. Morton was improperly trying to extend the scope of its patent to unpatented products (where it faced much stiffer competition). What exception No. 3 above in Patent Act 271(d) clarifies is that a tie in can be patent misuse only if the patentee has market power. In the past, some lawyers and courts believed that, because patents conferred something like a monopoly, patent holders always had market power. The prevailing law today (as evidenced by Section 271(d)) is that patents do not necessarily create market power because many patents are not so broad and so successful as to give their owners the power to raise prices. Just the same, licenses including tie ins of unpatented products are relatively risky if the patentee has a strong market position.
Another category of licensing terms that can be misuse are those that improperly extend the duration of a patentee’s exclusive rights. For example, in Brulotte v. Thys Co. (1964), the Supreme Court held a license unenforceable because it required the payment of royalties beyond the life of the patent. A licensor cannot argue that royalties of longer duration are “part of the price” of getting a license. This rule too depends on circumstances. For example, a licensor may require royalties before or after a patent issues if the license includes trade secret subject matter. See Aronson v. Quick Point Pencil (1979).
What is intriguing in the wake of Princo is the question of whether other practices can amount to misuse when they are not necessarily antitrust violations and/or that do not expand the “physical or temporal” reach of the patent, but that undermines the purposes of the Patent Act. The question has special significance for those doing business in the Fourth Circuit (which includes Virginia), because the Fourth Circuit has twice ruled (once in a patent case, and another in a copyright misuse holding) that licensor-imposed post-license noncompete clauses were unenforceable because they prevented the licensees from inventing their own devices or creating their own copyrighted software. In both cases (Compton v. Metal Products, Inc (1971), the patent matter, and Lasercomb America, Inc. v. Reynolds (1990), the copyright case), the licenses contained long-term noncompete clauses that prevented the licensees from creating and selling competitive offerings. The court held the practice undermined the policy inherent in both patent and copyright law of encouraging inventiveness and creativity, respectively.
The majority opinion in Princo distinguishes and criticizes the Compton decision, and it seems to require anticompetitive effects and, perhaps, an actual violation of antitrust law. Yet, it does not squarely hold that an IP policy-based misuse theory is invalid. Overall, Princo is not encouraging for those alleging misuse on the ground that a license subverts the policies in IP law (at least in patent matters, where the Federal Circuit is now the exclusive court of appeals). It said that misuse is not available “simply because a patentee engages in some kind of wrongful commercial conduct, even conduct that may have anticompetitive effects.” Princo appears to state that the practice must involve the patent, seek to extend the physical or temporal scope of the patent, and have anticompetitive effects. If the decision is as sweeping as the language suggests, then licensees facing some highly restrictive license terms that stifle creation but do not constitute antitrust violations (such as the noncompete in Lasercomb that barred the licensee from developing any competing software for the term of the license plus one year – the license had a 99 year term) may have no remedy. Aggressive licensors will be emboldened. –Christopher J. Mugel