Judging by the media coverage, the Supreme Court’s decision yesterday in Impression Products, Inc. v. Lexmark Inc. will have dramatic implications for producers and consumers of patented products around the world. The decision places “sharp limits on how much control patent holders have over how their products are used after they are sold,” says the New York Times’s Adam Liptak. The ruling is a “sure-to-be-landmark decision,” reports Ronald Mann at SCOTUSblog. It “takes away an important tool used by companies to control the marketplace,” according to Bloomberg.
Well, maybe. But the Court’s opinion, authored by Chief Justice Roberts, also opens the door for creative contract lawyers to draft licensing agreements that severely restrict resale of patented products. The full impact of the Supreme Court’s decision won’t be known for years, but much will depend on how courts view the newfangled licensing agreements that are almost certain to follow in the wake of Impression Products.
To see why, let’s start with a hypothetical: Suppose your firm, Company A, holds a U.S. patent covering a certain widget. You manufacture one such widget and sell it to B on the condition that B not resell the widget to anyone else. In violation of that condition, B resells the widget to C, who then uses the widget. Can you sue B and/or C for patent infringement?
Prior to yesterday, the answer under Federal Circuit precedent was yes: A could sue both B and C. Yesterday’s decision changes that. The Court holds that “a patentee’s decision to sell a product exhausts all of its patent rights in that item, regardless of any restrictions the patentee purports to impose.” This is true even if A’s sale to B occurs outside the United States. Thus, Company A now cannot sue B or C for patent infringement, though it still might be able to sue B for breach of contract.
But now reconsider the above scenario with the following modification: Company A—instead of selling the widget to B—licenses the widget to B with the proviso that B can do whatever she wants with the widget except resell it. If B violates the terms of the license and resells the widget to C, who then uses it, can Company A sue B and/or C for patent infringement?
The answer would seem to be yes. As the Chief Justice writes in yesterday’s decision, the Supreme Court’s 1938 ruling in General Talking Pictures v. Western Electric continues to “stand for the modest principle that, if a patentee has not given authority for a licensee to make a sale, that sale cannot exhaust the patentee’s rights.” That, says the Chief Justice, is a “fundamentally different situation” from the one encountered in Impression.
But just how different is it? The dispute in Impression involved a manufacturer of printer cartridges (Lexmark) that sold cartridges to certain customers in the United States and abroad on the condition that they not resell the cartridges to anyone else. Some of those customers nonetheless resold their cartridges to Impression, which then refurbished them and put them back on the retail market. Yesterday’s decision makes clear that Lexmark cannot sue Impression for patent infringement. But what if, tomorrow, Lexmark amends the terms of its arrangements with customers so that instead of “selling” cartridges subject to a resale restriction, it “licenses” the cartridges to customers for an indefinite term on the condition that they not resell the cartridges to anyone else? Under the “modest principle” of General Talking Pictures, it would seem that Lexmark’s grant of a license has not exhausted its patent rights. If Impression again tries to refurbish and resell those cartridges, Lexmark can again sue for infringement. While Lexmark lost in this case, now it knows what to do next time.
To be sure, General Talking Pictures applied to a license to make patented products, while our hypothetical printer cartridge arrangement involves a license to use a patented product. But nothing in yesterday’s decision suggests that the make/use distinction is a distinction with a difference. And such a distinction would effectively eliminate a patentee’s ability to ever license use of its products by converting all such licenses into outright sales.
Alternatively, courts in future cases might decide that licenses like the one we’ve imagined are really “disguised sales,” and they might decide to honor the substance of such transactions rather than the form. But the distinction between a sale and a license is a formal one, and any effort to draw a clear line will be a struggle. After all, what’s the difference between (1) a sale on the condition that the purchaser cannot resell the product and (2) a license that allows the licensee to do whatever she wants with the product except resell it? One might say that the latter is a disguised sale, but one might just as easily say that the former is a disguised license. Or one might say that neither is a disguised anything, because the only difference between a sale and a license is the contractual clothes that they wear.
Courts have confronted a similar question in the context of copyright’s first sale doctrine, under which the “first sale” of a copy exhausts the copyright owner’s right to restrict the purchaser’s disposition of that copy. Distinguishing between a “first sale” and a license is not always easy. In Vernor v. Autodesk, Inc., a 2010 case involving graphic design software packages, the Ninth Circuit set forth a multifactor test to help it draw that line. The Ninth Circuit’s Autodesk test looks to “whether the agreement was labeled a license,” “whether the copyright owner retained title to the copy,” and whether the contract between the copyright owner and the purchaser/licensee “required [the copy’s] return or destruction, forbade its duplication, or required the transferee to maintain possession of the copy for the agreement’s duration.” The presence of these factors indicates that the transaction is a license rather than a sale, though no single factor is decisive.
Now let’s say that instead of a copyrighted software package, we have a patent-protected smartphone. It should be easy enough for the smartphone manufacturer to write up a contract that will prevent the phone’s user from reselling it. Instead of buying your smartphone, you’ll “license” it from Apple or Samsung, which will retain title to the device. Your license will last indefinitely, but you’ll agree that when you’re done with the phone, you’ll either return it to the manufacturer or destroy it yourself. And bingo: Under the Autodesk test, we have a license rather than a sale, which means that General Talking Pictures rather than Impression Products supplies the applicable legal rule.
Contracting around Impression Products might be harder in the case of pharmaceutical products. Let’s say that a pharmaceutical company wants to sell an oral diabetes medication at a discounted price in India without cannibalizing its U.S. market. To do so, the pharmaceutical company needs to be able to ensure that the drug cannot be reimported and resold here. (We’ve argued elsewhere that price discrimination of this sort can make it easier for patients in less affluent countries to get access to potentially lifesaving medicines.) Instead of selling the drug in India, might the pharmaceutical company “license” the pill to Indian patients on the condition that they either return the pill or ingest it? And if so, would a court—following something like the Autodesk test—treat the transaction as a license rather than a sale?
We can’t say for sure. The notion of a pill license might strike some courts as too clever by half. The idea of licensing (essentially, leasing) a smartphone is not so farfetched; the idea of leasing a pill might be too hard for a court to swallow. The ability of a patent holder to restrict resale of a product may then come to depend largely on the product’s physical characteristics: Does this seem like the sort of things that parties might plausibly transfer by license/lease rather than by sale? And while we see no reason why the physical characteristics of a product ought to determine the patentee’s ability to restrict sale, that might well be the path that yesterday’s decision takes us.
Chief Justice Roberts writes in yesterday’s opinion that “[i]n sum, patent exhaustion is uniform and automatic.” We doubt it. In all likelihood, the ability of a patent holder to restrict resale of a product will be technology-specific—and will depend on the ingenuity of transactional lawyers as well as the inclinations of lower courts. The limits that Impression Products places on patent holders may turn out to be “sharp,” or they may turn out to be “soft.” All we can say for now is that it will be a long time before they are clear.
Lisa Larrimore Ouellette is an assistant professor at Stanford Law School.