United States: From Deepsouth Shrimp to Microsoft Windows: Exporting Components of Patented Inventions under 35 U.S.C. § 271(f)

For many years Congress and the courts have struggled to maintain the delicate balance between encouraging innovation via the patent laws on the one hand, and using those same laws to stifle the development and marketing of products in the United States on the other. A series of recent decisions from the U.S. Court of Appeals for the Federal Circuit demonstrate the difficulties the courts face in applying the patent laws to an increasingly digital world. These decisions, which will have an immediate impact on the U.S. software industry, will also be of interest to any U.S.-based company that uses offshore manufacturing facilities. This is particularly so given the increasing reliance on the use of software and/or digital information in manufacturing processes.

In 1984, the U.S. Congress enacted 35 U.S.C. § 271(f) to close a loophole in the patent laws. The U.S. Supreme Court had uncovered this loophole more than a decade earlier in Deepsouth Packing Co. v. Laitram Corp., when the court held that an infringer could avoid liability for patent infringement by exporting otherwise infringing products in component parts for assembly overseas.¹ Congress outlawed this practice with the enactment of § 271(f), which expanded the definition of infringement to include exporting components of patented inventions with the intent and/or knowledge that such components would be used in a manner that would otherwise infringe if the activity occurred in the United States. Although the statute was written with general language, the courts subsequently encountered some difficulty applying § 271(f) to inventions without clearly identifiable components.² An especially difficult question concerned computer software patents, which the Federal Circuit finally resolved in two cases during 2005, when the court held unequivocally that exporting software was a violation of § 271(f).³ The court’s holdings may foreshadow expanding liability for other complex inventions that are increasingly manufactured and distributed abroad.

Historically, the patent laws have not applied to activities outside U.S. borders.⁴ In 1849, the Supreme Court wrote that "the use of [a patent right] outside the jurisdiction of the United States is not an infringement."⁵ In 1915, the court wrote that "[t]he right conferred by a patent under our law is confined to the United States and its territories, and infringement of this right cannot be predicated of acts wholly done in a foreign country."⁶ In 1972, the Supreme Court showed how far it was willing to take this principle, refusing to find infringement where a manufacturer deliberately circumvented the patent laws by exporting its infringing machinery in parts that were assembled abroad.⁷ That case was Deepsouth Packing Co. v. Laitram Corp., which was a dispute over shrimp deveining machinery, as Justice White colorfully illustrated at the beginning of the court’s opinion: Shrimp, whether boiled, barbecued or fried, are a gustatory delight, but they did not evolve to satisfy man’s palate. Like other crustaceans, they wear their skeletons outside their bodies in order to shield their savory pink and white flesh against predators, including man. They also carry their intestines, commonly called veins, in bags (or sand bags) that run the length of their bodies. For shrimp to be edible, it is necessary to remove their shells. In addition, if the vein is removed, shrimp become more pleasing to the fastidious as well as more palatable.⁸

The plaintiff, Laitram Corporation, held two patents for machinery used in the process of deveining shrimp.⁹ The defendant, Deepsouth Packing Company, also manufactured deveining machinery, but because of Laitram’s patents, Deepsouth could not make or sell its machines in the United States.¹⁰ However, Deepsouth realized it could still sell its products to foreign customers if the machines were not fully "made" in the United States. Deepsouth thus manufactured its shrimp deveiners in components that were shipped overseas in three separate boxes, where customers could assemble the full machines in less than an hour.¹¹ In a 6-3 decision, the Supreme Court refused to find infringement on these facts, writing that the "patent system makes no claim to extraterritorial effect."¹² The court argued that "it would require a clear and certain signal from Congress before approving the position of a litigant who, as respondent here, argues that the beachhead of privilege is wider, and the area of public use narrower, than courts had previously thought."¹³

That "clear and certain signal" came in the Patent Law Amendments of 1984, when Congress enacted 35 U.S.C. § 271(f):

i) Whoever without authority supplies or causes to be supplied in or from the United States all or a substantial portion of the components of a patented invention, where such components are uncombined in whole or in part, in such manner as to actively induce the combination of such components outside of the United States in a manner that would infringe the patent if such combination occurred within the United States, shall be liable as an infringer.

ii) Whoever without authority supplies or causes to be supplied in or from the United States any component of a patented invention that is especially made or especially adapted for use in the invention and not a staple article or commodity of commerce suitable for substantial non-infringing use, where such component is uncombined in whole or in part, knowing that such component is so made or adapted and intending that such component will be combined outside of the United States in a manner that would infringe the patent if such combination occurred within the United States, shall be liable as an infringer.

As explained in the Congressional Record, § 271(f) "responds to the . . . decision in Deepsouth . . . concerning the need for a legislative solution to close a loophole in patent law."¹⁴ Congress wanted to avoid the obvious ramification of the Deepsouth decision, which was the potential loss of U.S.-based jobs due to the incentive to move manufacturing facilities off-shore in order to avoid the patent laws. According to the Congressional Record, Congress sought to "avoid encouraging manufacturers outside the United States" and to "prevent copiers from avoiding U.S. patents by supplying components of a patented product in this country so that the assembly of the components may be completed abroad."¹⁵

The statute does not create contributory liability for overseas infringement; a violator of § 271(f) directly infringes the patent through the act of supplying components.¹⁶ There is also no requirement of actual infringement abroad: "A party can intend that a shipped component will ultimately be included in an assembled product even if the combination never occurs."¹⁷ While actual combination of components is not necessary, actual supply of the components is required. An offer to supply components abroad is not sufficient to trigger a violation of § 271(f).¹⁸

For inventions similar to the machinery in Deepsouth , the courts applied § 271(f) with little controversy.¹⁹ However, the phrase "component of a patented invention" proved difficult to interpret for inventions that were not directed to easily separable mechanical parts.

In W.R. Grace & Co.-Conn. v. Intercat, Inc., a district court held that chemical compounds could be considered components.²⁰ The defendant argued that § 271(f) should be restricted to mechanical components "since the section was enacted specifically to overrule [Deepsouth]," but the court found no such limitation: "Nowhere in the statute or its legislative history is there a limitation to components of machines and other structural combinations."²¹ This principle was affirmed in Bristol-Myers Squibb Co. v. Rhone-Poulenc Rorer, Inc., where another district court found the defendant was "unable to point to any language in 271(f) or in its legislative history that supports excluding chemical patents from its reach."²²

While chemical patents have generally fallen within the purview of § 271(f), design and method patents have traditionally been exempt because of a perceived lack of component parts. In Aerogroup International, Inc. v. Marlboro Footworks, Ltd., a district court found that § 271(f) "does not apply to the patent at issue here, which has no ‘component parts’ but is rather a design patent for a shoe sole."²³ This reflected a general principle established by the Federal Circuit in Standard Havens Products, Inc. v. Gencor Industries, Inc., where the court declined to apply § 271(f) when a patented process for making asphalt was being used abroad.²⁴ In Enpat, Inc. v. Microsoft Corp., this principle led a district court to conclude that "the language and legislative history of § 271(f) demonstrate an exclusive focus on the sale of components patented in the United States for combination into a finished product, apparatus or invention abroad."²⁵ The Federal Circuit phrased this requirement another way in Pellegrini v. Analog Devices, Inc., when the court required that "components of a patent invention are physically present in the United States and then either sold or exported."²⁶ A strict reading of this case law suggested that § 271(f) specifically excluded method patents and required physical components; however, the Federal Circuit, when faced with these issues in the context of software-related patents, instead applied an expansive interpretation of the statute.

Computer software posed a particularly difficult question for the court-imposed limitations of § 271(f). In Enpat, the District Court for the Eastern District of Virginia found that Microsoft’s software products, which infringed a method patent, did not implicate § 271(f) because they had no components.²⁷ However, that same district court found, five years later, that exported copies of Microsoft’s NetMeeting software could be components under § 271(f).²⁸ The Federal Circuit agreed with this latter conclusion in a pair of cases decided adversely to Microsoft in 2005. In Eolas Technologies, Inc. v. Microsoft Corp., the court found the word "component" in § 271(f) expressly includes software.²⁹ In AT&T Corp. v. Microsoft Corp., the court found the term "supplied" includes the making of digital copies of software overseas from a single exported master disk. ³⁰ Moreover, in both cases, the court refused to limit the applicability of § 271(f) to only apparatus claims. There is no longer any question that the supply of computer software abroad may result in § 271(f) liability, regardless of whether the patented invention covers an apparatus (such as a computer) or a method (such as instructions for accomplishing a specific task on a computer).

Eolas and AT&T both focused on Microsoft’s practice of distributing software through "golden master" disks.³¹ These "golden masters" are produced by Microsoft and sent to manufacturers abroad who "replicate the code onto computer hard drives for sale outside the United States."³² Microsoft argued the intangible information on the disks could not be a "component," but the court in Eolas found "the software code on the golden master disk is not only a component, it is probably the key part of this patented invention."³³ In AT&T, Microsoft emphasized the fact that the golden masters themselves do not become a part of finished computers, but it is copies of the information on those exported disks that are installed on the computers. Microsoft drew an analogy to a master key that is sent abroad for mass replication, but the court found this hypothetical "unpersuasive and irrelevant."³⁴ Rejecting Microsoft’s attempts to distinguish "supplying" and "copying," the Federal Circuit considered "the nature of the relevant technology and business practices" and concluded for software, "the act of copying is subsumed in the act of supplying."³⁵ After Eolas and AT&T, there is no question that every copy of software replicated abroad using an original from the United States is a § 271(f) infringement.³⁶

Beyond rendering final judgment on computer software, the Federal Circuit clarified several other aspects of § 271(f) jurisprudence. The Eolas court effectively eliminated the method patent exception, writing the "court cannot construct a principled reason for treating process inventions different than structural products."³⁷ The court also removed the physical component requirement, clarifying "Pelligrini does not impose on section 271(f) a tangibility requirement that does not appear anywhere in the language of that section."³⁸ Freed from these rigid requirements, the Federal Circuit declared that "every component of every form of invention deserves the protection of section 271(f)."³⁹ The AT&T court was similarly expansive in its interpretation of the statute, refusing to "construe a statutory provision that was originally enacted to … clos[e] a loophole, in a manner that allows . . . technology . . . to subvert that intent."⁴⁰ In rejecting Microsoft’s argument that electronic transmission of software was outside the scope of § 271(f), the court found "[l]iability under § 271(f) is not premised on the mode of exportation, but rather the fact of exportation."⁴¹

In Eolas and AT&T, Microsoft was exploiting essentially the same loophole Deepsouth Packing had used thirty years ago. Instead of three boxes of parts, Microsoft exported "golden master" disks. Instead of assembling shrimp deveining machinery, Microsoft’s foreign manufacturers replicated and installed Windows on foreign computers. The Federal Circuit did not need to ask Congress for a signal this time because the intent of § 271(f) was already clear. Infringers should not be able to circumvent U.S. patents by finding novel ways to export their technology. By eliminating the tangibility requirement and the method patent exception, Eolas and AT&T essentially returned to the text of the statute, asking simply whether or not an exported item is intended to become a component of an infringing product. For Microsoft Windows on golden master disks, the answer was an unequivocal yes.

Although the specific question of computer software has been resolved, the application of § 271(f) to other technologies is not clear. The Federal Circuit case law relies on a distinction between infringing components, which can not be exported under § 271(f), and recipes or designs, which are still outside the scope of the patent law when communicated abroad.⁴² In the Microsoft cases, software copied onto foreign computers was a component that triggered liability. However, in Pelligrini, instructions for manufacturing a computer chip abroad were not components. Although the difference seems clear when phrased as "instructions" versus "copies," the reality of semiconductor manufacturing can be strikingly similar to installing computer software.

Foreign semiconductor manufacturing plants often receive instructions from the United States through electronic transmissions or even on computer disks similar to Microsoft’s golden masters. In this industry, the products are generally too complex for instructions to come over the phone or on paper, like the shoe sole design in Aerogroup or the asphalt-making process in Standard Havens. To make computer chips, complex instructions are commonly sent in a form that can only be read by automated fabrication machines. This process is known as "tape out," and like Microsoft’s software installation process, semiconductor fabrication often involves copying substantial quantities of software code (called "firmware") directly onto memory chips. Because of cheap labor overseas and stronger intellectual property protection in the United States, "tape out" is an increasingly common business practice for the American semiconductor industry. Under Pelligrini, these manufacturers should be protected from § 271(f) liability, but it is difficult to predict how the Federal Circuit might rule after Eolas and AT&T. This is not only a potential issue for the semiconductor industry but for an entire manufacturing sector that is becoming both increasingly automated and international. It may not be long before other high technology industries will face difficult questions concerning § 271(f), as patentees attempt to apply the Eolas and AT&T decisions to other manufacturing methods that rely on digital infor mation transferred from the United States.

§ 271(f) was enacted to close a loophole in the patent law, but the development of more advanced products and manufacturing techniques has created a new set of potential loopholes. Congress sought to prevent patent infringers from circumventing the law by separating patented machinery into components, but modern patents are often built with components of information rather than nuts and bolts. Manufacturers in the digital age are avoiding infringement liability by transferring information in intangible forms, loopholes that Congress did not anticipate in 1984. The Federal Circuit closed some of these loopholes in the Eolas and AT&T decisions, but the fates of other software and hardware manufacturers are still uncertain. The software industry rapidly responded by asking Congress to amend § 271(f) to specifically limit the definition of "component" to only tangible items that are physically combined with other components to create a patented combination. ⁴³ However, this proposed amendment has received a lukewarm response at best, and at any rate it could be years before Congress takes any affirmative action. Thus, given the growing density of patents in high technology, it will likely remain the task of the courts to refine the scope of § 271(f) in the coming years. Where the line is drawn will have multibillion dollar effects on future litigation and far-reaching consequences for international trade and technological development.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.