Starting point of the analysis is the presumption against extraterritoriality – “the general rule under United States patent law that no infringement occurs when a patented product is made and sold in another country.”2 “The presumption that United States law governs domestically but does not rule the world,” the Supreme Court explained, “applies with particular force in patent law.”3
That said, under the current case law, no single factor determines whether a sale is considered “within the United States” or not. As the Federal Circuit explained, “[t]he standards for determining where a sale may be said to occur do not pinpoint a single, universally applicable fact that determines the answer, and it is not even settled whether a sale can have more than one location.”4 “Places of seeming relevance,” the court stated, “include a place of inking the legal commitment to buy and sell and a place of delivery, and perhaps also a place where other ‘substantial activities of the sales transactions’ occurred.”5
The vague and open-ended standard of whether there were “substantial activities of a sales transaction” in the United States6 – recently called “the key question” by the Federal Circuit7 – allows for little predictability. Three cases involving the international supply chains of U.S. companies and similar kinds of related framework agreements illustrate the murky boundaries:
In Halo,8 defendant Pulse supplied electronic components to Asia-based contract manufacturers of U.S.-based Cisco. In addition to a “general agreement” on issues like manufacturing capacity, Cisco and Pulse periodically agreed upon component prices. Applying those prices, Cisco’s contract manufacturers would then order components from Pulse by sending purchase orders to Pulse’s sales offices outside the U.S. Pulse then delivered the products from its manufacturing facility in Asia to the contract manufacturers, also located in Asia. The Federal Circuit affirmed the district court’s summary judgment finding no sale within the United States. It held that “when substantial activities of a sales transaction, including the final formation of a contract for sale encompassing all essential terms as well as the delivery and performance under that sales contract, occur entirely outside the United States, pricing and contracting negotiations in the United States alone do not constitute or transform those extraterritorial activities into a sale within the United States for purposes of § 271(a).”9 “While Pulse and Cisco engaged in quarterly pricing negotiations for specific products,” the court explained, “the negotiated price and projected demand did not constitute a firm agreement to buy and sell, binding on both Cisco and Pulse. Instead, Pulse received purchase orders from Cisco’s foreign contract manufacturers, which then firmly established the essential terms including price and quantity of binding contracts to buy and sell. Moreover, Pulse was paid abroad by those contract manufacturers, not by Cisco, upon fulfillment of the purchase orders. Thus, substantial activities of the sales transactions at issue, in addition to manufacturing and delivery, occurred outside the United States.”10
In MediaTek,11 defendant Freescale Semiconductor had a “Standard Sales Agreement” with a U.S.-based subsidiary of Amazon (in the following “Amazon”), negotiated and executed in the U.S. The Agreement listed Amazon as the buyer, stated that it “governs all product purchases made by Authorized Purchasers … from Freescale,” and that “Freescale will sell to Authorized Purchasers, and Authorized Purchasers will buy from Freescale, products from time to time.” “Authorized Purchasers” included Amazon itself as well as “Designees,” such as Asia-based Foxconn. Prices were set forth in an attachment to the Agreement, and the Agreement stated that Freescale would not provide any Designee any rebates or discounts without Amazon’s consent. Certain of Freescale’s accused chips were manufactured outside the U.S., sold to Foxconn outside the U.S., and incorporated by Foxconn into Amazon Kindle products outside the U.S. Nevertheless, the Northern District of California denied summary judgment of non-infringement. The court pointed out that while Foxconn received the products in China for incorporation into Amazon Kindles manufactured there, Foxconn purchased those products pursuant to the Freescale–Amazon Agreement; Amazon controlled the pricing terms for all Designees and specifically restricted Freescale’s ability to negotiate pricing with those Designees directly.12 “The delegation of authority to issue purchase orders,” the court explained, “does not change the fact that every sale thereunder is subject to the terms and conditions in the Agreement.”13 The Agreement would provide “tangible evidence of a sales relationship between two U.S. companies, not merely ‘some pricing discussions’ as existed in Halo.”14
In Lake Cherokee,15 U.S.-based defendant Marvell Semiconductor competed for, and obtained, “design wins” with U.S.-based customers, whereupon the customer would buy all requirements for a particular chip from Marvell for a particular end-product; the agreement on specification and price would be memorialized in a product supply agreement. Based on such a product supply agreement, “companies abroad” would place purchase orders for specific quantities, with the ordered chips then being manufactured and delivered to third parties outside the United States. The Eastern District of Texas found, as a matter of law, that such sales could not be considered “within the United States” under § 271(a). The court explained that “[t]o hold otherwise would convert the act of entering into a product supply agreement within the United States into a springboard for liability each time a purchase order is subsequently placed and fulfilled entirely abroad and where the accused product at no time enters the domestic United States market.”[xv] A contract between two U.S. companies, negotiated and executed within the United States, but for delivery and performance outside the United States, would not constitute a sale under § 271(a), the court held.17
In all three cases, the commercial mechanisms were similar: Based on domestic pricing agreements between the respective defendant and the respective U.S. company – economically speaking the real customer the defendant was competing for – contract manufacturers abroad issued purchase orders to the defendant, who then delivered the accused products abroad. Yet, the courts focused on different aspects: The MediaTek-court emphasized that the domestic agreement governed the purchase orders and denied summary judgment of non-infringement. The Lake Cherokee-court emphasized that the domestic agreement was for delivery and performance outside the United States and therefore not a sale within the United States. And the Halo-court, in turn, emphasized that the domestically negotiated price did not constitute a “firm agreement to buy and sell,” finding no sale within the United States.
In the big picture, the guideposts for determining where a sale occurs remain fluid and elusive. The Federal Circuit never articulated a concise standard; to the contrary, as recently as 2015, it expressly declined to provide one: “At this point, we do not settle on a legal definition or even to say whether any sale has a unique location.”18
One fact pattern, however, was repeatedly and unambiguously – and, in my opinion, wrongly – addressed: Shipment of the allegedly infringing product “free on board” (“f.o.b.”) – and thus transfer of title – outside of the United States does not, in and of itself, constitute a sale outside of the United States and avoid liability. The genesis of this principle merits a closer look:
In North American Philips (1994),19 the Federal Circuit addressed the location of a sale in the context of personal jurisdiction, not liability under the Patent Act. The defendants sold the allegedly infringing products to buyers in Illinois, but shipped the products free on board in Texas and California. The court explained that “‘[f]ree on board’ is a method of shipment whereby goods are delivered at a designated location, usually a transportation depot, where legal title and thus the risk of loss passes from seller to buyer.”20 Finding that the Illinois district court had jurisdiction, the Federal Circuit emphasized the location of the buyers, holding that “to sell an infringing article to a buyer in Illinois is to commit a tort there (though not necessarily only there).” Citing the Supreme Court’s Burger King-decision on personal jurisdiction,21 the court stated that “[t]o hold otherwise would exalt form over substance in an area where the Supreme Court generally has cautioned against such an approach.”22 The court expressly rejected the proposition that the place where legal title passes controls, stating that “appellee has failed to explain why the criterion should be the place where legal title passes rather than the more familiar places of contracting and performance” (again citing Burger King), and that “[a]ppellees have pointed to no policy that would be furthered by according controlling significance to the passage of legal title here.”23
In Litecubes (2008),24 the Federal Circuit addressed a similar scenario in the context of liability under § 271 of the Patent Act. The Canadian defendant “sold and shipped the allegedly infringing products directly to customers located in the United States” but “the products were shipped f.o.b., and thus title over the goods were transferred while the goods were still in Canada.”25 Relying on North American Philips, the court found substantial evidence to support the jury’s conclusion that the defendant sold the accused products within the United States. It acknowledged that North American Philips dealt with personal jurisdiction, but summarily denied that that would make a difference: “While North American Philips dealt with the issue of where a ‘sale’ took place under § 271 in the context of personal jurisdiction, we see no basis for construing the location of a ‘sale’ differently when the issue is whether the plaintiff has established that the sale took place within the United States for the purposes of infringement.”26
In SEB (2010), the Federal Circuit relied on Litecubes to reject the argument that sales with product delivery f.o.b. in Hong Kong or China were overseas sales.27 Various district court decisions followed North American Philips or Litecubes, rejecting analogous “f.o.b.-defenses.”28
In light of the presumption against extraterritoriality, rejecting title transfer as the key factor for the location of a sale is not, in my opinion, consistent with the Patent Act.
On several occasions, the Federal Circuit noted that § 271(a) does not define the term “sells;” it therefore looked to the “ordinary meaning” of the term “sale,” citing the definition in Black’s Law Dictionary: “1. The transfer of property or title for a price. 2. The agreement by which such a transfer takes place.”29 The court also referred to Article 2 of the Uniform Commercial Code – “recognized as a persuasive authority on the sale of goods” – which “provides that ‘[a] ‘sale’ consists in the passing of title from the seller to the buyer for a price.’”30 But despite acknowledging these definitions, the court interpreted the term “sells” in § 271(a) much broader that what is reflected in the definitions; it included factors in the analysis that are not part of the ordinary meaning of “sale,” such as the location of the buyer in the “free on board” cases.
When the Federal Circuit refused to “accord controlling significance to the passage of legal title” in North American Philips, it did so in the context of personal jurisdiction – an area in which the Supreme Court cautioned against “mechanical” tests.31 In the context of liability under the Patent Act, however, the presumption against extraterritoriality, which “applies with particular force in patent law,”32 points in the other direction – a narrow reading of the statutory text. The Federal Circuit did not address this important difference when it decided Litecubes.33
Instructive is Microsoft, where the Supreme Court invoked the presumption against extraterritoriality in narrowly reading § 271(f) and reversing the Federal Circuit’s policy-driven, broad interpretation of that section. § 271(f) deems it an infringement when one “supplies … from the United States, for “combination” outside of the United States, a patented invention’s “components.” The Supreme Court expressly noted that it would “construe §271(f)’s terms ‘in accordance with [their] ordinary or natural meaning.’”34 Microsoft concerned the applicability of § 271(f) to computer software first sent from the United States to a foreign manufacturer on a master disk, or by electronic transmission, then copied by the foreign recipient for installation on computers made and sold abroad.35 The presumption against extraterritoriality, the Supreme Court explained, “tugs strongly against construing § 271(f) to encompass as a ‘component’ not only a physical copy of software, but also software’s intangible code, and to render ‘supplie[d] … from the United States’ not only exported copies of software, but also duplicates made abroad.”36
The rationale the Supreme Court expressed in Microsoft supports a narrow reading of the term “sells” in § 271(a): In light of the presumption against extraterritoriality, only the ordinary meaning of “sale,” as reflected in the mentioned definitions, should be relevant – to the exclusion of factors beyond the ordinary meaning.
Under the first Black’s Law Dictionary definition (“The transfer of property or title for a price”) as well as Article 2 of the UCC (“the passing of title from the seller to the buyer for a price”), it is the transfer of title to the allegedly infringing product (the alleged “patented invention”) that is the linchpin of a “sale” for purposes of § 271(a).
The second Black’s Law Dictionary definition refers to the “agreement by which such a transfer takes place,” which raises the question of what the location of a “sale” under that understanding is – the location where the agreement is executed or the location of the title transfer contemplated in the agreement? A useful scenario for that question is an agreement, executed in the U.S., for transfer of title to the accused product outside of the U.S. (and where the accused product may never be physically present in the U.S. at all). In light of the presumption against extraterritoriality, it would seem problematic to find a sale within the U.S. in this scenario – a point the Lake Cherokee-court made in the context of a product supply agreement.37 The reverse is not true. An agreement to transfer title to a product in the U.S. seems well within the boundaries of U.S. patent law, even if the agreement was executed abroad. Indeed, in Transocean, the Federal Circuit held that “a contract between two U.S. companies for the sale of the patented invention with delivery and performance in the U.S. constitutes a sale under § 271(a) as a matter of law,” even though the contract was negotiated and signed while the two U.S. companies were abroad.38 In light of the presumption against extraterritoriality, therefore, the better approach seems to be to focus on the location of the contemplated title transfer, not the one of the execution of the agreement.
In sum, the location of title transfer should be, in my opinion, the controlling factor in whether a sale occurred “within the United States” under § 271(a) of the Patent Act.
While there may be policy arguments for a more flexible, overall assessment of “substantial activities of a sales transaction” or a focus on the buyer’s location, nothing in § 271(a) calls for either approach. Under the current statute, the transfer of title is the critical factor. Moreover, it can serve that function well; transfer of title usually has a specific time and location under the applicable rules of contract law, allowing for consistent and predictable decisions.
If transfer of title controlled under § 271(a), courts would have likely found no direct infringement in any of the “free on board” cases described above. In the context of “f.o.b.” contracts, a distinction is made between, on the one hand, “f.o.b. place of shipment” contracts, in which the seller must put the goods in the hands of a carrier at the designated place, and on the other hand, “f.o.b. place of destination” contracts, in which the seller must transport the goods to that place and tender delivery of them there.39 While not expressly described as such, the “free on board” cases described above seemed to have involved f.o.b. place of shipment contracts. Unless otherwise agreed, in an f.o.b. place of shipment contract, title passes at the time and place of delivery to the carrier.40 Under the narrow reading of § 271(a) described above, the f.o.b. location is thus where the “sale” occurs. The carrier then transports the goods to the buyer “acting as an agent or bailee of the buyer.”41 And while some of the “free on board” cases emphasize that the goods were shipped “directly to customers located in the United States,”42 or that the invoices for the goods identified U.S. destinations for delivery,43 the mere fact that the seller directs the carrier to ship the goods to the customer (e.g., by way of “ship to” labels) does not turn a shipment contract into a destination contract.44
Of course, reading “sale” under § 271(a) narrowly to refer only to title transfer does face considerable policy objections. Indeed, in North American Philips, the Federal Circuit stated that “[a]ppellees have pointed to no policy that would be furthered by according controlling significance to the passage of legal title here.” There are certainly good arguments that the buyer’s location should be the key factor, since that would account for the economic reality that sellers compete for customers in the United States. But the wording of the statute, coupled with the presumption against extraterritoriality, leaves little room for such an interpretation of the law. Concerns that the statute is formalistic or creates loopholes should be addressed by the legislator.45
Furthermore, a narrow reading of § 271(a) does not leave the patent owner without remedies. First, if the patent owner desires to prevent sales of the patented invention in foreign countries, it can obtain and enforce foreign patents.46 Second, if the accused product enters the United States, the patent owner can enforce its patent against the importer. E.g., when the seller delivers the accused product f.o.b. outside the United States, and a shipping company then ships the product, as the buyer’s agent, from the f.o.b.-point to the buyer in the United States, it is usually the buyer who is the importer.47 And after importation, the patent owner can sue the buyer for any use or re-sale of the accused product in the U.S. Of course, in many constellations – in particular when patent owner and seller compete for the buyer’s business – the patent owner will be reluctant to sue not the seller (its competitor) but the buyer (its potential customer).48 But that is an issue of business strategy, not a factor in the legal assessment. Moreover, sellers and buyers can – if they do not agree on a title transfer within the United States in the first place – contractually shift the liability risk and duty to defend to the seller by way of an indemnification agreement. If § 271(a) is narrowly read as outlined above, the use of such agreements would likely increase, which in turn would reduce a patent owner’s reluctance to sue the buyer.
The Federal Circuit will soon encounter the above-described, narrow reading of § 271(a). In Pulse Electronics, Inc. v. U.D. Electronic Corp.,49 the Taiwan-based defendant sold allegedly infringing products to a U.S. company f.o.b. Hong Kong. Some of the invoices showed a U.S. “Bill to” address and, below listing the “freight forwarder,” a U.S. “Ship to” address. The district court found, in my opinion correctly, no sales within the United States. It mainly based its decision on the location of the transfer of title, as determined under the law it found applicable to the sales contract, the United Nations Convention of Contracts for the International Sale of Goods (“CISG”). The court stated that “determining the place of sale requires determining where title passed, which requires determining where the risk of loss passed. When a delivery term is F.O.B., the ‘goods are delivered at a designated location, usually a transportation depot, at which legal title and thus the risk of loss passes from seller to buyer.’”50 Noting that, in the case at hand, the f.o.b. invoices failed to indicate whether they were shipment or destination contracts, the court concluded that “[u]nder either a shipment contract or destination contract, title passed to the buyers of Defendant’s products either at the place of destination (i.e., Hong Kong) or the place of shipment (also Hong Kong). As a result, under the CISG (as well as UCC), the place of sale was in China, not the U.S., and Section 271 does not apply to create liability for Defendant for direct infringement.”51
Oral argument in the Federal Circuit in Pulse is scheduled for May 4, 2022.