TARANTO, Circuit Judge.

Constantine L. Trela, Jr., Sidley Austin LLP, Chicago, IL, argued for plaintiff-cross-appellant. Also represented by Robert N. Hochman; Benjamin Beaton, Joshua John Fougere, Washington, DC; Timothy Colin Meece, Bryan Medlock, Jr., Audra C. Eidem Heinze, Jason S. Shull, Banner & Witcoff, Ltd., Chicago, Il; Steven B. Loy, Stoll, Keenon & Park, LLP, Lexington, KY.

Edward F. O'connor, Avyno Law, P.C., Encino, CA, argued for defendant-appellant. Also represented by Jennifer Herbst Hamilton.

Andrew J. Pincus, Mayer Brown LLP, Washington, DC, argued for amici curiae LG Electronics, Inc., Dell Inc., Google Inc., Intel Corporation, L Brands Inc., Newegg Inc., Ninestar Image Tech Limited, QVC, Inc., Samsung Electronics Co., Ltd., SAS Institute, Inc., Xilinx, Inc. Also represented by Jamie B. Beaber, Kfir Levy, Paul Whitfield Hughes; James Suh, LG Electronics Inc., Seoul, Korea; Matthew R. Hulse, Intel Corporation, Santa Clara, CA. Melissa N. Patterson, Appellate Staff, Civil Division, United States Department of Justice, Washington, DC, argued for amicus curiae United States. Also represented by Benjamin C. Mizer, Mark R. Freeman.

Barbara A. Fiacco, Foley Hoag LLP, Boston, MA, argued for amici curiae Biotechnology Industry Organization, CropLife International. Also represented by Sarah S. Burg.

Margreth Barrett, University of California–Hastings College of Law, The Sea Ranch, CA, as amicus curiae pro se.

Frederick M. Abbott, Florida State University College of Law, Tallahassee, FL, as amicus curiae pro se.

Kristin Leigh Yohannan Moore, Cadwalader, Wickersham & Taft LLP, Washington, DC, for amicus curiae American Intellectual Property Law Association. Also represented by Tihua Huang; Lisa K. Jorgenson, American Intellectual Property Law Association, Arlington, VA.

Meenakshi Kala Sarvaiya, SoCal IP Law Group LLP, Westlake Village, CA, for amicus curiae Conejo Valley Bar Association. Also represented by Steven C. Sereboff.

Noah Leibowitz, Simpson Thacher & Bartlett, LLP, New York, NY, for amicus curiae New York Intellectual Property Law Association. Also represented by Walter E. Hanley, Jr., Kenyon & Kenyon LLP, New York, NY; David F. Ryan, Croton–On–Hudson, NY.

Charles Duan, Public Knowledge, Washington, DC, for amici curiae Public Knowledge, Electronic Frontier Foundation, Open Source Hardware Association, Digital Right to Repair Coalition, Public Citizen, Inc. Also represented by Mohammed Raza Panjwani, Sherwin Siy; Vera Ranieri, Electronic Frontier Foundation, San Francisco, CA.

Peter James Wied, Lee Tran Liang & Wang LLP, Los Angeles, CA, for amici curiae Quanta Computer, Inc., Acer, Inc. Also represented by Vincent K. Yip.

John R. Alison III, Winston & Strawn LLP, Washington, DC, for amici curiae HTC Corp., HTC America, Inc. Also represented by Owais Ahmed Siddiqui, San Diego, CA; Gino Cheng, Los Angeles, CA.

Charles Lifland, O'Melveny & Myers LLP, Los Angeles, CA, for amicus curiae SK Hynix Inc. Also represented by Susan Roeder, Susan Van Keulen, Menlo Park, CA.

James R. Klaiber, Pryor Cashman LLP, New York, NY, for amicus curiae The Association of the Bar of the City of New York. Also represented by Timothy P. Heaton, Troutman Sanders LLP, New York, NY; Aaron Ligoury Joseph Pereira, Buchanan Ingersoll & Rooney PC, New York, NY.

Steven A. Hirsch, Keker & Van Nest, LLP, San Francisco, CA, for amicus curiae SanDisk Corporation. Also represented by Christa M. Anderson, Robert A. Van Nest, Leo L. Lam.

Robert T. Haslam, Covington & Burling LLP, Redwood Shores, CA, for amicus curiae Texas Instruments, Inc. Also represented by Nathan Shaffer; Ranganath Sudarshan, Washington, DC.

Matthew J. Moore, Latham & Watkins LLP, Washington, DC, for amici curiae Costco Wholesale Corp., Retail Litigation Center, Inc. Also represented by James Scott Ballenger, Melissa Arbus Sherry.

Phillip R. Malone, Stanford Law School, Juelsgaard Intellectual Property and Innovation Clinic, Mills Legal Clinic, Stanford, CA, for amici curiae American Antitrust Institute, Jeremy W. Bock, Esq., Irene Calboli, Michael A. Carrier, Andrew Chin, Samuel Ernst, Shubha Ghosh, Ariel Katz, Mark A. Lemley, Yvette Joy Liebesman, Brian J. Love, Mark P. McKenna, Michael J. Meurer, Tyler T. Ochoa, Mark R. Patterson, Esq., Aaron Perzanowski, John A. Rothchild, Pamela Samuelson, Sharon K. Sandeen, Kurt M. Saunders, Christopher B. Seaman, Katherine J. Strandburg, Jennifer M. Urban, Esq., Ryan Vacca, Sarah R. Wasserman–Rajec. Also represented by Jeffrey Theodore Pearlman.

Robert Anthony Surrette, McAndrews, Held & Malloy, Ltd., Chicago, IL, for amicus curiae Association of Medical Device Reprocessors. Also represented by Christopher M. Scharff.

William Douglas Kari, Arbitech, LLC, Irvine, CA, for amici curiae Association of Service and Computer Dealers International, Inc., Owners' Rights Initiative.

Matthew A. Levy, Computer & Communications Industry Association, Washington, DC, for amicus curiae Computer & Communications Industry Association.

Daniel Stringfield, Steptoe & Johnson, LLP, Chicago, IL, for amicus curiae Licensing Executive Society (U.S.A. and Canada), Inc. Also represented by Katherine H. Johnson; Brian P. O'shaughnessy, Ratner Prestia, Washington, DC.

Merritt Blakeslee, The Blakeslee Law Firm, Washington, DC, for amicus curiae Recycling Times Media Corporation.

Mark Schonfeld, Burns & Levinson, LLP, Boston, MA, for amicus curiae Imaging Supplies Coalition. Also represented by Sara Beccia.

Garrard R. Beeney, Sullivan & Cromwell LLP, New York, NY, for amicus curiae Dolby Laboratories, Inc. Also represented by Adam R. Brebner.

Robert P. Taylor, Arnold & Porter, LLP, San Francisco, CA, for amicus curiae Intellectual Property Owners Association. Also represented by Herbert Clare Wamsley, Jr., Intellectual Property Owners Association, Washington, DC; Kevin H. Rhodes, 3M Innovative Properties Company, St. Paul, MN; Philip Staton Johnson, Johnson & Johnson, New Brunswick, NJ.

John D. Haynes, Alston & Bird LLP, Atlanta, GA, for amici curiae Nokia Technologies OY, Nokia USA Inc.

Kathi A. Cover, iBiquity Digital Corporation, Columbia, MD, for amicus curiae iBiquity Digital Corporation.

Joseph S. Cianfrani, Knobbe, Martens, Olson & Bear, LLP, Irvine, CA, for amicus curiae Medical Device Manufacturers Association. Also represented by Kent N. Shum.

Roger Brooks, Cravath Swaine & Moore LLP, New York, NY, for amicus curiae Qualcomm Incorporated. Also represented by David J. Kappos.

John Nilsson, Arnold & Porter LLP, Washington, DC, for amicus curiae Pharmaceutical Research and Manufacturers of America. Also represented by Kristan Lynn Lansbery, Samuel Drezdzon; Willow White Noonan, San Francisco, CA.

Theodore Lawrence Field, South Texas College of Law, Houston, TX, as amicus curiae pro se.

David S. Steuer, Wilson, Sonsini, Goodrich & Rosati, PC, Palo Alto, CA, for amicus curiae InterDigital, Inc. Also represented by Michael Brett Levin, Maura L. Rees.

Seth David Greenstein, Constantine Cannon LLP, Washington, DC, for amici curiae International Imaging Technology Council, Auto Care Association, Automotive Parts Remanufacturers Association.

Opinion

Opinion for the court filed by Circuit Judge TARANTO, in which Chief Judge PROST and Circuit Judges NEWMAN, LOURIE, MOORE, O'MALLEY, REYNA, WALLACH, CHEN, and STOLL join. Dissenting opinion filed by Circuit Judge DYK, in which Circuit Judge HUGHES joins.

TARANTO, Circuit Judge.

Congress has declared: “Except as otherwise provided in [the Patent Act], whoever without authority makes, uses, offers to sell, or sells any patented invention, within the United States or imports into the United States any patented invention during the term of the patent therefor, infringes the patent.” 35 U.S.C. § 271(a); see id. § 154(a) (granting patentee “right to exclude others” from itemized actions). The doctrine of patent exhaustion (or “first sale” doctrine) addresses the circumstances in which a sale of a patented article (or an article sufficiently embodying a patent), when the sale is made or authorized by the patentee, confers on the buyer the “authority” to engage in acts involving the article, such as resale, that are infringing acts in the absence of such authority. There is nothing “otherwise provided” on the issue in the Patent Act. In that respect, the Patent Act differs from the Copyright Act, whose infringement, importation, and exclusive-rights provisions, 17 U.S.C. §§ 501, 602, 106, are all subject to a separate, overriding statutory provision that grants owners of certain copyrighted articles a right to sell those articles “without the authority” of the copyright holder, id. § 109(a).

In this case, all of the initial sales at issue were made by the U.S. patentee, rather than by a licensee having authorization from the patentee. Some of the initial sales were made domestically, some abroad. All of the domestic sales, and an unknown portion of the foreign sales, were accompanied by clearly communicated restrictions on the buyer's reuse and resale.

We decided to hear this case en banc to consider whether two decisions of this court concerning the uncodified doctrine of patent exhaustion—one decision from 1992, the other from 2001—remain sound in light of later decisions of the Supreme Court. Today we reaffirm the principles of our earlier decisions.

First, we adhere to the holding of Mallinckrodt, Inc. v. Medipart, Inc., 976 F.2d 700 (Fed.Cir.1992), that a patentee, when selling a patented article subject to a single-use/no-resale restriction that is lawful and clearly communicated to the purchaser, does not by that sale give the buyer, or downstream buyers, the resale/reuse authority that has been expressly denied. Such resale or reuse, when contrary to the known, lawful limits on the authority conferred at the time of the original sale, remains unauthorized and therefore remains infringing conduct under the terms of § 271. Under Supreme Court precedent, a patentee may preserve its § 271 rights through such restrictions when licensing others to make and sell patented articles; Mallinckrodt held that there is no sound legal basis for denying the same ability to the patentee that makes and sells the articles itself. We find Mallinckrodt's principle to remain sound after the Supreme Court's decision in Quanta Computer, Inc. v. LG Electronics, Inc., 553 U.S. 617, 128 S.Ct. 2109, 170 L.Ed.2d 996 (2008), in which the Court did not have before it or address a patentee sale at all, let alone one made subject to a restriction, but a sale made by a separate manufacturer under a patentee-granted license conferring unrestricted authority to sell.

Second, we adhere to the holding of Jazz Photo Corp. v. International Trade Comm'n, 264 F.3d 1094 (Fed.Cir.2001), that a U.S. patentee, merely by selling or authorizing the sale of a U.S.-patented article abroad, does not authorize the buyer to import the article and sell and use it in the United States, which are infringing acts in the absence of patentee-conferred authority. Jazz Photo's no-exhaustion ruling recognizes that foreign markets under foreign sovereign control are not equivalent to the U.S. markets under U.S. control in which a U.S. patentee's sale presumptively exhausts its rights in the article sold. A buyer may still rely on a foreign sale as a defense to infringement, but only by establishing an express or implied license—a defense separate from exhaustion, as Quanta holds—based on patentee communications or other circumstances of the sale. We conclude that Jazz Photo's no-exhaustion principle remains sound after the Supreme Court's decision in Kirtsaeng v. John Wiley & Sons, Inc., ––– U.S. ––––, 133 S.Ct. 1351, 185 L.Ed.2d 392 (2013), in which the Court did not address patent law or whether a foreign sale should be viewed as conferring authority to engage in otherwise-infringing domestic acts. Kirtsaeng is a copyright case holding that 17 U.S.C. § 109(a) entitles owners of copyrighted articles to take certain acts “without the authority” of the copyright holder. There is no counterpart to that provision in the Patent Act, under which a foreign sale is properly treated as neither conclusively nor even presumptively exhausting the U.S. patentee's rights in the United States.

Background

The relevant facts are set forth in the limited record that the parties agreed was determinative of the result. Lexmark International, Inc. makes and sells printers as well as toner cartridges for its printers. Lexmark owns a number of patents that cover its cartridges and their use. The cartridges at issue here were first sold by Lexmark, some abroad and some in the United States. Some of the foreign-sold cartridges and all of the domestically sold cartridges at issue were sold, at a discount, subject to an express single-use/no-resale restriction. Impression Products, Inc. later acquired the cartridges at issue in order to resell them in the United States—the restricted ones after a third party physically modified them to enable re-use in violation of the single-use/no-resale restriction. Impression has resold the patented Lexmark cartridges at issue in the United States, and has imported those it acquired abroad. In each case, it has acted without affirmative authorization from Lexmark and, for the restricted cartridges, in violation of the express denial of authorization to engage in resale and reuse. Impression's actions infringe under 35 U.S.C. § 271—unless the fact that Lexmark initially sold the cartridges constitutes the grant of authority that makes Impression's later resale and importation non-infringing under the doctrine of exhaustion. Whether Lexmark's initial sales have that effect raises two questions—one regarding the single-use/no-resale restricted sales (wherever they occur), the other regarding the initial foreign sales of all cartridges, whether restricted or not.

A

Lexmark offers buyers a choice. A buyer may purchase a “Regular Cartridge” at full price, in which case the buyer is not subject to any sale terms restricting reuse or resale of the cartridge. Alternatively, a buyer may purchase a “Return Program Cartridge” at a discount of roughly 20 percent, subject to a single-use/no-resale restriction: the buyer may not reuse the cartridge after the toner runs out and may not transfer it to anyone but Lexmark once it is used, i.e., the buyer must “return” the cartridge “only” to Lexmark. J.A. 2559. Lexmark and Impression stipulated in this case that the reduced price “reflects the value of the property interest and use rights conveyed to the purchaser under the express terms of the conditional sale contract and conditional single-use license conferred by Lexmark.” Id. The stipulation adds that “Lexmark has an express and enforceable contractual agreement with each of its end-user customers.” J.A. 2562. And it is undisputed that all end users receive adequate notice of the restriction supporting the discounted price before they make their purchases.

The distinctness of the options for buyers, which produce different revenues for Lexmark, is not just a matter of different terms of sale. It also is reflected in a microchip in the cartridges that, among other things, communicates with the printer. For a Return Program cartridge, the chip and printer, by monitoring toner levels, prevent use of a refilled cartridge. For a Regular cartridge, the toner can be replenished and the cartridge reused. J.A. 2559–60. “To circumvent this technological measure,” however, “third parties have ‘hacked’ Lexmark's microchips and created their own ‘unauthorized replacement’ microchips” that, when installed in a Return Program cartridge, fool the printer into allowing reuse of that cartridge. J.A. 2560. It is undisputed that various companies gather spent cartridges, replace the microchips, refill and “remanufacture” the cartridges, and sell them to resellers like Impression for marketing to consumers for use with Lexmark printers.

Lexmark sells its cartridges in two channels of distribution. It sells directly to end users, and it sells to “re-sellers” (including wholesalers, dealers, and distributors). Lexmark offers the options of Return Program and Regular cartridges in both channels; the resellers pay less for the Return Program cartridges; and the single-use/no-resale restriction applies to the resale by resellers. J.A. 2564. There is no dispute about the adequacy of notice to resellers as well as end users or the binding nature of the Lexmark-reseller agreements. J.A. 2562–64. When Lexmark sells its cartridges to end users, that sale is the first sale; when it sells to resellers, that sale is the first sale. When a reseller subsequently sells to end users, that sale is not the first sale.

B

Lexmark sued Impression, among other companies, for infringement under 35 U.S.C. § 271. It alleged that Impression acquires spent cartridges, including some Return Program cartridges that have been altered by chip replacement and toner refilling, then sells them in the United States and, for the foreign-bought ones, imports them into the United States. For a large number of patents directly covering the cartridges, Lexmark alleged direct infringement under § 271(a). For a few patents that only the end user directly infringes, Lexmark alleged that Impression is liable for contributory infringement under § 271(c). The operative complaint states the infringement allegations in a single count (Count I), covering past and continuing activity.

More specifically, the infringement allegations are limited to two groups of cartridges. One group consists of Return Program cartridges that Lexmark sold in the United States under the restriction denying authority for resale and reuse. As it later made clear, Lexmark did not allege infringement by Impression's actions involving Regular cartridges Lexmark had first sold domestically. J.A. 1895–97, 2557. The second group consists of all cartridges that Lexmark sold abroad, including Return Program and Regular cartridges. It is undisputed that Lexmark never granted anyone permission to import those cartridges into, or sell or use them in, the United States.

C

The litigation progressed to the point at which no defendant remained except Impression, and only the single count of infringement remained against Impression. Impression came to agree that the patents covered the cartridges it was importing and selling, and it did not dispute the validity or enforceability of the patents. It contested liability for infringement on just one ground, namely, that Lexmark had exhausted its U.S. patent rights in the cartridges by its initial sales of them.

Three defining aspects of Impression's contention to the district court, and presentation to us, are worth noting here, because they narrow our focus. First, we discuss only Lexmark's sales to end users (and the resales and reuses deriving from those sales), because neither party has made an argument for distinguishing Lexmark's sales to resellers. Second, we take as a premise that both the first purchaser and Impression as a re-purchaser had adequate notice of the single-use/no-resale restriction before they made their purchases; the adequacy of that notice is unchallenged. Thus, we do not have before us the questions that would arise, whether under principles governing bona fide purchasers or otherwise, if a downstream re-purchaser acquired a patented article with less than actual knowledge of such a restriction. Third, Impression has not contended that the particular restriction at issue gives rise to a patent-misuse defense, constitutes an antitrust violation, or exceeds the scope of the Patent Act's express grant of exclusive rights over patented articles, 35 U.S.C. §§ 154, 271. Rather, Impression contends that, although there is no other illegality or breach of statutory limits identified, the single-use/no-resale restriction is to be disregarded for exhaustion purposes. According to Impression, it has the authority to resell despite the known denial of such authority by Lexmark for the Return Program cartridges.

Impression presented its exhaustion defense by filing motions to dismiss the infringement count, one motion for each of the two groups of cartridges at issue. For each motion, Impression did not contest that, under this court's governing law, its exhaustion defense must fail: Mallinckrodt for the cartridges initially sold in the United States, Jazz Photo for the cartridges initially sold abroad. But it argued that the Supreme Court's more recent decisions had made Mallinckrodt and Jazz Photo no longer good law. In a pair of opinions issued the same day, the district court agreed with Impression about Mallinckrodt but disagreed about Jazz Photo.

1

The district court granted Impression's motion to dismiss Lexmark's claim of infringement involving the single-use cartridges Lexmark had first sold in the United States. Lexmark Int'l, Inc. v. Ink Techs. Printer Supplies, LLC, No. 1:10–CV–564, 2014 WL 1276133 (S.D.Ohio Mar. 27, 2014) (Domestic Sale Opinion ), modified at J.A. 34–35 based on a joint stipulation of the parties, J.A. 2554–66. Like Impression, the court recognized that there is no exhaustion here under this court's decision in Mallinckrodt, which rejected an exhaustion defense in circumstances similar to those presented here—namely, where a patentee sold a patented article subject to an otherwise-unobjectionable single-use restriction. Id. at *4, *6. And the court recognized this court's post-Mallinckrodt decisions as reiterating that a “ ‘conditional sale’ ” of that type does not cause exhaustion of the patentee's preserved rights in the article. Id. at *4, *6 (quoting Princo Corp. v. Int'l Trade Comm'n, 616 F.3d 1318, 1328 (Fed.Cir.2010) (en banc)).

In nevertheless finding exhaustion here, the district court examined a number of Supreme Court decisions on patent exhaustion. It noted the Court's explanation in Bloomer v. McQuewan, 55 U.S. (14 How.) 539, 549–50, 14 L.Ed. 532 (1853), that a patentee's grant of a license to another to make and sell a patented article is not the same thing as the patentee's sale of the article itself. Domestic Sale Opinion, 2014 WL 1276133, at *3. It noted, too, the Court's rejection of an exhaustion defense in General Talking Pictures Corp. v. Western Electric Co., 304 U.S. 175, 58 S.Ct. 849, 82 L.Ed. 1273, opinion on rehearing at 305 U.S. 124, 59 S.Ct. 116, 83 L.Ed. 81 (1938), which held that a buyer of a patented article infringed when it used the article in a way forbidden by a known use restriction, having bought the article from a manufacturer licensed by the patentee to make and sell the article only to buyers who complied with the use restriction. Domestic Sale Opinion, 2014 WL 1276133, at *3. The district court also noted that the Supreme Court in Quanta found exhaustion where “the Supreme Court determined that the agreements [at issue] broadly authorized Intel [the seller] to sell the licensed products without restrictions or conditions.” Id. at *5 (emphasis added).

Despite that recognition of what Quanta involved, the district court concluded “that Quanta overruled Mallinckrodt sub silentio.” Id. at *5, *6. Although Return Program cartridges were sold under post-sale restrictions on reuse and resale, the district court held that “those post-sale use restrictions do not prevent patent rights from being exhausted given that the initial sales were authorized and unrestricted.” Id. The court thus dismissed the infringement claim regarding Impression's actions involving Return Program cartridges Lexmark had sold in the United States. Id. at *7.

2

As to cartridges Lexmark had sold abroad, the court held that exhaustion did not apply, i.e., did not render Impression's imports and domestic resales of those cartridges non-infringing. Lexmark Int'l, Inc. v. Ink Techs. Printer Supplies, LLC, 9 F.Supp.3d 830 (S.D.Ohio 2014) (Foreign Sale Opinion). The court recognized, and Impression did not dispute, that “under Jazz Photo, an initial authorized sale of a patented product outside of the United States would not exhaust the patent rights of the patent holder.” Id. at 833. It then examined the Supreme Court's decision in Kirtsaeng and rejected Impression's contention that Kirtsaeng “overturns the Federal Circuit's decision in Jazz Photo, 264 F.3d 1094, such that Lexmark's patent rights were exhausted upon the first authorized sale abroad.” Foreign Sale Opinion, 9 F.Supp.3d at 834 (footnote omitted).

The court stated that “[t]he Supreme Court's decision was rooted in interpretation of a statutory provision of copyright law,” namely, 17 U.S.C. § 109(a). Foreign Sale Opinion, 9 F.Supp.3d at 833. The district court noted the absence from Kirtsaeng of any discussion of exhaustion in the patent field and the Supreme Court's longstanding recognition that copyright law and patent law are not interchangeable. Id. at 835 (citing Bobbs–Merrill Co. v. Straus, 210 U.S. 339, 346, 28 S.Ct. 722, 52 L.Ed. 1086 (1908)). It added that Kirtsaeng “is rooted in statutory and legislative interpretation of section 109(a) of the Copyright Act,” but “[n]oticeably absent from patent law is a codification of the exhaustion doctrine,” concluding: “the core statutory text that weighed in favor of a non-geographical interpretation is non-existent in the context of patent law.” Id. The Patent Act, the court concluded, calls for its own analysis of “context, history and practical considerations.” Id. at 836.

For those reasons, while recognizing that this court might reconsider Jazz Photo in light of Kirtsaeng, the district court held that Jazz Photo remains good law. Id. at 837–38. The court therefore denied Impression's motion to dismiss Lexmark's claim of infringement involving the Foreign–Sold Cartridges. Id. at 838.

3

Soon thereafter, with the parties' agreement, the court entered a “Stipulated Final Judgment.” J.A. 1. The judgment was (a) for Impression (i.e., Impression does not infringe) as to the Return Program cartridges whose precursors Lexmark had sold in the United States and (b) for Lexmark (i.e., Impression infringes) as to cartridges whose precursors Lexmark had initially sold abroad. Id. The parties agree that the judgment is final under 28 U.S.C. § 1295(a)(1), even as to cartridges found to infringe.

In agreeing to the final judgment of infringement as to the foreign-sold cartridges, which remained an open issue after the Rule 12(b)(6) rulings, Impression reasonably construed the district court's Jazz Photo ruling to foreclose its exhaustion defense, even though all the district court had done was to deny Impression's request for judgment in its favor based on that defense. In particular, the district court's rationale as to the unavailability of exhaustion did not depend on the facts in the record that Lexmark identifies as suggesting the “regional” character of its foreign-sold cartridges, facts that therefore went unexplored in the district court. And, notably, when Impression agreed to a judgment of infringement as to foreign-sold cartridges, it did not preserve an implied-license defense, even though the Supreme Court made clear in Quanta the distinctness of implied-license and exhaustion defenses. 553 U.S. at 637, 128 S.Ct. 2109.

D

Impression appealed and Lexmark cross-appealed. This court has jurisdiction under 28 U.S.C. § 1295(a)(1). The parties submitted briefs and presented oral argument to a panel of this court, focused on whether Quanta had stripped Mallinckrodt of its controlling force and whether Kirtsaeng had stripped Jazz Photo of its controlling force.

Shortly after oral argument, this court sua sponte took the case en banc. Lexmark Int'l, Inc. v. Impression Prods., Inc., 785 F.3d 565 (Fed.Cir.2015). We directed the parties to address the following issues:

(a) The case involves certain sales, made abroad, of articles patented in the United States. In light of Kirtsaeng v. John Wiley & Sons, Inc., 133 S.Ct. 1351 (2013), should this court overrule Jazz Photo Corp. v. International Trade

Commission, 264 F.3d 1094 (Fed.Cir.2001), to the extent it ruled that a sale of a patented item outside the United States never gives rise to United States patent exhaustion[?]

(b) The case involves (i) sales of patented articles to end users under a restriction that they use the articles once and then return them and (ii) sales of the same patented articles to resellers under a restriction that resales take place under the single-use-and-return restriction. Do any of those sales give rise to patent exhaustion? In light of Quanta Computer, Inc. v. LG Electronics, Inc., 553 U.S. 617, 128 S.Ct. 2109 (2008), should this court overrule Mallinckrodt, Inc. v. Medipart, Inc., 976 F.2d 700 (Fed.Cir.1992), to the extent it ruled that a sale of a patented article, when the sale is made under a restriction that is otherwise lawful and within the scope of the patent grant, does not give rise to patent exhaustion?

Id. at 566.

Discussion

I

The Patent Act's language defines the framework within which the two exhaustion questions arise. In 1952, based on pre-existing uncodified understandings, Congress set forth a statutory prescription of what constitutes patent “infringement.” See Aro Mfg. Co. v. Convertible Top Replacement Co., 377 U.S. 476, 483–84, 84 S.Ct. 1526, 12 L.Ed.2d 457 (1964) (Aro II); Aro I, 365 U.S. at 341–42 & n. 8, 81 S.Ct. 599. In its current form, which includes a bar on importation and offers to sell added by a 1994 enactment, § 271(a) states that, unless another provision of the Act provides otherwise, whoever “without authority” during the term of a patent commits certain acts—“makes, uses, offers to sell, or sells any patented invention, within the United States or imports into the United States any patented invention”—“infringes the patent.” 35 U.S.C. § 271(a).

Section 271(a) connects “make,” “sell,” “use,” and the other terms with the disjunctive “or,” as does the related provision granting the patentee various rights to exclude others from the same activities, id. § 154(a). Congress has thus prescribed that whoever, “without authority,” does any one of the listed acts—“the making, using, offering to sell, selling, or importing of a patented invention,” Global–Tech Appliances, Inc. v. SEB S.A., 563 U.S. 754, 131 S.Ct. 2060, 2065, 179 L.Ed.2d 1167 (2011) (emphasis added)—is an infringer. See 5 Donald S. Chisum, Chisum on Patents § 16.01 (2015) (“The exclusive rights are disjunctive: one may infringe by (1) making without selling or using, (2) using without making or selling or (3) selling without making or using.”) (footnote omitted); William C. Robinson, The Law of Patents §§ 903–906 (1890). The government observes: “Nothing in the text of the Patent Act expressly prevents a patentee from demanding compensation from each downstream user or reseller of an article embodying his invention.” U.S. Br. 5.

Section 271(a)'s language embodies an understanding of “infringement” that was long recognized even before Congress enacted § 271 as part of the 1952 recodification of the patent laws. The pre–1952 statute included a right-to-exclude provision comparable to § 154, which, in language that varied over time, gave the patentee a right to exclude others (not a right to practice the invention). See 35 U.S.C. § 40 (1946); Rev. Stat. § 4884; Bauer & Cie. v. O'Donnell, 229 U.S. 1, 9–10, 33 S.Ct. 616, 57 L.Ed. 1041 (1913); 5 Chisum § 16.02[1]. But while the pre–1952 statute provided for actions for “infringement,” e.g., 35 U.S.C. §§ 67, 70 (1948); Rev. Stat. §§ 4919, 4921, there was no provision prescribing what constitutes infringement. Nevertheless, the courts consistently understood infringement to mean what § 271 came to say—committing the identified acts without authority (synonymously, without consent or permission): “The infringement of a patent, being the invasion of this exclusive right, therefore consists in the manufacture, use, or sale of the invention protected by the patent within the area and time described in the patent, by any person not duly authorized to do so by the patentee.” Robinson, § 890, at 43–44 (emphasis added); see 3 Anthony William Deller, Walker on Patents § 450, at 1681 (1937) (“An infringement is the unauthorized making or using or selling of the patented invention.”). Thus, the 1952 Act's “without authority” language simply codifies an authority requirement long recognized to be the meaning of “infringement” of the enumerated rights to exclude. See Warner–Jenkinson Co. v. Hilton Davis Chem. Co., 520 U.S. 17, 26–27, 117 S.Ct. 1040, 137 L.Ed.2d 146 (1997) (§ 271(a) left direct-infringement law intact); Aro II, 377 U.S. at 483, 84 S.Ct. 1526; Aro I, 365 U.S. at 342, 81 S.Ct. 599; Giles S. Rich, Infringement Under Section 271 of the Patent Act of 1952, 21 Geo. Wash. L.Rev. 521, 537 (1953).

The requirement of “authority” in order to avoid infringement, in its natural meaning, refers to a grant of permission. Logically, permission might come from Congress, whether outside the Patent Act or within the Patent Act itself, as reflected in § 271(a)'s “[e]xcept as otherwise provided in [the Patent Act]” language, which explicitly bows to other contrary sections of the Patent Act. But it is undisputed that no other statutory provision applies in this case. See U.S. Br. 5; compare 35 U.S.C. § 262 (each joint owner of a patent may engage in making, using, selling, offering to sell, and importing without authority from other owners). Nothing in the Act supersedes the § 271 requirement of authority from the patentee before a person in Impression's position may engage in the itemized acts without infringing.

In this respect, the Patent Act differs from the Copyright Act. In the copyright statute, Congress included a provision giving a right of sale to certain article owners, 17 U.S.C. § 109(a), and made the infringement, importation, and exclusive-rights provisions all subservient to that express guarantee. The Patent Act does not contain a congressionally prescribed exhaustion rule, let alone a provision that makes the express definition of infringement and rights to exclude (both of which now encompass importation) subservient to any congressionally expressed exhaustion rule.

In the Patent Act, then, as relevant here, it is a conferral of “authority” by the patentee that is needed in order for the actions listed in § 271(a) not to constitute infringement. As the government says, noting the parallelism of § 271(a) and the § 154(a) grant of rights to exclude, what § 271(a) means is that “[w]hoever does any of these acts ‘without authority’ from the patentee infringes the patent.” U.S. Br. 1 (emphasis added). In brief: § 271(a) by its terms requires that whoever engages in the enumerated acts receive permission from the patentee (directly or indirectly) for the acts being performed, which otherwise are infringing; and nothing in § 271(a) constrains the patentee's choices about whom to grant the required authority, if anyone, or about which acts (of manufacture, use, sale, etc.) to authorize, if any.

Congress defines the existence and scope of patent rights. See, e.g., Octane Fitness, LLC v. ICON Health & Fitness, Inc., ––– U.S. ––––, 134 S.Ct. 1749, 1755–56, 188 L.Ed.2d 816 (2014); Crown Die & Tool Co., 261 U.S. at 40, 43 S.Ct. 254; Continental Paper Bag Co. v. Eastern Paper Bag Co., 210 U.S. 405, 423, 28 S.Ct. 748, 52 L.Ed. 1122 (1908). Unless Congress has directed the courts to fashion governing rules in a particular statutory context (as in, e.g., the Sherman Act), “once Congress addresses a subject, even a subject previously governed by federal common law, the justification for lawmaking by the federal courts is greatly diminished. Thereafter, the task of the federal courts is to interpret and apply statutory law, not to create common law.” Northwest Airlines, Inc. v. Transp. Workers Union of Am., 451 U.S. 77, 95 n. 34, 101 S.Ct. 1571, 67 L.Ed.2d 750 (1981); see City of Milwaukee v. Illinois, 451 U.S. 304, 315, 101 S.Ct. 1784, 68 L.Ed.2d 114 (1981) (“Our commitment to the separation of powers is too fundamental to continue to rely on federal common law by judicially decreeing what accords with common sense and the public weal when Congress has addressed the problem.”) (internal quotation marks omitted); Am. Elec. Power Co. v. Connecticut, 564 U.S. 410, 131 S.Ct. 2527, 2537, 180 L.Ed.2d 435 (2011).

If ordinary congressional supremacy is to be respected, exhaustion doctrine in the Patent Act must be understood as an interpretation of § 271(a)'s “without authority” language. And so it has been understood: some sales confer authority on the purchaser to take certain actions—such as selling or using the purchased article in the United States or importing it into the United States—that would otherwise be infringing acts. See 5 Chisum § 16.03 [2][a], at 16–362.8; U.S. Br. 1 (tying exhaustion to “authority” language of § 271(a)). We decide here (a) whether a sale, even though accompanied by a clearly communicated and otherwise-lawful denial of such authority, nonetheless has the legal effect of conferring such authority and (b) whether a foreign sale has the legal effect of conferring such authority where (as we must assume at present in this case) neither a grant nor a reservation of § 271(a) rights was communicated to the purchaser before the foreign sale.

II

The Mallinckrodt issue has been framed for us in clear terms. Suppose that Lexmark had granted another firm a nonexclusive license to make and sell Return Program cartridges. It is undisputed and clear under Supreme Court precedent—most prominently, the 1938 decision in General Talking Pictures—that Lexmark would not have exhausted its patent rights in those cartridges, upon the manufacturing licensee's sale (the first sale), if a buyer with knowledge of the restrictions resold or reused them in violation of the restrictions. Impression and the government contend that a different result is required—that Lexmark automatically lost its patent rights—simply because Lexmark sold the Return Program cartridges itself, subject to the same communicated restriction, rather than having left the manufacture and sale to others under license. See U.S. Br. 7, 8, 10, 11 (case turns on distinction between patentee sale and non-patentee licensee sale). (Impression has left the en banc briefing on this issue largely to the government.)

We conclude otherwise, as we did in Mallinckrodt and subsequent decisions. A sale made under a clearly communicated, otherwise-lawful restriction as to post-sale use or resale does not confer on the buyer and a subsequent purchaser the “authority” to engage in the use or resale that the restriction precludes. And there is no sound reason, and no Supreme Court precedent, requiring a distinction that gives less control to a practicing-entity patentee that makes and sells its own product than to a non-practicing-entity patentee that licenses others to make and sell the product.

A

Mallinckrodt involved a patentee's sale of its medical device to hospitals, subject to a “single use only” restriction. The device consisted of a nebulizer and associated components for delivering to a patient, for diagnosis or treatment of lung diseases, a mist of radioactive or therapeutic material. It also trapped radioactive or toxic material when the patient exhaled. Mallinckrodt sold it under the single-use condition and with instructions for post-use disposal in a lead-shielded container. But some hospital purchasers instead sent used devices to Medipart for reconditioning and for replacement of certain components. When Mallinckrodt sued Medipart for direct and indirect infringement by virtue of the reuse in violation of the single-use restriction, the district court granted Medipart summary judgment of non-infringement, concluding that Mallinckrodt's sale of the devices exhausted its ability to assert its patent rights in the units sold. 976 F.2d at 701–02.

This court reversed. Id. at 709. The court stated its ruling: “The restriction here at issue does not per se violate the doctrine of patent misuse or the antitrust law. Use in violation of a valid restriction may be remedied under the patent law, provided that no other law prevents enforcement of the patent.” Id. at 701.

In explanation, the court observed that the patent grant of § 154 is a “right to exclude,” which “may be waived in whole or in part,” “subject to patent, contract, antitrust, and any other applicable law, as well as equitable considerations such as are reflected in the law of patent misuse.” Mallinckrodt, 976 F.2d at 703. It noted that the Supreme Court had held that two particular types of restrictions in sales exceeded the legitimate scope of patent rights, so that the patentee did not retain its patent rights against a buyer's sale or use of a patented article in violation of those particular conditions: resale-price-maintenance conditions, Bauer, 229 U.S. 1, 33 S.Ct. 616; Straus v. Victor Talking Machine Co., 243 U.S. 490, 37 S.Ct. 412, 61 L.Ed. 866 (1917); Boston Store of Chicago v. American Graphophone Co., 246 U.S. 8, 38 S.Ct. 257, 62 L.Ed. 551 (1918), and tying arrangements requiring use of non-patented articles with the patented one, Motion Picture Patents Co. v. Universal Film Mfg. Co., 243 U.S. 502, 37 S.Ct. 416, 61 L.Ed. 871 (1917). Mallinckrodt, 976 F.2d at 704. The Mallinckrodt court explained that those cases “did not hold, and it did not follow, that all restrictions accompanying the sale of patented goods were deemed illegal.” Id.

The court then described the key Supreme Court precedent, General Talking Pictures. As the Mallinckrodt court observed, in General Talking Pictures “the patentee had authorized the licensee to make and sell amplifiers embodying the patented invention for a specified use (home radios),” and “[t]he defendant had purchased the patented amplifier from the manufacturing licensee, with knowledge of the patentee's restriction on use,” but used it contrary to the restriction. Id. at 705. The Supreme Court held that the defendant was liable for infringement; it “observed that a restrictive license to a particular use was permissible, and treated the purchaser's unauthorized use as infringement of the patent.” Id. This court noted that the Supreme Court in General Talking Pictures “did not decide the situation where the patentee was the manufacturer” and seller. Id. This court then held that to distinguish the patentee sale (at issue in Mallinckrodt ) from the licensee sale (at issue in General Talking Pictures ) would be to make “formalistic distinctions of no economic consequence.” Id.

Finally, the court described “a group of cases in which the Supreme Court considered and affirmed the basic principles that unconditional sale of a patented device exhausts the patentee's right to control the purchaser's use of the device; and that the sale of patented goods, like other goods, can be conditioned.” Id. at 706. This court noted that, insofar as several cases ruling against patentees' claims discussed or involved sales, the sales were not made under restrictions as to use. Id. at 707 (discussing Bloomer, 55 U.S. (14 How.) 539; Adams v. Burke, 84 U.S. (17 Wall.) 453, 21 L.Ed. 700 (1873); Keeler v. Standard Folding Bed Co., 157 U.S. 659, 15 S.Ct. 738, 39 L.Ed. 848 (1895)). The court also noted the statement in Mitchell v. Hawley that “[s]ales of the kind may be made by the patentee with or without conditions,” 83 U.S. (16 Wall.) 544, 548, 21 L.Ed. 322 (1873), and the holding of Mitchell that a purchaser of a patented machine “licensed for use only during the original term of the patent” was liable for infringement for using the machine past that term when the patent term was extended. 976 F.2d at 707.

The court in Mallinckrodt concluded that “[u]nless the condition violates some other law or policy (in the patent field, notably the misuse or antitrust law, e.g., United States v. Univis Lens Co., 316 U.S. 241, 62 S.Ct. 1088, 86 L.Ed. 1408 (1942)), private parties retain the freedom to contract concerning conditions of sale.” 976 F.2d at 708. Thus, unless a sale restriction is improper under some other body of law, whether within the Patent Act or outside it, a patentee's own sale of its patented article subject to a clearly communicated restriction does not confer authority to sell or use the article in violation of that restriction, i.e., does not exhaust the patentee's § 271 rights against such conduct involving that article. Since Mallinckrodt, we have followed that principle. B. Braun Med., Inc. v. Abbott Labs., 124 F.3d 1419, 1426 (Fed.Cir.1997); Princo Corp. v. Int'l Trade Comm'n, 616 F.3d 1318, 1328 (Fed.Cir.2010) (en banc).

B

The district court concluded in this case that Quanta overturned the Mallinckrodt rule as to a patentee's sale of a patented article subject to a clearly communicated single-use/no-resale restriction. But no issue as to such a sale was presented for decision or decided in Quanta. And the Supreme Court in Quanta did not address the distinction between patentee sales and licensee sales on which the argument for overturning Mallinckrodt rests.

Quanta did not involve a patentee's sale at all, let alone one subject to a restriction or, more particularly, a single-use/no-resale restriction. Quanta involved a sale made (to computer maker Quanta) not by the patentee (LGE) but by a manufacturing licensee (chip maker Intel), which the patentee had authorized to make and sell the articles at issue (chips for installation in computers that would then be covered by LGE's patents). 553 U.S. at 623–25, 128 S.Ct. 2109. And the patentee's authorization to the licensee to make (the first) sales was not subject to any conditions, much less conditions to be embodied in those sales. Id. at 636–38, 128 S.Ct. 2109. While Intel had certain other contractual obligations to LGE regarding notice to Intel's purchasers, neither party contended that Intel breached those obligations, and in any event, the Court repeatedly stated that the relevant LGE–Intel contract gave Intel an unrestricted authorization to sell the articles. Id. at 636–37, 128 S.Ct. 2109 (“Intel's authority to sell its products embodying the LGE Patents was not conditioned on the notice or on Quanta's decision to abide by LGE's directions in that notice.”); id. at 637, 128 S.Ct. 2109 (“The License Agreement authorized Intel to sell products that practiced the LGE Patents. No conditions limited Intel's authority to sell products substantially embodying the patents.”); id. at 638, 128 S.Ct. 2109 (“Nothing in the License Agreement limited Intel's ability to sell its products practicing the LGE Patents.”).

In short, Quanta did not involve the issue presented here. The facts defining the issues for decision, and the issues decided, were at least two steps removed from the present case. There were no patentee sales, and there were no restrictions on the sales made by the licensee.

The two main issues decided by the Court in Quanta have no bearing on the issue of restricted sales by a patentee. The Court decided that exhaustion applies to method claims. Id. at 628–30, 128 S.Ct. 2109. And the Court decided “the extent to which a product must embody a patent in order to trigger exhaustion.” Id. at 630, 128 S.Ct. 2109; see id. at 630–35, 128 S.Ct. 2109.

Only the third issue addressed by the Court in Quanta concerns restrictions on sales—though not patentees' sales—and the Court's discussion of that issue does not undermine Mallinckrodt's ruling that a patentee can preserve its patent rights through restrictions on its sales. As just described, when LGE invoked precedent such as General Talking Pictures that make clear that patentees are able to preserve their patent rights through restrictions on the sales they authorizes licensees to make, Quanta, 553 U.S. at 636, 128 S.Ct. 2109, the Court's response was to conclude that there simply were no such restrictions on LGE's grant to Intel of the authority to sell. Id. at 636–38, 128 S.Ct. 2109. The Court thus had prominently in view the principle that, through at least one path, a patentee can reserve its patent rights through sale restrictions. The Court found the principle inapplicable to the case before it because of the absence of any restriction, and the Court said nothing to cast doubt on the principle. Indeed, the Court indirectly underscored the principle when it quoted Motion Picture Patents as stating “the rule” of exhaustion in terms expressly based on an “ ‘unconditional sale.’ ” Quanta, 553 U.S. at 626, 128 S.Ct. 2109 (quoting Motion Picture Patents, 243 U.S. at 516, 37 S.Ct. 416). And the Court did not consider whether or decide that the principle was limited to contracted-out, as distinguished from in-house, manufacturing and sales, or even recognize and discuss such a distinction.

Inferring disapproval of Mallinckrodt by the Supreme Court in Quanta is unwarranted for another reason. The decision of this court under review in Quanta relied centrally on Mallinckrodt and its successor case, B. Braun Medical. See LG Elecs., Inc. v. Bizcom Elecs., Inc., 453 F.3d 1364, 1370 (Fed.Cir.2006). In the Supreme Court, the government prominently featured an argument that Mallinckrodt was incorrect and should be repudiated, U.S. Amicus Brief at 18–24, Quanta (No. 06–937), 2007 WL 3353102, and Quanta presented similar criticisms of Mallinckrodt, Brief for Petitioner at 13, 30–33, Quanta (No. 06–937), 2007 WL 3276505. Yet the Supreme Court said nothing about Mallinckrodt or B. Braun Medical. The evident explanation is that, at a minimum, no question was presented for decision—or was being decided—as to the effect a restriction on the first sale, whether made by a patentee or by a manufacturing licensee, would have on preservation of § 271 rights.

C

For the foregoing reasons, the challenge to Mallinckrodt in the present case cannot rest on what the Supreme Court in Quanta ruled about the issues it said it was addressing or the facts and issues presented for decision. The challenge asserts that any post-sale restriction in a patentee's own sale fails, as a matter of law, to preserve the patentee's § 271 rights against unauthorized sales or uses. The argument rests ultimately on language the Court used in Quanta in introducing the exhaustion doctrine before defining the specific issues for decision—as it has done elsewhere. The Court in Quanta said: “The longstanding doctrine of patent exhaustion provides that the initial authorized sale of a patented item terminates all patent rights to that item.” 553 U.S. at 625, 128 S.Ct. 2109. More recently, the Court used similar “authorized sale” language in introducing the exhaustion doctrine in Bowman v. Monsanto Co., which held that a patentee, by selling patented seeds, does not lose its § 271 right to prevent the buyer from making new seeds. The Court said: “Under the doctrine of patent exhaustion, the authorized sale of a patented article gives the purchaser, or any subsequent owner, a right to use or resell that article.” ––– U.S. ––––, 133 S.Ct. 1761, 1764, 185 L.Ed.2d 931 (2013). See also, e.g., Univis, 316 U.S. at 249, 62 S.Ct. 1088 (“authorized sale”); Dissent at 774–76.

The challenge to Mallinckrodt asserts that any sale of a patented article by a patentee, even when the rights granted are expressly restricted, is automatically an “authorized sale,” causing the patentee to lose all § 271 rights in the item sold. That consequence follows, the argument goes, no matter how clearly the patentee states an otherwise-lawful restriction on what authority is being conferred and what authority is being withheld. In this view, exhaustion law embodies a sharp distinction between a sale by a patentee (for which restrictions are to be disregarded) and a sale made by another person authorized by the patentee to sell, i.e., a licensee as in General Talking Pictures (for which a patentee may preserve its § 271 rights by restricting the licensee's authorized sales).

That is an extraordinary doctrinal consequence to find established by the Supreme Court's use of the phrase “authorized sale.” No one suggests that Bowman (concerning new articles) intended such consequences. And there are good reasons not to find any such implied meaning in Quanta either.

1

Most obviously, as discussed above, the Court was not addressing the patentee-sale/licensee-sale distinction. Among other things, the Court did not consider the issues presented by making such a distinction, such as where the line would sensibly be drawn along the spectrum that includes original patentees, assignees, exclusive licensees, and non-exclusive licensees. Such distinctions matter for determining who may bring infringement suits, but they can involve detailed inquiries into the contractual relationships between an original patent owner and others (here, a first seller), based on information a buyer may have no ability or reason to acquire. See, e.g., Independent Wireless Telegraph Co. v. Radio Corp. of Am., 269 U.S. 459, 464–69, 46 S.Ct. 166, 70 L.Ed. 357 (1926); Crown Die & Tool, 261 U.S. at 40–41, 43 S.Ct. 254; E.W. Bliss Co. v. United States, 253 U.S. 187, 192, 40 S.Ct. 455, 64 L.Ed. 852 (1920); Pope Mfg. Co. v. Gormully & Jeffery Mfg. Co., 144 U.S. 248, 12 S.Ct. 641, 36 L.Ed. 423 (1892); Waterman v. Mackenzie, 138 U.S. 252, 255–56, 11 S.Ct. 334, 34 L.Ed. 923 (1891); 8 Chisum § 21.03. Nothing in Quanta suggests that the Court either considered such issues or intended to build such inquiries into the exhaustion doctrine by making the distinction the government now urges.

2

One cannot infer the contrary from the immediate context of the “authorized sale” phrase, i.e., from the several decisions that Quanta briefly describes in the paragraph it introduces with the “authorized sale” shorthand. 553 U.S. at 625, 128 S.Ct. 2109. Those decisions did not involve restricted patentee sales of patented articles. And the Quanta Court, in describing those cases, said nothing to indicate adoption of a patentee-sale/licensee-sale distinction.

Thus, Bloomer v. McQuewan identifies no sale of a patented article as involved in the case. During an initial patent term, a license granted to the defendants (via Collins and Smith, then Barnet, then Warner and McQuewan) the right to make patented machines and use them “without any limitation as to the time for which they were to be used.” 55 U.S. (14 How.) at 553; id. at 540–41, 548. When the defendants made and used such machines, and the patent term was later extended, the Supreme Court held that the unrestricted right to use did not end when the earlier patent term ended, because the right to use did not come from the patent statute, which grants only rights to exclude, not rights to practice. Id. at 549–50. The Court discussed purchased articles, but with no restrictions at issue, it did not decide the effect that any restrictions would have. Id.

Bloomer v. Millinger, 68 U.S. (1 Wall.) 340, 17 L.Ed. 581 (1864), is similar in its facts to McQuewan. The Court held that Millinger, who “constructed the machines and put them in operation under the authority of the patentee or his assigns” during a first term extension, was not subject to an infringement action for continued use during a second term extension. Id. at 349, 350. The Court made no distinction between construction and purchase, or between purchases “from the patentee ... or from any other person by him authorized to” make the sale, on the issues addressed. Id. at 350, 351. In referring to a constructor or purchaser of a patented machine having “also acquired the right to use and operate it during the lifetime of the patent,” id. at 350 (emphasis added), the Court implicitly recognized that a purchaser might not acquire a full right to use an acquired article.

And in Adams v. Burke, the Boston regional assignee (Lockhart & Seelye) sold patented coffin lids to Burke, “without condition or restriction.” 84 U.S. (17 Wall.) at 455 (emphasis added). When Burke used the lids outside the Boston territory, for burials within a different territory where Adams was the regional assignee, Adams sued. The Court held that, for such an unrestricted sale, “there is no restriction on their use to be implied for the benefit of the patentee or his assignees or licensees.” Id. at 457 (bold italics added).

When the Supreme Court in Quanta moved beyond the Bloomer cases and Adams, it likewise did not advance a patentee-sale/licensee-sale distinction. 553 U.S. at 625–26, 128 S.Ct. 2109. The Court's next paragraph recounts the developments of the 1910s: The Court initially adopted a broad greater-includes-the-lesser approach allowing preservation of patent rights through even otherwise-unlawful restrictions, such as tie-ins, Henry v. A.B. Dick Co., 224 U.S. 1, 32 S.Ct. 364, 56 L.Ed. 645 (1912); but the Court quickly rejected that broad principle and its application to resale price maintenance, Bauer, 229 U.S. 1, 33 S.Ct. 616 (1913), and then to tie-ins, Motion Picture Patents, 243 U.S. 502, 37 S.Ct. 416 (1917), the latter expressly overruling A.B. Dick. The Court in Quanta, summarizing that history, said nothing about the patentee-sale/licensee-sale distinction; and it recognized that the “rule” that emerged, as quoted from Motion Picture Patents, was one based on “ ‘a single, unconditional sale.’ ” 553 U.S. at 626, 128 S.Ct. 2109 (emphasis added).

The Court in Quanta then proceeded to a longer discussion of Univis. That discussion made no patentee-sale/licensee-sale distinction. Rather, it recounted Univis's application of exhaustion to sales of articles that may not be strictly covered by the patent but that sufficiently embody the patent. 553 U.S. at 627–28, 128 S.Ct. 2109.

In no part of the Court's discussion did the Court consider which cases involved “patentee” sales and which “licensee” sales. Indeed, it is not always easy to tell where the facts of each case fall on the spectrum from original patentee through non-exclusive licensee. See, e.g., Bauer, 229 U.S. at 8, 33 S.Ct. 616 (seller was either “agent” or “licensee”). The Court in Quanta did not say that the inquiry mattered. And in Univis, the Court did the opposite of suggesting that the distinction matters: it affirmatively stated that it was treating the patent owner (Univis Corporation) and the licensee-seller (Lens Company) as the same for purposes of its analysis. 316 U.S. at 243, 62 S.Ct. 1088.

3

In these circumstances, it would read too much into the Court's use of the phrase “authorized sale” to draw the government's conclusion—making the sharp patentee-sale/licensee-sale distinction—without full analysis of statutory, precedential, and other considerations. Full analysis of the relevant legal context is necessary.

Such analysis would be required regardless, but it is important to note that the phrase “authorized sale” does not, by its words alone, compel the government's conclusion. It has long been a familiar feature of our legal landscape that property rights in a particular thing—like the separate interests in making, selling, using, etc., an invention—are viewed as a “bundle” of rights (or sticks) that can generally be transferred separately. Of course, particular legal regimes, for various purposes, commonly identify the transfer of particular rights as determinative of a “sale.” In determining which transfers are decisive, context matters. And here the context is a statute that identifies separate rights of manufacture, sale, use, etc., and the precedential setting is one that expressly recognizes the possibility of separating the rights in a patented article. See, e.g., Mitchell, 83 U.S. (16 Wall.) at 547 (exhaustion exists when a patentee “has himself constructed a machine and sold it without any conditions, or authorized another to construct, sell, and deliver it, or to construct and use and operate it, without any conditions ...”; “the owner of the machine, whether he built it or purchased it, if he has also acquired the right to use and operate it during the lifetime of the patent, may continue to use it”) (emphases added); Bloomer v. Millinger, 68 U.S. (1 Wall.) at 350. Moreover, the statutory purpose of the inquiry here is to identify what sales confer “authority” on the buyer to engage in distinct, otherwise-infringing acts. 6 Context is particularly important where, as here, the phrase being interpreted comes from judicial opinions not directly deciding the point at issue. Chief Justice Marshall wrote for the Court almost 200 years ago: “It is a maxim not to be disregarded, that general expressions, in every opinion, are to be taken in connection with the case in which those expressions are used.” Cohens v. Virginia, 19 U.S. (6 Wheat.) 264, 399–400, 5 L.Ed. 257 (1821); see Armour & Co. v. Wantock, 323 U.S. 126, 132–33, 65 S.Ct. 165, 89 L.Ed. 118 (1944) (per Jackson, J.) (“[W]ords of our opinions are to be read in the light of the facts of the case under discussion.... General expressions transposed to other facts are often misleading.”). We bear that maxim in mind in applying the body of Supreme Court case law on exhaustion: that body of precedent contains no decision against a patentee's infringement assertion in the present circumstances, and the decisions on related circumstances require careful reading to determine the best understanding of what issues the Court actually decided.

D

We conclude that a patentee may preserve its § 271 rights when itself selling a patented article, through clearly communicated, otherwise-lawful restrictions, as it may do when contracting out the manufacturing and sale.

1

That conclusion follows naturally from the statute. Congress straightforwardly prescribed, in § 271(a), that a sale or use of a patented article “without authority” is an infringement. Under that language, a clear denial of authority leaves a buyer without the denied authority.

The exhaustion rule for unrestricted sales readily fits the language of § 271(a). It is reasonable for the courts to treat a patentee-made or patentee-authorized sale of a patented article (without distinction) as presumptively granting “authority” to the purchaser to use it and resell it. Such an approach recognizes the utility of having a default rule for determining whether authority has been conferred in the many circumstances where an express conferral is missing. And it chooses as the default a principle that sensibly accords with parties' likely expectations as to a domestic sale.

But it is quite a different matter to treat a sale as conferring on the buyer the very authority that is being denied through clearly communicated restrictions. Mallinckrodt sensibly rejects that counter-textual result. Unless granting “authority” is to be a legal fiction, a patentee does not grant authority by denying it. And that is so for patentee sales and licensee sales alike, i.e., whether the patentee denies the authority to its direct purchaser or to one purchasing through a manufacturing licensee. For the same reason, the result does not reasonably reflect market participants' expectations: a buyer cannot reasonably expect that the seller is conferring authority that the seller is expressly denying, whether the seller is the patentee or a manufacturing licensee. The statutory question of authority neither allows denials to be grants nor logically depends on whether there is an intermediary between the patentee and the first buyer.

In short, the government's position would create a rule of court-made law that runs counter to, rather than accords with, the statutory definition of actionable infringement. An exhaustion rule should fit rather than contradict the statutory text.

2

Under longstanding Supreme Court precedent, a patentee may preserve its patent rights against downstream buyers by arranging with someone else, even a nonexclusive licensee, to make and sell patented articles, under clearly stated restrictions on post-sale activities. There is no good reason that a patentee that makes and sells the articles itself should be denied the ability that is guaranteed to a non-practicing-entity patentee. No precedent requires a contrary conclusion.

a. The Supreme Court has recognized that a patentee may preserve its rights against infringement by establishing restrictions accompanying the sale of the patented article (communicated at the time of sale), including restrictions on the buyer's post-acquisition use. That recognition is implicit in the Motion Picture Patents statement of the exhaustion rule as based on an “unconditional” sale, as quoted in Quanta, 553 U.S. at 626, 128 S.Ct. 2109. More affirmatively, the Court upheld a claim of infringement on that basis in Mitchell v. Hawley, where the licensee seller was under a restriction as to the (temporal) scope of rights it could and did convey when selling the patented machine. Mitchell involved a licensee seller, but the Court stated the exhaustion principles in terms keyed to the absence of conditions and applicable to patentee sales:

[W]hen [the patentee] has himself constructed a machine and sold it without any conditions, or authorized another to construct, sell, and deliver it, or to construct and use and operate it, without any conditions, and the consideration has been paid to him for the thing patented, the rule is well established that the patentee must be understood to have parted to that extent with all his exclusive right, and that he ceases to have any interest whatever in the patented machine so sold and delivered or authorized to be constructed and operated.

83 U.S. (16 Wall.) at 547; see id. at 548 (sales “may be made by the patentee with or without conditions, as in other cases, but where the sale is absolute, and without any conditions, the rule is well settled that the purchaser may continue to use the implement or machine purchased until it is worn out”). Although Mitchell involved a licensee sale, its language did not distinguish licensees' sales from patentees' sales in the respects discussed, and it was not long before the Court invoked those formulations in a patentee-sale case. Keeler, 157 U.S. at 662–63, 15 S.Ct. 738.

Most importantly, the Court squarely held in the General Talking Pictures case that a patentee could preserve its infringement rights against unauthorized uses by restricting manufacturing licensees' authority to sell for such uses. General Talking Pictures Corp. v. Western Elec. Co., 304 U.S. 175, 58 S.Ct. 849, 82 L.Ed. 1273, opinion on rehearing at 305 U.S. 124, 59 S.Ct. 116, 83 L.Ed. 81 (1938). Companies holding patent rights in certain amplifiers (AT & T, Western Electric, RCA) licensed the American Transformer Company to make and sell amplifiers. The Transformer Company “was a mere licensee under a nonexclusive license, amounting to no more than ‘a mere waiver of the right to sue.’ ” 304 U.S. at 181, 58 S.Ct. 849 (quoting De Forest Radio Telephone & Telegraph Co. v. United States, 273 U.S. 236, 242, 47 S.Ct. 366, 71 L.Ed. 625 (1927)). The license permitted sales only “for radio amateur reception, radio experimental reception, and home broadcast reception,” not “for use in theaters as a part of talking picture equipment.” Id. at 180, 58 S.Ct. 849. The Transformer Company nevertheless sold amplifiers to General Talking Pictures for theater use in violation of the restriction, and General Talking Pictures “had actual knowledge that [the Transformer Company] had no license to make such a sale.” Id. When the patentees sued General Talking Pictures (the buyer) for infringement, the Supreme Court, based on Mitchell, affirmed the judgment of infringement. Id. at 181–82, 58 S.Ct. 849.

The Court came to the same conclusion when it reconsidered the question on rehearing. It said: “Any use beyond the valid terms of a license is, of course, an infringement of a patent.” 305 U.S. at 126, 59 S.Ct. 116. Moreover, “[t]hat a restrictive license is legal seems clear.” Id. at 127, 59 S.Ct. 116 (citing Mitchell ). The use restriction was a lawful one, the Court observed, and “[a]s the restriction was legal and the amplifiers were made and sold outside the scope of the license, the effect is precisely the same as if no license whatsoever had been granted to Transformer Company.” Id. General Talking Pictures, knowing of the restriction, was “liable because it ha [d] used the invention without license to do so.” Id. Thus, General Talking Pictures was held to be liable for patent infringement. It was not held liable for breach of contract; indeed, it had no contractual relationship with the patentees.

b. The Supreme Court thus held that a patentee can preserve its patent rights by authorizing a manufacturing licensee to make and sell a patented article under an otherwise-proper restriction, including a restriction on the buyer's post-purchase use. When the buyer, knowing of the restriction at the time of purchase, subsequently uses the article in violation of the restriction, the buyer is infringing.

To distinguish the present patentee-sale situation, the government must contend that a patentee cannot preserve its patent rights against uses of a patented article contrary to known use restrictions if, instead of licensing someone else to make and sell the article, it chooses to make and sell the article itself. That contention would draw a sharp line between practicing-entity patentees (those who themselves make and sell the articles at issue) and non-practicing-entity patentees (those who do not). Non-practicing-entities would have greater power to maintain their patent rights than practicing entities.

The government points to no basis in the policy of the patent statute for making that distinction. Nothing in the patent statute suggests that patentees should have to contract out their manufacture and sale of patented articles to preserve their § 271 rights. The essential tradeoff of the patent system—to provide a market-based reward in exchange for disclosure—is equally applicable whether the patentee sells or licenses another to make and sell. The Federal Trade Commission made the fundamental point as follows: “A patentee can obtain a financial reward for its patent by producing a product that incorporates the invention or by transferring the technology through a patent license or sale to a manufacturer who develops and produces a product. The market reward earned by the patentee in either case will depend upon the extent to which consumers prefer the patented invention over alternatives and prior technology, which helps determine the invention's economic value.” See Fed. Trade Comm'n, The Evolving IP Marketplace: Aligning Patent Notice and Remedies with Competition 139 (2011) (footnote attached to “market reward” states: “The ‘market reward’ defined here is the amount the patentee could have earned by either selling a patented product or licensing the patented technology in the absence of infringement.”). The distinction urged on us appears to be unjustifiably formalistic, not founded in relevant economic substance. Mallinckrodt, 976 F.2d at 705.

The proposed distinction would also introduce practical problems. Where would the line be drawn along the spectrum from original patentees to assignees (e.g., regional assignees) to exclusive licensees (exclusivity being possible as to some but not all of the § 154 rights) to nonexclusive licensees? As we already have noted, patent law makes those distinctions for purposes of identifying who may bring infringement actions, but the distinctions are sometimes difficult to pin down and dependent on detailed inquiries into contractual provisions. When purchasing a patented article from a particular seller under specified restrictions that are not independently improper, how is the buyer to know where the seller falls along the spectrum—and, hence, whether the buyer may ignore the restrictions without fear of patent infringement?

c. The government advances a doctrinal defense of the patentee-sale/licensee-sale distinction on which it rests its challenge to Mallinckrodt. The government begins its argument on the Mallinckrodt issue by first noting the absence of statutory text supporting its position and then turning, in the next sentence, to a passage from Bloomer v. McQuewan to supply a foundation for its argument. U.S. Br. 5. It asserts that “[t]his distinction stems from the fact that licensees exercise a portion of the patentee's rights.” Id. at 8 (citing Bloomer v. McQuewan, 55 U.S. (14 How.) at 549–50). But that key distinction is wrong as a matter of basic patent law, misreads Bloomer v. McQuewan, and cannot distinguish General Talking Pictures from this case. A mere non-exclusive licensee, as in General Talking Pictures, possesses no portion of the rights granted by Congress in the patent.

Patent rights are only rights to exclude, not rights to practice. See 5 Chisum § 16.02[1]. Among the “clearly established principles” of patent law, as the Supreme Court described them in Crown Die & Tool Co., are that “the government did not confer on the patentee the right himself to make, use or vend his own invention” and “in its essence all that the government conferred by the patent was the right to exclude others from making, using or vending his invention.” 261 U.S. at 35, 43 S.Ct. 254; see Bauer, 229 U.S. at 10, 33 S.Ct. 616 (“The right to make, use, and sell an invented article is not derived from the patent law.... The [patent statute] secured to the inventor the exclusive right to make, use, and vend the thing patented, and consequently to prevent others from exercising like privileges without the consent of the patentee.”); Continental Paper Bag, 210 U.S. at 425, 28 S.Ct. 748; Bloomer v. McQuewan, 55 U.S. (14 How.) at 549 (“The franchise which the patent grants, consists altogether in the right to exclude every one from making, using, or vending the thing patented, without the permission of the patentee. This is all that he obtains by the patent.”); Western Elec. Co. v. Pacent Reproducer Corp., 42 F.2d 116, 118 (2d Cir.1930); see also Dawson Chem. Co. v. Rohm & Haas Co., 448 U.S. 176, 215, 100 S.Ct. 2601, 65 L.Ed.2d 696 (1980) (“long-settled view that the essence of a patent grant is the right to exclude others from profiting by the patented invention”). It is for that reason, for example, that a patentee may be prevented from practicing its own patent by another's patent. See Cantrell, 117 U.S. at 694, 6 S.Ct. 970; Blake, 94 U.S. at 733; Smith, 88 U.S. (21 Wall.) at 118–19.

A patentee exercises its congressionally granted rights only when it invokes its power to exclude others, not when it sells its product. Similarly, the congressionally granted right to exclude may be viewed as being shared by certain exclusive licensees, who, in appropriate circumstances (e.g., with joinder of the patent owner), may bring infringement actions against others to enforce exclusivity. See Independent Wireless, 269 U.S. at 464–69, 46 S.Ct. 166 (discussing cases); 8 Chisum § 21.03[2]. But an exclusive licensee, in merely selling (or making, using, etc.) a patented article, is not exercising any power conferred by the patent statute. That is a fortiori true of a non-exclusive licensee, like the licensee in General Talking Pictures, which has no exclusivity protections at all. Thus, although the government's assertion that “licensees stand in patentees' shoes” in sharing certain patent-granted rights is true in a significant respect as to exclusive licensees, U.S. Br. 8, it is not true as to non-exclusive licensees—like the licensee in General Talking Pictures. Accordingly, a patentee's ability to preserve its patent rights (rights to exclude) by arranging for sales to be made by a non-exclusive licensee, like the Transformer Company, cannot rest on a premise that “licensees exercise a portion of the patentee's rights.” U.S. Br. 8.

Bloomer v. McQuewan, on which the government relies from the outset of its argument, U.S. Br. 5, does not say otherwise. The government states that in Bloomer v. McQuewan “the Court explained that a purchaser of a patented article ‘stands on different ground’ than one who obtains a license under the patent,” because the latter “ ‘obtains a share in the monopoly ... derived from, and exercised under’ the patent.” Id. (emphasis added) (quoting Bloomer v. McQuewan, 55 U.S. (14 How.) at 549). But the Court did not say that about simply “one who obtains a license,” like the licensee in General Talking Pictures.

What the Court said in Bloomer v. McQuewan—immediately after the sentences stating that the patent gives the patentee only “the right to exclude”—is that “when [the patentee] sells the exclusive privilege of making or vending [the invention] for use in a particular place,” the privilege ends with the patent that creates it. 55 U.S. (14 How.) at 549 (emphasis added). That statement is about the exclusivity right, i.e., the right to exclude others, which an assignee or exclusive licensee obtains in whole or in part. By its terms, and consistent with the just-stated definition of the limited nature of the patent franchise, it says nothing about a non-exclusive licensee obtaining a share in the monopoly. The Court then contrasted, as “stand[ing] on different ground,” one who simply buys a patented device, which “[t]he inventor might lawfully sell to him, whether he had a patent or not, if no other patentee stood in his way.” Id. The buyer, as mere owner and user of the device, lacks any patent-granted right to exclude, i.e., “exercises no rights created by the act of Congress,” so whatever rights it has do not end with the patent's expiration. Id. The distinction was not between a sale from a patentee and a sale from a bare (non-exclusive) licensee. In short, General Talking Pictures cannot be distinguished on the doctrinal basis the government invokes.

d. No Supreme Court decision compels adoption of the distinction the government urges. As Impression noted at oral argument, it is undisputed that no Supreme Court decision has involved a single-use/no-resale restriction on a patentee's sale and found the restriction insufficient to preserve the patentee's infringement rights against a buyer engaging in the forbidden reuse or resale. More generally, no Supreme Court precedent denies a patentee the ability to preserve its § 271 rights, by a clear communication of an otherwise-permissible restriction, when it sells the patented article itself, just as the patentee may do, under the General Talking Pictures principle, when contracting out the making and selling of the patented article.

We have already noted the limitations on what was presented and decided in the Bloomer cases (rejecting implied temporal use restrictions) and Adams (rejecting implied geographic use restrictions). In 1881, the Supreme Court summarized the relevant law: “The right of the owner of a patented machine, without any conditions attached to his ownership, to continue the use of his machine during an extended term of the patent, is well settled.” In re Paper–Bag Cases, 105 U.S. at 770–71 (emphasis added) (citing Bloomer v. McQuewan, Mitchell, Adams, and Chaffee v. Boston Belting Co., 63 U.S. (22 How.) 217, 16 L.Ed. 240 (1859)). In so stating the law, in terms that tied exhaustion to the absence of conditions, the Paper–Bag Cases Court cited at least one case, Adams, involving a sale made by an assignee (a patentee under patent law).

The Court in Hobbie v. Jennison, 149 U.S. 355, 13 S.Ct. 879, 37 L.Ed. 766 (1893), described and followed the holding of Adams that, when an assignee sold machines, “there was no restriction on their use to be implied, for the benefit of the patentee or his assignees or licensees.” 149 U.S. at 362, 13 S.Ct. 879 (emphasis added). The Court held that an unrestricted sale of pipe in Michigan by the Michigan assignee (Jennison's firm), with full rights to sell it, allowed the buyer to use it in Connecticut free of the patent rights belonging to the Connecticut assignee (Hobbie).

In its 1895 decision in Keeler, the Court applied its existing precedents, especially Adams and Hobbie. The Massachusetts assignee (Standard Folding–Bed) sued Keeler when he bought patented beds from the Michigan assignee, subject to no restrictions, and brought them to Massachusetts for sale. 157 U.S. at 660, 15 S.Ct. 738 (stating facts before opinion starts). The Court noted that in Adams the patented articles “were sold to [Burke] without condition or restriction,” id. at 663, 15 S.Ct. 738, and it concluded that Adams “was applicable,” id. at 665, 15 S.Ct. 738. The Court also quoted as defining the law (in this patentee-sale case) the passage set out above from Mitchell (a licensee-sale case). Id. at 663, 15 S.Ct. 738. In any event, with no restriction on the sale present, the Court followed Adams' refusal to find one implied by the patent law.

It was against that background that the Court then noted: “Whether a patentee may protect himself and his assignees by special contracts brought home to the purchasers is not a question before us,” but “such a question would arise as a question of contract, and not as one under the inherent meaning and effect of the patent laws.” Id. at 666, 15 S.Ct. 738 (emphasis added). That language lends itself to use in the government's argument that sale restrictions never preserve patent rights, but give only contract rights. But even in Keeler the language does not compel that reading, and the later decisional law—most importantly, General Talking Pictures—undermines the contract-only interpretation.

Thus, in Keeler itself, the word “inherent” naturally ties the language to the modest point based on Adams that actually decided Keeler: with no contract restriction as part of the sale, an implied one cannot be found in patent law itself. That says nothing about the irrelevance of an actual contract restriction to preservation of patent rights. Moreover, in the licensee-sale context, General Talking Pictures establishes that contract restrictions can indeed preserve a patentee's infringement rights and are not merely enforceable under contract law: General Talking Pictures, which had no contract with the patentees, was held liable for patent infringement, not breach of any contract with the patentees. Notably, then, in Quanta, a licensee-sale case like General Talking Pictures, the Court quoted the language from Keeler after discussing General Talking Pictures (without a hint of disapproval) and concluding that, as in Keeler, Adams, and Hobbie, there simply was no patentee-imposed contractual restriction applicable to the sales at issue. 553 U.S. at 637 n. 7, 128 S.Ct. 2109. And in the present case, as in General Talking Pictures, there is no reason to think that the patentee has a contract remedy available as a substitute for patent infringement: as far as the record before us shows, Lexmark has no contractual relationship with Impression.

In United States v. General Electric Co., 272 U.S. 476, 47 S.Ct. 192, 71 L.Ed. 362 (1926), the Supreme Court rejected the government's antitrust challenge to (among other things) General Electric's licensing of Westinghouse to make and sell patented lamps under terms controlling resale prices. See Fed. Trade Comm'n v. Actavis, Inc., ––– U.S. ––––, ––––, 133 S.Ct. 2223, 2232, 186 L.Ed.2d 343 (2013) (describing General Electric ). In making a general point about patentee sales (not involved in the case), the Court said: “It is well settled, as already said, that where a patentee makes the patented article, and sells it, he can exercise no future control over what the purchaser may wish to do with the article after his purchase. It has passed beyond the scope of the patentee's rights.” 272 U.S. at 489, 47 S.Ct. 192 (citing cases). We read that language to deem “settled” only what was settled in the cited precedents—a patentee's sales without restrictions exhaust patent rights in the item sold. The cited cases are Adams, Bloomer v. McQuewan, Hobbie, Keeler, and Mitchell. They do not go beyond that proposition.

The prominent exhaustion decisions from the 1910s do not make the patentee-sale/licensee-sale distinction urged by the government here. After A.B. Dick adopted a broad principle that preserved a patentee's patent-law rights against restriction-violating sale, use, etc., even if the restrictions were otherwise unlawful—in A.B. Dick, a tie-in requiring purchase of unpatented products—the Supreme Court repudiated A.B. Dick's broad principle and held particular restrictions improper and therefore not effective at preserving patent rights against actions contrary to those restrictions. It did so as to tie-ins in Motion Picture Patents, overruling A.B. Dick and relying on the 1914 Clayton Act, 38 Stat. 730. See 243 U.S. at 517, 37 S.Ct. 416. And it did so as to resale price maintenance in Bauer, Straus, and finally Boston Store, relying on the judicially adopted per se antitrust condemnation of that practice in Dr. Miles Medical Co. v. John D. Park & Sons Co., 220 U.S. 373, 31 S.Ct. 376, 55 L.Ed. 502 (1911), overruled by Leegin Creative Leather Prods., Inc. v. PSKS, Inc., 551 U.S. 877, 127 S.Ct. 2705, 168 L.Ed.2d 623 (2007). See Bauer, 229 U.S. at 12, 33 S.Ct. 616; Straus, 243 U.S. at 498, 37 S.Ct. 412; Boston Store, 246 U.S. at 21, 38 S.Ct. 257. But the Court did not rule that all restrictions on a patentee's sale were ineffective to preserve the patentee's patent-law rights. Instead, it called for an inquiry—in accord with what Mallinckrodt later said—into whether a patentee's restrictions were otherwise improper, as by “extend[ing] the scope of its patent monopoly.” Motion Picture Patents, 243 U.S. at 516, 37 S.Ct. 416. And the Court did not adopt the line the government suggests between patentee sales and licensee sales.

In its 1942 decision in Univis, the Supreme Court rejected a patent-based defense to the government's antitrust challenge to resale-price-maintenance restrictions in the licensing and selling of eyeglass lenses. The Court said that it had two questions before it: first, whether the restrictions were “excluded by the patent monopoly from the operation of the Sherman Act,” i.e., whether if the restrictions were illegal under the Sherman Act, they were saved by the patent law; and second, whether the restrictions were illegal under the Sherman Act. 316 U.S. at 243, 62 S.Ct. 1088. The Court said no on the first question, id. at 249–52, 62 S.Ct. 1088, and yes on the second, id. at 252–54, 62 S.Ct. 1088.

The second answer (regarding the substance of antitrust law) is immaterial here, and the first answer is in accord with Mallinckrodt's ruling—which expressly recognizes that, as Univis held, restrictions that are otherwise unlawful do not preserve patent rights. Moreover, although some language in Univis, like language in other decisions in the area, can be taken out of context and read as going beyond the specific restrictions involved, id. at 249–51, 62 S.Ct. 1088, the most the Court ruled, even as to patent law all by itself, was that a vertical price-control restriction was ineffective to preserve patent rights after sales of articles embodying the patents. While Univis is controlling on what it decided on the issues before it, we do not think it appropriate to give broad effect to language in Univis, taken out of context, to support an otherwise-unjustified conclusion here on a question not faced there. And that is particularly so today, given that the Univis opinion relied in part on strongly restrictive patent-misuse decisions that were repudiated by Congress after Univis was decided. Most pointedly for present purposes, Univis does not support the distinction between patentee sales and licensee sales the government urges in this case. The Univis case was decided just four years after General Talking Pictures, which confirmed that a patentee may preserve its patent rights by imposing otherwise-lawful restrictions on sales by its manufacturing licensees. Yet Univis says nothing to limit that prominent, recent ruling. Moreover, as we have already noted, Univis is explicit that it made no difference to the Court's analysis whether the patentee (Univis Corporation) or the manufacturing licensee (Univis Lens Company) was doing the selling: the Court stated that the two companies “may for the purposes of this suit be treated as though they were a single corporation.” 316 U.S. at 243, 62 S.Ct. 1088. The Court also stated its point about the particular sale as equally applicable to a “[s]ale of a lens blank by the patentee or by his licensee.” Id. at 249, 62 S.Ct. 1088 (emphasis added).

For the foregoing reasons, we think that the best lesson to draw from the Supreme Court's precedents, as applied to the question before us, is that a patentee may preserve its patent rights by otherwise-proper restrictions when it makes and sells patented articles itself and not only when it contracts out manufacturing and sales.

3

We see no basis for a different conclusion in Lord Coke's description in 1628 of a British common-law principle, as quoted in Kirtsaeng, 133 S.Ct. at 1363. Lord Coke described what British courts, in the absence of an overriding legislative prescription, would treat as an impermissible anti-alienation restriction on a seller's disposition of “ ‘his whole interest’ ” in a chattel. Id. Lord Coke's formulation was part of the judicial formulation of background law for personal property generally, and it neither addressed possible differences among particular kinds of personal property nor suggested that a judicial rule would override specific legislative grants. Lord Coke's formulation was a pertinent reference point in Kirtsaeng, because, as we have noted, the copyright statute, 17 U.S.C. § 109(a), states a rule that itself over-rides the otherwise-applicable statutory bars on infringement and importation and grants of exclusive rights. In stating one common-law jurisdiction's general judicial policy at one time toward anti-alienation restrictions, Lord Coke's description confirmed that the otherwise-supported reading of § 109(a) fit a legal tradition.

Lord Coke's quote does not purport to address the effect of a legislative prescription of broad rights to control sale and use. That is what is present in the Patent Act, but not the copyright law. Sections 154(a) and 271(a) legislatively establish a patentee's rights over sale and use, without subservience to a superseding grant of rights to one who owns a particular article. They grant those rights separately as to making, selling, using, etc., thus recognizing different sticks in the bundle of rights; they grant them without an exception keyed to the patentee's prior ownership of a particular article embodying the invention; and they grant them unless, as relevant here, the patentee confers “authority.” Lord Coke's quote does not address the Patent Act situation or suggest that federal courts may treat a denial of authority as a conferral of authority.

Different policy choices can readily be made and justified in this area, even as to background rules applicable to personal property generally. Some of the numerous, distinct common-law jurisdictions, including Lord Coke's, have departed at various times from the background rule expressed by Lord Coke. See De Mattos v. Gibson, (1859) 45 Eng. Rep. 108 (Ch. App.); Waring v. WDAS Broad. Station, Inc., 327 Pa. 433, 194 A. 631, 637–38 (1937); Metro. Opera Ass'n v. Wagner–Nichols Recorder Corp., 199 Misc. 786, 101 N.Y.S.2d 483, 494–95 (N.Y.Sup.Ct.1950), aff'd, 279 A.D. 632, 107 N.Y.S.2d 795 (N.Y.App.Div.1951); Pratte v. Balatsos, 99 N.H. 430, 113 A.2d 492, 494–95 (1955); Nadell & Co. v. Grasso, 175 Cal.App.2d 420, 346 P.2d 505, 509–10 (1959); Clairol Inc. v. Cosmetics Plus, 130 N.J.Super. 81, 325 A.2d 505, 508 (N.J.Sup.Ct.1974); Zechariah Chafee, Jr., Equitable Servitudes on Chattels, 41 Harv. L.Rev. 945, 1007–13 (1928); Zechariah Chafee, Jr., The Music Goes Round and Round: Equitable Servitudes and Chattels, 69 Harv. L.Rev. 1250, 1254–56 (1956); Glen O. Robinson, Personal Property Servitudes, 71 U. Chi. L.Rev. 1449, 1455–60 (2004). In 1918, then-Dean Harlan Fiske Stone wrote: “The tendency in the United States has been to apply the doctrine of restrictive agreements to personal property when not regarded as an unlawful restraint of trade or in violation of public policy.” The Equitable Rights and Liabilities of Strangers to a Contract, 18 Colum. L.Rev. 291, 310 (1918).

In any event, and more specifically, whatever considerations might go into a jurisdiction's choice as to the background rule for personal property in general, lawmaking authorities may reasonably make different choices for particular kinds of property. Notably, as to intellectual property in its various forms, Congress, implementing the Constitution, has long deemed it important to incentivize creation and disclosure through grants to the creator of rights to exclude others for a time, the duration and scope based on features of the particular kind of intellectual property (e.g., patent terms are much shorter than copyright terms). The Patent Act expressly does so regarding patent rights, specifically giving separate rights to exclude others from making, using, selling, etc. That overriding legislative prescription removes the patented-article sale from the scope of Lord Coke's 1628 description of his country's general judicially fashioned property law, as British tribunals recognized long ago. See A.B. Dick, 224 U.S. at 42–43, 32 S.Ct. 364 (quoting the judicial committee of the Privy Council, speaking through Lord Shaw in 1911: “the general doctrine of absolute freedom of disposal of chattels of an ordinary kind is, in the case of patented chattels, subject to the restriction that the person purchasing them, and in the knowledge of the conditions attached by the patentee, which knowledge is clearly brought home to himself at the time of sale, shall be bound by that knowledge and accept the situation of ownership subject to the limitations. These limitations are merely the respect paid and the effect given to those conditions of transfer of the patented article which the law, laid down by statute, gave the original patentee a power to impose.”) (quoting Nat'l Phonograph Co. v. Menck, [[1911] A.C. 336, 349 (P.C.1911 appeal taken from Aus.)]); Incandescent Gas Co. v. Cantelo, [1895] 12 R.P.C. 262, 264 (Eng.).

In short, notwithstanding Lord Coke's description of English general personal-property judge-made law, the patent-specific statutory analysis must govern here.

4

Finally, following the analytical method of Kirtsaeng, we consider what we can reliably gauge about the likely real-world consequences of one answer or another to the exhaustion question presented here. As indicated at the front of this opinion, we have received numerous amicus briefs making competing arguments, with varying degrees of reliable factual support, for the effect of Mallinckrodt on their interests or the interests they promote. We cannot assess those contentions and make policy choices in the way Congress can. We can say only that the amicus presentations give us no reason to depart from the application of § 271 we derive from the statute and precedent.

In particular, we see no basis for predicting the extreme, lop-sided impacts the Court found plausible in Kirtsaeng in different circumstances. Mallinckrodt has been the governing case law since 1992 and has been reiterated in subsequent precedent. And yet we have been given no reliable demonstration of widespread problems not being solved in the marketplace. Given General Talking Pictures, the only question is about patentees' ability to do for their own sales what they already can do by contracting out their manufacturing and sales. Regarding the specific scenario we are addressing today—in which the patentee has sought to preserve its patent rights by conditioning its first sale on a single-use/no-resale restriction of which the accused infringer had adequate notice at the time of purchase—we have been given no proof of a significant problem with enforcing patent rights.

At the same time, the conduct challenged here can have benefits. Lexmark's Return Program provides customers an immediate up-front benefit: a choice between two options, one offering them a lower price in exchange for the single-use/no-resale limitation. And a company in Lexmark's position could have a plausible legitimate interest in not having strangers modify its products and introduce them into the market with the quality of modifications (including ink refills) not subject to Lexmark's control: lower quality of remanufactured cartridges could harm Lexmark's reputation. See Chafee, 41 Harv. L.Rev. at 946–47. A medical supplier in Mallinckrodt's position plausibly may have similar reason to believe that reuse, when not under its own control, carries a significant risk of poor or even medically harmful performance, to the detriment of its customers and its own reputation. Such interests are hardly unrelated to the interests protected by the patent law—the interests both of those who benefit from inventions and of those who make risky investments to arrive at and commercialize inventions. See Robinson, 71 U. Chi. L.Rev. at 1480–1515 (surveying reasons for restrictions, particularly in intellectual-property area).

We do not have a record on such interests in this case, as Impression has not claimed that the restrictions at issue violate antitrust, patent-misuse, or similar constraints. And it is not our function to assess the strength of such interests against those which might pull the other way. Nor can we fairly assume the illegitimacy of the conduct here. Such an assumption would run counter to the large-scale changes in antitrust law and patent-misuse law, especially over the last four decades, that have displaced the strict condemnation of various vertical restrictions that characterized both areas of law in the first half of the twentieth century.

Thus, the Supreme Court broadly held that non-price vertical restraints are to be judged by a rule of reason. See Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 57–59, 97 S.Ct. 2549, 53 L.Ed.2d 568 (1977), overruling United States v. Arnold, Schwinn & Co., 388 U.S. 365, 87 S.Ct. 1856, 18 L.Ed.2d 1249 (1967). The Court abandoned its “strong disapproval of tying arrangements” by insisting on market power in the tying product as a precondition to condemnation. See Illinois Tool Works Inc. v. Independent Ink, Inc., 547 U.S. 28, 35–38, 126 S.Ct. 1281, 164 L.Ed.2d 26 (2006); Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2, 104 S.Ct. 1551, 80 L.Ed.2d 2 (1984). It overturned both the per se ban on vertical agreements setting maximum resale prices, State Oil Co. v. Khan, 522 U.S. 3, 7, 118 S.Ct. 275, 139 L.Ed.2d 199 (1997), overruling Albrecht v. Herald Co., 390 U.S. 145, 88 S.Ct. 869, 19 L.Ed.2d 998 (1968), and the per se ban on vertical agreements setting minimum resale prices, Leegin Creative Leather Prods., Inc. v. PSKS, Inc., 551 U.S. 877, 900–01, 127 S.Ct. 2705, 168 L.Ed.2d 623 (2007), overruling Dr. Miles Medical Co. v. John D. Park & Sons Co., 220 U.S. 373, 31 S.Ct. 376, 55 L.Ed. 502 (1911).

The absence of a general basis for finding market harm from vertical restrictions is recognized specifically in the patent area, too. Field-of-use restrictions in patent licenses have long been common, as Mallinckrodt points out and General Talking Pictures shows. In 1988, building on an initial relaxation of patent-misuse standards in 1952, Congress made clear that tying arrangements involving a non-patented product do not constitute patent misuse where the patentee lacks market power. 35 U.S.C. § 271(d)(5). The Supreme Court, citing that determination, subsequently overturned its own longstanding antitrust presumption that patent (and copyright) owners have market power. Illinois Tool Works, 547 U.S. at 41–42, 126 S.Ct. 1281. And the two federal antitrust agencies have recognized that restrictions in intellectual-property licenses can be procompetitive. See U.S. Dep't of Justice & Fed. Trade Comm'n, Antitrust Guidelines for the Licensing of Intellectual Property § 2.3 (1995) (“Field-of-use, territorial, and other limitations on intellectual property licenses may serve procompetitive ends by allowing the licensor to exploit its property as efficiently and effectively as possible.”).

For those reasons, we see no basis for departing from the legal analysis set out above. A patentee already may preserve its patent rights against downstream buyers (with notice) through otherwise-lawful restrictions, by licensing others to make and sell its patented articles. We conclude that the law does not forbid the patentee to do the same when making and selling the articles itself.

III

The second question presented to us is whether Lexmark's sales of its cartridges abroad conferred authority on its buyers, and derivatively on Impression, to import the cartridges into, and sell and use them in, the United States, which would be infringing acts in the absence of authorization. The question was decided by the district court, and is presented here, on the premise that Lexmark made the foreign sales without communicating a reservation of U.S. patent rights. And the question is presented only as an exhaustion question, because Impression did not press any implied-license defense, despite the fact that Quanta made clear that the doctrines are distinct.

The absence of an implied-license defense in this case sharpens the definition of the issue presented. There is no doubt that a U.S. patentee, when selling a U.S.-patented article abroad, could give the buyer permission, expressly or by implication from the circumstances, to import the purchased article into the United States and sell and use it here. Such a license would make those acts non-infringing. The question for decision is whether, if there is no proof of any such license (express or implied), there is nonetheless a legal rule that such a foreign sale confers authority on the overseas buyer to import the patented article into the United States and sell and use it here. If there is such a legal rule of authorization, the next question is whether the authorization is conclusive, effective even in the face of the U.S. patentee's reservation of U.S. rights, or only presumptive, with the U.S. patentee able to reserve its U.S. rights if it can demonstrate with adequate certainty that it has taken the steps needed to do so.

We conclude, as we did in Jazz Photo, that there is no legal rule that U.S. rights are waived, either conclusively or presumptively, simply by virtue of a foreign sale, either made or authorized by a U.S. patentee. The government, we note, agrees that a conclusive-exhaustion rule should be rejected but argues for a presumptive-exhaustion rule regarding a U.S. patentee's foreign sales. In the government's view, a U.S. patentee can reserve its U.S. rights when selling abroad (but not if selling domestically, under the government's view that Mallinckrodt is wrong). We conclude that neither a conclusive- nor a presumptive-exhaustion rule is legally justified.

A

This court's 2001 decision in Jazz Photo reviewed the International Trade Commission's finding that Jazz Photo (and others) infringed patents of Fuji Photo Film by importing refurbished disposable cameras originally sold by or with the authorization of Fuji Photo. 264 F.3d 1094, 1098. Two groups of cameras sold by or with the authorization of Fuji Photo were at issue: those sold initially in the United States, refurbished abroad, and imported back into the United States; and those initially sold abroad, refurbished abroad, and imported into the United States. This court drew different conclusions about infringement regarding those two groups of imported cameras.

Disagreeing with the Commission on the central issue in the case, the court held that a specifically described set of refurbishment changes (involving insertion of new film into the used casings) were mere repairs, not reconstructions that amounted to creation of new articles. Id. at 1102–07. Therefore, the court ruled, for the “used cameras whose first sale was in the United States with the patentee's authorization,” and that were subjected to only those changes (with disclosure to the Commission), Jazz Photo's importations were not infringing: repair maintained the identity of the article initially sold, and the domestic sale exhausted the patentee's rights in the article sold (but not in newly created articles). Id. at 1098–99, 1102–05 (discussing, for example, Aro I, 365 U.S. at 346, 81 S.Ct. 599); see Bowman, 133 S.Ct. at 1766. As to those used cameras, the court reversed the Commission. 264 F.3d at 1099.

The court held, however, that a different result was required for any of the imported cameras that previously had been “sold only overseas,” even if the changes in them amounted only to repair. Id. at 1105 (emphasis added). Relying on Boesch v. Graff, 133 U.S. 697, 701–03, 10 S.Ct. 378, 33 L.Ed. 787 (1890), the court ruled that “United States patent rights are not exhausted by products of foreign provenance,” i.e., products previously sold only abroad. 264 F.3d at 1105. Thus, the court's non-infringement ruling (and reversal of the Commission) applied only to used cameras “for which the United States patent right has been exhausted by first sale in the United States” and whose refurbishing was limited as described. Id. The imported cameras not previously sold in the United States, in contrast, “are not immunized from infringement of United States patents by the nature of their refurbishment.” Id. There is no suggestion that Jazz Photo argued that Fuji Photo had expressly or impliedly licensed importation in making or authorizing the foreign sales, and the court said nothing to foreclose such a defense to infringement. Accordingly, as to cameras “whose prior sale was not in the United States,” the court affirmed the Commission's infringement finding and remedies against Jazz Photo. Id. at 1111.

The court followed Jazz Photo in Fuji Photo Film Co., Ltd. v. Jazz Photo Corp., 394 F.3d 1368, 1370 (Fed.Cir.2005), which affirmed a district court's judgment of infringement by Jazz Photo (and others) in favor of Fuji Photo in litigation involving the same dispute as Jazz Photo. This court rejected Jazz Photo's argument for exhaustion as to first sales made abroad. The court in Fuji Photo explained that in Jazz Photo “this court expressly limited first sales under the exhaustion doctrine to those occurring within the United States.” Id. at 1376. Accordingly, exhaustion of U.S. rights is not triggered by “[t]he patentee's authorization of an international first sale.” Id. In Fuji Photo, as in Jazz Photo, the court did not foreclose any argument about express or implied licenses conferred in particular foreign sales.

In short, this court has held since 2001 that the foreign sale of a U.S.-patented article, when the sale is either made or authorized by the U.S. patentee, does not, standing alone, confer on the buyer “authority” to import the item into the United States or to sell and use it here, and so does not save those acts from being infringing under § 271(a). See Ninestar Tech. Co. v. Int'l Trade Comm'n, 667 F.3d 1373, 1378 (Fed.Cir.2012). The court has not curtailed the ability of an accused infringer to show that the patentee conferred such authority by words or implications. Exhaustion cannot rest on a foreign first sale, but an express or implied license might be found based on the circumstances of particular foreign sales.

B

The Supreme Court's decision in Kirtsaeng does not undermine the no-exhaustion conclusion of Jazz Photo. In Kirtsaeng, the Court interpreted § 109(a) of the Copyright Act, which states that “the owner of a particular copy ... lawfully made under this title ... is entitled, without the authority of the copyright owner, to sell or otherwise dispose of the possession of that copy....” 17 U.S.C. § 109(a). The Court held that § 109(a)'s guarantee is not limited to copies manufactured in the United States, but applies regardless of the place of manufacture, as long as the maker of the copies had permission from the copyright owner to make them. Kirtsaeng, 133 S.Ct. at 1355–71.

For various reasons, that ruling does not answer the question presented under the Patent Act. Kirtsaeng says nothing about patent law; and it does not address, even in the context of copyright law, the exhaustion question presented by the Patent Act. Whether a sale constitutes a grant to a buyer of (conclusive or presumptive) “authority” to engage in otherwise-prohibited acts of importation, sale, and use is the question here. It is not the question presented by the Copyright Act. That Act contains no right to exclude anyone from “use,” and § 109(a) of the Act expressly overrides the copyright holder's rights to exclude from importing or selling copies, permitting acts “without the authority” of the rights owner—in circumstances that undisputedly have nothing to do with the place of sale. The Court in Kirtsaeng merely interpreted § 109(a) and resolved the dispute about whether the place of manufacture matters under § 109(a), holding—for a number of reasons “taken together”—that it does not. 133 S.Ct. at 1358, 1371. The Kirtsaeng question thus is several steps removed from the question presented under the Patent Act, which requires a quite different analysis.

To elaborate: In Kirtsaeng, the Supreme Court did not advert to the foreign-exhaustion issue under patent law. Nor did it cite, even to distinguish, its own leading case on exhaustion and foreign sales in the patent area, namely, Boesch—which has no counterpart in the copyright area. More generally, the Court nowhere relied on the wealth of exhaustion cases in the patent area. The absence of such references to patent law, even at a general level, reinforces the need for a distinct patent-law analysis.

The Court has long recognized the distinctness of the copyright and patent regimes and observed that particular questions require separate analysis for each body of law. For example, the Court has noted that the patent right is broader in scope than the copyright right in at least one important respect: the patent statute gives a right to exclude others from “use,” whereas the copyright statute does not. Bauer, 229 U.S. at 13–14, 33 S.Ct. 616. In a decision relied on in Kirtsaeng, 133 S.Ct. at 1363, the Court stated more generally that copyright-law conclusions and patent-law conclusions do not necessarily align, so that a conclusion about copyright law does not automatically carry over to patent law. Bobbs–Merrill, 210 U.S. at 345–46, 28 S.Ct. 722. In Sony Corp. of America v. Universal City Studios, Inc., the Court, noting that patent and copyright law “are not identical twins,” required “caution ... in applying doctrine formulated in one area to the other.” 464 U.S. 417, 439 n. 19, 104 S.Ct. 774, 78 L.Ed.2d 574 (1984). In Eldred v. Ashcroft, the Court explained that “patents and copyrights do not entail the same exchange.” 537 U.S. 186, 216, 123 S.Ct. 769, 154 L.Ed.2d 683 (2003).

The answer to a particular question therefore requires analysis of the specifics of the relevant statute. The Court in Kirtsaeng conducted just such an analysis for the copyright-law question before it. It analyzed a copyright-specific text, namely, § 109(a), and stressed that it was determining “the best reading of § 109(a).” 133 S.Ct. at 1370 (emphasis in original). See id. at 1371 (“we do no more here than try to determine what decision Congress has taken”); id. at 1357 (“We must decide whether the words ‘lawfully made under this title’ restrict the scope of § 109(a)'s ‘first sale’ doctrine geographically.”). And the structure of the Court's analysis confirms the primacy of the statutory text: The Court began its analysis with an extensive consideration of the text of § 109(a). 133 S.Ct. at 1358–60. It concluded that “[t]he language of § 109(a) says nothing about geography”; reading “made under this title” to mean made with the rights holder's permission, not to mean made in the United States, “is simple, ... promotes a traditional copyright objective (combatting piracy), and ... makes word-by-word linguistic sense”; while the made-in-the-United-States interpretation “bristles with linguistic difficulties.” 133 S.Ct. at 1358. The Court then found various contextual, historical, and practical considerations to support that textual conclusion. Id. at 1358–62.

The text construed in Kirtsaeng has no counterpart in the Patent Act. And that text presents a sharply different question from the statutory question presented by the Patent Act. By its terms, far from calling for a determination of whether any kind of sale constitutes the conferring of “authority” from the rights holder, § 109(a) defines the circumstances (ownership of a copy lawfully made) that, when present, give a copy owner a right to sell or dispose of the owned copy “without the authority of the copyright owner.” 17 U.S.C. § 109(a) (emphasis added). Moreover, as we have explained, and as the Court ruled in Kirtsaeng and Quality King, the Copyright Act makes the provisions on exclusivity, infringement, and importation all subservient to § 109(a). In the Copyright Act, the § 109(a) grant to copy owners overrides other requirements of authority from the rights holder, specifically those governing importation and sale. That is not so in the Patent Act, under which exhaustion textually can be nothing but an answer to a statutory question of when a patentee has, by a sale, conferred such authority.

Section 109(a)'s language, which gives an owner an entitlement to resell “without the authority of the copyright owner,” does not make that entitlement depend on an assessment of whether a first sale made or authorized by the copyright holder confers resale authority on the buyer. The right to resell is given to the “owner” of an article “lawfully made” under the Act. The “owner” language—whose meaning was not at issue in Kirtsaeng—says nothing about where ownership is acquired and does not require a prior sale at all: a copy made by the person who owns it, as long as the making was authorized, is within § 109(a)'s language. See 133 S.Ct. at 1361 (discussing 17 U.S.C. § 115); 2 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 8.12[B][3][c] (2015). Even when ownership comes from a sale, moreover, the provision prescribes the result for the owner's resale entitlement in terms not dependent on “authority” from the copyright holder but as independent of any such authority. And the remaining requirement stated by the language at issue in Kirtsaeng—“lawfully made under this title”—undeniably refers only to the manufacture of the copy and whether that manufacture was lawful. That language, like the “owner” language, leaves no place for consideration of the location of a prior sale (if there was one), which is the issue here.

Years before deciding Kirtsaeng, the Court in Quality King had made clear that the language of § 109(a) makes sale location irrelevant: under that language, “the owner of goods lawfully made under the Act is entitled to the protection of the first sale doctrine in an action in a United States court even if the first sale occurred abroad.” 523 U.S. at 145 n. 14, 118 S.Ct. 1125. The Court in Kirtsaeng confirmed the point. 133 S.Ct. at 1371 (noting the “holding in Quality King that § 109(a) is a defense in U.S. courts even when ‘the first sale occurred abroad’ ”). Not surprisingly, neither party in Kirtsaeng argued that the provision could be read to refer to the sale location. And the Court readily dismissed a sale-location view as “not defensible” and facing “linguistic and other hurdles that ... are insurmountable.” Id. at 1366. With sale location off the table, the dispute in Kirtsaeng was simply between two different interpretations of the statutory reference to manufacture (“lawfully made under this title”)—whether it referred to where manufacture occurred (the “geographic” interpretation) or to whether the manufacture was lawful under the standards of the Copyright Act (e.g., with the copyright owner's permission, as opposed to pirated). For that reason, whether an “implied license” would arise from sales in some circumstances was immaterial to the statutory question being debated, and the Court did not comment on that notion, despite its mention in the dissent, 133 S.Ct. at 1389 & n. 25 (Ginsburg, J., dissenting), and the government's brief, Brief for the United States as Amicus Curiae Supporting Respondent at 21, Kirtsaeng (No. 11–697), 2012 WL 3902599.

In short, given the nature of the question framed by § 109(a), the Court in Kirtsaeng did not have occasion to decide, and did not decide, whether a foreign sale (by or authorized by the U.S. rights holder) is properly treated as conferring on the buyer, conclusively or presumptively, the authority to resell. The Court did not decide, even for copyrights, the question presented here for patents.

The nature of the statutory question decided in Kirtsaeng also shows why the Court's discussion of considerations supporting its textual conclusion cannot be transposed to the patent-law setting at issue here. The discussion of the statutory history and certain provisions of the Copyright Act, 133 S.Ct. at 1360–62, 1370, is statute-specific. And the Court's discussion of the absence of any constitutional history or congressional action permitting market division was limited to the copyright area. Id. at 1370–71. Compare 35 U.S.C. § 261 (patentee may assign rights “to the whole or any specified part of the United States”).

Similarly, the Court's discussion of Lord Coke's 1628 description of his country's general judicial personal-property law, id. at 1363–64, is inapplicable here. That description, as already explained, is apt background for a provision, like § 109(a), that is superior to any legislative grant of rights that cover post-purchase activities of a buyer. The Patent Act contains no such override of the Act's grant of rights to patentees. And the Court in Kirtsaeng drew from Lord Coke's description only a general recognition of “the importance of leaving buyers of goods free to compete with each other when reselling or otherwise disposing of those goods,” adding that American antitrust law recognizes a similar point. 133 S.Ct. at 1363. That observation merely confirms that the result of the § 109(a) analysis is sensible because it fits one policy found in aspects of American and British law containing no specific statutory override. And the Court then cited Bobbs–Merrill, which construed the pre–1909 copyright statute not to contain an override in the circumstances at issue, 210 U.S. at 349–51, 28 S.Ct. 722, and noted that the next year, Congress adopted that result by enacting the statutory predecessor of § 109(a). Kirtsaeng, 133 S.Ct. at 1363–64.

In addition, the Court's account of the potential real-world consequences of the statutory interpretation it was rejecting, though it mentions sale location a few times, is pervasively tied to the issue actually in dispute—whether a foreign manufacture location makes § 109(a) inapplicable. 133 S.Ct. at 1364–67. Under that interpretation, the Court stated, the rights holder would have “permanent” control, id. at 1362, “perpetual downstream control,” id. at 1371, of copies circulating in the United States as long as those copies had been made abroad: § 109(a) would not kick in to give resale rights to purchasers of those copies even if the copyright holder sold the article in the United States. See also id. at 1366 (referring to “the absurd result that the copyright owner can exercise downstream control even when it authorized the import or first sale”). As the government notes to us, the Patent Act, which lacks a provision like § 109(a), is quite different: “there is no concomitant risk of ‘perpetual downstream control’ over patented goods.” U.S. Br. 24. At the very least, an unrestricted patentee-made or - authorized sale in the U.S. triggers exhaustion as to the article sold.

Moreover, the “copyright-related consequences” emphasized by the Court in Kirtsaeng, 133 S.Ct. at 1367, were to a large extent, though not entirely, tied to the distinctive problems of museums, libraries, and booksellers. Id. at 1364–67. To that extent, the Court's overall analysis of plausible practical effects—of an interpretation keeping every foreign-made copy forever outside § 109(a)—was copyright-specific. The Court in Kirtsaeng also concluded that circuit-court precedent on § 109(a) was too fractured to give meaningful comfort that the practical problems from the Court's adoption of Wiley's view of § 109(a) were unlikely to materialize. Id. at 1366. In contrast, our exclusive jurisdiction and clear rule since 2001, together with the pre–2001 precedents from other courts, provide considerably more reason to discount predictions that adhering to a territorial line to make exhaustion unavailable based on a foreign sale will result in serious practical problems.

For all of those reasons, Kirtsaeng is not controlling in this case. The patent-law issue presented here requires a separate analysis in its own legal setting.

C

The Patent Act question is whether a foreign sale of a U.S.-patented article made or authorized by a U.S. patentee, standing alone, confers on the buyer authority to import the article into the United States and sell and use it here, even though such an act would be infringing in the absence of authority. The best answer to that question, we conclude, is that such a foreign sale does not confer such authority. A U.S. patentee, simply by making or authorizing a foreign sale of an article, does not waive its U.S. rights to exclude regarding that article, either conclusively (no matter how clear the reservation of U.S. rights) or only presumptively (subject to sufficiently clear preservation of U.S. rights).

1

The combined logic of the statutory grant of patent rights and the long-recognized basis for exhaustion leads naturally to rejecting exhaustion based on a foreign sale. The statute gives patentees the reward available from American markets. A patentee cannot reasonably be treated as receiving that reward from sales in foreign markets, and exhaustion has long been keyed to the idea that the patentee has received its U.S. reward.

Thus, what the statute expressly provides to a U.S. patentee is the reward available from the right to exclude “in the United States.” See 35 U.S.C. §§ 154(a)(1), 271(a). The reward is inherently a market reward: “it is one of the legal beauties of the system that what is given by the people through their government—the patent right—is valued automatically by what is given by the patentee. His patent has value directly related to the value of his invention, as determined in the marketplace.” In re Kirk, 54 CCPA 1119, 376 F.2d 936, 964 (1967) (Rich, J., dissenting). And the market reward, under the statute, is explicitly the reward available from American markets subject to American laws, a reward obtained by selling or authorizing sales in those markets.

At the same time, the Supreme Court has “explained the basis for [exhaustion] doctrine” in terms of the patentee's receipt of the reward given by the statute. Bowman, 133 S.Ct. at 1766. “ ‘The purpose of the patent law is fulfilled with respect to any particular article when the patentee has received his reward ... by the sale of the article.’ ” Id. (quoting Univis, 316 U.S. at 251, 62 S.Ct. 1088, itself citing earlier authorities). Only when it is appropriate to assume the receipt of that reward does the sale support an inference of conferral of authority on an article's buyer (in the absence of clearly communicated restrictions on the authority conferred).

Whatever other issues may be presented by determining when a patentee has “received his reward,” the territorial nature of the statutory guarantee supplies a simple, strong reason to exclude foreign sales. The guarantee is the reward from sales in American markets, not from sales in foreign markets. A sale in a foreign market therefore does not furnish “the basis for” exhaustion—even for a presumption that authority is being conferred on the buyer to exploit the article in American markets by the actions (importation, sale, use, etc.) that are infringing in the absence of patentee-conferred authority.

American markets differ substantially from markets in many other countries, and not just because of disparities in wealth that can lead to dramatically different prices (especially for low-marginal-cost products). Government policies differ dramatically, including policies on price regulation and, most particularly, policies on the availability and scope of patent protection. Patents involve costly government-approval processes, and the standards vary. The government explains:

The independence of national patent systems ... is one of the defining principles of the international legal regime governing the protection of inventions. The United States has ratified the Paris Convention for the Protection of Industrial Property, originally adopted more than a century ago, which specifically provided in Article 4bis that “Patents applied for in the different contracting States ... shall be independent of the

patents obtained for the same invention in the other States....” 32 Stat.1936 (Aug. 25, 1902). While international agreements facilitate the ability of inventors in one country to seek patent protection in others, the patents laws of each country are not reciprocal in their protections for particular inventions. As every patent attorney knows, the United States may issue a patent while another country denies protection for the same invention, or approves claims significantly different in scope.

U.S. Br. 15–16.

Copyrights are different. They generally spring into being without any government approval, and standards hardly vary compared to patenting standards. As the government says, “patent law is different from copyright law, under which authors automatically ‘enjoy copyright protection in nations across the globe’ pursuant to the Berne Convention for the Protection of Literary and Artistic Works.” U.S. Br. 16 (quoting Golan v. Holder, –––U.S. ––––, 132 S.Ct. 873, 878, 181 L.Ed.2d 835 (2012)); see Golan, 132 S.Ct. at 878 (referring to “Berne's 164 member states”). And there is no reason to think that the costs of copyright registration (not even a prerequisite to all copyright protection) are generally comparable to those of securing patent protection.

Given the varying standards, and the separate examination processes and fees, a U.S. patentee may choose not even to seek patent protection in particular foreign countries. And those seeking protection may not obtain it, or may not obtain protection comparable to that of the U.S. patent. In either event, foreign sales in such circumstances may not occur under protections likely to produce market returns comparable to the reward contemplated by our patent law. Such country-to-country differences in patent law, moreover, are only part of the likely differences affecting foreign sales, supplementing differences in economic circumstances and in governments' price and other non-patent regulations bearing on sales.

For those reasons, a foreign sale, standing alone, is not reasonably viewed as providing the U.S. patentee the reward guaranteed by U.S. patent law. Such a sale is not reasonably viewed as itself a waiver by the patentee of its U.S. patent rights to prevent the buyer or others from bringing that article into the United States and selling or using it to satisfy a U.S.-market demand that the patentee could otherwise help satisfy at U.S.-market prices, as guaranteed by the Patent Act.

2

The only Supreme Court decision directly addressing the effect of a foreign sale on U.S. patent rights is the 1890 decision in Boesch v. Graff, 133 U.S. 697, 10 S.Ct. 378. Albert Graff and J.F. Donnell, by assignments, held an 1883 U.S. patent on certain lamp burners. Without their permission, Emile Boesch and Martin Bauer sold patent-covered burners in the United States. Boesch and Bauer had bought those burners from a supplier in Germany, without permission from the holders of the rights under German patents (dated 1879–1880), which had been issued to the same individuals as the U.S. patent. The German supplier was authorized to make the sale to Boesch and Bauer, not by the German patentees, but by a German law that allowed continuation of certain preparatory activities that began before the application for the German patents was filed. 133 U.S. at 698–99, 701–02, 10 S.Ct. 378. When Graff and Donnell sued Boesch and Bauer for infringement, the single-judge circuit court for the Northern District of California found infringement, and the Supreme Court affirmed that holding. Id.

The Court rejected Boesch and Bauer's defense that they “could not be held for infringement, because they purchased the burners in Germany from a person having the right to sell them there, though not a licensee under the German patents.” Id. at 699, 10 S.Ct. 378. The Court stated “the exact question presented [a]s whether a dealer residing in the United States can purchase in another country articles patented there, from a person authorized to sell them, and import them to and sell them in the United States, without the license or consent of the owners of the United States patent.” Id. at 702, 10 S.Ct. 378. In answering the question, the Court recited the then-leading decisions on domestic exhaustion, culminating in Adams, which, the Court observed, held that, within the confines of the United States, the courts would not add an unexpressed territorial limit to the rights conferred by an unrestricted sale by a regional assignee. Id. at 703, 10 S.Ct. 378. The Court then concluded that the cross-border situation was different. Its full rationale was as follows:

The right which [the German seller] had to make and sell the burners in Germany was allowed him under the laws of that country, and purchasers from him could not be thereby authorized to sell the articles in the United States in defiance of the rights of patentees under a United States patent. A prior foreign patent operates under our law to limit the duration of the subsequent patent here, but that is all. The sale of articles in the United States under a United States patent cannot be controlled by foreign laws.

Id.

That rationale by its terms does not make relevant whether the foreign sale was made under a foreign patent. Indeed, the second sentence says that a “prior foreign patent” does not cause loss of U.S. patent rights. Rather, the rationale turns only on the fact that the foreign sale was made under foreign law. The last sentence states the territorial principle: “The sale of articles in the United States under a United States patent cannot be controlled by foreign laws.” Id.

That principle does not preclude an accused infringer from establishing that the U.S. patentee actually gave it a license, expressly or by implication. It means that the exhaustion doctrine does not treat a foreign sale, lawful abroad for whatever reason, as having the cross-border legal effect of authorizing otherwise-infringing U.S. acts involving the purchased article. And that is how the principle has been understood by the Supreme Court. See

Keeler, 157 U.S. at 664–65, 15 S.Ct. 738 (“The exact question presented was whether a dealer residing in the United States could purchase in another country articles patented there from a person authorized there to sell them, and import them to and sell them in the United States without the license or consent of the owners of the United States patent, and the court held that the sale of articles in the United States under a United States patent cannot be controlled by foreign laws. In this case neither the patentee nor any assignee had ever received any royalty or given any license to use the patented article in any part of the United States.”); A. Bourjois & Co. v. Katzel, 260 U.S. 689, 692, 43 S.Ct. 244, 67 L.Ed. 464 (1923) (trademark case, explaining: “Ownership of [particular] goods ... does not necessarily carry the right to sell them at all in a given place. If the goods were patented in the United States a dealer who lawfully bought similar goods abroad from one who had a right to make and sell them there could not sell them in the United States. Boesch....”).

The principle of Boesch, precluding foreign control of U.S. rights, has a mirror-image counterpart in the territoriality principle of U.S. patent law that broadly denies projection of U.S. patent rights to cover foreign conduct. In Brown v. Duchesne, 60 U.S. (19 How.) 183, 15 L.Ed. 595 (1857), the Supreme Court, referring to the patent statutes, said that “these acts of Congress do not, and were not intended to, operate beyond the limits of the United States.” Id. at 195. The Court also explained the guaranteed reward from domestic markets: the patent laws “secure to the inventor a just remuneration from those who derive a profit or advantage, within the United States, from his genius and mental labors.” Id.; see, e.g., Dowagiac, 235 U.S. at 650, 35 S.Ct. 221 (“The right conferred by a patent under our law is confined to the United States and its territories (Rev.Stat. § 4884, Comp.Stat.1913, § 9428), and infringement of this right cannot be predicated of acts wholly done in a foreign country.”).

The Supreme Court relied on the strength of the territorial principle in Deepsouth Packing Co. v. Laitram Corp., 406 U.S. 518, 92 S.Ct. 1700, 32 L.Ed.2d 273 (1972), which rejected a claim of infringement against Deepsouth just because the ultimate combination covered by the patent was made abroad, after Deepsouth shipped all the parts from the United States. The Court explained:

Our patent system makes no claim to extraterritorial effect; “these acts of Congress do not, and were not intended to, operate beyond the limits of the United States,” Brown v. Duchesne, 19 How., at 195 (1856), and we correspondingly reject the claims of others to such control over our markets. Cf. Boesch v. Graff, 133 U.S. 697, 703 (1890).

406 U.S. at 531, 92 S.Ct. 1700.

In Microsoft Corp. v. AT & T Corp., 550 U.S. 437, 127 S.Ct. 1746, 167 L.Ed.2d 737 (2007), the Supreme Court quoted the foregoing passage from Deepsouth as support for “[t]he traditional understanding that our patent law ‘operate[s] only domestically and do[es] not extend to foreign activities,’ ... [which] is embedded in the Patent Act itself, which provides that a patent confers exclusive rights in an invention within the United States. 35 U.S.C. § 154(a)(1) (patentee's rights over invention apply to manufacture, use, or sale ‘throughout the United States' and to importation ‘into the United States').” Id. at 455, 127 S.Ct. 1746. And the Court added:

As a principle of general application, moreover, we have stated that courts should “assume that legislators take account of the legitimate sovereign interests

of other nations when they write American laws.” F. Hoffmann–La Roche Ltd. v. Empagran S. A., 542 U.S. 155, 164 [124 S.Ct. 2359, 159 L.Ed.2d 226] (2004); see EEOC v. Arabian American Oil Co., 499 U.S. 244, 248 [111 S.Ct. 1227, 113 L.Ed.2d 274] (1991). Thus, the United States accurately conveyed in this case: “Foreign conduct is [generally] the domain of foreign law,” and in the area here involved, in particular, foreign law “may embody different policy judgments about the relative rights of inventors, competitors, and the public in patented inventions.” Brief for United States as Amicus Curiae 28.

Id. (emphasis added). The Court then applied the presumption of congressional respect for territorial limits in patent law to interpret the very provision, § 271(f), that Congress had enacted (in 1984) to supersede Deepsouth, explaining that the presumption “remains instructive in determining the extent of the statutory exception” to the strict territorial limits elsewhere stated in the statute. Id. at 456, 127 S.Ct. 1746.

The principles thus expressed, perhaps especially the sentence highlighted in the quote just above, recognize what we noted above: Patent law is especially territorial, and laws vary considerably from country to country. The Supreme Court's recognition of those points reinforces our conclusion that foreign markets are not the predictable equivalent of the American markets in which the U.S. patentee is given a right to exclude and the rewards from that exclusivity. The Court's closest patent-law precedents thus support our holding that sales in foreign markets should not be presumed to confer on the buyer authority to displace sales in American markets.

3

Congress has not enacted a general provision of the Patent Act specifically addressed to foreign-sale exhaustion of U.S. patent rights. Congress has left the general issue to judicial resolution.

When the subject arose in the Uruguay round of multilateral negotiations that led to the Agreement on Trade–Related Aspects of Intellectual Property Rights (TRIPS), the parties agreed not to address the subject, stating, in article 6, that “nothing in this Agreement shall be used to address the issue of the exhaustion of intellectual property rights.” TRIPS, Apr. 15, 1994, 33 I.L.M. 1197, 1200, quoted in Kirtsaeng, 133 S.Ct. at 1383 (Ginsburg, J., dissenting). And when Congress implemented the international agreement through the 1994 legislation that (among other things) added the new importation ban to § 271(a), the accompanying Statement of Administrative Action—which Congress deemed authoritative, 19 U.S.C. § 3512(d)—stated that “[t]he Agreement ... does not affect U.S. law or practice relating to parallel importation of products protected by intellectual property rights.” H.R.Rep. No. 103–316, at 633, 981 (1994), reprinted in 1994 U.S.C.C.A.N. 4040, 4280. The recent Trans Pacific Partnership agreement includes a similar disclaimer, reserving the parties' rights to make other international agreements on the subject. See Trans–Pacific Partnership art. 18.11 & n. 8, Oct. 5, 2015, https://ustr.gov/trade-agreements/free-trade-agreements/trans-pacificpartnership/tpp-full-text.

Congress did act in three specific instances formally to guarantee a U.S. patentee the right to retain its U.S. rights despite selling abroad. Congress so provided through legislation, adopted by both houses and signed by the President, that approved three international agreements. United States–Morocco Free Trade Agreement Implementation Act, Pub.L. No. 108–302, 118 Stat. 1103 (2004); United States–Australia Free Trade Agreement Implementation Act, Pub.L. No. 108–286, 118 Stat. 919 (2004); United States–Singapore Free Trade Agreement Implementation Act, Pub.L. No. 108–78, 117 Stat. 948 (2003). In doing so, Congress did not provide the promised rights other than through the existing Patent Act provisions of §§ 154, 271.

Those congressionally approved guarantees would be negated if Impression's view of the Patent Act were adopted: U.S. patentees would lose their U.S. patent rights by selling abroad. An interpretation of a statute that produces such a contradiction with other enactments is to be avoided, at least where other considerations already point against such an interpretation. See FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 143, 120 S.Ct. 1291, 146 L.Ed.2d 121 (2000); Vimar Seguros y Reaseguros, S.A. v. M/V Sky Reefer, 515 U.S. 528, 539, 115 S.Ct. 2322, 132 L.Ed.2d 462 (1995); W. Va. Univ. Hosps., Inc. v. Casey, 499 U.S. 83, 100, 111 S.Ct. 1138, 113 L.Ed.2d 68 (1991). The three congressional enactments thus provide a further reason to reject Impression's view. At the same time, they leave to our internal law—the Patent Act, as judicially interpreted—whether even a presumptive-exhaustion rule governs. The agreements say nothing to undermine our reasons for rejecting a presumptive-exhaustion rule.

The only other legislative enactment presented for our consideration is 21 U.S.C. § 381(d)(1)-(2). Paragraph (1) of that subsection states a general rule that “no drug subject to section 353(b) of this title [concerning prescription-necessitating drugs] or composed wholly or partly of insulin which is manufactured in a State and exported may be imported into the United States unless the drug is imported by the manufacturer of the drug.” The general rule is subject to one exception, stated in paragraph (1), for certain prescription drugs imported by pharmacists and wholesalers from Canada, as regulated under 21 U.S.C. § 384. And it is subject to a second exception stated in paragraph (2), which authorizes the Secretary of Health and Human Services to permit importation otherwise within the paragraph (1) ban “if the drug is required for emergency medical care.” 21 U.S.C. § 381(d)(2).

That provision does not alter our conclusion. The provision does not purport to limit patentees' rights regarding importations under 35 U.S.C. §§ 154, 271. It adds an express government-enforced ban on certain importations, and it makes certain exceptions to the added ban, authorizing the Secretary to allow certain importations. Perhaps where the Secretary does so, an injunctive remedy might be unavailable to the patentee under 35 U.S.C. § 283, for public-interest reasons. See eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388, 126 S.Ct. 1837, 164 L.Ed.2d 641 (2006). But nothing in 21 U.S.C. § 381(d) makes non-infringing any conduct that otherwise would be infringing.

Congress may modify patentees' rights under the Patent Act. It may do so with respect to particular articles, without modifying the general exhaustion rule for foreign sales under the Patent Act—though § 381(d) does not do even that. Or it may more generally prescribe a general exhaustion rule for patented articles, specifying the conditions for exhaustion, as it did in the Copyright Act for copyrighted works. But it has not done that either.

4

Our no-exhaustion conclusion—which leaves undisturbed the availability of an express- or implied-license defense to infringement—is broadly consistent with the decisions of courts other than the Supreme Court, with the apparent exception of a trial-court decision that pre-dates Boesch.

The pre-Boesch decision is Holiday v. Mattheson, 24 F. 185 (C.C.S.D.N.Y.1885), in which few facts are set out. The defendants bought some U.S.-patented article in England from “a vendee of the patentee,” “without restriction or conditions.” Id. at 185. The court denied the patentee's motion for a preliminary injunction against U.S. activities involving the article. It reasoned that, whether or not an article is patented, “[w]hen the owner sells an article without any reservation respecting its use, or the title which is to pass,” “[t]he presumption arising from such a sale is that the vendor intends to part with all his rights in the thing sold”; and a patentee-seller “parts with his monopoly” as to that article—“unless by the conditions of the bargain the monopoly right is impressed upon the thing purchased,” i.e., unless “the owner of a patent sells the patented article under circumstances which imply that the purchaser is not to acquire an unqualified property in the thing purchased.” Id. at 185–86. That description, with its emphasis on the absence or presence of patentee-conveyed restrictions on post-purchase use, is taken entirely from domestic exhaustion law. The court said nothing to recognize that a distinct issue is presented when the sale was made abroad; and the opinion, describing few facts, does not make clear even indirectly if the circumstances would have given rise to an implied-license defense. In any event, just a few years after the trial-court decision in Holiday, the Supreme Court in Boesch made clear how much the crossing of international boundaries matters.

After Boesch, the Second Circuit in Dickerson v. Matheson, 57 F. 524 (2d Cir.1893), affirmed a finding of infringement against Matheson & Co., which had acquired from a German seller in 1887, and brought into the United States for sale and use here, batches of a coloring agent that was subject to a U.S. patent and a German patent, both assigned to the Bayer Company. The German seller (the Berlin Company) was a licensee of the Bayer Company, with the right to sell both in Europe and the United States, and it made clear that importation into the United States was prohibited. Id. at 525–26. In the suit brought by Dickerson, to whom the Bayer Company assigned the U.S. patent in 1888, the Second Circuit rejected Matheson's defense to infringement. It read Boesch to establish that “[a] purchaser in a foreign country of an article patented in that country and also in the United States, from a licensee under the foreign patent only, does not give the purchaser a right to import the article into, and to sell it in, the United States, without the license or consent of the owner of the United States patent.” Id. at 527.

The Eighth Circuit reached a similar result in Dickerson v. Tinling, 84 F. 192 (8th Cir.1897), involving Bayer & Co.'s phenacetine product. The court noted that “it appears that no patent [on the product] had ever been issued in Germany” and that “every package of phenacetine that had ever been sold by Bayer & Co. in a foreign country had a prohibition against its importation into and sale within the United States printed upon it, and was sold subject to that prohibition.” Id. at 193. It was unclear whether Tinling bought the phenacetine at issue from Bayer & Co. (or its vendees) or from “others,” but it did not matter to the outcome. Id. at 194. “If he bought it of others than Bayer & Co. or their vendees, he bought with it no right to sell it in the United States, because no one but Bayer & Co. and their vendees had that right in this country.” Id. “On the other hand, if [Tinling] bought the phenacetine he is selling in a foreign country from Bayer & Co., or from its vendees, subject to the express condition that it should not be imported into the United States, or sold within their limits, the exclusive right to sell the patented article within the United States which was granted to Bayer & Co. by the patent was not abridged by that purchase.” Id.

The Eighth Circuit pointedly noted that it did not have to decide what the result would be if no restrictions attended a sale made or approved by Bayer. It said: “Conceding—but not deciding—that one who buys a patented article without restriction in a foreign country from the owner of the United States patent” is clear of the U.S. patent for domestic sale and use, id. at 195 (citing Holiday and Matheson ), “there can be no doubt that a patentee has the same right and power to sell the patented article upon conditions or with restrictions that he has to sell it at all,” id. With Bayer having “sold on the express condition that [the phenacetine] should not be imported into or sold within the United States,” Tinling's domestic sale of the purchased product was infringing. Id. The Eighth Circuit thus reversed the trial court's denial of an injunction and ordered an injunction to issue. Id.

The Second Circuit likewise reversed the denial of infringement relief in Daimler Manufacturing Co. v. Conklin, 170 F. 70 (2d Cir.1909). The U.S. holder of certain automobile-component patents (Daimler) sought to enjoin the use in the United States of a vehicle containing such components. Conklin had bought the vehicle in Europe, under no restrictions as to importation into or use within the United States, from a company licensed to sell it in Europe by the holder of European patent rights—a company distinct from the U.S. patent-holding company, though with common origins and some overlapping ownership involving the inventor Maybach, see Daimler Manufacturing Co. v. Conklin, 160 F. 679 (C.C.S.D.N.Y.1908). The Second Circuit, based on Boesch, concluded: “The use of articles covered by a United States patent within the United States can no more be controlled by foreign law than the sale can. The sale by a German patentee of a patented article may take it out of the monopoly of the German patent, but how can it take it out of the monopoly of the American patentee who has not sold?” 170 F. at 72.

In 1920, the Second Circuit affirmed the denial of relief where it was clear from the circumstances that the U.S. patentee had granted permission for otherwise-infringing U.S. activities with airplanes bought in Canada. Curtiss Aeroplane & Motor Corp. v. United Aircraft Eng'g Corp., 266 F. 71 (2d Cir.1920). Curtiss was the holder of U.S. patents on various airplane-engine technologies. During World War I, an 83–percent–owned Canadian subsidiary of Curtiss (which the court treated as indistinguishable from Curtiss) granted a license—covering its Canadian patents and applications and any further inventions it owned or controlled involving changes to the engines at issue—to an entity created by the British government, authorizing the latter to make airplanes for sale to and use by the British government. Id. at 72–74. The British government bought planes during the war and, after the war ended, sold some of them to United Aircraft, which brought them into the United States for sale and use here. Id. at 72, 74. When Curtiss sued, the dispute was over whether “the authorization to make was general and unrestricted or subject to qualification and conditions, as to the disposition of the planes by the British government.” Id. at 77; id. at 75.

The Second Circuit, agreeing with the district court, concluded that the authorization gave the British government freedom from U.S. patent constraints on what it could do with the planes. The court relied on “the very nature of things” and “the language used in the agreements.” Id. at 75. It explained that “[a]n aeroplane has been said to be the most mobile article manufactured, and it is not confined by geographical boundaries,” id.; that the British had used airplanes in numerous countries during the war, id.; and that “the aviation fields in Texas and in other states were placed at the disposal of the British authorities and were actually used by them as training fields for Canadian aviators,” id. It concluded: “[Curtiss] and the British government alike understood and intended that the aeroplanes to be manufactured by that government as well as those to be supplied to it by [Curtiss] were to become the absolute property of the government, and were to be disposed of as the latter should see fit. The express language of the contract is that the aeroplanes and other articles should ‘become and be the absolute property of the British government.’ ” Id.

Some decisions of district courts from decades later round out the picture of case law predating Jazz Photo. Judge Lord rejected an exhaustion defense in Griffin v. Keystone Mushroom Farm, Inc., 453 F.Supp. 1283 (E.D.Pa.1978). Griffin was the owner of the U.S. patent, as well as Italian patents, covering certain machinery. Keystone bought several machines in Italy from Griffin's exclusive licensee in Italy and brought them into the United States, one for use, two for sale. Griffin sued Keystone for infringement, and Keystone sought summary judgment based on exhaustion. Judge Lord rejected the defense.

He read Boesch to apply, because in Boesch “[t]he source of the alleged infringer's authorization under foreign law ... was without significance in the Court's reasoning.” Id. at 1285. Therefore, it did not matter whether Griffin “owned concurrent United States and Italian patents and had entered into analogous licensing agreements concerning the same inventions,” giving Griffin a share of royalties from the Italian and American exclusive licensees. Id. “[T]he basic thrust of the Boesch decision” was “that the ‘sale of articles in the United States under a United States patent cannot be controlled by foreign laws.’ ” Id. at 1286 (quoting 133 U.S. at 703, 10 S.Ct. 378).

In Sanofi, S.A. v. Med–Tech Veterinarian Products, Inc., 565 F.Supp. 931 (D.N.J.1983), Judge Sarokin denied a preliminary injunction to U.S. patentee Sanofi, S.A., but granted one to U.S. exclusive licensee American Home Products. A Sanofi subsidiary in France sold to an American processor certain pharmaceutical products covered by Sanofi's U.S. patent; the subsidiary “placed no restrictions in the sales contract,” and Sanofi had no French patent. Id. at 938; see also id. at 934–35. When the buyer brought the products to the United States for sale, both Sanofi and American Home Products sued.

The court concluded first that “if Sanofi were permitted to impose restrictions upon the resale of its patented product, the expectations of the purchaser would be defeated.” Id. at 938 (emphasis added). “[W]here the owner of a patent exhibits conduct from which one dealing with him may properly infer that the owner consents to his use of the patent, an implied license will arise.” Id. at 940 (citing De Forest Radio, 273 U.S. at 241, 47 S.Ct. 366). But a different conclusion was required as to American Home Products, the U.S. exclusive licensee, the court reasoned, which did not cede its patent rights. “Because the purchaser is under an obligation to inquire of the seller as to the existence of any outstanding licenses, the purchaser cannot claim that his expectations have been frustrated if he fails to make the necessary inquiry and later discovers that an outstanding license interferes with his right of enjoyment.” Id. “If the court were to hold that Sanofi's sale of the product exhausted the patent, it would be crediting Sanofi with greater rights than the patentee actually had. Sanofi had no right to allow its product to enter this country without the permission of its exclusive licensee.” Id. at 941.

All of the foregoing decisions after Boesch reflect both (a) the Boesch principle that foreign laws do not control domestic patent rights and (b) some assessment of the particular circumstances and language of foreign sales to determine if the U.S. patentee gave permission for importation. The pre-Boesch decision in Holiday aside, the results accord with the Jazz Photo no-exhaustion rule coupled with the availability of a defense based on an express or implied license. That combination of principles, supported in the statute and Supreme Court doctrine, provides a clear doctrinal statement that fits the pre-Jazz Photo case law from outside the Supreme Court.

5

Finally, we consider what we can reliably gauge about the likely real-world consequences of one answer or another to the exhaustion question presented here. As on the first issue before us, the amicus briefs filed here present competing arguments about the effect of one foreign-sale exhaustion rule or another on their interests and the interests they promote, offering varying amounts of empirical support. Such arguments necessarily play a much more limited role for us than they might for Congress. As on the first issue, all that we can conclude is that we see no basis for altering the conclusion we think warranted by the legal sources already considered.

a. We have not been shown that substantial problems have arisen with the clear rule of Jazz Photo, which has been in place since 2001, or with the comparable legal understandings based on a century of case law in the area. There is, of course, the possibility—noted by the Dissent at 787, citing amici's assertions—of unintended infringement by buyers of goods in foreign countries who bring them into the United States, whether to use them as components in new goods they make, to sell them, or to use them as consumers. But that possibility is limited by the availability of an implied-license defense from the circumstances of a sale (perhaps, e.g., an unrestricted patentee sale at a seaport or airport to a buyer loading or boarding a vessel or plane bound for the United States). In addition, a large share of such possible unintended infringement, according to the most common policy complaint by electronics-industry amici, is by definition immaterial to any exhaustion—namely, infringement of patents asserted by non-practicing entities that have neither made nor authorized the sale of patent-covered articles. The only scenario relevant to exhaustion is one involving patentee-made or -authorized foreign sales, and we simply have no reliable evidence that the possibility of unintended infringement in that scenario is actually a significant issue in practice. The absence of such evidence in the many years since Jazz Photo, and the still longer period since Boesch, provides good reason to think otherwise.

Indeed, it has long been a feature of the patent-law landscape that there can be instances of innocent infringement, because § 271(a) sets a “strict-liability” standard. Commil USA, LLC v. Cisco Systems, Inc., –––U.S. ––––, ––––, 135 S.Ct. 1920, 1926, 191 L.Ed.2d 883 (2015). Thus, even domestic purchasers of products from domestic sellers who have not obtained authority from the owners of patents covering the products' components could find themselves in that position. But Congress has left strict liability in place, even in light of the scenario not relevant to exhaustion, i.e., patent infringement claimed by non-practicing-entity patentees that have neither made nor authorized the sales at issue. In any event, despite the law in place since Jazz Photo and for decades earlier, there is no reason to think that this is a distinctive problem for foreign-purchased goods, much less a problem affecting a meaningful share of foreign sales leading to imports.

In this respect, we have no reason to think that the most serious real-world problems described in Kirtsaeng carry over to the patent arena. Prominent among the problems in Kirtsaeng were those that would be faced, under the rejected interpretation of § 109(a), by libraries, museums, and bookshops. Those entities often would be dealing on a regular basis with changing inventories of large numbers of individually distinct long-shelf-life works subject to copyrights that have multiple owners and that last for periods far longer than the terms of patents (and variable with the life of the authors). See Eldred, 537 U.S. at 194–96, 123 S.Ct. 769 (describing copyright terms). And there was good reason to think that they built up “deeply embedded” reliance interests in the absence of clear law pointing against § 109(a)'s applicability just because a work was made abroad. Kirtsaeng, 133 S.Ct. at 1366. If there is a counterpart to such situations in the patent arena, it has not been shown to loom large in the full range of circumstances governed by the answer to the question of foreign-sale-exhaustion.

b. Overturning Jazz Photo would plausibly cause significant disruption of existing practices adopted under the contrary law established by Jazz Photo and decades of prior case law. Such disruption is most likely if exhaustion of U.S. rights were held to follow from a foreign sale without the U.S. patentee having the ability to reserve its U.S. rights. While a conclusive-exhaustion rule is opposed by the government, it is the rule urged by Impression and certain amici that stress the possibility of unintended infringement we have just discussed.

An example of likely disruption involves pharmaceutical products. There seems to be no dispute that U.S.-patented medicines are often sold outside the United States at substantially lower prices than those charged here and, also, that the practice could be disrupted by the increased arbitrage opportunities that would come from deeming U.S. rights eliminated by a foreign sale made or authorized by the U.S. patentee. One official recognition of both the fact of low prices abroad and the linkage of such prices to territorial resale protection appears in a 2003 World Trade Organization decision made with the agreement of the United States. The WTO there waived certain TRIPS patent-recognition provisions in order to allow certain countries to import generic versions of needed medicines. The WTO took care, however, to condition the waiver on agreement by the importing countries “to control re-exportation of drugs they import in this fashion.” Ganslandt & Maskus, at 1036 (discussing WTO General Council, Implementation of paragraph 6 of the Doha Declaration on the TRIPS Agreement and Public Health, Aug. 30, 2003, www.wto. org/english/tratop_e/trips_e/implem_para6_e.htm). Reversing Jazz Photo and replacing it with a conclusive-exhaustion rule would likely upset such established practices.

c. A presumptive-exhaustion rule, subject to some kind of preservation of U.S. rights by the U.S. patentee when making or authorizing a foreign sale, would be less consequential. After all, to try to negate a potential implied-license defense, U.S. patentees would have an incentive to make express reservations of U.S. rights in making or authorizing foreign sales, simply to make clear that no license was being conferred. But even for the U.S. patentees that recognize the incentive and try to act on it, whether there is a presumptive loss of U.S. rights makes a difference. In particular, it makes a difference—though we cannot say just how significant—who has the burden of proof on the issue: must the patentee prove a reservation (communicated to the accused infringer) to avoid exhaustion, or must the accused infringer prove a license?

A U.S. patentee that wishes to reserve its U.S. rights may not be able to do so. For a foreign sale, the required reservation is an act in a foreign country. And the foreign sovereign, or local governments in the country, may prohibit sellers from stating reservations of rights that would make importation into and sale in the United States more difficult.

A presumptive-exhaustion rule would place a U.S. patentee's preservation of U.S. rights within foreign sovereign control. For doctrinal reasons already emphasized, we should avoid attributing to Congress such a ceding of control over domestic rights to foreign sovereigns without clearer reason than we have seen here. The Supreme Court's final statement of its rationale in Boesch says as much: “The sale of articles in the United States under a United States patent cannot be controlled by foreign laws.” 133 U.S. at 703, 10 S.Ct. 378. Indeed, such foreign control of U.S. rights is a mirror image of projecting U.S. patent rights into foreign sovereigns' territories. The Supreme Court has long recognized that the latter is strongly disfavored in reading the Patent Act. See pages 763–65, supra. And since Boesch, the Court has twice recognized the symmetric impropriety of reading the Patent Act to allow projection of foreign sovereigns' decisions to control rights in U.S. territory: “Our patent system makes no claim to extraterritorial effect; ‘these acts of Congress do not, and were not intended to, operate beyond the limits of the United States,’ Brown v. Duchesne, 19 How., at 195; and we correspondingly reject the claims of others to such control over our markets. Cf. Boesch v. Graff, 133 U.S. 697, 703, 10 S.Ct. 378 (1890).” Deepsouth, 406 U.S. at 531, 92 S.Ct. 1700; see Microsoft, 550 U.S. at 455, 127 S.Ct. 1746.

In practical terms, moreover, there is a plausible problem with adopting a presumptive-exhaustion rule, compared to leaving the matter to express- or implied-license doctrine. Intermediary companies between the foreign purchase and the importation into the United States may be created that make it difficult for the U.S. patentee to carry an affirmative burden of proving adequate notice of reservations attached to a foreign-sold article. Once the article leaves the hands of the initial seller (the U.S. patentee or its authorized seller), the U.S. patentee seems likely to have limited knowledge about the movement of the article to U.S. markets, through what may be multiple hands. On the other hand, if the burden is on the U.S. importer/seller to establish a conferral of authority, as it is under the express- or implied-license doctrine, there would be incentives to communicate a conferral of authority reliably throughout the chain of custody on the way to the U.S. importer and seller. That is because the latter, at the end of supply chain, would have the incentive to insist on ultimately receiving such information in order to establish the license defense.

A related point may be made about the reasonable expectations of a potential U.S. reseller of goods acquired abroad in sales made or authorized by a U.S. patentee. As to the reseller's freedom from the patentee's U.S. rights, the difference between a rule leaving the matter to the reseller's affirmative proof of a license (express or implied) and a rule of presumptive exhaustion (subject to disproof by the U.S. patentee) is significant just when there are genuine uncertainties about whether a license could be established. But in that situation the reseller is not entitled to a strong expectation that it has permission to conduct its otherwise-infringing activities in the United States.

Conclusion

We hold that, when a patentee sells a patented article under otherwise-proper restrictions on resale and reuse communicated to the buyer at the time of sale, the patentee does not confer authority on the buyer to engage in the prohibited resale or reuse. The patentee does not exhaust its § 271 rights to charge the buyer who engages in those acts—or downstream buyers having knowledge of the restrictions—with infringement. We also hold that a foreign sale of a U.S.-patented article, when made by or with the approval of the U.S. patentee, does not exhaust the patentee's U.S. patent rights in the article sold, even when no reservation of rights accompanies the sale. Loss of U.S. patent rights based on a foreign sale remains a matter of express or implied license.

Under our first holding, we reverse the district court's judgment of non-infringement as to the Return Cartridges first sold in the United States. Under our second holding, we affirm the district court's judgment of infringement as to the cartridges first sold abroad. The case is remanded for entry of a judgment of infringement for Lexmark and for any further proceedings necessary upon entry of such judgment.

Costs awarded to Lexmark.

AFFIRMED IN PART, REVERSED IN PART, AND REMANDED

DYK, Circuit Judge, dissenting, with whom Circuit Judge HUGHES joins.

I respectfully dissent from the majority's holding that Mallinckrodt, Inc. v. Medipart, Inc., 976 F.2d 700 (Fed.Cir.1992), and Jazz Photo Corp. v. International Trade Commission, 264 F.3d 1094 (Fed.Cir.2001), remain good law. First, I agree with the government that Mallinckrodt was wrong when decided, and in any event cannot be reconciled with the Supreme Court's recent decision in Quanta Computer, Inc. v. LG Electronics, Inc., 553 U.S. 617, 128 S.Ct. 2109, 170 L.Ed.2d 996 (2008). We exceed our role as a subordinate court by declining to follow the explicit domestic exhaustion rule announced by the Supreme Court.

Second, I would retain Jazz Photo insofar as it holds that a foreign sale does not in all circumstances lead to exhaustion of United States patent rights. But, in my view, a foreign sale does result in exhaustion if an authorized seller has not explicitly reserved the United States patent rights.

I. Domestic Exhaustion

A

Both here and in Mallinckrodt the patentee itself sold the patented item to the purchaser. In Mallinckrodt, “the device [was] manufactured by [the patent owner], who [sold] it to hospitals as a unitary kit.” 976 F.2d at 702. Here, as the majority recognizes, “Lexmark sells its cartridges ... directly to end users, and [ ] to ‘resellers' (including wholesalers, dealers, and distributors).” Maj. Op. at 728. Lexmark's sales of so-called “Return Program Cartridges” were subject to a single-use/no-resale restriction that barred the purchaser from reusing the cartridge, or transferring a used cartridge to anyone besides Lexmark. See Maj. Op. at 727 & n. 1. Those sales were authorized by the patent holder and transferred title to the purchaser.

Beginning in the 1850s, the Supreme Court recognized that such authorized sales exhaust the patentee's patent rights in the items sold. The patentee's right to exclude under the Patent Act expires with an authorized sale. The question of whether the seller has “authorized” the buyer to use or resell the item is simply irrelevant. The Court's language is unequivocal:

• “[W]hen the machine passes to the hands of the purchaser, it is no longer within the limits of the monopoly. It passes outside of it, and is no longer under the protection of the act of Congress.... Contracts in relation to it

are regulated by the laws of the State, and are subject to State jurisdiction.”

Bloomer v. McQuewan, 55 U.S. 539, 549–50 (1852).

• “[W]hen [patentees] have made and vended to others to be used one or more of the things patented, ... they have parted with their exclusive right.... By a valid sale and purchase the patented machine becomes the private individual property of the purchaser, and is no longer specially protected by the laws of the United States, but by the laws of the State in which it is situated.... [I]f a person legally acquires a title to that which is the subject of letters patent, ... he may repair it or improve upon it as he pleases, in the same manner as if dealing with property of any other kind.”

Bloomer v. Millinger, 68 U.S. 340, 350–52 (1863).

• “[W]hen [the patented article] rightfully passes from the patentee to the purchaser, [it] ceases to be within the limits of the monopoly.”

Mitchell v. Hawley, 83 U.S. 544, 548 (1872).

• “The true ground on which [McQuewan, Millinger, and Mitchell ] rest is that the sale by a person who has the full right to make, sell, and use such a machine carries with it the right to the use of that machine to the full extent to which it can be used in point of time.... [I]t is open to the use of the purchaser without further restriction on account of the monopoly of the patentees.”

Adams v. Burke, 84 U.S. 453, 455–56 (1873).

• “[W]hen the patentee ... sells a machine or instrument whose sole value is in its use, he receives the consideration for its use, and parts with the right to restrict that use.... [I]t is open to the use of the purchaser, without further restriction on account of the monopoly of the patentee....”

Hobbie v. Jennison, 149 U.S. 355, 361–62 (1893).

• “[O]ne who buys patented articles of manufacture from one authorized to sell them becomes possessed of an absolute property in such articles, unrestricted in time or place.”

Keeler v. Standard Folding Bed Co., 157 U.S. 659, 666 (1895).

• “[B]y virtue of the patent law one who had sold a patented machine and received the price and had thus placed the machine so sold beyond the confines of the patent law, could not by qualifying restrictions as to use keep under the patent monopoly a subject to which the monopoly no longer applied.”

Bos. Store of Chi. v. Am. Graphophone Co., 246 U.S. 8, 25 (1918).

• “[W]here a patentee makes the patented article, and sells it, he can exercise no future control over what the purchaser may wish to do with the article after his purchase. It has passed beyond the scope of the patentee's rights.”

United States v. Gen. Elec. Co., 272 U.S. 476, 489 (1926).

• “The first vending of any article manufactured under a patent puts the article beyond the reach of the monopoly which that patent confers.”

United States v. Univis Lens Co., 316 U.S. 241, 252 (1942).

Thus, by the mid–1850s and continuing for the next century, even before Quanta, the Supreme Court repeatedly held that the authorized sale of a patented article exhausted all of the patentee's patent rights in that article, and freed the article from any restrictions on use or sale based on the patent laws. Post-sale restrictions were enforceable only as a matter of state contract law.

B

The sole Supreme Court case to depart from that principle, Henry v. A.B. Dick Co., 224 U.S. 1 (1912), was explicitly overruled five years later by Motion Picture Patents Co. v. Universal Film Manufacturing Co., 243 U.S. 502, 518 (1917). See Quanta, 553 U.S. at 625–26 . In Henry v. A.B. Dick Co., the A.B. Dick Company sold a rotary mimeograph, and affixed to it a restriction stating that it could only be used with stencil paper, ink, and other supplies made by the patentee. 224 U.S. at 11 . The Supreme Court in A.B. Dick upheld that restriction, and, more broadly, held that

[t]he property right to a patented machine may pass to a purchaser with no right of use, or with only the right to use in a specified way, or at a specified place, or for a specified purpose. The unlimited right of exclusive use which is possessed by and guaranteed to the patentee will be granted if the sale be unconditional. But if the right of use be confined by specific restriction, the use not permitted is necessarily reserved to the patentee. If that reserved control of use of the machine be violated, the patent is thereby invaded.

Id. at 24–25, 32 S.Ct. 364 (emphasis added). The Court reasoned, in part, that the patent owner's “larger right” of excluding all others from using the patent “embraces the lesser of permitting others to use upon such terms as the patentee chooses to prescribe.” Id. at 35, 32 S.Ct. 364.

The holding of A.B. Dick, that a patent owner has the right to impose post-sale restrictions under the patent law, provided the purchaser has sufficient “notice that he buys with only a qualified right of use,” id. at 26, 32 S.Ct. 364, is the same as the panel's holding in Mallinckrodt and the majority's holding in this case.

A.B. Dick was quickly overruled in Motion Picture Patents, 243 U.S. at 518, 37 S.Ct. 416, which stands as compelling authority against the majority's conclusion. There, the licensee-manufacturer sold film projectors subject to an attached notice restricting their use to unpatented films made by the Motion Pictures Patents Company, and other restrictions “not stated in the notice, but which are to be fixed, after sale.” Id. at 505–09, 37 S.Ct. 416. When a purchaser used the projector to display films made by another company, the Motion Picture Patents Company sued for infringement. Id. at 508, 37 S.Ct. 416. The question was whether the restrictions were enforceable after the sale. The Court rejected the basic rationale of A.B. Dick that, since the “patentee may withhold his patent altogether from public use, he must logically and necessarily be permitted to impose any conditions which he chooses upon any use which he may allow of it,” id. at 514, 37 S.Ct. 416, and concluded that A.B. Dick “must be regarded as overruled,” id. at 518, 37 S.Ct. 416. Instead, the Court reaffirmed that “the right to vend is exhausted by a single, unconditional sale, the article sold being thereby carried outside the monopoly of the patent law and rendered free of every restriction which the vendor may attempt to put on it.” Id. at 516, 37 S.Ct. 416.

The majority attempts to distinguish Motion Picture Patents, on the ground that it only “held particular restrictions improper ... relying on the 1914 Clayton Act,” but “did not rule that all restrictions on a patentee's sale were ineffective to preserve the patentee's patent-law rights.” Maj. Op. at 749. That is not accurate. Motion Picture Patents did not leave behind the remnants of A.B. Dick—minus tie-ins and resale price maintenance. To the contrary, the Court in Motion Picture Patents found that “[t]he patent law furnishes no warrant for” the restrictions imposed by the patent owner. 243 U.S. at 516, 37 S.Ct. 416. The passage of the Clayton Act only “confirmed” the Patent Act holding reached in Motion Picture Patents. Id. at 517, 37 S.Ct. 416.

In later cases, the Court characterized Motion Picture Patents as having broadly settled the ineffectiveness of all post-sale restrictions under the patent law. In Boston Store of Chicago v. American Graphophone Co., Motion Picture Patents was viewed as “concern[ing] whether the monopoly of the patent law can be extended beyond the scope of that law or, in other words, applied to articles after they have gone beyond its reach.” 246 U.S. at 26, 38 S.Ct. 257 (emphasis added). The Court stated that Motion Picture Patents accordingly settled “the general question of the power of the patentee to sell and yet under the guise of license or otherwise to put restrictions which in substance were repugnant to the rights which necessarily arose from the sale which was made.” Id. at 24, 38 S.Ct. 257. Resting on patent exhaustion principles, Motion Picture Patents “decided that as by virtue of the patent law one who had sold a patented machine and received the price, and had thus placed the machine so sold beyond the confines of the patent law, could not by qualifying restrictions as to use keep under the patent monopoly a subject to which the monopoly no longer applied.” Id. at 25, 38 S.Ct. 257 (emphasis added).

In Quanta, the Court reiterated the broad patent exhaustion rule and left no room for a resurrection of A.B. Dick. LG Electronics (“LG”) owned system and method patents related to computer technology. Quanta, 553 U.S. at 621–22, 128 S.Ct. 2109. LG licensed Intel to manufacture microprocessors and chipsets that used the LG patents. Id. at 623, 128 S.Ct. 2109. The licensing agreement stipulated that no license was given to Intel's customers to combine the licensed Intel products with non-Intel components in ways that practiced the LG patents. Id. A separate master agreement required Intel to provide a notice to its customers that they were not licensed to practice the LG patents by combining Intel products with non-Intel products. Id. at 623–24, 128 S.Ct. 2109. Quanta purchased microprocessors and chipsets covered by the LG patents from Intel but combined them with non-Intel products to manufacture computers. LG filed suit against Quanta for patent infringement. Id. at 624, 128 S.Ct. 2109.

The Court found that the Intel products embodied the LG patents and that Intel had authority to sell its products to Quanta. Id. at 635, 636–37, 128 S.Ct. 2109. It then expansively held that “[t]he authorized sale of an article that substantially embodies a patent exhausts the patent holder's rights and prevents the patent holder from invoking patent law to control postsale use of the article.” Id. at 638, 128 S.Ct. 2109. Significantly, Quanta described Motion Picture Patents as having “reiterated the rule that ‘the right to vend is exhausted by a single, unconditional sale, the article sold being thereby carried outside the monopoly of the patent law and rendered free of every restriction which the vendor may attempt to put upon it.’ ” 553 U.S. at 626, 128 S.Ct. 2109 (quoting Motion Picture Patents, 243 U.S. at 516, 37 S.Ct. 416).

After Quanta, the Court confirmed again that the “doctrine of patent exhaustion limits a patentee's right to control what others can do with an article embodying or containing an invention. Under the doctrine, ‘the initial authorized sale of a patented item terminates all patent rights to that item.’ ” Bowman v. Monsanto Co., ––– U.S. ––––, 133 S.Ct. 1761, 1766, 185 L.Ed.2d 931 (2013) (quoting Quanta, 553 U.S. at 625, 128 S.Ct. 2109).

The patent exhaustion doctrine, as stated by Quanta, admits of no exception. Authorized sales “prevent[ ] the patent holder from invoking patent law to control postsale use.” Quanta, 553 U.S. at 638, 128 S.Ct. 2109.

Contrary to the majority, Quanta's reference to an “unconditional sale,” id. at 626, 128 S.Ct. 2109, a reference appearing as well in other exhaustion cases, can hardly be read to contradict the Court's central holding that post-sale restrictions are unenforceable under the patent laws. The language referring to “conditions” imposed on sale or “unconditional” sales is used in these cases in two different senses. On the one hand, there are cases in which such language is used to denote the existence of post-sale restrictions imposed by the patent holder. A.B. Dick and Motion Picture Patents fall into this category. A.B. Dick stated that exhaustion applied only if the sale was “unconditional[ ],” i.e., free of post-sale restrictions. 224 U.S. at 19, 32 S.Ct. 364. Motion Picture Patents, in overruling A.B. Dick, rejected the notion that a seller could impose “conditions,” i.e., restrictions on post-sale use. 243 U.S. at 514–15, 37 S.Ct. 416. The use of such language in those cases refutes the majority's theory, since Motion Picture Patents holds that conditions (i.e., restrictions) are not permissible under the patent laws.

In the few other cases that use the “unconditional sales” language, the reference to an “unconditional” sale is to a sale in which title passes, not to a sale in which no restrictions are imposed. The contemporaneous understanding of “conditional sale” was as a security device, i.e., an “agreement to sell upon a condition to be performed.” Harkness v. Russell, 118 U.S. 663, 665, 7 S.Ct. 51, 30 L.Ed. 285 (1886); see also Motion Picture Patents, 243 U.S. at 520–21, 37 S.Ct. 416 (Holmes, J., dissenting) (“[A] conditional sale retaining the title until a future event after delivery has been decided to be lawful again and again by this court.”).

That the use of the term “unconditional” in those cases is not referring to a sale without restrictions is crystal clear from Quanta itself, where the Court stated that Motion Picture Patents “reiterated the rule that ‘the right to vend is exhausted by a single, unconditional sale, the article sold being thereby carried outside the monopoly of the patent law and rendered free of every restriction which the vendor may attempt to put upon it.’ ” 553 U.S. at 626, 128 S.Ct. 2109 (quoting Motion Picture Patents, 243 U.S. at 516, 37 S.Ct. 416). In other words, a sale with restrictions could nonetheless be an “unconditional” sale in which title passes, with the restrictions invalid under the patent laws because of exhaustion.

C

The rule articulated in the Supreme Court's cases is consistent with the common law rule against restraints on the use or alienation of chattels, which formed the background of the patent statute. In Kirtsaeng v. John Wiley & Sons, Inc., 133 S.Ct. 1351, 1363 (2013), the Court noted, in the context of copyright law, that the “ ‘first sale’ doctrine is a common law doctrine” traceable to “the common law's refusal to permit restraints on the alienation of chattels.” The Court cited Lord Coke's 17th century observation that

[If] a man be possessed of ... a horse, or of any other chattel ... and give or sell his whole interest ... therein upon condition that the Donee or Vendee shall not alien[ate] the same, the [condition] is voi[d], because his whole interest ... is out of him, so as he hath no possibilit[y] of a Reverter, and it is against Trade and Traffi[c], and bargaining and contracting betwee[n] man and man....

Id. (quoting 1 Edward Coke, Institutes of the Laws of England § 360, at 223 (1628)). Kirtsaeng concluded that “[a] law that permits a copyright holder to control the resale or other disposition of a chattel once sold is similarly ‘against Trade and Traffi[c], and bargaining and contracting.’ ” 133 S.Ct. at 1363.

So too a rule permitting a patent holder to enact post-sale restraints would be contrary to the general common law. Post-sale restraints would “cast a cloud of uncertainty over every sale,” Tessera, Inc. v. Int'l Trade Comm'n, 646 F.3d 1357, 1370 (Fed.Cir.2011). The Supreme Court has repeatedly instructed us not to ignore traditional legal principles to fashion rules “unique to patent disputes.” eBay Inc. v.

MercExchange, L.L.C., 547 U.S. 388, 393, 126 S.Ct. 1837 (2006); see also Global–Tech Appliances, Inc. v. SEB S.A., 131 S.Ct. 2060, 2069 (2011); MedImmune, Inc. v. Genentech, Inc., 549 U.S. 118, 132 n. 11, 127 S.Ct. 764, 166 L.Ed.2d 604 (2007). We should decline to do so here. There is no indication in the patent laws that there should be a special exception for patent holders to the general longstanding common law doctrine that promotes free competition in the resale market and certainty in commercial transactions. Allowing the patent holder to impose conditions on the sale of a patented item would indeed largely eviscerate the exhaustion doctrine, by permitting the imposition of all manner of post-sale restrictions except for tie-ins, price-fixing, and other violations of the patent misuse and antitrust law.

D

The majority's justifications for refusing to follow Supreme Court authority establishing the exhaustion rule misconceive our role as a subordinate court.

First, the majority characterizes the statement of the exhaustion rule in the Supreme Court cases as mere dictum because in those cases there was either no restriction imposed or the restriction would otherwise violate the antitrust laws. But the cases impose no such qualification on the rule announced. The Supreme Court has repeatedly advised the courts of appeals that our task is to follow the rules proclaimed by the Court, and not to attempt to distinguish Supreme Court cases on their facts. See, e.g., Rivers v. Roadway Express, Inc., 511 U.S. 298, 312, 114 S.Ct. 1510, 128 L.Ed.2d 274 (1994) (“[O]nce the Court has spoken, it is the duty of other courts to respect that understanding of the governing rule of law.”); Thurston Motor Lines, Inc. v. Jordan K. Rand, Ltd., 460 U.S. 533, 534–35, 103 S.Ct. 1343, 75 L.Ed.2d 260 (1983) (per curiam) (A court of appeals must not “confus[e] the factual contours of [a Supreme Court decision] for its unmistakable holding” in an effort to reach a “novel interpretation” of that decision.).

Previously we have faithfully adhered to this rule. See Ariad Pharm., Inc. v. Eli Lilly & Co., 598 F.3d 1336, 1347 (Fed.Cir.2010) (en banc) (“As a subordinate federal court, we may not so easily dismiss [the Supreme Court's] statements as dicta but are bound to follow them.”); Stone Container Corp. v. United States, 229 F.3d 1345, 1349–50 (Fed.Cir.2000) (“[W]e do not share the Supreme Court's latitude in disregarding the language in its own prior opinions.”). We cannot appropriately depart from it here. Second, the majority relies on 35 U.S.C. §§ 271(a) and 154(a)(1) to suggest that a broad reading of the exhaustion doctrine is inconsistent with statutory language making an act of infringement, inter alia, any use or sale of a patented invention “without authority” of the patent owner, and providing the patent owner with a “right to exclude.” Maj. Op. at 732, 734. That reliance is misplaced. Patent exhaustion is a limit on the patentee's statutory right to control what purchasers can do with an article embodying or containing a patented invention. See Bowman, 133 S.Ct. at 1766 & n. 2 (recognizing that patent exhaustion removes restrictions imposed by §§ 271(a) and 154(a)(1)). The focus of patent exhaustion is not whether the buyer has been expressly or impliedly authorized to sell or use a product in a certain way after the sale. Instead, it begins and ends with an inquiry of whether the seller had authorization to make a sale. The exhaustion doctrine is simply a limit on the scope of the patent monopoly, that is, a limit on the exclusive rights of the patentee. The right to exclude expires (or is “exhausted”) by an authorized sale.

Third, the majority claims that giving full sweep to the articulation of the exhaustion doctrine in Quanta and other cases would be inconsistent with the Supreme Court's decision in General Talking Pictures Corp. v. Western Electric Co., 304 U.S. 175, 58 S.Ct. 849 (1938), aff'd on reh'g, 305 U.S. 124, 59 S.Ct. 116 (1938). The majority asserts that General Talking Pictures “held that a patentee can preserve its patent rights by authorizing a manufacturing licensee to make and sell a patented article under an otherwise-proper restriction, including a restriction on the buyer's post-purchase use.” Maj. Op. at 744. The majority suggests it would be incongruous if “a patentee cannot preserve its patent rights against uses of a patented article ... if, instead of licensing someone else to make and sell the article, it chooses to make and sell the article itself.” Id. The majority urges there is “no sound legal basis” for distinguishing restrictions on a purchaser from restrictions on a licensee. Id. at 726.

In General Talking Pictures, a patent owner granted a non-exclusive license to a licensee to manufacture and sell patented sound amplifier products. 304 U.S. at 180, 58 S.Ct. 849. The license contained a field-of-use restriction: the licensee could only make and sell amplifiers for non-commercial use. Id. Nonetheless, in violation of the license terms, the licensee made and sold the products knowing that they were to be used in a commercial theater, and the buyer had actual knowledge that the licensee lacked authority to make such a sale. Id. The Court stated the “controlling facts” as, “[t]he patent owner did not sell to petitioner the amplifiers in question or authorize [the licensee] to sell them or any amplifiers for use in theaters or any other commercial use. The sales made by the [licensee] were outside the scope of its license and not under the patent.” Id. at 180, 58 S.Ct. 849. There had been no authorized first sale, for the licensee “could not convey to [the ultimate purchaser] what both knew it was not authorized to sell,” and thus both were liable for infringement. Id. at 181–82, 58 S.Ct. 849.

There is nothing anomalous about General Talking Pictures. The Supreme Court has clearly distinguished between sales and licenses, holding that while a patentee cannot impose post-sale restrictions on an authorized sale, it can impose restrictions on a licensee. See Gen. Elec., 272 U.S. at 489–90, 47 S.Ct. 192; McQuewan, 55 U.S. at 549–50; 6A Donald S. Chisum, Chisum on Patents § 19.04[3][h] (2015).

That the exhaustion of rights applies only to sales and not licenses was clear in Kirtsaeng, which stated that under the copyright “first sale” doctrine, 17 U.S.C. § 109(a), because many movie theater owners “were lessees, not owners, of their copies [of copyrighted films], ... they (like bailees and other lessees) cannot take advantage of the ‘first sale’ doctrine.” 133 S.Ct. at 1361.

Thus, in Quanta, the Court stated that General Talking Pictures “held that exhaustion did not apply because the manufacturer [licensee] had no authority to sell the amplifiers for commercial use.” 553 U.S. at 636, 128 S.Ct. 2109. But Quanta held that where the licensee does have authority to sell, the authorized sale results in exhaustion. In Quanta, Intel, a licensee, did have authority to make sales to purchasers, and “exhaustion turns only on Intel's own license to sell products practicing the [patentee's patents].” Id. at 637, 128 S.Ct. 2109.

The majority makes much of the fact that the sale from the licensee to the ultimate purchaser in General Talking Pictures did not result in exhaustion. See Maj. Op. at 744–45, 747–49. But this is not surprising. The licensee infringed the patent by its manufacture and sale of the item. The sale of the amplifier by the infringer to the ultimate purchaser was the antithesis of an authorized sale, and it is hardly surprising that an infringer's unauthorized sale did not result in exhaustion.

In any event, even if there were some tension between the Supreme Court's broad statement of the exhaustion rule and General Talking Pictures, it is not our task to ignore Supreme Court rulings as “unjustifi[ed]” or “[un]sound” because they are purportedly inconsistent with other Supreme Court cases. The distinction between restrictions on sales (impermissible) and restrictions on licensees (permissible) exists in the Court's precedent, and it is not for us to decide if it is a sound distinction. “If a precedent of th[e] Court has direct application in a case, yet appears to rest on reasons rejected in some other line of decisions, the Court of Appeals should follow the case which directly controls, leaving to th[e] Court the prerogative of overruling its own decisions.” Rodriguez de Quijas v. Shearson/Am. Express, Inc., 490 U.S. 477, 484, 109 S.Ct. 1917, 104 L.Ed.2d 526 (1989); see also State Oil Co. v. Khan, 522 U.S. 3, 20, 118 S.Ct. 275 (1997) (Even if a Supreme Court precedent contains many “infirmities” and rests upon “wobbly, moth-eaten foundations,” it remains the “Court's prerogative alone to overrule one of its precedents.”).

Finally, the majority proposes that we should somehow sustain the restriction here because it may be pro-competitive. Exhaustion does not turn on whether a particular post-sale restriction is desirable or undesirable, pro-competitive or anti-competitive, but whether the sale was authorized and the item has passed beyond the scope of the patent monopoly. In any case, the Court has suggested that a prohibition on resale is “manifestly anti-competitive.” Kirtsaeng, 133 S.Ct. at 1363.

There is, in sum, no colorable basis for the majority's failure to follow the exhaustion rule for domestic sales as articulated by the Court in Quanta and numerous other cases.

II. Foreign Exhaustion

The second issue here concerns foreign exhaustion. Lexmark sold patented ink cartridges outside the United States to foreign purchasers. As the majority recognizes, “Lexmark made the foreign sales without communicating a reservation of U.S. patent rights.” Maj. Op. at 754. These were, in other words, authorized sales by the holder of United States patent rights, and the sales of so-called Regular Cartridges did not contain “any sale terms restricting reuse or resale.” Maj. Op. at 727. If those latter sales had been made in the United States, even under the majority's cramped view of exhaustion, there is no question that the sales would have exhausted Lexmark's domestic patent rights. The issue is whether the foreign location of the sale should lead to a different result, as we previously held in Jazz Photo, 264 F.3d at 1111. Like the majority I would retain Jazz Photo insofar as it holds that a mere foreign sale does not in all circumstances lead to exhaustion of United States patent rights. But the government argues, and I agree, that the foreign sale should result in exhaustion if the authorized seller does not explicitly reserve its United States patent rights.

A

Let us first consider the centerpiece of the majority's holding that there is a doctrinal blanket ban on foreign exhaustion, namely the Supreme Court's decision in Boesch v. Graff, 133 U.S. 697, 10 S.Ct. 378, 33 L.Ed. 787 (1890). Boesch announced no such blanket ban. It did not even involve an authorized sale by the holder of U.S. patent rights but rather a sale by a third party under a foreign law's prior use exception.

In that case, a seller in Germany sold patented lamp burners to two individuals, Boesch and Bauer. Id. at 701, 10 S.Ct. 378. The seller was not the U.S. patent holder, or a German patent holder, nor was he even a licensee. Id. Under German law, the seller could make and sell the burners because he had made preparations to manufacture them prior to the filing of the German patent by the holder of the U.S. patent rights. Id. When Boesch and Bauer imported and sold the lamp burners in the United States, the American assignees sued for infringement. Id. at 698, 10 S.Ct. 378. The Court affirmed the holding of infringement, finding that Boesch's and Bauer's sales were “in defiance of the rights [of] patentees under a United States patent.... The sale of articles in the United States under a United States patent cannot be controlled by foreign [ (i.e., German)] laws.” Id. at 703, 10 S.Ct. 378.

Thus Boesch does not apply here because the foreign sales were made by Lexmark—the U.S. patent rights holder—itself. The accused infringer does not rely on foreign law as the source of its authority but the doctrine of exhaustion resulting from an authorized sale by a U.S. rights holder.

Just as Boesch is inapposite, so too is the doctrine of extraterritoriality, reflected in Deepsouth Packing Co. v. Laitram Corp., 406 U.S. 518, 92 S.Ct. 1700 (1972); Dowagiac Manufacturing Co. v. Minnesota Moline Plow Co., 235 U.S. 641, 35 S.Ct. 221 (1915); and Brown v. Duchesne, 60 U.S. 183 (1856). See Maj. Op. at 763–64. The question here is not whether the manufacture or use of a patented product wholly outside of the United States is patent infringement under U.S. law, see Deepsouth, 406 U.S. at 527, 92 S.Ct. 1700, or whether foreign law creates a defense to infringement in the United States, see Boesch, 133 U.S. at 703, 10 S.Ct. 378. Rather, the question is whether United States patent law recognizes exhaustion that occurs abroad from an authorized foreign sale by the holder of the U.S. patent rights and without reservation of U.S. rights. The majority itself admits that foreign activity, such as express or implied license, can have an impact on the rights of a United States patent owner. See Maj. Op. at 727.

B

Strikingly, every one of the lower court decisions before Jazz Photo applied exactly the rule for which the government argues. When the sale was made by an entity not holding U.S. patent rights, as in Boesch, or when the authorized foreign seller clearly reserved U.S. rights, there was no exhaustion. See Sanofi, S.A. v. Med–Tech Veterinarian Prods., Inc., 565 F.Supp. 931, 934–35 (D.N.J.1983) (foreign sale not authorized by U.S. exclusive licensee); Griffin v. Keystone Mushroom Farm, Inc., 453 F.Supp. 1283, 1285, 1287 (E.D.Pa.1978) (foreign sale not authorized by U.S. exclusive licensee); Daimler Mfg. Co. v. Conklin, 170 F. 70, 70, 72–73 (2d Cir.1909) (foreign sale was not authorized by U.S. patent holder); see also Dickerson v. Tinling, 84 F. 192, 193 (8th Cir.1897) (foreign sale made with prohibition on import into and sale within United States); Dickerson v. Matheson, 57 F. 524, 525–26 (2d Cir.1893) (foreign sale with prohibition on import into United States).

But the cases uniformly recognize or assume that where the foreign sale was made by a seller holding U.S. patent rights without a contractual reservation of U.S. rights, exhaustion occurred as a result of an authorized foreign sale. In Holiday v. Mattheson, 24 F. 185, 185 (C.C.S.D.N.Y.1885), the U.S. patentee sold its patented article in England “without restriction or conditions” to a first purchaser. A second purchaser obtained the article from the first, and brought the article back to the United States. Id. The circuit court affirmed the trial court's judgment of noninfringement, stating, “[w]hen the owner sells an article without any reservation respecting its use ... the purchaser acquires the whole right of the vendor in the thing sold.... The presumption arising from such a sale is that the vendor intends to part with all his rights in the thing sold.” Id. In Dickerson v. Matheson, in 1893, the Second Circuit concluded that “[a] purchaser in a foreign country, of an article patented in that country and also in the United States, from the owner of each patent, or from a licensee under each patent, who purchases without any restrictions ... acquires an unrestricted ownership in the article, and can use or sell it in this country.” 57 F. at 527. Similarly in Dickerson v. Tinling, in 1897, the Eighth Circuit “[c]onced[ed,] [but did not decide,] that one who buys a patented article without restriction in a foreign country from the owner of the United States patent has the right to use and vend it in this country.” 84 F. at 195. The Second Circuit also found foreign exhaustion in Curtiss Aeroplane & Motor Corp. v. United Aircraft Engineering Corp., 266 F. 71 (2d Cir.1920). There, the U.S. patent owner licensed a corporation to build airplanes in Canada with “no restriction or limitation as to time, or place, or manner of use of the aeroplanes.” Id. at 80. A buyer who purchased the airplanes in Canada and then brought them back to the United States was not liable for infringement. See id. In Sanofi, S.A. v. Med–Tech Veterinarian Products, Inc., in 1983, the district court found exhaustion because even “assuming that Sanofi had a right to enjoin the reselling of the goods in [the United States], it waived that right by not placing any written restrictions upon the purchaser at the time of sale.” 565 F.Supp. at 938.

This uniform approach, existing well before the 1952 Patent Act and continuing thereafter, strongly supports the government's position. There is indeed a strong argument that the 1952 Act should be read as adopting these earlier cases. See SCA Hygiene Prods. Aktiebolag v. First Quality Baby Prods., LLC, 807 F.3d 1311, 1321 (Fed.Cir.2015) (en banc) (well-established doctrine of laches codified by 1952 Patent Act).

C

So too congressional legislation described by the majority, far from contradicting the government's approach, confirms it. Each bilateral trade agreement cited by the majority requires preservation of U.S. patent rights only where the U.S. rights have been expressly reserved. This is illustrated by the U.S.–Australia agreement, where the patentee's domestic rights must be preserved “where the patentee has placed restrictions on importation by contract or other means.” United States–Australia Free Trade Agreement, Aus.–U.S., art. 17.9.4, May 18, 2004, KAV 6422 (2005). Likewise the U.S.–Singapore agreement requires recognition of an action to prevent or redress the unauthorized procurement of a patented pharmaceutical product, including where it was first sold abroad, but only where someone “knows or has reason to know that such product is or has been distributed in breach of a contract between the right holder and a licensee.” United States–Singapore Free Trade Agreement, Sing.–U.S., art. 16.7.2, May 6, 2003, 42 I.L.M. 1026 (2003). And the U.S.–Morocco agreement permits the United States to limit foreign exhaustion, as it did previously with Australia and Singapore, “to cases where the patent owner has placed restrictions on importation by contract or other means.” United States–Morocco Free Trade Agreement, Morocco–U.S., art. 15.9.4, n. 10, June 15, 2004, 44 I.L.M. 544 (2005).

D

This brings us to the Supreme Court's decision in Kirtsaeng v. John Wiley & Sons, Inc., 133 S.Ct. 1351 (2013). I agree with the majority that Kirtsaeng does not compel identity between the “first sale” doctrine in copyright and patent exhaustion, due to the differences between copyright and patent law.

But unlike the majority, I think that Kirtsaeng provides significant guidance and cannot be dismissed as simply a copyright case, or as limited to the “first sale” provision of the Copyright Act. The policies that animated Kirtsaeng are in large part applicable to patent exhaustion. The Court emphasized the importance of leaving purchasers free to resell goods to enhance competition in the marketplace. Id. at 1363. The Court found that the “first sale” doctrine “frees courts from the administrative burden of trying to enforce restrictions upon difficult-to-trade, readily movable goods.” Id. The Court also found significant the plea of technology companies, who informed the Court that “automobiles, microwaves, calculators, mobile phones, tablets, and personal computers contain copyrightable software programs or packaging.” Id. at 1365 (internal quotation marks omitted). “A geographical interpretation [of the ‘first sale’ doctrine] would prevent the resale of, say, a car, without the permission of the holder of each copyright on each piece of copyrighted automobile software.... Without that permission a foreign car owner could not sell his or her used car.” Id.

Those commercial consequences are equally applicable to patent exhaustion. Automobiles, microwaves, calculators, mobile phones, tablets, and personal computers also contain patented components. To paraphrase, “a geographical interpretation [of patent exhaustion] would prevent the resale of, say, a car, without the permission of the holder of each [patent] on each piece of [patented] automobile [software or hardware].... Without that permission a foreign car owner could not sell his or her used car.”

Refusing to find presumptive exhaustion by foreign sales would have serious adverse consequences in the patent area, just as in the area of copyright. Technology companies have echoed the concerns in Kirtsaeng and report that “modern devices include components from dozens—if not hundreds—of suppliers.” Brief for LG Electronics, Inc., Dell Inc., Google Inc., Intel Inc., et al. as Amici Curiae 2. The majority's rule would require a manufacturer to “trace the patent rights of every component it purchases and then negotiate appropriate license arrangements with the component manufacturer (as well as any sub-component manufacturer),” and ultimately “it is consumers who suffer most directly through higher prices.” Id. at 5, 8. A major retailer informs us that it “often sells patented products that, although genuine, were not purchased directly from the patent holder” and that “[s]ome of those products were first sold outside of the United States.” Brief for Costco Wholesale Corp. et al. as Amici Curiae at 1. A domestic-only patent exhaustion rule would seriously impair international trade.

Kirtsaeng emphasized the “ever-growing importance of foreign trade to America,” 133 S.Ct. at 1367, which includes trade not just in artwork and books but also automobiles, appliances, mobile phones, tablets, and personal computers. The Court concluded:

[T]he fact that harm has proved limited so far may simply reflect the reluctance of copyright holders so far to assert geographically based resale rights. They may decide differently if the law is clarified in their favor. Regardless, a copyright law that can work in practice only if unenforced is not a sound copyright law. It is a law that would create uncertainty, would bring about selective enforcement, and, if widely unenforced, would breed disrespect for copyright law itself.

Id. at 1366. So too with patent law.

E

Despite these significant policy considerations favoring foreign exhaustion for both copyright and patent, there are significant differences between copyright and patent law that cut the other way. The premise of exhaustion is that the rights holder has been compensated for its efforts. See Univis, 316 U.S. at 251, 62 S.Ct. 1088 (“The reward he was demanded and received is for the article and the invention which it embodies.... He has thus parted with his right to assert the patent monopoly with respect to it....”). In the area of copyright, given the uniform international protection of copyrights, it is reasonable to assume that the rights holder will receive compensation for a foreign sale. But patent law is different. It is not uniform from country to country. Indeed, there are typically significant differences from country to country. Many countries offer no realistic protection or very little protection for items patented under U.S. law. In other words, there is reason to doubt that the rights holder has been fully compensated for a foreign sale. This suggests an accommodation between the interests of the rights holder and the unsuspecting buyer must be found.

Even the majority recognizes the need for such an accommodation. The majority acknowledges that the law should accommodate the potential of “unintended infringement by buyers of goods in foreign countries who bring them into the United States,” but believes that problem could be solved by the availability of an express or a vague implied license defense. See Maj. Op. at 771, 773–74. That defense provides little comfort, however, because it places the burden on the purchaser to obtain a statement from each patentee of a patented component in a product that it has permission to import the component into the United States, or else prove in court that the circumstances of each patentee's sale of its component to the manufacturer constituted an implied license to import into the United States.

In my view, the necessary accommodation between the interests of the rights holder and the unsuspecting buyer can only be achieved by the government's proposal to put the burden on the U.S. rights holder to provide notice of a reservation of U.S. rights to the purchaser, an approach supported by the earlier lower court decisions and legislative action.

In other words, the country-to-country differences in patent laws, and the different economic choices patentees must make as a result, suggest that patentees should be able to reserve their U.S. patent rights when making or authorizing foreign sales. But Kirtsaeng's policy concerns indicate that that right should not extend to situations where the patentee is silent or unclear. If a patentee wishes to reserve its U.S. rights, it should be required to do so unmistakably. The patentee is in a better position to reserve its rights than the purchaser is to inquire into any reservation. A rule requiring reservation would protect both the interests of the authorized seller and the unsuspecting buyer.

In conclusion, I would overrule our decision in Mallinckrodt as inconsistent with governing Supreme Court authority and overrule Jazz Photo to the extent that it imposes a blanket ban on foreign exhaustion. I would recognize foreign exhaustion where the U.S. rights holder has not notified the buyer of its retention of the U.S. patent rights. I respectfully dissent from the majority's contrary holdings.