Misappropriation on a Global Scale: Extraterritoriality and Applicable Law in Transborder Trade Secrecy Cases

In recent years, trade secrecy has come to the fore, both domestically and internationally. For technological firms, trade secrecy protection has always served an important role: it is cheaper than patents, can last longer than the 20-year term of patent protection, and it covers intellectual contributions that are not advanced enough or sufficiently inventive to be considered patentable. 1 Many of the changes of the last few decades have made trade secrecy rights even more critical. A series of cases restricting the reach of patent law has put new emphasis on finding alternatives to patenting. 2 Because jobs no longer last a lifetime, more employees move from firm to firm, possibly taking valuable information with them. 3 Modern business practices, including especially value chain production methods, spread information around the globe from creators to manufacturers, distributors, sellers, and maintenance organizations. 4 Most importantly, developments in computer technology, robotics, and manufacturing create more situations where information can be feasibly protected by secrecy. 5 At the same time, these technologies can also make it easier to take valuable information without authorization. U.S. losses due to trade secrecy theft are thought to be in the neighborhood of $300 billion per year; 6 one source estimates these losses to represent between 1 and 3 percent of U.S. GDP. 7 The United States has a long tradition of trade secrecy protection, initially through state common law, 8 and later through the states' adoption of the Uniform Trade Secrets Act (UTSA), 9 their enactment of criminal misappropriation statutes, 10 and their creation of systems to exclude goods made with misappropriated information. 11 At the federal level, the Unfair Import Trade Practices Act of 1930 (Tariff Act) empowers, through so-called section 337 actions, the International Trade Commission (ITC) to order the exclusion from the U.S. market of goods made through unfair practices, including the marketing of goods embodying misappropriated trade secrets. 12 In 1996, Congress passed the Economic Espionage Act (EEA) to impose strong criminal penalties for unauthorized takings benefiting foreign governments ("espionage") or private furtherance of the offense was committed in the United States; 18 in fact, a great number of EEA prosecutions involve foreign entities, especially in China. 19 While few cases under the DTSA amendment to the EEA have been fully adjudicated, courts could easily consider this provision as approving an equally (or more) extensive reach on the civil side.

It is not surprising that U.S courts would apply domestic laws in this farreaching way. In the past, most cases were largely about local activity, where the law of that locality (usually the forum) was assumed to apply, even if the case included transnational elements. 20 Furthermore, until fairly recently, other nations were under no compulsion to recognize rights in trade secrets. While international law has long included intellectual property obligations, the oldest agreement regulating technological information-the Paris Convention-never required more than protection against "unfair competition." That term was not defined and the Convention elucidated it with particulars that involved only acts more akin to 18 18 U.S.C. § 1837; See generally Robin J. Effron, Secrets and Spies: Extraterritorial Application of the Economic Espionage Act and the TRIPS Agreement, 78 N.Y.U. L. REV. 1475, 1477 (2003). 19 See Rochelle Cooper Dreyfuss & Orly Lobel, Economic Espionage as Reality or Rhetoric: Equating Trade Secrecy with National Security, 20 LEWIS & CLARK L. REV. 419, 428 (2016). 20 Cf. Commission Regulation 864/2007 art. 8.1, of the European Parliament and of the Council of 11 July 2007 on the Law Applicable to Non-Contractual Obligations (Rome II), 2007 O.J. (L199) 40, 45 (EC) ("The law applicable to a non-contractual obligation arising from an infringement of an intellectual property right shall be the law of the country for which protection is claimed."); id. Recital (26) ("Regarding infringements of intellectual property rights, the universally acknowledged principle of the lex loci protectionis should be preserved."); Berne Convention for the Protection of Literary and Artistic Works art. 5, July 24, 1971, 1161 U.N.T.S. 31. [hereinafter Berne Convention] ("[T]he extent of protection, as well as the means of redress afforded to the author to protect his rights, shall be governed exclusively by the laws of the country where protection is claimed."). trademark infringement than trade secret misappropriation. 21 Even after the TRIPS Agreement went into force in 1995, 22 its provision on trade secrets (labeled "undisclosed information" in TRIPS) was not widely implemented. 23 With other countries lacking robust protection, courts may have seen reliance on U.S. law as the only way to protect information developers. 24 Alternatively, U.S. courts may have applied U.S. law on the (erroneous) theory it was identical to foreign law. 25 After all, the TRIPS Agreement essentially tracks the main features of the UTSA right of action by requiring protection for information that has "commercial value because it is secret" against "being disclosed to, acquired by, or used by others without their consent in a manner contrary to honest commercial practices." 26 For many years, the Supreme Court seemingly approved the application of U.S. intellectual property law to foreign activities when there was a U.S. link of some type. In a 1952 decision, Steele v. Bulova Watch Co., the Court held that U.S. trademark law (the Lanham Act) 27 could be applied to the unauthorized use of the Bulova mark on goods that were manufactured and sold in Mexico. The Court explained-in a sentence susceptible to a rather expansive interpretationthat acts committed abroad legally "lose that character when they become part of an unlawful scheme." 28 To be sure, as other countries have begun to protect trade 21 Paris Convention for the Protection of Industrial Property art. 10bis, July 14, 1967, 21 U.S.T. 1583, 828 U.N.T.S. 305 [hereinafter Paris Convention] (mentioning "in particular" acts that create confusion, mislead the public, or discredit the right holder). 22 Agreement on Trade-Related Aspects of Intellectual Property Rights art. 39.2, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1C, 1869 U.N.T.S. 299, 33 I.L.M. 1197 (1994) [hereinafter the TRIPS Agreement or TRIPS]. 23 Id. art. 39. For example, even as late as 2013, a study by the European Commission revealed that in the EU, national trade secrecy laws had developed unevenly, see Proposal for a Directive of the European Parliament and of the Council on the Protection of Undisclosed Know-How and Business Information (Trade Secrets) Against Their Unlawful Acquisition, Use and Disclosure, at 4, COM (2013) 813 final (Nov. 28, 2013). 24 Cf., e.g., Sheldon v. Metro-Goldwyn Pictures Corp., 106 F.2d 45, 52 (2d Cir. 1939), aff'd, 309 U.S. 390 (1940); Los Angeles News Serv. v. Reuters Television Int'l, Inc., 149 F.3d 987 (9th Cir. 1998); Curb v. MCA Records, Inc., 898 F. Supp. 586, 596 (M.D. Tenn. 1995) (adopting various approaches to apply U.S. copyright law to global infringements). 25 See, eg., TianRui Grp. Co. v. Int'l Trade Comm'n 661 F.3d 1322, 1332 (Fed. Cir. 2011). ("TianRui has failed to identify a conflict between the principles of misappropriation that the Commission applied and Chinese trade secret law."). 26 The negotiation history of the provision is recounted in SHARON K. SANDEEN, The Limits of Trade Secrecy Law: Article 39 of the TRIPS Agreement and the Uniform Trade Secrets Act on Which it is Based, in THE LAW AND THEORY OF TRADE SECRECY: A HANDBOOK OF CONTEMPORARY RESEARCH 537 (Rochelle C. Dreyfuss & Katherine J. Strandburg, eds. 2011). 27 Lanham (Trademark) Act, Pub. L. 79-489, 60 Stat. 427 (1946); Steele v. Bulova Watch Co., 344 U.S. 280, (1952). 28 Bulova, 344 U.S. at 287. secrets in the new millennium, the need for the United States to apply its own law to police the world has diminished. In 2012, the United States began using its Trade Representative's annual report on global intellectual property deficiencies to draw attention to the inadequacy of trade secrecy protection in other countries and to threaten the withdrawal of trade preferences from nations that failed to enact laws the United States regards as sufficiently protective. 29 When it also began to insert trade secrecy obligations into regional free trade agreements, 30 other countries took note. Whereas the members of the EU once had extremely disparate trade secrecy laws, 31 by 2016, the EU complied with U.S. wishes to improve trade secret protection and promulgated a Directive on trade secrecy for the member states to implement over the next two years. 32 China is likewise reviewing its position. 33 Significantly, however, none of these countries were required to enact laws identical to those of the United States. Although these laws are most likely TRIPS compliant, TRIPS is a minimum standard agreement. 34 Thus, countries can, and do, take different approaches to such questions as adjusting the balance between trade secrecy and patent rights, nourishing the public domain, protecting employee mobility, facilitating whistleblowers, and dealing with good faith users of misappropriated information.

In this changed atmosphere, the practice of viewing U.S. law as available to protect trade secrets worldwide demands further examination. Arguably, an analogy could be made to the way that applicable law is handled in the copyright and patent context. There, the focus is on the product (creative expression or invention), and each country uses its own law to determine the rights that attach to use of that product in its territory, irrespective of where it was developed. 35 However, that traditional territorial approach works poorly for trade secrets. Trade secrecy protection does not so much focus on the product (the information) because almost anything can be a trade secret. Rather, the emphasis is on specific activities: how well the developer maintained the information (whether reasonable efforts were made to preserve secrecy) and how the user obtained the information (whether it was lawfully acquired or misappropriated). Applying the law of one country to activities that arise in another raises considerable conceptual difficulties when the laws of the two countries are different-could, for example, information that is accessible in one country nonetheless be regarded as secret in another? And because the place of use may not be foreseeable at the time that secrecy must be maintained or the information appropriated, this approach also poses problems for information developers and users. 36 Significantly, the Supreme Court has in recent years embarked on a vigorous agenda to prevent, as it stated in Kiobel v. Royal Dutch Petroleum Co., "unintended clashes between [U.S.] laws and those of other nations," 37 and has warned that U.S. intellectual property law "does not rule the world." 38 The Supreme Court has stressed that courts can apply legislation to acts outside the United States only when the statute evinces clear congressional intent that it be accorded extraterritorial effect or the focus of the legislation is on domestic activity (meaning the application is not, in fact, extraterritorial). 39 Even so, courts consider other aspects of prescriptive comity before applying a U.S. statute in a particular case. 40 If U.S. law is not applied, the case is dismissed. The outcome 35 See TRIPS Agreement, supra note 22, art. 27.1 (requiring WTO countries to recognize patent rights without discrimination as to the place of invention); id. art 3. (requiring national treatment); Berne Convention, supra note 20, art. 3.1 (extending protection to authors who are nationals of the Berne Union and to works first published in the Union). 36 Trademark law shares some of the same features: protection hinges largely on specific activities (establishing the meaning of a mark, differentiating it from other meanings, registering the mark, using it to confuse customers). In general, the applicable law is the place where these actions occur. Rights to well-known marks are something of an exception, but even there, the law of the place of use is applied and protection depends on whether the mark is known in that jurisdiction, see Paris Convention, supra note 21, art. 6bis; TRIPS Agreement, supra note 22, art. 16. Tellingly, transnational cases pose especially difficult questions for trademarks as well, see Graeme B. Dinwoodie, Developing a Private International Intellectual Property Law: The Demise of Territoriality?, 51 WM. & MARY L. REV. 711, 775-76 (2009). 37 Kiobel v. Royal Dutch Petroleum Co., 569 U.S. 108 (2013) (quoting Equal Emp't Opportunity Comm'r v. Arabian Am. Oil Co., 499 U.S. 244, 248 (1991). 38 Microsoft Corp. v. AT&T Corp., 550 U.S. 437, 454 (2007). 39 RJR Nabisco v. European Cmty., 136 S. Ct. 2090, 2100 (2016). 40 Courts have used prescriptive comity to limit application of a statute, even when the presumption has been rebutted, see F. Hoffman-LaRoche Ltd. v. Empagran, S.A., 542 U.S. 155, 164-65 (2004). Courts have used a similar concept to limit application of a statute, even when a transaction falls within the domestic application of claims that were "predominantly foreign. " in cases such as Sino Legend raise such clashes-as can be discerned from the decision of Chinese Ministry of Commerce to file an amicus brief in that case seeking to overturn the decision. 41 Given the Supreme Court's concerns and the emerging differences among national intellectual property legislation, the time is ripe to consider the appropriate extraterritorial reach of section 337, the DTSA, and state law.

This paper, authored jointly by an intellectual property professor and a conflicts scholar, is among the first to do so. There is considerable case law and a growing literature on the substantive provisions of trade secrecy law, 42 but very little analysis of either the extraterritorial application of U.S. trade secrecy law or the related question of what law should apply in a transnational trade secrecy case. To the extent that these issues have been discussed in intellectual property, the focus has mainly been on patent, trademark, and copyright law. 43 For example, the World Intellectual Property Organization (WIPO) has considered private international law questions arising in connection with transnational patent, copyright, and trademark disputes. 44 Additionally, the American Law Institute (ALI), in parallel with groups in Europe and Asia, has developed principles focused on the resolution of only these categories of cases. 45 In some jurisdictions, conflict issues in trade secrecy cases may have received short shrift because it is often unclear whether these cases sound in tort, contract, or property. 46 In the United States, trade secrecy also may have received lesser attention because its importance has only recently emerged. In addition, U.S. courts have apparently taken the position that they need to provide global protection when other countries do not. Alternatively, they have tended to assume that the laws of countries with trade secrecy protection are so similar, the application of U.S. law is not controversial.

To begin the discussion on extraterritoriality and applicable law in trade secrecy cases, Part I provides a short survey of national trade secrecy laws. It demonstrates the many ways in which even countries that all operate under the TRIPS Agreement and generally agree on the availability of protection can nonetheless make divergent policy choices and enact laws that differ in critical detail. In short, this Part exposes how clashes among national trade secrecy laws can easily occur. Part II reviews the limits that modern courts have imposed on the extraterritorial reach of federal legislation and elucidates the Supreme Court's twostep approach to the question whether Congress intended an enactment to reach activity outside the United States. Part III applies this analysis to trade secrecy cases: Section A considers section 337 actions, such as TianRui, and Section B looks at the fact patterns that arise under the DTSA. These Sections demonstrate how the Supreme Court's extraterritoriality jurisprudence delimits the reach of federal statutory law. Section C deals with state trade secrecy law and the question of whether states can fill any gaps by furnishing a cause of action in situations that federal law does not reach. We note that states do not tend to approach the issue as one of extraterritoriality, as does the Supreme Court with federal statutes; instead, they frame the question as one of the choice of applicable law. In Section D, we revisit the DTSA. Given its roots in state trade secrecy law, we suggest that a traditional choice of law approach should be considered in federal trade secret 45 misappropriation cases.

Our extraterritorial analyses often turn on the existence of a "U.S. trade secret." 47 This statutory interpretation will often benefit U.S. trade secrecy holders more than it will benefit foreigners. Thus, it may have implications regarding the antidiscrimination provisions of the TRIPS Agreement. 48 For those situations in which we propose a choice of law rule-whether it is the law of the place where a trade secret is developed or where the relevant conduct takes place-goods from different countries will be treated differently, arguably in violation of the nondiscrimination provisions of the General Agreement on Tariffs and Trade (GATT). 49 For all these cases, we offer interpretive guidance that endeavors to soften the tension with these international obligations. However, a full examination of the international issues awaits a future publication.

Were all countries to agree on the law, policies, and importance of enforcing trade secrecy protection, the application of one country's law to events that happen elsewhere would be of little moment: no matter which country entertained a case, the results would be identical and the same policies furthered. As noted above, trade secrecy easily appears to be an area of law where conflicts are absent, because the general contours of the law vary little as between the states of the United States, as reflected in the Restatements and the UTSA. The TRIPS Agreement and the EU Directive are similar as well. 50 The details, interpretation, and application can, however, vary considerably. For simplicity, the focus here is on (1) the TRIPS Agreement because all WTO members must conform to it; (2) the UTSA because so many U.S. states have adopted it and because it furnished the basis for the DTSA; 51 and (3) the EU Directive because it suggests the approach other countries may consider as they come up to (what the U.S. regards as) international standards. 52 47 We define a U.S. trade secret infra at notes 244-250. 48 TRIPS Agreement, supra note 22, art. 3 (guaranteeing that foreign nationals will be treated no less favorably than each country's own nationals); see, e.g., Panel Report, European Communities-Protection of Trademarks and Geographical Indications for Agricultural Products and Foodstuffs, WTO Doc. WT/DS174/R (adopted Mar. 15, 2005). 49 General Agreement on Tariffs and Trade Art. I, Art. III:4, Oct. 30, 1947, 61 Stat. A-11, 55 U.N.T.S. 194 [hereinafter GATT] (prohibiting discrimination based on the geographic source of goods). 50 TianRui, 551 F.3d, at 1331 & 1333. 51 See, e.g., DTSA Senate Rep., supra note 6, at 10 (noting intent to bring the terms used in alignment with the UTSA). 52 For a more detailed comparison between the DTSA and the EU Directive, see Sharon K. Sandeen, Implementing the EU Trade Secret Directive: A View from the United States, 39 EUR. INTELL. PROP.

The TRIPS Agreement is fairly stark in its mandate. It requires protection against unfair competition as defined in the Paris Convention, and then goes on to provide (in a section the EU Directive largely tracks 53 ) that:

Natural and legal persons shall have the possibility of preventing information lawfully within their control from being disclosed to, acquired by, or used by others without their consent in a manner contrary to honest commercial practices so long as such information:

(a) is secret in the sense that it is not, as a body or in the precise configuration and assembly of its components, generally known among or readily accessible to persons within the circles that normally deal with the kind of information in question;

(b) has commercial value because it is secret; and (c) has been subject to reasonable steps under the circumstances, by the person lawfully in control of the information, to keep it secret. 54 The elements of the cause of action are thus (1) the information must be secret-or as the UTSA puts it, the information "must derive independent economic value… from not being generally known. . . and not being readily ascertainable by proper means""; 55 (2) it must be subject to reasonable steps to maintain its secrecy; 56 (3) it must be taken in a manner contrary to honest business practices-or in UTSA terminology, it must be "taken by improper means" or "misappropriated"; 57 and (4) it must be disclosed, acquired, or used by others. 58 53 EU Directive, supra note 32, arts. 2(1) & 4. 54 TRIPS Agreement, supra note 22, art. 39.2. A footnote to "honest commercial practices" provides:

For the purpose of this provision, "a manner contrary to honest commercial practices" shall mean at least practices such as breach of contract, breach of confidence and inducement to breach, and includes the acquisition of undisclosed information by third parties who knew, or were grossly negligent in failing to know, that such practices were involved in the acquisition.

In addition, art. 39.3 requires WTO members to regard undisclosed data used to obtain premarketing approval in each country. These cases are uniquely local and beyond the scope of this article. 55 UTSA, supra note 9, § 1(4)(i) (amended 1985). 56 Id. at § 1(4(ii). 57

The similarity among these provisions is not coincidental. The United States was instrumental in adding trade secrecy to the agenda for TRIPS negotiations and the U.S. experience with domestic protection created a useful template for TRIPS drafters. 59 In turn, the EU relied on TRIPS and the UTSA in framing its Directive. From a theoretical perspective, the law is nicely drawn. By its terms, it targets valuable advances that do not (or do not yet) qualify for other intellectual property rights. As with other intellectual property, it protects those who invested in their creation from free riders, who might otherwise take the information and use it to compete down the price to the point where the developer would be unable recover its costs. By providing the assurance of a legal right of action if reasonable (but not comprehensive) measures are taken to protect secrecy, the law also allows innovators to save money on extensive self-help measures and likely results in greater output than would complete secrecy. With trade secrecy protection, innovators can transmit technical information to investors, licensees, fabricators, distributors, and potential customers with assurance that legal action can be taken in case of loss. 60 Furthermore, entrepreneurs can more safely enter into joint ventures and spin off businesses that depend on sharing know-how. Start-ups lacking substantial resources can rely on the automatic protection afforded by trade secrecy law instead of absorbing the expense of obtaining patent protection.

But there is another side to trade secrecy protection. 61 Most obviously, the stronger the protection, the more likely that a creator will prefer it to patenting. 62 From a societal perspective, however, patenting is superior for it requires disclosure of the technical information protected. 63 That allows others to avoid the cost of recreating it. Moreover, it gives others the opportunity to take the advance in directions the inventor did not consider and permits researchers to 58 The UTSA envisions enjoining even threatened misappropriation, UTSA, supra note 9, § 2(a). It does, however, require that the acquirer "can obtain economic value from its disclosure or use," § 1(4)(i). The EU Directive, supra note 32, adds the fourth point in art. 4(1). 59 build on the underlying science and to publish their findings, safe in the knowledge that valuable applications can be protected with patents. In addition, disclosure allows government to regulate use of the invention in order to protect safety, health, and the environment. Patent protection is also limited to a finite term of years. When the patent expires, the information is available to the public for free use. By contrast, information that is secret cannot be freely repurposed, advanced, regulated, or published. The secret can last far longer than the term of patent protection. To be sure, legal protection for secrets leads to more output and lower prices than requiring innovators to keep "real" secrets (i.e., to not reveal the information to anyone-if that were possible). However, prices are higher and output is less than if information regularly spills into the public domain.

More troubling, many trade secrecy cases involve situations where an employee leaves one job and takes another in the field of his expertise, only to find that the move is challenged on trade secrecy grounds. 64 In many cases, documentation is taken, but there are cases that concern information that is in the employee's head-information that the employee may regard as part of his or her skill set, knowledge, and qualifications, but which the employer (and the law) deem to be proprietary. 65 Zealous protection for trade secrets can thus interfere significantly with employee mobility. Employers may be reluctant to hire employees from positions where they learned secret information, lest they are later accused of using it in their own operations. With less opportunity to leverage job offers, salaries in high tech industries could decline, and with it, the inclination to invest in human capital (for example, to enroll in advanced programs in STEM subjects 66 ). Paradoxically, stringent trade secrecy laws could also impinge upon a firm's ability to restructure its operations and merge, divest, or spin off divisions. In the end, too much protection could harm innovation, for it could prevent employees from putting their talents to their highest and best uses, inhibit the flow of information within technical fields, and lock firms into suboptimal organizational structures. 67 Given the competing policies at stake in trade secrecy law, the delicate balance between secrecy and patenting, and the ambiguous effects of trade secrecy 64 James Pooley, TRADE SECRETS §5.01(2)(a) (2013) ("Most trade secret lawsuits involve employees allegedly using their former employer's secrets to benefit themselves or a competitor."). See, e.g., General Motors Corp. v. Lopez, 948 F.Supp. 684 (E.D.Mich. 1996) (concerning an employee who left GM in Michigan for employment with VW in Germany, taking along 20 cartons of GM documents). 65 on social welfare and scientific and technical progress, variations in protection are to be expected. And that is true even among states that have adopted the UTSA (or as James Pooley calls it, the "non-Uniform Trade Secrets Act"foot_2 ) and certainly among countries subject to the barebones framework set out in TRIPS. Indeed, each element of protection offers jurisdictions opportunities to tailor the law to their own conceptions of the proper balance between private and public interests. 69 Consider, to start, the question of what information can be protected. TRIPS, the UTSA, and the EU Directive require that the information be secret. But what sorts of information are covered? The first Restatement of Torts took a hard line on this issue and excluded protection for "ephemeral events." 70 Likewise, the examples in the UTSA include "a formula, pattern, compilation, program, device, method, technique or process" 71 -in other words, technical information. 72 Nonetheless, U.S. states and the EU Directive tend to protect a broad range of materials, including customer lists and marketing strategies. 73 There is also a question whether the information that is protected must by positive or whether negative information-blind alleys (knowing what not to try)-are also protectable. Many U.S. states protect negative information, but such protection is not explicitly required by either TRIPS or contemplated by the UTSA. 74 The EU Directive does not mention negative information. Indeed, its focus on "infringing goods" and the acquisition of information, suggests that it does not require member states to protect knowledge of dry holes. 75 Clearly, the more expansive the protection, the more it interferes with employee mobility. A salesman, for example, is more able to change jobs if he does not have to worry about litigation over knowledge of his prior customers; it is easier for a scientist to take a new position if she is able, without incurring legal problems, to avoid redoing experiments she already knows do not work.

Moving beyond the type of information, there are different approaches to how "secret" the information must be to merit protection. The TRIPS Agreement and the EU Directive use as a benchmark knowledge in circles that normally deal with the information, suggesting that if a group within an industry is using similar information, it cannot be protected. 76 But the UTSA (in conformity with what was regarded as industry practice) considers a secret to be information that is not known to "other persons who can obtain economic value from its disclosure or use." 77 That approach is more protective of innovators as it permits small groups to safeguard their knowledge from new entrants. Furthermore, it allows plaintiffs to protect information even if they suspect that there are others in the industry who are relying on the same tacit knowledge. Jurisdictions that are interested in strong intellectual property rights are thus likely to adopt that sort of standard, while those that believe innovation is best promoted through vigorous competition will-like the EU-reject the idea that established participants can consider the information they are all using to be secret.

TRIPS, the UTSA, and the EU Directive also specify that the information protected must not be "readily accessible." 78 TRIPS, however, fails to define the term. Presumably it means that the information cannot be independently invented easily, but it is not clear whether it also means that it cannot be, perhaps with some difficulty, reverse engineered-figured out from public embodiments. 79 To be sure, the EU Directive clarifies that both independent discovery and "observation, study, and disassembly" are lawful means of acquisition. 80 U.S. law is in accord: the commentary to section 1 of the UTSA defines "proper means" to include such behaviors and the Restatement of Unfair Competition explicitly permits reverse engineering, 81 as does the DTSA. 82 Moreover, there is authority suggesting that the 76 EU Directive, supra note 32, art. 2 (1)(a). See also Pooley, supra note 68 at 1185 (noting that availability on the internet could destroy a trade secret, even if there is no evidence anyone has accessed the site). 77 UTSA, supra note 9, § 1 (4)(i) (amended 1985). The DTSA puts this in the singular: another "person," §1839(3)(B). 78 UTSA, supra note 9, § 1(4)(i) (amended 1985); EU Directive, supra note 32, art. 2 (1)(a). 79 Reichman, supra note 34, at 378. permissibility of reverse engineering is constitutionally required in the United States. 83 However, these provisions do not say when the effort to figure out the invention is so great, the information will not be regarded as readily accessible. Moreover, it is not clear whether the ability to acquire the information lawfully means it is to be considered public. Thus, there is a split among jurisdictions on whether misappropriation is actionable if the information could be reverse engineered, but that is not how the defendant acquired it. 84 Nor is it evident whether the inability to readily access the information is part of the plaintiff's affirmative case or a defense that the defendant bears the burden to prove. 85 At the margin, these factors make a difference as to how much information is considered in the public domain.

To many commentators, the most puzzling aspect of trade secrecy protection is the requirement that the secret be subject to reasonable efforts to protect it. 86 As we saw, the TRIPS Agreement, the EU Directive and the UTSA envision such efforts. However, if one goal of the law is to relieve innovators of the burden of taking costly self-help measures, why should any effort be required? Arguably, it is socially useful to require employers to notify employees of trade secrets because that would alert the employees to potential mobility problems. But if that is the effort the law requires, it would make more sense to enforce trade secrets only against persons in a confidential relationship with the plaintiff. If, instead, the idea is to make sure the information is genuinely valuable and secret, the plaintiff could be required to prove that directly. The difficulty in justifying the reasonable efforts element adds another reason why jurisdictions differ on the specifics of trade secrecy protection. Some do not, in fact, require proof of such efforts-the Restatement of Unfair Competition, for example, does not. 87 Some jurisdictions might look to a constellation of factors to determine the adequacy of the measures, but some might demand a showing that a confidentiality agreement was in place (and even then, there could be disagreement on whether the agreement must be explicit or can be implied from the circumstances). 88 83 See, e.g., Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470, 476, 490 (1974); Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U.S. 141, 160 (1989). 84 See, e.g., Hecny Transp., Inc. v. Chu, 430 F.3d 402, 404-05 (7th Cir. 2005) ("[I]t is unimaginable that someone who steals property, business opportunities, and the labor of the firm's staff would get a free pass just because none of what he filched is a trade secret."). 85 Since the method of acquisition is uniquely in the defendant's knowledge, it is likely a defense in many jurisdictions. However, the EU Directive suggests that the means of acquisition is part of the plaintiff's case, art. 11(1)(c). 86 See generally Robert G. Bone, Trade Secrecy, Innovation and the Requirement of Reasonable Secrecy Precautions, in Dreyfuss and Strandburg, supra note 42, at 46. 87 RESTATEMENT (THIRD) OF UNFAIR COMPETITION, § 39 (Am. Law Inst. 1995). 88 Max Planck Comments, supra note 46, ¶ 19 (noting that under TRIPS, some countries require explicit measures regarding confidentiality). In China, for example, most employees are liable only if there is an explicit contract, but senior managing staff have a statutory duty to keep secrets. See Xiong, supra note 33, at 256.

The misappropriation element-acquisition by improper means-can also be subject to differing interpretations. First, the TRIPS provision and the EU Directive refer to "honest commercial practices."foot_4 If the standard is derived from local practices, there is obviously room for jurisdictional variation. Second, there is the question alluded to in connection with accessibility: whether independent invention and reverse engineering are proper methods of acquisition. For independent invention, the answer is clear: it is never considered improper. 90 In most cases, the second inventor will expend as many resources as the first. Thus, there will have been no free ride; because both parties must recover development costs, price competition will not undercut the originator. It is, however, somewhat harder for a jurisdiction to choose a position on reverse engineering. Reverse engineering does not usually cost as much as inventing anew, so if it is permissible, there is an element of a free ride. Jurisdictions intent on furnishing strong protection are thus more likely to find reverse engineering a form of misappropriation or allow reverse engineering to be restricted by contract. 91 Similarly, they may find otherwise lawful activity, such as flying over a plant and photographing it, to constitute misappropriation. 92 On the other hand, if technology is protected against reverse engineering, more innovators will prefer it to patenting. Jurisdictions seeking to promote disclosure and competition will therefore regard reverse engineering as proper. 93 Some may even go further and specifically require a heightened showing of intent. 94 Jurisdictions can similarly differ on what is known as "tippee liability." Under certain circumstances, the UTSA, the TRIPS Agreement, and the EU Directive impose liability on a person or firm (the tippee) that acquired a trade secret from the party who took it (the tipper). However, the circumstances are different in the various instruments. TRIPS requires liability only when the third party knew or was grossly negligent in failing to know that the information was misappropriated; 95 the UTSA and the EU Directive make actionable acquisition of information by a person who knew or had reason to know of the misappropriation. 96 The latter formulation gives states leeway to decide for themselves what the tippee should reasonably have known. Jurisdictions can similarly differ on when the tippee needs to have known of the misappropriation. The UTSA allows for the possibility that the tippee invested in the information in reliance on its free availability. Thus, it makes the tippee liable only if it knew of the problem before a material change in position. 97 TRIPS is silent on the matter, but the implication of the EU Directive, which considers the use of the information unlawful, suggests that the tippee is liable no matter when it learned of the problem. 98 Further, there is the question of what constitutes misappropriation ("infringement" in the parlance of the EU Directive). The TRIPS Agreement gives WTO members considerable flexibility. Protection for trade secrets appears in a section that, by its terms, is intended to expand on the Paris Convention's concepts of unfair competition and honest business practices. Thus, a member could interpret the provision as envisioning that the use must be rivalrous (i.e. head-tohead competition). 99 For example, a jurisdiction could consider it nonactionable if the defendant acquired the information for purely comparative purposes (to determine, for example, the acceptability of the acquirer's own product or whether it is worth buying from the trade secret holder). 100 But jurisdictions could equally decide to impose liability even when there is less obvious competition. Under the EU Directive, the mere acquisition of the trade secret seems to be enough to trigger liability. 101 But that is not clear. The legislation appears aimed at protecting the power of creators to recoup the costs of innovation and protecting incentives to innovate, 102 so it is possible the provision can be interpreted to require competitive use. 103 Likewise, U.S. law can be construed in different ways. The UTSA speaks of improper acquisition not followed by disclosure or use as sufficient for, at least, injunctive relief;foot_5 it only appears to require the secret to be of the sort that permits others to "obtain 96 UTSA, supra note 9, § 1 (2)(ii)(B) (amended 1985); EU Directive, supra note 32, art. 4(4). 97 UTSA, supra note 9, § 1(2)(ii)(C) (amended 1985). 98 EU Directive, supra note 32, art. 4(4). However, the Directive permits the court to alter the remedies in cases of good faith acquisition. Recital (29) and art. 13(3)(a). 99 See, e.g., United Golf Ass'n v. St. Andrews Sys. Data-Max, Inc., 749 F.2d 1028 (3d Cir. 1984) (no competition between the U.S.GA and a firm that sold handicaps using the U.S.G.A's formula). 100 See, e.g., Omitech Int'l Inc. v. The Clorox Co., 11 F.3d 1316 (5th Cir. 1994). 101 EU Directive, supra note 32, art. 4(1). 102 Id., Recitals (1), (2), and (4). 103 Much of Chinese trade secrecy law is similarly concerned with competitive behavior. Xiong, supra note 33, at 264. economic value from its disclosure or use." 105 Similarly, the Restatement of Unfair Competition refers to information that "can be used" and provides a "potential economic advantage." 106 Both phrases suggest that there is no need to show actual competitive use of the information; it is enough that such use is possible and not overly speculative. 107 Jurisdictions can have particularly divergent views on defenses to trade secrecy actions. The TRIPS Agreement provides no direction on this point (which is interesting, given that exceptions and limitations in patent, copyright, and trademark law are highly regulated). 108 Aside from the independent invention and reverse engineering possibilities, the text of the UTSA and Restatement of Unfair Competition (as opposed to the associated commentary) is similarly devoid of guidance on allowable defenses. The DTSA, however differs from the UTSA in this respect: it includes a specific reference to "reverse engineering" and "independent development" as proper means of acquiring trade secrets and creates immunity from criminal or civil federal or state law for whistleblowers, such as employees who reveal trade secrets to federal, state officials, or to an attorney in order to report violations of law. 109 Presumably, antitrust law could also affect the enforceability of provisions in trade secrecy agreements 110 and labor law could impose limits on restraining employee mobility. 111 In contrast, the EU Directive includes several explicit defenses. In addition to independent discovery and 105 Id., § 1(4)(i). See, e.g., InnoSys, Inc. v. Mercer 364 P.3d 1013, 1020 (Utah 2016) (secret revealed in a dispute about unemployment benefits). 106 RESTATEMENT (THIRD) OF UNFAIR COMPETITION, § 39 (emphasis added). Significantly, the first Restatement of Torts protected information "used in one's business." § 757. 107 The rights of non-practicing entities to relief have been in dispute for some time in patent law. (noting that the DTSA may lead to more litigation by so-called trolls). In addition, some states have adopted (or flirted with) the view that there are situations where it is so inevitable that the secret will be disclosed that injunctive relief is necessary. See, e.g., Pepsico, Inc. v. Redmond, 54 F. 3d 1262 (7th Cir. 1995). The DTSA does not, however, appear to have accepted the inevitable disclosure doctrine, see David Bohrer, Threatened Misappropriation of Trade Secrets: Making A Federal (DTSA) Case Out of It, 33 SANTA CLARA HIGH TECH. L.J. 506, 509 (2017).

108 TRIPS Agreement, supra note 22, arts. 13, 17, 30.

109 18 U.S.C. § 1833(2), which applies in both state and federal cases. For federal claims, the revised statute also includes limits on injunctive relief designed to protect employees, 18 U.S.C. § 1836(b)(3)(A)(i)(I). 110 See, e.g., Elizabeth A. Miller, Antitrust Restrictions on Trade Secret Licensing: A Legal Review and Economic Analysis, L. & CONTEMP. PROBS. (1989). 111 See, e.g., Grant R. Garber, Noncompete Clauses: Employee Mobility, Innovation Ecosystems, and Multinational R&D Offshoring, 28 BERKELEY TECH. L.J. 1079, 1102-07 (2013) (comparing restrictions on noncompete agreements in California, China, India, and Brazil); Max Planck Comments, supra note 46, ¶ 15 (noting that this permits each country to protect employees as it sees fit); see also ¶ 32 (suggesting this be made more explicit). reverse engineering, it protects a worker's ability to acquire information for bargaining purposes; it safeguards freedom of expression, including, in particular, freedom to reveal misdoing by the trade secret owner; and there is an open-ended provision allowing use "for the purpose of protecting a legitimate interest." 112 As to remedies, TRIPS requires WTO members to "prevent" disclosure, acquisition, or use-which presumably means that judges must have the authority to enjoin these activities. However, the Agreement does not say how long the injunction must last (in contrast, the UTSA and DTSA require the award be reasonable and preclude the possibility of enjoining the defendant in perpetuity). 113 A long-term injunction maximizes deterrence, but jurisdictions that are concerned about innovation, competition, and information flows could enjoin for shorter periods. For example until the information would have been made public, or they might dissolve the injunction when the information becomes public. 114 Jurisdictions are also free to differ on how they calculate monetary damages. The EU, for example, puts considerable emphasis on proportionate relief, but it provides only very general criteria for judges to use to make that determination. 115 To sum up, despite considerable similarity in the basic contours of all trade secrecy laws-which reflects the influence of U.S. law on TRIPS and TRIPS on the rest of the WTO-there is nevertheless considerable scope for variation in outcome, reflecting different balances among competing policies. Jurisdictions that wish to offer trade secrecy as a less costly alternative to patents can define the subject matter of the law expansively, take a hard line on what is considered secret or readily accessible, and adopt a relaxed approach to the efforts required to maintain secrecy, to what constitutes misappropriation and infringement, and to who is considered an infringer. A jurisdiction that believes trade secrecy is an important component of innovation policy but is nonetheless concerned about information flows and the patent/trade secrecy balance can take actions, such as calibrating its view on accessibility, that increase the innovator's risk that the information will be considered unprotectable. Jurisdictions concerned with protecting employees can limit the scope of protection to technical information, require explicit notice at the time an employee is hired or becomes exposed to sensitive information, refuse to adopt the inevitable disclosure doctrine, and enact defenses protective of employment.

Chinese trade secrecy law has served as an example of a highly nuanced approach. At least until its latest iteration, it has included a list of information that 112 EU Directive, supra note 32, art. 5. 113 See UTSA, supra note 9, § 2(b); DTSA § 1836(b)(3)(A)(i). 114 For the competing rules on the length of injunctions in the United States prior to the adoption of the cannot be regarded as nonpublic (including customs of the industry, information that can be observed from inspection, and information in the published literature). It appeared to require explicit confidentiality agreements and also specified when ex-employees can use their former employer's customer lists. Most interesting is the treatment of tippee liability: the law included what is, in effect, a compulsory license-a provision that permits bona fide purchasers of information to continue to use that information upon payment of reasonable royalty to the owner of the trade secret. 116 Under this approach, the owner of the trade secret recovers its investment, but a tippee (including, perhaps an employer who hires an employee who uses confidential information in his new position) can exploit that knowledge for the benefit of society and build and improve upon it. 117

Part I demonstrated that as foreign countries implement their international obligations to enact trade secrecy protection, they take divergent approaches to key policy issues, including questions on how to foster inventive activity and protect creative investments. In addition, they differ in their views on the patent/trade secrecy trade off, ensuring the mobility of the workforce, and the importance of a robust public domain. U.S. courts can therefore no longer proceed as though there are no foreign interests at stake in trans-border trade secrecy cases. This Part analyzes the Supreme Court's current approach to the extraterritorial application of U.S. law, generally, and in the context of intellectual property litigation. The next Part takes up the question of applying the Court's approach to section 337 exclusion actions and to the DTSA, and compares that to the approach traditionally used in state law claims.

116 For an account of Guangdong Taike Electronic Co. Ltd. v. Wang Xiaowei, where a Chinese appellate court affirmed a ruling that an employee could not be dismissed for taking trade secrets when the only evidence was in illegally monitored email, see Michael The approach of the Supreme Court to the application of U.S. statutes in international cases has had a long and somewhat inconsistent history, reflected in both court opinions and the ALI's successive revisions of the Restatement of Foreign Relations Law. A presumption of territorial application based on the location of the conduct at issue was invoked by U.S. courts throughout the nineteenth and early twentieth century and confirmed by the Supreme Court's 1909 decision in American Banana Company v. United Fruit Co. 118 However, notions of extraterritoriality expanded, with the Supreme Court retreating from the conduct test of American Banana and embracing a broader view of territoriality that included effects within the United States of activity that occurred abroad. In a series of cases, the move to expand conduct to include the acts or the effects of the acts permitted a significantly broader reach for U.S. antitrust laws; 119 But even with respect to statutes that were interpreted to have an extraterritorial reach, their application to foreign conduct was often restrained in light of foreign interests. Of particular significance was the Ninth Circuit's invocation of a comity framework in the antitrust context in Timberlane Lumber v. Bank of American National Trust and Savings Ass'n, 121 where the court identified a list of factors to consider in order to balance U.S. interests in regulating the activity in question against the interests of relevant foreign nations. That approach was later reflected in the second iteration of the Restatement, the Restatement (Third) of Foreign Relations Law. The Restatement (Third) did not include a statement about interpretation of statutes or a rule about extraterritorial application. Instead it identified a variety of legitimate bases for the exercise of prescriptive jurisdiction along with a provision stating that any assertion of prescriptive jurisdiction was nonetheless limited by a principle of "reasonableness" that required both U.S. and foreign interests to be considered. 122 "Reasonableness" in this context was to be understood not merely as the application of discretion, but rather as an obligation among states. Notwithstanding this limiting principle of reasonableness, the approach resulted in the broad extraterritorial application of a number of federal statutes. 123 In its 1990 decision in Equal Employment Opportunity Commission v. Arabian American Oil Co. (Aramco), 124 the Supreme Court, in holding that the Civil Rights Act did not extend to conduct abroad, appeared to return to a presumption against extraterritorial application of U.S. law. The Court emphasized the need to protect against unintended clashes between U.S. and foreign law in the absence of a clear intention by Congress to extend coverage "beyond places over which the United States has sovereignty or has some measure of legislative control." 125 Two years later, in Hartford Fire Insurance Co. v. California, 126 the Supreme Court did not invoke the presumption in a case brought by U.S. plaintiffs under the Sherman Act to conduct that took place abroad but with effects in the 121 549 F.2d 597 (9th Cir. 1976).

122 RESTATEMENT (THIRD) OF FOREIGN RELATIONS LAW OF THE UNITED STATES (Am. Law Inst.1987), §403-Limitations on Jurisdiction to Prescribe ("Even when one of the bases for jurisdiction under § 402 is present, a state may not exercise jurisdiction to prescribe law with respect to a person or activity having connections with another state when the exercise of such jurisdiction is unreasonable"). 123 See, e.g., Hartford Fire Insurance Co. v. California, 509 U.S. 764 (1993) (stating that because the defendant could comply with both U.S. and British law, there was no conflict and the application of the U.S. antitrust laws was not unreasonable. In the securities area, the federal courts applied both a "conduct" and "effects" test to permit extraterritorial application of the U.S. securities laws. In the 2015 Term, the Supreme Court took up the issue of extraterritorial application of a federal statute in the context of the Racketeer Influenced and Corrupt Organizations Act (RICO) in its RJR Nabisco v. European Community decision. 133 The European Community (EC, now the European Union (EU)) had alleged a cigarette smuggling scheme resulting in competitive harm to state-owned businesses, harm to European financial institutions, and lost tax revenue. Much like the amended EEA, RICO includes both criminal and civil remedies. As regards the substantive provisions of RICO, 134 a unanimous Court determined that a number of the predicate acts required for a RICO violation were by their terms intended to apply to conduct occurring abroad, and to the extent that a RICO criminal complaint included such acts, the provisions could have extraterritorial application. However, with respect to a private plaintiff's claims for damages, under a different section of the statute, 135 a divided Court (4-3) held that a civil plaintiff must prove a domestic injury in order to establish liability. The Court explained that without the "check imposed by prosecutorial discretion . . . providing a private civil remedy for foreign conduct creates a potential for international friction beyond that presented by merely applying U.S. substantive law to that foreign conduct." 136 Accordingly, "clear direction from Congress is required." 137 The Court failed to find that direction in the civil provisions of the statute. 136 Nabisco, 136 U.S. at 2107 (internal citation omitted). 137 Id. at 2107. 138 Section 1964 provides a damages action to "[a]ny person injured in his business or property" by a violation of § 1962. First, the Court reasoned that § 1964 was not co-extensive with § 1962, because the phrase "business or property" excluded some injuries, such as personal injuries. Then, the Court reasoned that the presumption must be applied to each provision, and rejected the Second Circuit (and the dissent's) alternative view that § 1964's extraterritorial effect "flows from" § 1962's. Thus, finding no clear indication that the remedial provision applied extraterritorially, the Court held that it did not.

Significantly, the invocation of the presumption against extraterritoriality in this new era of prescriptive jurisdiction does not end the inquiry with respect to the application of a federal statute. Rather, in the first step, where the presumption may be rebutted by congressional intent identified by language or clear purpose of the statute, the Court has identified prescriptive comity as a limitation to avoid unreasonable interference with the interests of other countries. For example, in F. Hoffmann-LaRoche Ltd. v. Empagran S. A., 139 the Court held that application of the Foreign Trade Antitrust Improvements Act of 1962 (the FTAIA) 140 to foreign anti-competitive conduct was nonetheless constrained by principles of prescriptive comity and thus would not extend to foreign plaintiffs who suffered independent foreign harm.

A recent decision of the Second Circuit, In Re Vitamin C Antitrust Litigation, 141 illustrates how prescriptive comity operates as a limitation on the reach of the U.S. antitrust laws. In the Vitamin C case, U.S. purchasers of vitamin C brought an antitrust claim against Chinese defendants, alleging price fixing and supply manipulation of vitamin C exported from China. Although the presumption against extraterritoriality has generally been held not to limit application of U.S. antitrust statutes, the defendants argued that the case should be dismissed because they were acting at the express direction and mandate of the Chinese government. Indeed, the Chinese Ministry of Commerce filed an amicus brief in the district court confirming that the output levels and export price were regulated and approved by the Chinese Government. The district court refused to dismiss the action, but the Court of Appeals for the Second Circuit reversed, stating that the case should be dismissed on comity grounds, referencing unique international concerns implicated by this type of regulatory action by a foreign government.

In situations where a statute is found not to have extraterritorial application, a further inquiry is required. As a second step, the question is whether the application of a particular provision of the statute is nonetheless considered domestic. This inquiry turns on the "focus" of the statute. If the focus is on activity (such as a particular act, injury, or an effect) that occurred in the United States, then the application of the statute is not regarded as extraterritorial and can be applied as a domestic application. However, if the focus is on particular activity that takes place abroad, then it is an impermissible extraterritorial application. For example, in the context of the 1934 Securities Exchange Act, the Supreme Court held in Morrison that the focus of Section 10(b)(5)-the object of the statute's solicitude-was on conduct on an "exchange," and because the purchases of securities were made on a foreign exchange, application of this provision of the Act was impermissibly extraterritorial. Once again, a recent decision of the Second Circuit-now before the Supreme Court-illustrates this approach. In Microsoft v. United States, 143 the court looked to the "focus" of the Stored Communication Act 144 in order to determine whether a search warrant issued to Microsoft, the U.S.-based internet server, was impermissibly extraterritorial because the information was stored in Ireland. Accepting that there was no showing that the presumption against extraterritorial application of the statute had been rebutted, the court turned to the "focus of the statute." The Court of Appeal concluded that the "focus" of the SCA was the privacy of the stored communication (which was in Ireland) and not the disclosure to the government (which would be made in the United States). Thus the court concluded that the warrant did not suffice as a domestic application of the statute.

This general framework is now reflected in the ALI's current revision of the Foreign Relations Law of the United States, Restatement (Fourth). Section 203 is entitled "Presumption Against Extraterritoriality" and provides: "U.S. courts interpret federal statutory provisions to apply only within the territorial jurisdiction of the United States unless there is clear indication of congressional intent to the contrary." 145 Comment c to section 203, titled "Focus of the statutory provision," explains that if the presumption has not been rebutted, a court will look to the focus of the provision, to determine if the application of the provision would be domestic or extraterritorial. 146 An additional section, section 204, titled "Reasonableness in Interpretation," further provides that: "As a matter of prescriptive comity, a U.S. court may interpret federal statutory provisions to include other limitations on their applicability." However, the Restatement (Fourth) insists that section 204 and the reasonableness prong are limited. This prong will not apply when Congress has made its intent to apply a particular provision clear, even if doing so would interfere with the sovereign authority of other states." 147 The Restatement (Fourth) states the principle that courts may avoid only "unreasonable" interference with the authority of other states, and "[i]nterference with the sovereign authority of a foreign state may be reasonable if such application would serve the legitimate interests of the United States." Finally, this principle does not give courts discretion to decline to apply federal law, but rather serves as an approach to statutory interpretation. 148 Thus it appears that the Restatement (Fourth) resists any type of case-specific interest balancing. 149 Nonetheless, it is important to note that case law suggests that reasonableness may have a more robust role to play.

Applying these principles on extraterritoriality to intellectual property is complicated by the fact that the Supreme Court has, in these cases, sent somewhat mixed signals. As noted earlier, the presumption against extraterritorial application has been applied to patent cases, 150 including by lower courts in process patent cases. 151 In Microsoft v. AT&T Corp., 152 the Supreme Court was asked to determine whether a provision of the Patent Act (section 271(f)(1)) 153 that deemed a supplier liable for patent infringement for actively inducing infringement by sending "a substantial portion of the components of a patented invention" abroad was applicable to Microsoft's sending software on a master disk for reproduction outside the United States. Admonishing that the presumption against extraterritoriality "applies with particular force" in patent law, Justice Ginsburg stated:

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. 154 Thus, even though section 271(f) was expressly designed to deal with parties who evade U.S. patent rights by off-shoring the production of patented inventions, 155 the Court refused to extend U.S. patent law to hold Microsoft liable for infringing AT&T's patented software. The Court reasoned that software is not a "component" within the meaning of the statute. Moreover, it noted that the software was reproduced abroad, not in the United States. 149 William S. Dodge, Jurisdiction in the Fourth Restatement of Foreign Relations Law, Yearbook of Private International Law (forthcoming 2017), https://ssrn.com/abstract=2972612 ("the Fourth Restatement concludes that state practice does not support a requirement of case-by-case balancing to establish reasonableness as a matter of international law.") (internal quotations omitted). 150 154 550 U.S. at 454-55 (internal quotations omitted). 155 The statute overruled Deepsouth Packing Co., Inc. v. Laitram Corp., 409 U.S. 902 (1972), which refused to apply the Patent Act when components were sent abroad for assembly there. See also § § 271(f)(2) and 271(g).

At the same time, Justice Ginsburg suggested that the rule for copyright is different. She deduced from the fact that Congress had, in the Digital Millennium Copyright Act (DMCA), 156 protected the encryption of digital content to prevent unlawful reproduction, that the legislature had recognized the need to use U.S. law to protect works from global piracy. 157 As to trademarks, we have seen that the Court in Steele v. Bulova seemingly took a rather expansive view of the reach of trademark law's Lanham Act. 158 It applied the statute to the activities of the defendant (Steele), who had lawfully purchased watch components from the United States and Switzerland and lawfully assembled them in Mexico, where the watches were lawfully stamped with the "Bulova" name, and sold. 159 The Court recited the presumption against extraterritoriality, but held it was rebutted by the jurisdictional provision in the Lanham Act, which asserts Congress's intent "to regulate commerce within the control of Congress." 160 It then held Steele's conduct was unlawful under the Lanham Act. 161 These cases suggest that the type of intellectual property might matter to the extraterritoriality analysis. If so, the question is how trade secrets should be classified. On the one hand, one might view trade secrets as falling on the patent side of the line, where the Court in Microsoft v. AT&T found dispositive the absence of a pointed statement of congressional intent regarding the specific conduct at issue. 162 As Part I suggested, foreign trade secrecy laws, like patent laws, embody different policy judgments about the relative rights of inventors, competitors, and the public." 163 In contrast, there is generally more international agreement on trademark and copyright laws. 164 It is also possible that the Court understands patents as protecting intellectual contributions incorporated mainly in 156 17 U.S.C. § 1201 et seq. 157 550 U.S. at 458. 158 344 U.S. 280 (1952). 159 Id. at 281-82. 160 Id. at 283-284 (citing 15 U.S.C. § 1127. As noted later, reliance on a jurisdictional provision relating to foreign commerce has not generally been sufficient to overcome the presumption. See infra notes 203-204.). 161 Id. at 289. 162 550 U.S. at 458. 163 Id. 164 The TRIPS Agreement, for example, incorporates the highly detailed Berne Convention by reference, TRIPS Agreement, supra note 22, art. 9.1, and adds considerably to the Paris Convention's substantive rules on trademark protection, Paris Convention, supra note 21, arts. 15-21. By contrast, the TRIPS Agreement is the first multinational agreement to impose minimum standards of substantive protection for patents and trade secrets; the trade secrecy section contains few details. tangible materials, which are usually difficult to manufacture and transport across borders, but sees trademark and copyright law as protecting information (the cognitive impact of marks; the works embodied in digitized books, films, and recordings) that are readily reproduced and can be instantaneously distributed globally. 165 Because the risk of infringement abroad is so much greater in the latter cases, it may be that the Court is more willing to find extraterritorial intent in the copyright and trademark context. Technological trade secrets tend to be embodied in goods; in this respect, they are similar to patents. On the other hand, trade secrets are, in essence, information. As the Introduction suggests, modern business practices and technological developments have increased the risk that trade secrets (that is, secret information) will be transferred around the world, taken without authorization, and exploited in multiple markets. Once known outside of the putative trade secret owner's business, the value of the information is significantly diminished. Thus, although the risk of infringement may not be as great as in copyright or trademark cases, the injury suffered is likely to be more severe. In that sense, perhaps trade secrecy statutes should be treated more like copyright and trademark measures for purposes of the extraterritoriality analysis.

Factors other than the nature of the intellectual property right in question may also explain why the outcome in Steele v. Bulova is quite different from the result in Microsoft v. AT& T. Perhaps it is not the language in the jurisdictional grant in Bulova that necessarily led the Court to approve the extraterritorial reach. In addition to the reference to the "broad [prescriptive] jurisdictional grant," the Court in Bulova listed a set of facts justifying application of the statute in the particular case. 166 Some of the defendant's infringing watches had "filtered" into the United States. U.S. consumers in the border area attributed problems with the watches to the U.S firm; accordingly, there were potential adverse effects on the Bulova brand. 167 In addition, the defendant was a U.S. citizen; parts had been purchased in the United States; and because the defendant's Mexican trademark had been cancelled prior to Supreme Court consideration of the case, the possibility of conflict with foreign law had been eliminated. 168 165 Indeed, the copyright cases Justice Ginsburg was likely considering all involved an unauthorized reproduction (called a "predicate act" or a "root copy") that occurred in the United States, prior to further reproduction abroad. See, e.g., Jane C. Ginsburg, Global Use/territorial Rights: Private International Law Questions of the Global Information Infrastructure, 42 J. COPYRIGHT SOC'Y U.S.A. 318, 335 (1995) ("In the U.S., some courts have simplified the choice of law problem by applying U.S. law to the entirety of a multinational infringement claim, when the root act of copying occurred in the U.S."), comparing Subafilms, Ltd. v. MGM-Pathe Communications Co., 24 F.3d 1088 (1994) with Update Art v. Modiin Publishers, 843 F.2d 67 (2d Cir. 1988). By contrast, the software at issue in Microsoft was reproduced abroad. In this respect, claims regarding trade secrets misappropriated abroad come closer to the claim asserted by AT&T than to the claims in the root copy cases. 166 344 U.S. at 284. 167 Id. at 285. 168 Id. at 285-288.

It is uncertain which of the many factors cited in Bulova were considered determinative by the Court. 169 In the heavily cited case, Vanity Fair Mills, Inc. v. T. Eaton Co., 170 the Court of Appeals for the Second Circuit distilled the Bulova factors into a three-part framework that considered: (1) whether the defendant is a U.S. citizen; (2) whether the conduct has a substantial effect on U.S. commerce; and (3) the prospect of a conflict with foreign laws. 171 In subsequent lower court cases, courts differed as to the relative importance of particular factors, 172 as well as their precise formulation. 173 However, recent decisions take a different tack: they find the Lanham Act to have extraterritorial effect, but then place considerable emphasis on concerns of comity in determining the Act's reach. 174 For example, in Trader Joe's v. Hallatt, 175 the Ninth Circuit Court of Appeals invoked its earlier Timberlane comity framework to justify application of the Lanham Act to reach a Canadian vendor. In that case, the defendant Hallatt operated a store in Vancouver, Canada, where he sold authentic Trader Joe's products albeit at inflated prices, using, without authorization, marks that were potentially confusingly similar to those of Trader Joe's. 176 When Trader Joe's subsequently refused to sell its products to him, Hallatt donned disguises to buy from the stores and also hired agents to make bulk purchases for him. 177 Trader Joe's sued under the Lanham Act, and the Ninth Circuit, applying the Timberlane factors, held that the Lanham Act applied extraterritorially because there were effects on American commerce, a cognizable injury to the plaintiffs under the Act, and the interests of and links to American commerce were sufficiently strong in relation to those of the foreign nation. 178 Although the infringing goods did not 169 See Curtis A. Bradley, Territorial Intellectual Property Rights in an Age of Globalism, 37 VA. J. INT'L L. 505, 528 (1997) ("[T]he [Bulova] Court did not explain which of the facts of the case, if any, were essential to liability or how such facts were to be weighed"). 170 234 F.2d 633 (2d Cir. 1956). 171 Id. at 642. 172 See, e.g., McBee v. Delica Co., 417 F.3d 107, 121 (1st Cir. 2005) ("[W]e first ask whether the defendant is an American citizen, and if he is not, then we use the substantial effects test as the sole touchstone to determine jurisdiction."). 173 See, e.g., Nintendo of Am., Ltd. v. Aero Power Co., 34 F.3d 246, 250-51 (4th Cir. 1994) (requiring a "significant effect" rather than a "substantial effect" on United States commerce). 174 176 Id. at 964-65. 177 Id. 178 Id. at 975. flow back into the United States, as they did in the Bulova case, the court nonetheless found that the reputational harm suffered by Trader Joe's created the necessary effect on U.S. commerce. In addition, the court emphasized that the defendant had lawful permanent status in the United States (thereby "subject[ing] himself to the laws of [the United States]" 179 ), and engaged in commercial activity in the United States as part of his infringing scheme. Accordingly, the defendant's actions involved some domestic (as well as foreign) conduct. 180 In assessing the U.S. and foreign intereststhe comity factorsthe court did not find reason to limit the application of the Lanham Act. 181 There was no pending adversarial proceedings between Trader Joe's and Hallatt in Canada and because Hallat had permanent resident status, orders made by the court (damages and an injunction) were easily enforceable; Trader Joe's U.S. trademarks were affected; and an essential, albeit small, part of defendant's conduct took place in the United States. 182

Since the Supreme Court appears to distinguish among intellectual property rights, and in view of the significant policy difference between trademark and trade secrecy law, the Lanham Act extraterritoriality cases are not a perfect analogy for trade secrecy cases. Nonetheless, the factors developed in the aftermath of Bulova suggest the importance of context in determining the extraterritorial reach of federal statutes. Thus, they inform the application of the Supreme Court's current two-step approach in analyzing section 337 actions, the DTSA, and state law. We look at each statutory scheme in turn, and then reconsider federal law in light of state law experience.

The Tariff Act of 1930 gives the International Trade Commission (ITC), a quasi-judicial federal agency with roots in the Tariff Commission of 1916, authority to exclude from the United States goods made abroad unlawfully. Section 337 of the Tariff Act defines "unlawful" conduct to include "unfair methods of competition and unfair acts in the importation of articles . . . into the United States." 183 Section 337 complaints are brought by private individuals and adjudicated by the ITC, which in recent years, has become an increasingly 179 Id. at 973. 180

181 Trader Joe's, 835 F.3d at 975. 182 Id. at 973-75. important venue for protecting trade secrets. 184 The ITC has consistently construed section 337(a)(1)(A) to reach trade secret misappropriation, 185 an interpretation that is supported by the Federal Circuit's view that the statute gives the ITC "broad authority to address every type and form of unfair trade practice." 186 Upon finding a violation of section 337, the ITC "shall direct that the articles concerned . . . be excluded from entry into the United States" subject to limited exceptions on public interest grounds. 187 The resulting nationwide exclusion order is reviewable by the President, who may "disapprove[]" the ITC's determination for policy reasons, with the result that the exclusion order thereafter "ha[s] no force or effect." 188 Although the ITC has apparently excluded goods based on trade secret misappropriations occurring abroad for some time, 189 the question of the extraterritorial reach of section 337 first came before the Federal Circuit in TianRui Group Co. Ltd. v. International Trade Commission. 190 In TianRui, Amsted-a U.S. railway wheel manufacturer-had developed multiple secret processes for manufacturing cast steel railway wheels. 191 A Chinese firm (TianRui) acquired the secret wheel-making technology by poaching employees from one of Amsted's licensees in China after its own negotiations with Amsted over licensing terms broke down. TianRui used the information thus acquired to manufacture wheels in China, which it then sought to distribute in the United States. 192 Amsted filed a complaint with the ITC seeking an exclusion order for a violation of section 337. The ITC assigned the dispute to an Administrative Law Judge (ALJ) who conducted an evidentiary hearing and issued a recommended determination. 193 With the DTSA not yet enacted, the ALJ looked to the law of the state where Amsted was headquartered-Illinois-on the question of whether trade secret misappropriation had occurred. 194 As for the question of the prescriptive reach of section 337, the ALJ rejected TianRui's argument that the application of the statute was limited to misappropriation in the United States and emphasized the statute's focus on "the nexus between the imported articles and the unfair methods of competition." 195 On appeal, the Federal Circuit rejected the application of Illinois law to determine whether the activity was unfair, instead finding the question of what conduct constitutes "unfair methods of competition and unfair acts" to be one of federal law, as reflected in, for example, the Restatement (Third) of Unfair Competition and the UTSA. 196 As to the reach of the statute, the Federal Circuit undertook an extended discussion of extraterritoriality, addressing both the presumption as applied to this statute and the question whether the exclusion order operated as a domestic or extraterritorial application. 197 In a somewhat blurred analysis of the two issues, the court held that the section 337 exclusion order was not an improper extraterritorial application. 198 192 Id. at 1324-25. 193 When the ITC receives a complaint, it first decides whether to investigate by vote of the Commissioners. If the investigation proceeds, the ITC assigns the dispute to an ALJ who conducts a hearing that culminates in the issuance of an "recommended determination." See 19 C.F.R. 210(b) (2012). The ITC may subsequently review the ALJ's recommendation and reaches a final determination as to whether there is a violation of section 337. Upon finding a violation of section 337, the ITC "shall direct that the articles concerned . . . be excluded from entry into the United States" subject to various public interest considerations. 194 TianRui, 661 F.3d at 1325. 195 Id. at 1325. 196 Id. at 1327-1328. 197 Id. at 1326-27.

A case can be made either that the language of the statute supports an extraterritorial application of section 337 or that the facts of TianRui constitute a domestic application of the statute; however, the Federal Circuit's opinion does not make the most convincing case for either position. 199 Thus, the Federal Circuit identified several reasons why the presumption against extraterritoriality "does not govern" the case at hand. 200 First, in finding that section 337 applies extraterritorially, the Federal Circuit effectively grounded its entire analysis on the fact that section 337(a)(1)(A)'s phrase "in the importation of articles" indicates that the law targets "an inherently international transaction-importation." 201 The primary inference it drew from the transnational nature of importation was that Congress did not "ha[ve] only 'domestic concerns in mind.'" 202 To the extent that the Federal Circuit understood importation to imply foreign commerce, its analysis would be incomplete, in that the Supreme Court has "repeatedly held that even statutes that . . . expressly refer to foreign commerce do not apply abroad." 203 In fact, the case the Federal Circuit cited, Pasquantino v. United States, 204 drew the opposite conclusion from similar language. However, the Federal Circuit may have understood importation as a reference to extraterritorial scope. 205 But because 199 For what is in our view a better account favoring section 337's extraterritorial reach, see notes infra notes 239-240 and accompanying text. For the argument that TianRui could be said to represent a domestic application, see text accompanying note 241 infra. 200 TianRui, 661 F.3d at 1329. This analysis will address the first two reasons. The third reason related to legislative history, and is less important for present purposes. 201 Id. (internal quotations omitted). 202 Id., citing Pasquantino v. United States, 544 U.S. 349, 371-72 (2005). The irony of citing Pasquantino is that although the Court in Pasquantino did say "this is surely not a statute in which Congress had only domestic concerns in mind," the Court explicitly did not find that the wire fraud statute applied extraterritorially. See id. at 371 ([O]ur interpretation of the wire fraud statute does not give it "extraterritorial effect."). At that time Pasquantino was decided, there was not yet a clearly delineated two-step inquiry associated with the presumption against extraterritoriality. Thus, as the court considered whether it was appropriate to apply the statute to activities that encompassed some domestic and some foreign conduct, it found it relevant to note that Congress had in mind, borderspanning schemes. However, it is a clear misreading of Pasquantino to invoke it for the principle that congressional cognizance of border-spanning schemes constitutes a clear intent for a law to apply extraterritorially. 203 Morrison v. Nat'l Australia Bank Ltd., 561 U.S. 247, 262-3 (2010) (quoting E.E.O.C. v. Arabian American Oil Co., 499 U.S. 244, 251 (1991)). 204 544 U.S. at 372. 205 661 F.3d at 1329. The Federal Circuit did emphasize that section 337 was similar to immigration statutes that permit barring the entry of non-citizens who make false statements to obtain entry to the United States (thereby arguably considering foreign conduct). However, the analogy probably does not survive recent Supreme Court precedent. In Nabisco, the Court flatly rejected an argument that the extraterritorial reach of RICO should be interpreted in the same fashion as that of the Clayton Act because RICO was modeled after the Clayton Act, RJR Nabisco, Inc. v. European Cmty., 136 S. Ct. 2090, 2109 (2016). it did not offer any support for that view, it failed to identify the "clear indication" of extraterritorial application as required by Morrison just one year earlier. 206 A case for the extraterritorial application of section 337 can be made more persuasively. Nabisco emphasized that context is an important factor in determining whether a statute has a "clear indication" of extraterritorial application. 207 There, context was "dispositive": although Congress had "not expressly said that section 1962(c) applies to patterns of racketeering activity in foreign countries," the fact that racketeering activity was defined to include predicate acts that do apply abroad was sufficient. 208 Although RICO-as the Court noted-has a "unique structure," the point about context is a more general one. 209 To the extent that the context of section 337 actions includes actions related to imported articles, which almost by definition have had a trade secret incorporated abroad, the Federal Circuit could have taken up Morrison's invitation to consult context.

Further strengthening the case for the result-if not the reasoning-in TianRui is the fact that actions before the ITC differ in important ways from the civil actions the Supreme Court has considered in other types of cases. Although section 337 proceedings are initiated by private parties bringing complaints, the ITC is itself an Article I tribunal that exercises more control over the proceeding than an Article III court. 210 The limited remedy (exclusion), along with these procedural differences, would seem to lower the bar for finding congressional intent that the statute applies extraterritorially, at least to some extent. 211 In light of that limited remedy, the argument that a statute targeting unfairness "in the importation of articles" 212 begins to present a stronger case for its extraterritorial application. But the reason is not because the statute "refers" to transnational conduct. Rather, congressional intent regarding section 337(a)(1)(A)'s geographic 206 Morrison, 561 U.S. at 255 ("[U]nless there is the affirmative intention of the Congress clearly expressed to give a statute extraterritorial effect, we must presume it is primarily concerned with domestic conditions. When a statute gives no clear indication of an extraterritorial application, it has none."). 207 Nabisco, 136 S.Ct. at 2102-03. 208 Id. 209 Id. at 2103. 210 Elizabeth A. Rowe & Daniel M. Mahfood, Trade Secrets, Trade, and Extraterritoriality, 66 ALA. L. REV. 63, 92 (2014) (describing ITC procedures following receipt of a complaint, which include voting on whether to commence an investigation, conducting its own investigation following the evidentiary hearings that take place under the supervision of an ALJ). 211 Cf. Morrison v. Nat'l Australia Bank Ltd., 561 U.S. 247, 269 (2010) ("The probability of incompatibility with the applicable laws of other countries is so obvious that if Congress intended such foreign application 'it would have addressed the subject of conflicts with foreign laws and procedures.'") (citation omitted). scope begins to resemble the Court's analysis in Bulova: 213 acts abroad that would otherwise be lawful (because they are beyond the reach of U.S. law) become unlawful (i.e., may be reached by U.S. law) when they are part of a scheme that involves importation into the United States. 214 Such a construction of congressional intent-unlike one that places great weight on "references" to foreign commerce-is not plainly in tension with the Court's "presumption against extraterritoriality" doctrine.

When one further considers that the President may veto ITC determinations for incompatibility with foreign relations or trade policy, 215 the extraterritorial application of section 337 seems basically consistent with modern Supreme Court jurisprudence; it is, in effect, similar to Nabisco, where the civil and criminal provisions were distinguished in part on the basis of the prosecutorial discretion present in the criminal context. Similarly, the limitation in section 337 to situations in which there is a domestic injury further reflects comity concerns. In short, the extraterritorial reach of section 337 is supported by: (1) the geographically limited remedy of exclusion; (2) the fact that importation takes place in the context of transnational commerce; (3) presidential review, traditionally exercised to safeguard foreign or economic policy concerns, is available; and (4) the requirement of a domestic injury.

However, this interpretation of section 337 is not without its own contradictions. Section 337(a)(1)(B)-a provision in the same statute that was explicitly adopted to provide extraterritorial reach to one specific situationindicates that Congress knows how to draft a measure to achieve an extraterritorial effect. That provision covers "importation into the United States . . . of articles" that are made "by means of [] a process covered by the claims of a valid and enforceable United States patent." 216 The language in that provision was modified after the Federal Circuit's predecessor court, the Court of Customs and Patent Appeals (CCPA), held in In re Amtorg that section 337 could not be used to exclude from importation products made abroad by a process patented in the United States (a scenario very similar to the question whether to exclude a product made abroad with information that would have been unlawful to appropriate in the United States). 217 The language at issue in Amtorg was in relevant respects 213 See supra note 166-174 and accompanying text. 214 It is, of course, an independent question whether the foreign acts in question might constitute unlawful conduct under the laws of the state where they occurred. 215 See 19 U.S.C. 1337(j)(2) ("If. . . the President, for policy reasons, disapproves such determination and notifies the Commission of his disapproval, then, effective on the date of such notice, such determination and the [remedial] action . . . with respect thereto shall have no force or effect."). The President's appeal power is controversial, both as to what is meant by "policy reasons" and from a normative perspective, see Nicolaas T. Bressers, Comment, A Presidential Remedy Under Administrative Control: Why Section 337(j) Should be Repealed, 19 MARQ. INTELL. PROP. L. REV. 99, 117 (2015) (noting that the authority is considered broad in scope). identical to that in section 337(a)(1)(A) today: "unfair methods of competition and unfair acts in the importation of articles." 218 Congress thereafter enacted what is now section 337(a)(1)(B)(ii) because "owners of American process patent[s] are helpless to prevent the infringement abroad of their patent rights." 219 The new language necessarily indicates the provision has extraterritorial reach because articles imported into the United States are "made" outside the United States. Thus, section 337(a)(1)(B)(ii) contains-both textually and as a matter of legislative history-a clear indication of extraterritorial application, and would meet Morrison's standard. 220 By contrast, section 337(a)(1)(A) was not amended. Thus, it is not evident that the Morrison "clear indication" standard has been met. 221 While reading section 337(a)(1)(A) to have only domestic application is difficult to square with the language of "all unfair methods" or "unfair acts in the importation of articles," it could be interpreted as envisioning exclusion only when a trade secret is taken without authority from the United States for use in manufacturing abroad items destined for the U.S. market. 222 There was, however, a second reason why the TianRui court found that the presumption against extraterritoriality "d[id] not govern" the case. 223 The court held the presumption posed no obstacle because it found the ITC had not applied section 337 to "purely extraterritorial conduct," given that there was importation into the United States and domestic injury. 224 The court initially asserted that "importation and the resulting domestic injury" constitutes "the statute's focus," suggesting that it considered the case to represent a domestic application of section 337. 225 However, the ensuing discussion revealed confusion about how the "focus" inquiry operates in the context of the presumption against extraterritoriality. First, the court described Morrison as "focusing the extraterritoriality analysis on the 218 The majority and the dissent in TianRui disagreed about the extent to which Amtorg reflected unique historical factors. However, somewhat oddly, the majority noted that the decision by Congress to amend the language at issue in Amtorg supported the result it reached in the case-despite the fact that the language found in Amtorg to be insufficient to support extraterritorial application continues to exist in the contemporary version of section 337(a)(1)(A). See TianRui Group Co. Ltd. v. Int'l Trade Comm'n, 661 F.3d 1322, 1333-34 (Fed. Cir. 2011). 219 See H.R. REP. NO. 76-1781, 3d Sess. 4 (1940). 220 19 U.S.C. § 1337(a)(1)(B)(ii). 221 See also 35 U.S.C. § 271(g), which creates infringement liability when products made abroad by processes patented in the United States are sold, offered for sale, or imported into the U.S. market. 222 Cf. General Motors Corp. v. Lopez, 948 F.Supp. 684 (E.D.Mich. 1996) (secret information taken from General Motors in the United States for use by its rival, Volkswagen, in its German facility, for use in the United States in competition with GM). 223 TianRui, 661 F.3d at 1329. 224 Id.

'objects of the statute's solicitude'." 226 Of course, this is only true in a limited way: as Morrison makes clear, the effort to identify the "objects of the statute's solicitude" only becomes relevant at the second step, after it is determined that the law in question does not apply extraterritorially. At that point, the "objects of the statute's solicitude" (i.e., the focus) must be identified in order to classify the proposed application as either domestic or (impermissibly) extraterritorial. 227 Thus, it is only correct to say that the analysis of extraterritoriality focuses on the "objects of the statute's solicitude" if it was already determined that the statute lacks extraterritorial application. Yet in the very next sentence, after characterizing Morrison in this fashion, the court concluded, "the presumption against extraterritorial application does not apply" because the foreign conduct simply establishes one "element of a claim alleging a domestic injury and seeking a wholly domestic remedy." 228 The Federal Circuit thus blurred the analytically distinct inquiries into: (1) whether the presumption applies to the statute; (2) whether the presumption has been rebutted; and (3) what the focus of the statute is (and whether, in light of the focus, a given application is domestic or extraterritorial).

The lack of clarity in the analysis renders it difficult to discern what limiting principle would cabin the majority's application of section 337, as Judge Moore observed in a vigorous dissent:

The The majority responded that it was not interpreting section 337 to "police Chinese business practices," but simply recognizing that the statute "sets the conditions under which products may be imported into the United States." 230 But it is not entirely accurate to say that an exclusion order does not "police Chinese business practices." Because the U.S. market is very large, closing it to imports can have a significant effect on foreign producers. In some cases, the loss of potential revenue may be large enough to make continued global operations uneconomical. 231 To avoid the threat of a section 337 action, foreign producers may well choose to guide their activity by the dictates of U.S. law rather than the law of the country in which they operate. Given the policy bases for differences among national trade secrecy laws, exclusion could arguably produce exactly the sort of clash that has concerned the Supreme Court in other contexts.

The prospect of regulatory overreach was presented in an even sharper fashion in a recent section 337(a)(1)(A) case in which Chinese courts had already adjudicated the trade secret misappropriation claim, and had absolved the defendant (the respondent in the section 337 action) of liability. Sino Legend v. Int'l Trade Commission 232 concerned a New York-based chemical manufacturer (SI Group) that brought a section 337 claim against a Chinese company, Sino Legend, for trade secret misappropriation. The secrets in question related to processes for producing a resin used in the production of synthetic rubber tires containing multiple layers; they were transferred by the complainant, SI Group, a New York corporation, to China under a confidentiality agreement expressly governed by New York law. 233 Sino Legend allegedly poached employees from a Chinese subsidiary of SI Group, who worked with Sino Legend to take and use SI Group's trade secrets. 234 As in TianRui, all the essential conduct constituting misappropriation occurred abroad. Nonetheless, the ITC determined that it was bound by TianRui, and consequently issued a limited exclusion remedy for a period of ten years. In the subsequent appeal, the judges at oral argument framed the case as a referendum on TianRui. 235 Two days after the oral argument, the 230 Id. at 1330. 231 234 Id. at *52. 235 Oral Argument at 1:42, Sino Legend v. ITC, 623 F. App'x 1016 (Fed. Cir. 2015) ("Are you asking us to overrule TianRui?"); id. at 6:16 ("So where that leaves you is an argument as to whether TianRui was correct or not."); id. at 6:48 ("It seems to me that most of the briefing and most of the argument is based on TianRui, that it was wrong, it was wrongly decided.").

Federal Circuit summarily affirmed the ITC's determination without an opinion. 236 Despite framing the appeal as an attempt to re-open legal questions settled in TianRui, important factual differences distinguished the two cases. In TianRui, the court discussed at some length whether there was any tension between Chinese and U.S. law, ultimately concluding that it "could not discern any relevant difference" on the record below. 237 By contrast, in Sino Legend, SI Group had brought multiple actions in China over a period of four years regarding the alleged misappropriation, all of which were unsuccessful. The Chinese Trade Remedy and Investigation Bureau of the Ministry of Commerce (TRB) filed two amicus briefs-one in support of rehearing en banc at the Federal Circuit, and another in support of the subsequent certiorari petition once the petition for rehearing was denied. In the former, the TRB elaborated:

[T]he ITC does not have jurisdiction over conduct that occurs entirely in China, especially . . . where the very same issues have been resolved by China's competent courts. The TRB expresses its disappointment and displeasure with this aspect of the adjudication to date. It is also the TRB's view that the astonishing ruling in this case-that the decisions of Chinese courts on the identical issue between the same parties are totally irrelevant and, therefore, can simply be ignored by the ITCfrustrates the respect properly due to the judicial sovereignty of any nation and treaty partner. 238 Thus, the summary affirmance without opinion might suggest that the comity (or "sovereignty") concerns discussed in the TRB brief were wholly without significance in the eyes of the Federal Circuit. However, even when comity concerns are taken into account, the results in TianRui and Sino Legend can be supported. The application of section 337 to foreign conduct can be justified on the basis of the nexus created by importation into the United States and the development of the trade secrets in the United States. Even in Sino Legend, exclusion is not necessarily inconsistent with the Chinese judgments absolving the ITC respondent of liability. Section 337 is a limited remedy; it is not a civil action for a trade secrecy violation, does not provide for an injunction or damages, and protects only the U.S. market. 239 Moreover, as cases such as Empagran, Vitamin C, and the 50 years of trademark cases following Bulova indicate, prescriptive comity may still operate in a particular case as a limitation on the application of a statute with extraterritorial application. 240 That said, the decision would have been more persuasive if the Federal Circuit had not relied on the TianRui's assumption of no "conflict between the principles of misappropriation that the Commission applied and Chinese trade secret law" 241 and instead engaged in a comity analysis that took the Chinese judgments into account and inquired whether the threat of section 337 exclusion orders would affect conduct in China and interfere with Chinese policy regarding employee mobility and the tippee's expectations.

We have offered an analysis of extraterritorial reach in TianRui and Sino Legend that attempts to reconcile the decision with recent Supreme Court jurisprudence and concerns of comity like the ones implicitly raised by Judge Moore and squarely presented in Sino Legend. 242 As we demonstrated, the cases can be reconciled with the step-one presumption against extraterritoriality by interpreting section 337(a)(1)(A) as evincing congressional intent to reach foreign conduct. However, that interpretation sits uneasily with the express language of other parts of the statute. Thus, we believe that the court was on firmer ground when it moved to the second stepan interpretation of section 337 that emphasized its "focus." It should, in our view, have reasoned that when section 337(a)(1)(A) is invoked to exclude goods embodying misappropriated trade secrets that causes injury, the focus of the statute-the "object of its solicitude"- 240 Cf. Elizabeth A. Rowe & Daniel M. Mahfood, Trade Secrets, Trade, and Extraterritoriality, 66 ALA. L. REV. 63, 92 (2014) ("TianRui's reasoning does not foreclose the possibility that extraterritorial application of section 337 could be upheld based solely on the court's first reason-i.e., the jurisdictional nexus to the United States of attempted importation."). Vitamin C, however, is a somewhat unusual case. The Second Circuit framed the question as one of whether U.S. courts should "abstain from exercising jurisdiction," 837 F.3d at 179, despite the fact that the doctrine of international comity-based abstention only applies while parallel proceedings are pending abroad. Second, the court used an interest-balancing approach associated with Timberlane Lumber Co. v. Bank of America, 549 F.2d 597 (9th Cir. 1976), when arguably the case should have been analyzed under the rubric of foreign compulsion because the defendants argued the violations were compelled as a matter of foreign law. For a discussion of this question, see William S. Dodge (excluding products made using a secret method of manufacturing stainless steel stolen in Italy and used in Germany on the theory it interfered with the U.S. business of the Italian firm's U.S. subsidiary-the defendants were convicted of a crime in Italy and a civil case was pending). 242 For a wide-ranging account of international comity, see William S. Dodge, International Comity in American Law, 115 COLUM. L. REV. 2071 (2015) (discussing various doctrines associated with prescriptive comity, adjudicative comity, and sovereign party comity as reflecting principles of restraint and principles of recognition).

is the threat of "destroy[ing] or substantially injur[ing] an industry in the United States" through the loss, in the United States, of a competitive advantage conferred by its secret technology. 243 Accordingly, preventing the importation of articles that incorporate (or were produced using) what we term "U.S. trade secrets"-meaning information developed in the United States by a U.S. or foreign companyrepresents a domestic application of the statute. Had the court followed that approach, it would have arrived at an analysis better tailored to the interests that Congress sought to protect in enacting section 337. In particular, a "focus" analysis avoids the potential broad extraterritorial application that troubled Judge Moore in applying section 337 to other types of cases. 244 While this interpretation would mean that information developed abroad will often not qualify for protection, 245 support for this approach can be found in Congress's focus on the importation of articles that threaten to "destroy or substantially injure [a domestic] industry" and on the concern section 337 evinces for promoting innovation and creating employment and training opportunities in the United States by preserving the U.S. market for holders of U.S. trade secrets. It is consistent with Congress's decision not to provide for a global injunction, or another far-reaching remedial scheme. 246 And it is consistent with international law, which requires WTO members to offer copyright and patent protection to works developed abroad, 247 but does not impose the same detailed obligation for trade secrets. 248 In sum, both the restriction on triggering application of section 337 (that there must be unlawful conduct in connection with importation and a domestic injury) and the scope of the remedy (national exclusion) support the view 243 19 U.S.C. § 1337(a)(1)(A)(i) (2016). 244 Cf. InterDigital Comm'ns, LLC v. ITC, 707 F.3d 1295, 1301, 1303 (Fed. Cir. 2013) (noting, in the context of interpreting the domestic injury requirement of section 337(a)(2), that the goal of section 337 is to protect companies that "have a substantial stake in the United States" and have made significant investments in "engineering, research, and development"). 245 See In the Matter of Certain Stainless Steel Products, discussed in note 241, supra. 246 Cf. Google, Inc. v. Equustek Solutions, (2017) S.C.R. 36602 (Can.), see infra notes 302. 247 See TRIPS Agreement, supra note 22, art. 27.1 (requiring WTO countries to recognize patent rights without discrimination as to the place of invention): Berne Convention, supra note 20, art. 3.1 (extending protection to authors who are nationals of the Berne Union and to works first published in the Union). 248 See Berne Convention, supra note 20, art. 3.1 and TRIPS Agreement, supra note 22, art. 39. In this respect, trade secrets are similar to trademarks, where international law contemplates local understanding or registration as a prerequisite to recognition, see TRIPS Agreement, arts. 15 and 16. that a trade secret developed in the United States is the focus of the statute. 249 Furthermore, this approach emphasizes the salience of the U.S. regulatory interest in protecting U.S. trade secrets to justify application of section 337(a)(1)(A). To be sure, the cases do not always include information on where the secret at issue was developed. Presumably, if the place of development became an issue, evidence on the place of development would be introduced at trial.

To put this another way, interpreting section 337 as focusing on the concept of a "U.S. trade secret" allows courts to confine the scope of the statute to situations where U.S. interests are paramount. Furthermore, it puts those exposed to these secrets on notice that U.S. law will apply to the question of importation. For example, in TianRui, the information was developed in the United States and then licensed to a firm in China, from which it was stolen; in Sino Legend, the transfer to China from a New York firm was even governed by a confidentiality agreement specifying the application of New York law. In such cases, U.S. interests would be frustrated if the secret could be freely used on products sold in the United States. Furthermore, the licensee (and, arguably, the misappropriator, if not a tippee) would have been aware of the provenance of the information and would therefore have known to conform their behavior to the U.S. approach to balancing interests in confidentiality, employee mobility, and investment. And because the potential application of U.S. rather than Chinese law may have been an important factor in the licensor's initial decision to transfer the technology to China, the application of U.S. law furthers the licensor's interests and promotes broad dissemination of new knowledge.

In contrast, when the information is developed abroad and taken abroad for incorporation into products placed on the global market, it is difficult to see how the appropriation riggers the policies that support U.S. trade secrecy protection. Such cases are more akin to the situations that Judge Moore was concerned with in her dissent: they apply U.S. law to activity occurring entirely abroad; in many cases, the parties involved in the technology transfer may have had no reason to believe that U.S. law might apply and thus may have conformed their behavior on information sharing and use to the laws of the country in which they acted. U.S. law should, in short, have no place in policing their behavior. To be sure, there will also be intermediate cases. The trade secret may have been developed by a foreign subsidiary of a U.S. firm, which might trigger a U.S. interest in promoting innovation, albeit in an attenuated way. Or the information may have been developed for customers in the United States. 250 In that situation, there is a U.S. interest in the sense that promoting the advance would give U.S consumers more choice, but it is more difficult to consider these "U.S. trade secrets" because it is harder to classify that interest as the "focus" of the statute. Moreover, a party can protect itself by contracting for the application of U.S. law. It can also negotiate for an indemnity clause to cover situations where the forum state does not honor the choice of law provision or the information is transferred without alerting subsequent transferees to the obligation to keep the information secret.

It is important to emphasize that in all these situations the statute permits the ITC or the court to apply a comity analysis, and the President to intervene on policy grounds to prevent the sorts of clashes that have concerned the Supreme Court. It is also worth noting that if section 337 were to be interpreted as focusing on only the secrets of U.S.-owned firms, it would likely violate the TRIPS obligation to accord to right holders in all WTO countries protection that is the same as that offered to U.S. right holders or the antidiscrimination requirements of the GATT. 251 For that reason, we have noted the possibility of adopting a somewhat more generous interpretation to cover trade secrets developed in the United States, regardless of the nationality of the right holder or the source of the goods.

The Defend Trade Secrets Act of 2016 (DTSA) is an amendment to the 1996 Economic Espionage Act (EEA) that became effective May 11, 2016. As originally enacted, the EEA imposed criminal penalties for unauthorized takings to benefit foreign governments ("espionage") or private parties ("theft"). 252 The DTSA was enacted, principally, to provide a private right of action. The relevant provision - § 1836(b) -states that: "An owner of a trade secret that is misappropriated may bring a civil action under this subsection if the trade secret is related to a product or service used in, or intended for use in, interstate or foreign commerce." 253 Thus, it adds to the criminal statute a private federal cause of action for trade secret misappropriation when interstate or foreign commerce is involved, and provides for federal court jurisdiction in such actions. The federal remedies may include injunctive relief, damages, and the possibility of exemplary damages for willful and malicious appropriation up to two-times the actual damages proven. 254 Ostensibly, the extraterritorial reach of the DTSA is easily analyzed because section1837 of the original enactment, the EEA, includes a clear statement of legislative intent. It specifies:

This chapter also applies to conduct occurring outside the United States if (1) the offender is a natural person who is a citizen or permanent resident alien of the United States, or an organization organized under the laws of the United States or a State or political subdivision thereof; or (2) an act in furtherance of the offense was committed in the United States. 255 Since section 1836 is in "This chapter," presumably the provision on extraterritoriality applies as well to the civil claims the statute creates. Indeed, section 5 of the DTSA states that it is the sense of Congress that "trade secret theft occurs in the United States and around the world" and that "trade secret theft, wherever it occurs, harms the companies that own the trade secrets and the employees of the companies." It also adds that the Act "applies broadly to protect trade secrets from theft." 256 Thus it is possible that Congress meant to extend trade secrecy protection to acts of a U.S. national outside of the United States and to all situations where some act in furtherance of the misappropriation occurs in the United States.

Upon closer inspection, however, the case resembles Nabisco in that the relationship between the civil provision and the original criminal statute is not straightforward. Moreover, the ramifications of engrafting the extraterritorial provision of the criminal statute on to the civil cause of action leads to an astonishingly broad reach. The DTSA authorizes trade secret owners to bring civil claims for misappropriation of a trade secret. However, it does not supply autonomous definitions of these terms. Thus, in considering the extraterritorial reach of federal trade secret law, one possibility is that courts will use the definitions of wrongdoing found in sections 1831 and 1832 of the EEA. Alternatively, DTSA actions might be brought based on activity abroad as defined by state laws such as the UTSA. Because sections 1831 and 1832 cover such diverse actions as taking, copying, communicating, receiving, or possessing a trade secret, the DTSA could be interpreted to apply not only to unauthorized appropriations in the United States for use abroad, but also to: appropriations abroad for use in the United States; takings abroad for use abroad so long as part of the transmission of the information occurred in the United States; or any taking abroad by a U.S. national. Indeed, because these provisions include attempts and conspiracies," the DTSA could arguably apply to inchoate activity. It might, for instance, be possible to bring a claim in the United States against a foreign firm whose employee took information from a former foreign employer and stored it on a U.S. server for use in his new position. Even if state-law definitions are used, 257 the UTSA also includes a broad range of activity without any specificity as to where the actions take place. 258 Thus, under either interpretation, U.S. trade secrecy law could indeed "rule the world" 259 and police global business practices.

As a practical matter, this is very different from how section 1837 of the EEA operates in criminal cases, where there are several mechanisms that limit the ambit of prosecution. First, prosecution is mediated through prosecutorial discretion, which as we saw, was an important factor in the Nabisco Court's decision to distinguish between the civil and criminal sections of the RICO statute. Second, the reach of the EEA is cabined by the limits of criminal jurisdiction, where there is no such thing as long arm jurisdiction or the ability to satisfy judgments with assets found in the United States, as in civil cases. 260 To be sure, alleged criminal misappropriators can be indicted even if not physically present in the United States. But to be tried, they must be extradited, and extradition is difficult. It requires a treaty with the country in which the misappropriator is found. Furthermore, virtually all U.S. extradition treaties apply only when the conduct charged would also be a crime under local law. 261 The employee in a criminal analogue to Sino Legend, for example, could not be extradited (even if the United States had an extradition treaty with China) once the Chinese courts decided the activities at issue did not constitute a trade secrecy violation. Third, although the criminal provisions of the EEA include the possibility of an award of injunctive relief to protect the trade secrecy holder, they do not authorize an award of damages. 262 Even the availability of an injunction is highly constrained: it must be obtained by the Attorney General (thus, it is subject to prosecutorial discretion) and the statute says it must be "appropriate," suggesting that the Attorney General should consider comity factors. In practice, it is notable that criminal prosecution under the EEA has generally involved purely domestic misappropriators or foreign nationals who have taken information acquired in the United States for use abroad. 263 The absence of mechanisms to cabin the reach of the federal trade secret statute is not the only reason to think that Congress did not fully consider the interaction between section 1837, on extraterritoriality, and section 1836(b), on civil liability. While it is true that there is a reference in section 1836(b) to "foreign commerce," we have seen that such language has generally not been sufficient to overcome a presumption against extraterritorial application. 264 Furthermore, section 1837 was not amended when section 1836 was changed by the adoption of the DTSA. Indeed, the references in section 1837 to "offender" and "offense," not "infringer" or "infringement," suggest that Congress had not affirmatively considered the relationship between the two sections, as envisioned by Nabisco. 265 Also, to the extent that Congress has addressed concerns about conflicts in the remedial provisions in paragraph (b)(3) of section 1836, it is puzzling that it considered only potential conflicts with state laws. The statute and legislative history refer to U.S. state law (such as the Uniform Trade Secrets Act) but do not mention foreign laws. 266 For example, the provision includes limitations on the issuance of an injunction; one of those limitations refers to "an applicable State law prohibiting restraints on the practice of a lawful profession, trade, or business." Significantly, the concern about "conflict" does not deal with the interplay between trade secrecy and patent policy, which would be a problem if foreign misappropriation were included, but is not necessary in the state context because states do not have their own patent law or policy. Finally, the statute includes a requirement that the Under Secretary of Commerce report to Congress on whether U.S. companies can adequately protect their trade secrets abroad, and recommend ways to assist them in reducing the risk of loss. 267 If the statute were applicable to foreign thefts and harms, the need for such a report would be greatly diminished.

In short, the absence of clarity within the statute leads to the same problem that we saw in TianRui-a failure to provide a sharp analysis of the extraterritorial reach of the statute. A scope that includes protection for any trade secret, irrespective of where the information was developed, or who developed it, or where it was stolen or used, would be exorbitant and would raise Judge Moore's concern that the United States has no right "to determine what business practices, conducted entirely abroad, are unfair." 268 Moreover, unlike section 337 actions, which are limited to exclusion orders, or EEA injunctions, which are within the scope of prosecutorial discretion, a claim for injunctive relief under the DTSA depends on the decision of private parties, and it is available "to prevent any actual or threatened misappropriation … on such terms as the court deems reasonable," so long as it does not conflict with certain State business laws. 269 Analogously, damages can be awarded for "actual loss caused by the misappropriation" as well as for unjust enrichment." 270 Thus it appears courts can award injunctions with extraterritorial effect and damages to compensate for losses in foreign markets.

Courts might react to the ambiguities in section 1837's application to the DTSA in one of two ways. They could decide that the presumption against extraterritoriality is overcome by the inclusion of the DTSA in the EAA. While that approach would give the DTSA a broad reach, courts could consider prescriptive comity limits to avoid serious clashes with foreign law. One consequence of such comity concerns would be to limit injunctive relief to the U.S. market and monetary relief to losses suffered in the United States. 266 See DTSA House Report, supra note 257, at 12. Alternatively, courts might determine that the lesson of Morrison, Kiobel, Nabisco, and Microsoft is that the presumption of extraterritoriality is strong and must be applied strictly and with specificity to each provision in a legislative scheme, that it can be overcome only by an express indication of intent, and that no such expression can be found in the civil provisions of the DTSA. To determine whether the DTSA has a domestic application in a particular case, the court would then turn to the focus of the statute. The trademark cases starting with Bulova and refined in Vanity Fair and Trader Joe's could be read to suggest that the focus of that statute is the protection of a U.S. company from injury caused by the infringement of its U.S. trademark. The analogy could prove helpful here and suggest a similar domestic focus in trade secret cases. The House Report on the DTSA states that Congress was motivated to deal with theft "harmful to United States companies" and notes that "the trade secrets of American companies are increasingly at risk," 271 suggesting that the focus of the statute is the misappropriation of a U.S. company's trade secret. Prescriptive comity would then allow consideration of any serious conflicts with foreign law, particularly with regard to remedial provisions that extend beyond the territory of the United States. Under this interpretation, the statute would be of much more limited scope. However, as with section 337, the interpretation might then raise concerns under international law, for it would arguably discriminate between foreign and domestic right holders. 272 Once again, a broader interpretation U.S. trade secret, to include trade secrets developed in the United States by foreign or domestic firms may avoid this problem.

The DTSA states expressly that it does not "preempt any other provision of law," whether civil or criminal, and thus trade secret owners can usually choose whether to seek a remedy under federal or state law. 273 The fact that state law may offer an alternative remedy invites an inquiry whether state trade secret law can cover claims that federal law cannot reach, such as claims involving conduct that mainly takes place abroad. Curiously, courts hearing state trade secrecy cases have said very little about extraterritoriality. In some states, protection remains a matter of common law (or was a matter of common law when key precedents were decided). In those cases, courts have (or had) no occasion to consider legislative intent concerning the projected scope of a statute. Most states have now adopted the UTSA, which does not contain an express statement of territorial reach. Although there are certainly theoretical reasons and case law suggesting that courts should adopt a presumption that state statutes do not extend beyond the boundaries of the enacting jurisdiction, 274 as well as reason to think judges should 271 DTSA Senate Rep., supra note 6, at 12; DTSA House Rep., supra note 257, at 3. 272 See supra note 49. 273 18 U.S.C. § 1838 (f). (There is one exception for whistleblowers. 18 U.S.C. § 1833(b).) 274 See, e.g., N. Alaska Salmon Co. v. Pillsbury, 174 Cal. 1, 4 (Cal. 1916) ("Ordinarily the statutes of a state have no force beyond its boundaries."). be even more circumspect when applying state law to conduct outside the United States,foot_19 courts rarely reason in these terms. 276 Not all courts agree that there is a presumption against the extraterritorial application of state law. 277 Furthermore, unlike federal cases, which are dismissed once a determination is made that the scope of the statute in question does not extend to the facts of the case, in state cases courts consider "multiple potentially applicable laws and ultimately . . . select one to resolve the case." 278 Thus courts hearing state cases will, in the end, most likely look to traditional choice of law principles to determine the appropriate law and then adjudicate the case under that law (or laws).

The few transnational trade secret cases that address choice of law follow this pattern. Both state courts and federal courts hearing state law cases generally say nothing about a presumption against extraterritorial application or the prescriptive scope of the statute. 279 Indeed, some barely consider the choice of law question. As noted in the Introduction, in E.I. DuPont de Nemours and Co. v. Kolon Industries, Inc., 280 a claim was made that the defendant, a Korean firm, received information on the secret methods for manufacturing Kevlar (the fiber in bullet-proof vests), which had been developed by an American company, Dupont. The district court applied Virginia law (the Uniform Trade Secrets Act) without much analysis. 281 Other courts have analyzed the question in somewhat more traditional choice of law terms; they have, for example adopted the choice of law analysis used in tort cases and debated whether to apply the law of the place of conduct or injury. One such example is BP Chemicals, Ltd. v. Formosa Chemical & Fibre Corporation, 282 in which Formosa, a Taiwanese corporation was alleged to have entered into a contract with a New Jersey company to manufacture parts based on trade secrets taken from BP, a British corporation, and ultimately to be used in the construction of an acetic acid plant in Taiwan. At the time, New Jersey took a common law approach to trade secrecy cases and the court applied a choice of law analysis, identifying the relevant issues as (1) whether BP's proprietary information was protectable and had not returned to the public domain by the time it was taken and (2) whether acquisition of the alleged trade secret in Taiwan was wrongful. 283 To decide what law to apply, the court focused on where the trade secret was taken and where it was licensed. It held that Taiwan had the greater interest in setting the standards as to whether information had been sufficiently safeguarded and whether it had entered the public domain. The court explained that the issue implicated policy judgments regarding the appropriate balance between protecting trade secrets, thereby encouraging the development of new technology, the willingness of a foreign company to share their technology with Taiwanese business, and promoting free interchange and access to information, which had profound implications for the health of the Taiwanese economy. Similarly, the court determined that Taiwan had the greater interest in determining whether a Taiwanese company acted tortiously in acquiring BP's information in Taiwan, and in particular, the circumstances under which a Taiwanese company shared information with its employees. Although BP alleged that the misconduct took place in New Jersey-the fabrication of equipment using misappropriated technical specifications-the court stated that such conduct had no relevance to the analysis. Rather the court found that the majority of the relevant conduct occurred in Taiwan where the trade secrets were licensed and where the information was acquired. It then instructed the trial court to take evidence on Taiwanese law. As to New Jersey law, the court noted that New Jersey was not the principal situs of either the direct or indirect injury inflicted; and as to English law, the court did not believe that the law at the place of injury should be given great weight in cases arising out of claims of misappropriation of trade value. 284 282 229 F.3d 254, 257 (3rd Cir. 2000). 283 Id. at 265.

The emphasis on where the trade secret was taken is arguably consistent with the position taken by the Restatement (Second) of Conflicts. Section 145 of the Restatement adopts the law of the state that has the "most significant relationship to the parties and the transaction" with respect to the particular issue in question. 285 Although in many tort cases it is the place of injury that is identified as the state with the "most significant relationship," particularly when that law permits recovery, comment f to section 145 addresses the tort of "misappropriation of trade values" more specifically, stating that it is the "principal location of the defendant's conduct" that usually will be given the greatest weight in determining the state whose local law determines the rights and liabilities that arise from the tort. Comment f is, however, somewhat ambiguous in transnational cases. First, "conduct" can occur in more than one location. The defendant might have taken the secret where it was developed, but used it in another location or taken it in a place where the trade secrecy holder licensed it and used it in a third country. In those circumstances, it is not clear which place should be regarded as the "principal location" of the conduct. Moreover, as described in Part I, the tort of trade secret misappropriation involves legal issues on which there is substantial disagreement among jurisdictions. In BP Chemicals, the bulk of the activities occurred in Taiwan so once the court decided to use a conduct rule, it was straightforward to apply Taiwanese law on all the issues relevant to that case. But when activities span several jurisdictions, it is not clear whether comment f includes a dépeçage approach. For example, comment f might be understood as pointing to the law of the place of acquisition to determine whether, at the time of acquisition, the information was secret, subject to reasonable efforts to maintain secrecy, and taken by improper means; to the law of the place where a tip was communicated to determine tippee liability; and to the law of the place of use to determine questions regarding the lawfulness of the use. 286 From the perspective of innovation policy, coupling the dépeçage approach with some version of the conduct rule has a great deal to recommend it. If the goal of the law is to promote scientific and technological progress, it is important that new scientific developments and technical advances can be fully exploited. To avoid chilling legitimate uses, those in contact with information must know whether it can be freely taken and used. Applying the law of the state where conduct relevant to that inquiry takes place arguably accomplishes that purpose. Of course, trade secrecy holders may wish to develop, use, and license information in places that lack adequate protection. They can, however, often protect themselves with agreements that provide for the application of the law of a different location. As long as those who learn the information know the terms under which they are in possession of it, they can understand the limits of its use.

On the other hand, if one considers the policies underlying the field of trade secrets and certain practical realities of global business, a strong argument can be made that the applicable law on most issues should be that of the place where the trade secret was developed and the loss of the opportunity to capture revenue on account of the innovation might be said to be incurred. 287 If the law of the place of development permits recovery, it is presumably because the state has a policy of encouraging innovation and business development by protecting trade secrets. Of course, the state where the information was taken may have a different policy: it might deny recovery (by determining that the information was not a trade secret, that it was not adequately protected, or that it was not wrongly shared) in order to encourage employee mobility and information exchange, thereby allowing others to build on existing information and push forward the frontiers of knowledge. Such cases present the paradigm of the "true conflict" and the question remains as to which of these policies should give way to the other. 288 For reasons explained below, we favor the place of development over the place of conduct. 287 See PETER HAY, PATRICK BORCHERS & SYMEON SYMEONIDES, CONFLICT OF LAWS 997 (5th ed. 2010) (arguing that applicable law should be the place where the claimant was injured in its business). 288 We use the term "true conflict" as it is used in the contemporary conflicts of laws literature, meaning that the relevant states have different policies and more than one state has a policy that is furthered in the particular factual pattern presented. A converse situation to the one presented in the text also presents a true conflict. When the place of development does not provide protection (for example, if it favors employee mobility over providing incentives to innovate) but the state where the information was used does provide protection (for example, because it wants to encourage technology transfer), we believe the law of the place of development should nonetheless apply. Presumably, the availability of trade secret protection was not what motivated the developer to invest in the technology at issue. Admittedly, the policy in technology transfer is then subordinated to the policy of employee mobility. Such a rule can be defended in order to adopt a coherent choice of law principle through an evaluation of the competing policies. To the extent to which there is disagreement about this evaluation of competing policies, one could argue for a rule that the law of the state with the greater protection always applies, subject to a foreseeability requirement. Cf. RESTATEMENT (THIRD) CONFLICT OF LAWS (AM. LAW INST., Preliminary Draft No. 3, 2017) (providing that as to issues of conduct regulation when the conduct in one state causes injury in another, the law of the state of conduct governs the issue, but if the location of the injury was foreseeable, the injured person may select the law of the state of injury); Hague Convention of 2 October 1973 on the Law Applicable to Products Liability (providing for the law of the State of the victim's habitual residence to apply when it is also the State of the principal place of business of the person claimed to be liable or the place where the product was acquired; if those criteria are not met the applicable law is the internal law of the State of the place of injury if it is also the State of the victim's habitual residence or State of the principal place of business of the personal claimed to be liable or the place where the product was acquired; and if none of these criteria are met, the applicable law is that of the State of the principal place of business of the person claimed to be liable, unless the plaintiff bases the claim upon the law of the State of the place of injury. In all cases the applicable law is subject to a limitation if the person claimed to be liable could not reasonably have foreseen that the produce would be available in that State). There are other permutations where under contemporary choice of law analysis only one state may be said to have a relevant policy to be furthered and is thus considered a "false conflict." Here, however, we consider only the paradigm examples.

First, one must consider the context in which trade secret exploitationand theft-now takes place. Today, businesses are subjected to a greater volume of, and more varied, threats to their trade secrets from both inside and outside the company. 243 More than an employee who takes a physical box of files (or walks off the job with knowledge that constitutes a trade secret), a 21 st century trade secret misappropriator may well use an electronic medium-perhaps even accessing the relevant files remotely. 289 In such a technological environment, it is not clear that the conduct rule works as intended. 290 For example, take the case of files containing trade secrets being stored on Dropbox or a similar cloud computing service. 291 In one such case, one defendant ex-employee had installed Dropbox on their company computer, where it was used to store and access files containing trade secrets, and also linked three personal devices to the Dropbox account (an Android phone, an iPad, and a personal iMac). 292 Another defendant ex-employee accessed the same files via personal email, while traveling on the company's behalf and working remotely. 293 The defendants in this case left a clear trail: they transferred large numbers of files between and to themselves near their dates of resignation. 294 But the facts of the case illustrate a broader point: under a conduct rule that emphasizes the place of appropriation, the applicable law might change depending on which jurisdiction the employee was in when she accessed the protected files for the purpose of appropriation, rather than for working remotely. Worse still for the conduct rule, the plaintiff might provide log-in credentials for the cloud service to a third party who themselves accessed the files from a different jurisdiction. Under such a scenario, would the act of appropriation be the provision of log-in credentials, or the downloading of the files? Given the reality of data access and transmission in the world of cloud computing, with multiple devices and access points, a savvy appropriator could "shop" for the least protective trade secret law, and then misappropriate strategically from within that jurisdiction. Indeed, the "migratory appropriator" is not merely speculative: in some cases, employees have accessed cloud storage accounts after departing the company, once they reside in another jurisdiction and work for a new company located there. 295 Second, we disfavor the conduct rule in this context for reasons that are more fundamental than the application challenges posed by modern technological realities. Just as the archetypal threats (e.g. hackers, remote access via cloud computing) are now global, so too are the activities of the archetypal business involved in trade secret litigation. In the domestic tort context, there is something to be said for the idea that a party bears some risk when they leave behind the more favorable policy of their own jurisdiction-for example, by driving into an adjacent sister state. 296 However, that rationale is less persuasive when the default modus operandi and typical case is one involving transnational business activity. In the transnational business context, it is not useful to think in terms of the trade secret leaving the protection of the corporation's domicile to enter another jurisdiction. Instead, a corporation develops a trade secret to exploit it globally in its business operations. For these reasons, we favor a place of development rule over a place of conduct rule, at least on the questions of whether the information qualifies as a trade secret and was adequately protected and taken unlawfully. 297 This rule assures firms that their secrets will be protected and creates the strongest incentive to invest in information development. The downside is that it is not always easy to determine where information was developed or exactly who developed it. For example, it may have been developed in increments, in several places where the firm does business; it may have been developed jointly by a group of employees working in different places or by several firms acting in concert. This uncertainty poses difficulties for adjudicators. Moreover, the inability to determine what law applies presents a problem for anyone seeking to use the information. In such a situation, a subsidiary rule may be necessary. Accordingly, the applicable law should be the law of the place where the secret was developed and thus where the trade secrecy holder suffered injury, provided that the defendant could foresee, at the time it acquired or used the information, that this law would be applied. In cases where the place of injury could not have been foreseen, courts should apply, at least on the issues mentioned above issues, the law of the place where the information was taken. Once again, the trade secrecy holder will often be able to protect itself with a choice of law provision 295 Estes Forwarding Worldwide LLC v. Cuellar, 239 F. Supp. 3d 918, 921 (E.D. Va. 2017) (discussing Virginia-based employee who left Virginia employer and moved to Washington State, whereinafter he accessed an account containing protected information of his previous employer). Although the suit in this case was brought under computer fraud laws, the scenario resembles a common trade secret case fact pattern. 296 For example, under Cavers' principles of preference, a defendant from state A, who entered a state B with more protective tort laws and injured a plaintiff domiciled there, would be subject to the more protective law; by contrast, a plaintiff from state C, who entered state D with less protective tort laws and was injured by a defendant domiciled there, would not be able to subject the defendant to the more protective laws of state C. See DAVID CAVERS, THE CHOICE-OF-LAW PROCESS 114-224 (1965). 297 Cf. Innovia v Frito-Lay [2012] EWHC 790 (Pat) (breach of confidence case with contacts in England, Delaware, and Texas; applicable law determined to be England, the place of development, albeit chosen on a different theory than the one proposed here). Note that, at least in some cases, the law applicable to the question of whether the information was improperly used would be decided under the law of the place of use. An example might be tippee liability, where an independent policy of the place of use might include encouraging the tippee's investment.

The choice of any single law that permits recovery and permits global damages and/or broad injunctive relief appears to permit that law to rule the world with respect to the relevant issues. However, because the choice-of-law analysis represents a balance of competing policies, global damages do not seem unwarranted. 298 There is also authority in the U.S. cases for broad injunctive relief. For example, in the Kolon case, where the court applied Virginia law to permit recovery by a U.S. company against a Korean company where the Korean firm had received information about a secret method apparently developed in the United States, the court permitted broad injunctive relief and prohibited the Korean firm from selling materials made with Kevlar worldwide, including in South Korea. In authorizing the injunction, the court noted that the loss of trade secrets harmed an American corporation, the defendant had tried to sell its products in the United States, that some aspects of the misappropriation occurred in the United States; and that both the Virginia Uniform Act and South Korean law permitted injunctive relief, indicating lack of a conflict. 299 Although it is not clear that South Korea would have authorized trade secret protection on these facts and imposed relief for sales in South Korea, if the defendants knew the information was developed in the United States, we believe the court would be fully justified in ordering complete relief for infringement of the trade secret.

Global injunctive relief can also be defended because it may be the only way to fully protect the trade secret holder. For example, in Nordson Corp. v. Plasschaert 300 the defendant had breached a non-compete agreement by showing the Ohio plaintiff's secret plans to a firm in Europe. In awarding injunctive relief that extended to Western Europe and Canada, the court stated that "most confidential information is worthy of protection without geographic limitation because once divulged the information or the fruits of the information quickly can pass to competitors anywhere in the word. 301 Global injunctions may be especially necessary in situations where the information, or material embodying the information, is distributed on the internet. Putting aside the fact that this injunction was ordered against a third party and not the misappropriator of the trade secret, Google, Inc. v. Equustek Solutions 302 highlights the need for global injunctions in that context. In that case, the Supreme Court of Canada affirmed an interlocutory decision ordering Google to delist websites where goods designed and manufactured with a misappropriated trade secrets were being sold (and passed off). The court acknowledged the international comity concerns involved in an injunctive order that interferes with freedom of expression. The majority nonetheless reasoned that the injunction was appropriate:

Where it is necessary to ensure the injunction's effectiveness, a court can grant an injunction enjoining conduct anywhere in the world. The problem in this case is occurring online and globally. The Internet has no bordersits natural habitat is global. The only way to ensure that the interlocutory injunction attained its objective was to have it apply where Google operatesglobally. If the injunction were restricted to Canada alone or to google.ca, the remedy would be deprived of its intended ability to prevent irreparable harm, since purchasers outside Canada could easily continue purchasing from D's websites, and Canadian purchasers could find D's websites even if those websites were de-indexed on google.ca. 303 Significantly, the Canadian Supreme Court also envisioned the possibility that Google could obtain a variance of the injunction by showing that it conflicted with the laws or policies of another jurisdiction. Furthermore, it is quite possible that other jurisdictions would decline to enforce the judgment on public policy grounds. 304 Finally, because injunctive relief is an equitable remedy, there is room for the issuing court to take comity considerations into account.

No matter which of the conflict of law rules discussed above is adopted for state trade secrecy cases, it represents an approach that is preferable to imposing the extraterritoriality analysiswhereby courts decide either that U.S. law applies or they dismisson the DTSA. After all, the DTSA is itself heavily based on state law. Indeed, the Uniform Act codified a common law tort (the tort of misappropriation). The advantage in using the conflicts approach is that it would offer better protection to the competitive environment in the United States. As we saw in our previous discussion of the DTSA, 305 the extraterritoriality analysis permits recovery only when a US trade secret is misappropriated: otherwise, the case is dismissed. Even if the concept of a "U.S. trade secret" is defined expansively, information developed outside the United States will generally not qualify for protection. Under the applicable law rule we suggest for state cases, courts in the United States could retain the case and accord relief when development occurred in any place with a protective regime. In places with law 303 Id. 304 See, e.g., Google LLC v. Equustek Sols. Inc., No. 5:17-CV-04207-EJD, 2017 WL 5000834 (N.D. Cal. Nov. 2, 2017) (granting a preliminary injunction to prevent enforcement in the United States of the Canadian court order). akin to that of the United States (for example, EU states that have implemented the Directive), the case might be decided quite similarly. 306 Even under the residual rule that looks to the place of conduct, U.S. courts can accord relief when the behavior occurred in a protective state. Of course, this choice of law approach does mean that a court in the United States will render judgment for the defendant when the content of the applicable law is that no relief is available.

We do not believe that interpreting the DTSA in this way is foreclosed by the Morrison line of cases. The extraterritoriality approach appears to have evolved because U.S. courts do not consider themselves competent to decide cases under the public law of another country. 307 However, the public-law nature of intellectual property law is, in other contexts, giving way to the notion that private rights are at stake. 308 Thus, where at one time patent cases were considered nonarbitrable, 309 Congress has since provided for voluntary arbitration of infringement actions. 310 Moreover, private rights language is often found in intellectual property cases. For example, in United States v. Martignon, the Second Circuit distinguished between a criminal anti-bootlegging statute and copyright legislation on the ground that the latter was designed to "allocate rights," whereas the former regulated behavior. 311 In copyright, some federal courts have entertained foreign copyright claims under diversity jurisdiction or as supplemental to a claim under federal law and do not automatically dismiss foreign claims on forum non conveniens grounds, (suggesting they do not see them as public law cases). 312 Significantly, other countries-the EU and the UK-have concurred in this approach and have entertained foreign copyright actions when related to domestic claims. 313 As we have seen, U.S. courts have applied foreign trade secrecy laws in numerous situations.

A more conceptual objection to the use of conflicts principles rather than an extraterritoriality analysis has been voiced by William Dodge. 314 In his view, courts do not have an easy time determining which law ought to apply to a case with multiple national contacts. Rather than taking on that task (and requiring the parties to shoulder the burden of proving foreign law), the international system is better off if courts rely only on local law-either they should decide it applies extraterritorially, or they should refrain from adjudicating. If this creates problems, he posits that nations will negotiate international agreements. While this approach has much to recommend it in theory, it falls short with respect to intellectual property. As we have seen, it is often difficult to reify and localize intangible rights. Nor do international instruments solve the problem. Intellectual property has been the subject of international agreements for over a century. However, in all that time, the best that countries have done is to agree on minimum standards of protection. As Part I demonstrated, such standards leave a great deal of room for national variation. Countries could alternatively negotiate agreements on private international law, including choice of law. 315 Notably, the EU has adopted several conventions but they do not solve the problem for trade secrecy because countries differ on whether these claims sound in tort, contract, or property. 316 Significantly, the EU's new Directive on trade secrecy, although it clearly contemplates cross border cases, neither fully harmonizes the law (it is not a Regulation), nor does it provide rules on what law applies to such disputes. Given this vacuum, the DTSA offers the possibility for federal courts to develop a choice of law rule for federal trade secret cases and apply the appropriate law or laws to the global marketplace.

Because the social welfare effects of trade secrecy protection are indeterminate, it is not surprising that countries that are all bound by the TRIPS Agreement have reached different conclusions on the details of trade secrecy causes of action. Nonetheless, there was a period of time when U.S. courts tended to routinely apply its domestic trade secrecy laws to reach activity outside U.S. territory. In recent years, however, the Supreme Court has emphasized the importance of interpreting U.S. laws in ways that avoid clashes among national laws. In this article, we consider the choice of law principles that should guide decisions on applicable trade secrecy law based on our perceptions of innovation and conflicts policy. To be sure, much depends on context and there are three ways that trade secrecy claims can be brought: through actions for exclusion order in the International Trade Commission; under federal trade secrecy law; and under state law. We suggest that because the ITC provides a remedy that affects only U.S. markets, the United States can always apply its own law to U.S. trade secrets-trade secrets developed in the United States. In other contexts, we believe courts should have the ability to apply U.S. or foreign law based on the principle that the law of the place of development of the trade secret governs if it is foreseeable by the defendant. If not, the applicable law is the place where the defendant took the secret.