6 - Reconceptualization, Reinterpretation, and Typology

Our aim to establish harmony of laws can be seen as attainable if the colliding substantive laws are in a closer relationship, if their material fundaments consonantly point in the same direction. It will then be possible to identify a conflicts norm with a simple rule of attachment which, in all of the different legal orders, … can (not must!) be accepted, since none of these regimes must thereby sacrifice material substantive law interests in favor of certainty in choice of law. If this uniform conflicts norm is of a plausible expedience, of a convincing power, it will achieve regular international practice; over time, then, it may develop into a public international law standard.

Author’s translation from Franz Kahn, Über Inhalt, Natur und Methode des internationalen Privatrechts, 40 JherJB 1, 76 (1898)

Introduction

As Franz Kahn explained in 1898—and this still holds true for modern choice of law and conflicts law—the “material fundaments” of substantive laws must point in the same direction in order to make the promulgation of uniform conflicts rules a theoretically imaginable option.¹ Only if the underlying substantive law policies concordantly allow for and indicate a certain structure of conflicts resolution will a uniform system be acceptable and ultimately be successful as a rule of the “harmonious” choice of law that had been prophesized by Friedrich Carl von Savigny fifty years earlier. Indeed, both icons’ wisdom is fundamentally reflected in modern choice-of-law theory and its so-called functional method. Functionality actually lies at the crossroads of two disciplines—it is the ultimate connex between choice of law, or conflicts law, and comparative law.² My analysis in the foregoing chapters laid the foundation for such a functional approach. The earlier chapters revealed the history and status quo of doctrine in the United States and Europe: ever since Steele, US law has adhered to commercial effects as the indicator of Lanham Act subject-matter analysis. The German Bundesgerichtshof, by contrast, has only recently begun to refer to a similar paradigm of commercial effects, as demonstrated in HOTEL MARITIME. Finally, in European unfair competition choice of law, the marketplace effects rule has found its way into the Rome II Regulation. With respect to choice-of-law and conflicts law structure, therefore, the issue no longer seems to be whether a test of “commercial effects” or “marketplace effects” can be accepted as such. On the contrary, the instrument is virtually universally acknowledged. Yet a number of questions still await answers—these answers must be based on a functional analysis of the fields in both substantive law and choice of law. The comparative inquiry in the preceding two chapters provided the necessary groundwork. Comparing the different common law and civil law phenotypes of substantive law doctrine and of different variants of trademark protection and unfair competition prevention unveils a fundamental convergence of policies. At least with respect to the core policies in both trademark and unfair competition law, it can be said that—so to speak—a common genotype of trademark and unfair competition policies exists. Virtually everywhere, consumer decision making is acknowledged as the most essential element of the market mechanism. Protection of the information infrastructure thereby provides the architecture of a functioning system of free competition. This is the quintessence of a free and unmanipulated evolution of competition as a dynamic process of marketplace transacting. On the basis of these structural similarities across different jurisdictions’ systems, I will use this chapter to reconceptualize choice of law and conflicts law and present a practical guideline for implementing the results of my historical-comparative, theoretical, and doctrinal inquiry. I will begin by outlining the essential structure of choice of law in international trademark and unfair competition disputes (see infra p. 492 et seq.). On this basis, I will attempt to suggest some modest correctives and an according reformulation of the current rules in US and European law (see infra p. 521 et seq.). Finally, I will promulgate a typology of typical cross-border conflicts scenarios and thereby illustrate the reconceptualized conflicts resolution structure “in action” (see infra p. 548 et seq.).

Section 1 The New Conflicts Resolution Structure

The first issue of reconstruction, as we have seen, concerns the traditional dichotomy between trademark conflicts and unfair competition choice of law. In light of the two fields’ common core of policies aimed at protecting market information infrastructure, a jettisoning of the formal distinction is needed. Extending the view to this virtually universal functional architecture unveiled in chapter 4 further suggests the need to overcome the traditional fixation on conduct that still governs in European civil law doctrine in particular. But the pendulum must not swing too far into the domain of effects testing—notably not toward the US doctrine of an overextensively vague understanding of “commercial effects.” Therefore, the effects test must be transformed into a uniform system of protecting consumer decision making—this implies a new qualitative standard for determining effects relevance. In addition, we also need a metric for determining minimum effects quantity. Civil law doctrine calls it a de minimis limitation, while US law deals with the issue in light of the Bulova test, mostly with respect to the aim of avoiding “conflicts with foreign law.” This aspect requires drawing on what we have learned with respect to the doctrine of international comity.

I Trademark/Unfair Competition Uniformity: Core Policies

The previous chapters’ findings on substantive law policy indicate that the existing divergence between the two approaches to trademark and unfair competition conflicts must be overcome. Above all, both sectors’ core policies are homogeneous. Trademark protection and unfair competition prevention are the normative backbone of market information infrastructure.³ This implies that the uniform basis of conflicts attachment is marketplace effects—more concretely, effects on market information that are relevant for the consumer’s decision making.

Of course, black letter law in Europe has formally consolidated a dichotomy via articles 6 and 8 of the Rome II Regulation. But this does not preclude a uniform and homogeneous approach. Given the common core of trademark and unfair competition policies, conflicts resolution techniques will be structurally identical and yield consistent results if founded on a consolidated marketplace effects approach. In US trademark law, although the effects test seems to govern both trademark and unfair competition conflicts doctrine, a reconceptualization must be reconciled with the actual practice of subject-matter jurisdiction. The Steele doctrine constitutes a unilateral and quasi statutist rule. US courts, when confronted with an international dispute, will determine whether federal trademark law applies and, in doing so, will conflate subject-matter jurisdiction and choice of law into one test.⁴ Here, too, a consolidated marketplace or effects test can and should be integrated into existing structures.⁵

Such an implementation of the effects test into trademark and unfair competition conflicts law requires a foundation on a uniform effects-based lex loci protectionis rule. This rule still allows for a multilateral determination of the applicable law or laws. It is actually a choice-of-law technique in the sense initially proposed by Friedrich Carl von Savigny. In practice, a plaintiff must plead and specify the regime for resolution of the conflict at issue.⁶ She will thus reasonably tailor her pleading to the jurisdiction(s) where a right exists or where a violation of unfair competition rules has occurred or is about to occur; this concerns all locations where marketplace effects have already occurred or may occur in the future.

II Quality of Effects: A Rule of Alternatives

My analysis of substantive law has revealed that the protection of trademark rights, like the prevention of unfair competition at its core, is a question of market information regulation. Consumers’ minds determine the extension of rights and the issue of competitive fairness. The analysis, therefore, must take into account effects on the transmission of information, on the consumer’s decision-making process, and on her transacting.⁷ Under such a functionally qualified effects test, the point of attachment—that is, the locale of relevant effects’ occurrence—must be found in the last stage of the consumer’s decision making. This last stage of transacting is the focal point of the decision-making process.

In many cases, the point of attachment under such a perspective can be determined according to the consumer’s actual point of transacting. Quite often, this is the point of sale. Yet the multiangular structure of market competition requires more precision. This is due to the fact that trademark infringement, as unfairly competitive conduct, not only causes an actually or potentially improper transaction (between consumer and trademark infringer or competitor-violator) but also always causes the consumer to forego an alternative transaction—with the trademark owner or one of the competitors that was striving toward the transaction.⁸ It is the locale of these alternative transactions that represents the extension of the respective market and the place(s) where actual or potential competition existed.⁹ The point of sale can only then serve as a (but need not necessarily be the only) point of attachment if the consumer had at least one alternative to transact there. More generally, however, the localization of relevant effects requires considering more than one locale of unimplemented alternative transactions. The question is, Where did the alternatives to the actual transaction exist, or—if the consumer actually forwent a transaction—where could a transaction have been made absent the infringer’s or competitor-violator’s interference? In this way, it is also clear that the point of attachment is not necessarily the alleged infringer’s place of conduct (Werbemarkt) or the place of impact (Einwirkungsort). Nor is it the place where potentially ubiquitous “commercial effects” may occur (Auswirkungsort). My typology of conflicts attachment presented in the last section of this chapter will clarify many practical questions in this regard.¹⁰ For the moment, an example from Austria shall serve as an illustration of the basic principle.

In 1986, the Circuit Court of Innsbruck decided on a case in which a German airline company had advertised its services (flights from the United States to Austria) to US customers by offering free ski rental in Austria for each flight ticket purchased.¹¹ The court applied Austrian law, arguing that the Austrian market for ski rentals had been affected by the defendant’s US marketing. It thereby did not apply the rule of the advertising market, instead undertaking a result-oriented analysis. The collision of interests, the court explained, was to be found in the Austrian market for ski equipment and rentals.¹² This approach would arguably be questionable under current European choice of law. After all, “interests” existed in both the United States (i.e., consumers and competitor airlines) and Austria (i.e., consumers, competitor airlines, and ski-rental providers). Under the marketplace effects rule in unfair competition conflicts, notably under the Gran Canaria doctrine relying on the advertising market, US law rather than Austrian law should have been applied.¹³ And a transfer of the antitrust conflicts rule to unfair competition choice of law, considering largely unqualified effects in both markets, might have even called for an application of both American and Austrian law.¹⁴

Quite differently, avoiding the vagaries of interest analysis and unqualified effects testing, looking for the consumer’s decision making and the alternatives to her actual transaction allows the inquiry to be reduced to the core aspect of a market economy: the consumer must be able to freely decide how, when, and where to transact. Under a functional perspective, depending on the product at issue, the outcome may differ. If flights to Europe (or Austria) are at issue, alternative transactions existed only in the United States. Looking at ski-rental services provided for alternatives in Austria alone.¹⁵ Yet even though the ultimate detriment may have occurred in the Austrian market for ski-rental services, these effects were irrelevant with respect to consumers’ decision making. What was at issue in the defendant’s advertising campaign was the sale of flights to Europe. In this regard, the search for alternative transactions should have led to an application of American law. At best, therefore, the case could have been understood as a scenario of unfair competition by breach of a statutory norm in Austrian law. This, however, as we will see in more detail later, is not a question of “marketplace” effects.¹⁶ The court’s decision to apply Austrian law may have been ultimately correct with regard to the breach of a statutory duty in Austria—but its analysis was beside the point when based on “effects” on the Austrian market for ski-rental services.

Finally, with respect to cross-border trademark infringements, further clarification is indicated: it is important to notice that giving regard to territorial effects under a rule of alternative transactions does not “extend” domestic trademark rights.¹⁷ To correct this misunderstanding, we must distinguish between the protection of domestic (or territorial) goodwill and the protection of foreign-based (or extraterritorial) goodwill. The former concerns effects that occur within the national territory, no matter whether they have been caused by conduct inside or outside. One example is confusion of the forum’s consumers. Even if the conduct occurs abroad (e.g., through the upload of confusing information to a foreign website), the effect still materializes territorially. Conduct may have occurred abroad, but regulation and rights protection are domestic. Foreign-based goodwill, by contrast, will be protected if the effects that are regulated occur beyond the national territory. In this case, one could actually speak of an extraterritorial “extension” of rights. This is the case if foreign-based consumers’ confusion is prevented by the application of domestic law. The national regime will then address the impact on foreign markets in the sense of protecting goodwill and rights abroad.¹⁸

III Quantity of Effects: Jurisdictional Self-Restraint

Virtually all suggestions for choice of law in both trademark and unfair competition disputes that have been brought forward in the form of an effects test or a marketplace principle require a limitation: de minimis effects must be found within a state’s territory (or its market) in order to apply the respective state’s law.¹⁹ The precise demarcations of such a de minimis test, however, are far from clear. This is due to the fact that the unqualified reference to minimum effects unduly intermingles several factors that must be kept apart. First, many suggestions still rely on nominal or intent-based standards to determine minimum effects. But we have already seen that actual numbers, as well as an alleged infringer’s state of mind, must be disregarded in a system of market information regulation determined by modern trademark and unfair competition laws.²⁰ In addition (and this has yet to be subjected to a structured analysis), a well-balanced marketplace effects test requires an additional threshold below which effects—even if qualified and foreseeable—will not be deemed sufficient to trigger the application of the territorial regime. Unlike current theory and practice, the description of such a threshold requires a comity-based de minimis analysis in the interest of avoiding both over- and underregulation.

A A Word in Advance: Practical Relativity

Multistate infringements and the application of potentially innumerable national regimes to cross-border marketing activity appear to be a serious practical concern. The situation does not require extensive explanation: a website may be accessed from virtually anywhere on the planet—accordingly, an infringement of trademark rights or a violation of unfair competition laws may occur everywhere. Prior to the internet, similar problems existed with newspaper, radio, and television marketing activities. In all of these cases, the collision-of-interests approach could lead to a cumulative application of different national laws. In general, injunctive relief and damages will then be territorially segmented. Each jurisdiction will govern only those claims related to its national territory.²¹ The issue is less critical with regard to damages and compensation claims, where determining the relevant jurisdiction is largely a practical problem of computation and proof. But injunctive relief can be problematic. The concurrent application of numerous legal regimes, as is usually warned, could create serious problems if the marketing activity or method at issue is indivisible. In such cases, it will ultimately be the strictest law that governs the whole case, usually leading to a complete prohibition of the activities at issue.²² Suggestions for remedying this problem are legion.²³ And, of course, there are strong arguments in favor of setting limitations to the judicial overreach of regulatory laws.²⁴ In particular, concerns of procedural and court efficiency must not be undervalued. Nevertheless, a number of clarifications are necessary. First, it must be noted that any judicial self-limitation with respect to the applicable law is an issue of substantive law doctrine rather than a rule of choice of law. The question is inseparably connected to local lawmakers’ interest in regulating market communication.²⁵ Moreover, before developing a rule for judicial self-restraint, we must take a closer look at the circumstances of transnational and cross-border competition. This reveals that the situation is far less dramatic than is sometimes explained. Most fundamentally, as Jack L. Goldsmith points out in his provocative critique of cyberenthusiast suggestions on internet self-regulation, it is questionable whether the right to economic activity should be granted “without borders.” Many proponents of practicality approaches assume that right owners’ activities in foreign markets should always be supported by beneficial regulatory circumstances. Under this rather parochial perspective, foreign policies seem only to stand in the way of unlimited international transacting. Looking beyond the false front of free-market demands, however, reveals that foreign regulation is a logical obstacle. If cross-border activity extends marketplaces, the application of foreign laws’ limitations is attached to these new domains, just as are all other market parameters that determine the costs of a commercial venture. There is no reason to expect cross-border activities to receive special treatment with regard to regulatory circumstances.²⁶

In addition, warnings about legal uncertainty concerning the applicable law(s) and the risks for domestic entities involved in international economic activities are often exaggerated. We have already seen why and how enforcement capacities are limited in the international arena.²⁷ If individuals or entities have no local presence or assets within a certain jurisdiction, that state’s regime is seldom enforceable.²⁸ An individual’s expectation of conflict with foreign law is thus often more an issue of assessing the risk of being successfully haled into court (which is unlikely) than a question of whether the foreign law will actually prohibit the conduct at issue (which does not matter absent a realistic threat of litigation). A risk of litigation exists only where ties to the jurisdiction are sufficient. In other words, the more an activity is focused on a certain market, the less we should balk if the relevant legal regime is found applicable to the individual’s conduct.

In this regard, one more aspect is important: with respect to the oft-enunciated risk of having the strictest law applied to instances of “indivisible” marketing activity, the conundrum has evolved into a pseudo problem for many relevant scenarios, particularly in the online environment. The issue is usually debated with an eye on cases of internet advertising. Yet the technical possibilities for segmenting and stratifying online activities have been significantly enhanced since the creation of the internet. As Jack Goldsmith explained in 1998 already, cyberspace is anything but “borderless,” and territorial segmentation is possible.²⁹ Apart from geolocation techniques, which have significantly enhanced over time,³⁰ content providers have the option of conditioning access to their websites on the users’ presentation of information. Considering the progress made in the technical control of information flow, it is no longer impossible to reterritorialize online activity on a geographical basis. Effects within foreign territories can be avoided much more than used to be the case.³¹

In this light, we no longer face the tremendous problem of spillover effects (if we ever did). With regard to the volume of litigation, contrary to what has been prophesied, the number of international and multistate conflicts has not exploded during recent decades. In any case, a less hurried look at the costs and opportunities of international marketing reveals that a certain degree of concurrent regulation and regime overlap is not a phenomenon that must be avoided at any cost. Indeed, many scenarios cannot and need not be withdrawn from the reach of multiple concurrent laws. Rather than completely avoiding conflicts, therefore, the challenge for a modern conception of choice of law is to provide for rules of minimum conflict.

B Objective Foreseeability

One early example of a de minimis principle is the Institut de Droit International’s suggestion in its 1983 Conflict-of-laws Rules on Unfair Competition, which require applying the law of the state where effects occurred that “could reasonably have been expected.”³² The institute did not elaborate further on the structural foundations of this factor. It described foreseeability by reference to the “injury to the competitor’s business.” This was still rather individual-competitor focused than truly policy oriented.³³ Notwithstanding this narrow perspective, its choice of terminology illustrates a central aspect of quantitative analysis: de minimis limitation is a question of reasonable party expectations—more concretely—of foreseeable effects.

1 Party Expectations and the International Private Law Order

A requirement of foreseeability reflects the limitations of human cognition. In legal analysis, this mostly concerns the facts of a case. We may, for instance, ask, Could the defendant have foreseen or expected the plaintiff’s injuries? Yet foreseeability is about more than the factual realities. The background regime of legal norms also plays an important part of the reference frame for expectations to materialize. In fact, the direct correlation between our normative environment and our expectations ultimately underlies any conception of conduct-regulating legal policies. The order of private law forms the basis on which the individual can conduct her activities, guided by reasonable expectations of the rules that apply to this conduct.³⁴ Conversely, this means that commercial activity in particular cannot be conceived of as being devoid of at least rudimentary expectations that norms of just conduct must be complied with.³⁵ Expectations are no less important when it comes to activities touching on the international sphere. In fact, territoriality must generally be taken as axiomatic for conflicts law purposes.³⁶ This axiom particularly concerns choice of law with respect to regulatory norms. There, it is the individual parties’ territorial connex to the regulating state that determines the actual choice of law. Such a connection—in other words, submission of private individuals to the application of a certain regime—can be found in either the individual’s consent to submission or in her receiving of benefits upon contact with a state. Both aspects are reflected in the factors “presence within” or “doing business in” a territory.³⁷ They are necessarily founded on the knowledge and acknowledgment of the possibility that the state’s legal regime might apply. This aspect actually reflects a larger shift in modern private international law: from an initial focus on single jurisdictions and laws, private international law has turned into the internal private law of socioeconomic transacting in a multijurisdictional world.³⁸ More concretely, this means that with the growth of transnational activity, in addition to domestic law, legal norms of foreign origin may also constitute a part of the framework that shapes private individuals’ activities. Necessarily, therefore, with respect to conduct that may touch on interests beyond the national border, individual expectations will comprise not just domestic law but also limitations under possibly many foreign legal regimes.³⁹

In this regard, one more aspect is important, though often neglected: parties’ expectations must be assessed objectively. Foreseeability in choice-of-law doctrine is often still understood as depending on subjective perspectives. Legal certainty also appears to be founded on a consideration of individual expectations.⁴⁰ This may be a remnant of the Savignian era, where private law and choice of law were conceived of as systems of apolitical norms for national and international community members’ self-determination.⁴¹ The picture changes, however, if norm conflicts are seen in light of the underlying regulatory purposes and if one acknowledges that international private law has increasingly come to serve as an order for international socioeconomic transacting.⁴² Of course, an individual will act in accordance with her subjective expectations. The overall order, as a system of limiting all private individuals’ freedom‚ however, will not ask for the single individual’s state of mind. Rules of conduct are objectivized by definition. With regard to competition-related activities covered by trademark and unfair competition law, this concept of objectivity reflects an essential economic logic: the extension of a market not only provides opportunities to increase profits but also implies costs. One facet of these costs is compliance with foreign laws. Hence, whatever can be foreseen as an opportunity abroad should also be foreseen as being attached to a corresponding set of limitations. Accordingly, it is the foreseeability of foreign-based effects that implies the application of foreign law regulating these effects.⁴³

2 Technique and Factors of Market Analysis

Regarding the subject matter of protection, as we have seen, the analysis must be founded on a policy of protecting consumer decision making. The ultimate stage of the consumer’s decision should be protected from manipulation. The point(s) of attachment will thus be determined according to where alternatives to the transaction (or nontransaction) exist.⁴⁴ The analysis to be undertaken must be conducted from a dual perspective. First, it has to determine the relevant product market. Second, it must consider the specific instance of marketing.

With respect to a market’s geographical scope, the globalization of commerce and trade has dissolved former boundaries between “national” marketplaces. Looking at a product market requires giving regard to the bulk of factors discussed in my summary of scholarship on the issue.⁴⁵ Even though a marketing medium may be unlimited in its reach, the product at issue can still be geographically (and, at the same time, territorially) limited in its availability. One example is restaurant or movie-theater advertising that (even if online) usually solicits local customers only. Another example is a physician’s, dentist’s, or lawyer’s advertising that commonly focuses on potential customers within a local or regional reach.⁴⁶ The situation might differ, however, for advertising by an upscale gourmet restaurant, the New York Metropolitan Opera, an international law firm, or a renowned plastic surgeon’s clinic. In essence, the central issue for assessing market extension is the ratio between the costs of transporting a good or performing a service (or transporting the consumer to the provider’s place) and the value of the product. In addition, it may be relevant whether the goods or services at issue are prestigious or without an adequate local substitute.⁴⁷

Furthermore, the concrete presentation of the marketing activity at issue may lead to a geographical and territorial confinement of relevant product alternatives. I have already mentioned the advertising language and methods of payment or delivery options.⁴⁸ The context of an advertisement may also matter, particularly whether the advertisement is published in a local or an international context.⁴⁹ In this regard, it is important to remember that trademark and unfair competition laws are intended to regulate market information. If a specific instance of market communication cannot reach the consumer or if it lacks the capacity to influence her decision making, then it will not be relevant. The most evident example is internet advertising in a language that is practically unknown to the population in a certain jurisdiction.⁵⁰ Finally, of course, the capacity of market communication to reach the consumer or customer will always also depend on the target group. Commercial customers may be more versatile and willing to overcome language barriers or pay transportation costs than private consumers.⁵¹

3 Clarification: Defendant’s Intent and Actual Effects

Against this backdrop, two clarifying remarks are in order. The first one concerns the suggestion that so-called finality or the intentional targeting of marketing activities should determine the applicable law. As discussed earlier, this must be rejected: if we accept the idea that trademark and unfair competition law aims to protect consumer decision making, we must also avoid elements of subjectivity for conflicts law and choice of law. The field has evolved into a regime of market information regulation. The traditional tort paradigms, notably intent and other subjective elements, are thus largely irrelevant.⁵² There is one aspect, however, that requires further analysis. Of course, the burden of proof is an issue that each national regime is free to establish under its own substantive or procedural law. Sensibly, however, it should be common understanding that the plaintiff, as part of her claim, must assert (and prove) that the effects at issue were foreseeable. Once this has been established, mere assertions by the alleged violator that she did not intend to target a certain market must generally be disregarded; otherwise, this would provide bad-faith defendants with escape options.⁵³ With regard to the defendant’s assertions, therefore, intent is irrelevant. But the reverse scenario is different. If proven, the defendant’s intent may be a proxy for the foreseeability of marketplace effects. This is a rule of procedural efficiency: most of the time, the judge is a nonexpert in the sector of the litigant parties’ industry and trade. It is hence often difficult, if not impossible, for her to correctly assess and evaluate the conflict at bar with respect to the underlying short- and long-term economics. Quite differently, however, the parties must be presumed to have the necessary skills and to act rationally. They will thus undertake only those activities that promise favorable outcomes—in other words, their cost–benefit analysis has a higher probability of being correct than the judge’s evaluation.⁵⁴ Consequently, if intent on the side of the defendant has been established, effects on the relevant market should be deemed foreseeable.⁵⁵

In addition, a second clarification concerns the issue of seemingly insignificant actual effects. As we have seen, the German Bundesgerichtshof found sufficient effects in its 1970 decision in Tampax on the basis of a spillover of Swiss newspaper advertising into Germany.⁵⁶ The case illustrates that the problem is not instances of actual impact. Trademark protection, like unfair competition prevention, does not look at actual injury. Potential effects are what matter.⁵⁷ The Tampax reasons, read more closely, also highlight this point: the spillover may have had anywhere from no effect (if the newspaper had remained unread) to a significant effect (if the newspaper’s marketing message had been circulated and proliferated repeatedly). The court, however, saw no need for further inquiry. It did not care about actual numbers but correctly based its finding on the fact alone that the newspaper’s distribution had occurred in the course of usual business activity providing for the possibility of sufficient effects.⁵⁸ Another case illustrates how this point can be brought to an extreme: the Oberlandesgericht Hamburg, in a scenario concerning advertising for printing equipment in a French trade magazine, found a sufficient basis for German law to be applied even though the magazine had only two regular German subscribers.⁵⁹ The court’s correct finding highlights that market structures matter. If the market is small enough with regard to the buyer side, even a nominally minuscule impact will suffice to exert sufficient effects. There is no blanket de minimis rule to be applied.

C International Comity

But foreseeability analysis as such does not constitute a comprehensive rule. There is one more aspect that must be given regard to—the requirement of jurisdictional self-restraint beyond the framework of public international law limitations. This is given short shrift in current theory and practice both in the United States and in Europe. Most approaches thus share a defect: they conflate fundamentally different aspects of de minimis analysis and lack a precise structural guideline regarding when and how to limit the territorial scope of national laws.

1 Current De Minimis Standards

I have already explored the HOTEL MARITIME case.⁶⁰ An interesting counterpart in US practice can be found in the Second Circuit’s 1994 Sterling Drug v. Bayer AG opinion.⁶¹ The court had to decide on an American right owner’s claim of trademark infringement by a European drug company that was using the same trademark as the plaintiff—not only in Europe, but also in the United States. While the plaintiff had rights in the United States, the defendant relied on trademark rights in Germany. Even though the Second Circuit and the Bundesgerichtshof began with a different understanding of their respective law’s scope,⁶² both decisions are representative of a widely convergent technique of de minimis limitation.

a The Paradigm of “Shields” and “Swords”

Under the law as it stood in 1994, the plaintiff’s situation in Sterling Drug was precarious. The Second Circuit’s governing precedent at the time (Vanity Fair) was unfortunate for US trademark owners trying to fend off a foreign right owner’s use of a validly registered foreign trademark abroad.⁶³ As the Second Circuit explained:

[I]f we applied the Vanity Fair test mechanically to the instant case, we would forbid the application of the Lanham Act abroad against a foreign corporation that holds superior rights to the mark under foreign law. But such an unrefined application of that case might mean that we fail to preserve the Lanham Act’s goals of protecting American consumers against confusion, and protecting holders of American trademarks against misappropriation of their marks.⁶⁴

Even though Chief Judge Newman ultimately distinguished the case from Vanity Fair,⁶⁵ it was clear that a rigid adherence to a binary system of conflicts resolution—in other words, an all-or-nothing approach—had become obsolete. On the one hand, nonapplication of the Lanham Act would neglect the protection of US customers against confusion initiated by uses of foreign trademarks abroad. On the other, applying the Lanham Act indiscriminately would also be unreasonable. Required instead was a concretized and fact-specific tailoring of remedies in accordance with the instances of the trademark use at issue. As Newman further pointed out—and this is critical—a certain degree of consumer confusion within the national territory may have to be tolerated in order to prevent a breakdown of international commercial communication and advertising activities:

In today’s global economy, where a foreign TV advertisement might be available by satellite to U.S. households, not every activity of a foreign corporation with any tendency to create some confusion among American consumers can be prohibited by the extraterritorial reach of a District Court’s injunction.⁶⁶

A decade later, in Germany, the Bundesgerichtshof expressed the same concern in the context of the online dispute at the center of the HOTEL MARITIME case. As the court warned, if each instance of internet use were found to constitute a legally relevant effect in the protecting country, it would result in a problematic return to the nineteenth-century paradigm of trademark universality. Such an overextension of domestic rights would ultimately stifle international communication and transacting:

Not any use of a mark on the internet is subject to the national legal order’s protection of marks against confusion. Otherwise, protection of national rights would be extended shorelessly and would—contrary to the European freedom to provide services … —inadequately restrict self-expression of foreign enterprises. This would involve a significant limitation of opportunities to make use of rights on the internet since owners of confusingly similar marks, protected in different countries, could—irrespective of the priority of the conflicting marks—reciprocally demand forbearance of use from the other side.⁶⁷

Indeed, both opinions can be seen as expressions of an internationalist understanding in trademark conflicts law. Judicial self-restraint is a must. Newman’s allegoric reference to “swords” and “shields” is fitting:

Though Congress did not intend the Lanham Act to be used as a sword to eviscerate completely a foreign corporation’s foreign trademark, it did intend the Act to be used as a shield against foreign uses that have significant trademark-impairing effects upon American commerce.⁶⁸

While fending off an invasion or impairment of national policies is acceptable, there exists a certain threshold of minimum effects below which domestic rights protection becomes an offense rather than a mere defense. It is the appropriate balancing of offensive and defensive measures that enables a functioning environment for international transacting and commerce.

b Analysis: An Ad Hoc Rule of “International-Individual Equity”

What becomes evident from looking at both decisions in light of my findings on policy and comity is that both courts’ reasons are based on an underdeveloped structural concept of self-restraint. Both courts applied a similar technique of effects testing. The Second Circuit tested for significant trademark-impairing effects on US commerce.⁶⁹ The Bundesgerichtshof explained that the risk of mutually blocking trademarks in the international arena could be avoided only by requiring sufficient economic effects within the German territory.⁷⁰ Both courts’ tests, however, neglect two specific aspects: they suggest a widely unqualified effects analysis and they do not distinguish between private-party and public interests.

The Bundesgerichtshof’s arguments are particularly illustrative: even though starting from the World Intellectual Property Organization’s (WIPO) Joint Recommendation and its criterion of “commercial effects,” the court ultimately did not apply the recommendation’s factor list to its analysis of commercial effects.⁷¹ Instead, the judges roughly and without further qualification compared the parties’ interests, concluding that the plaintiff’s interest in receiving the requested injunction was less significant than the defendant’s interest in advertising for its hotel.⁷² Had the court considered the defendant’s conduct in more detail, however, the outcome might have been quite different. After all, the list of factors in the recommendation would have allowed for several of the case’s facts to be actually interpreted to predetermine a finding of “commercial effect”: the defendant not only served German customers in its hotel but also solicited online reservations on its website, which was fully available in German. In addition, the defendant had sent German-language advertising brochures to prospective customers in Germany.⁷³

The court’s direct reference to the WIPO Joint Recommendation may explain why it did not attempt to formulate a guideline for the qualitative determination of what should be given regard to when analyzing and evaluating effects. The Senate did not undertake a policy-oriented analysis. This left its interest-balancing approach in a rather ambiguous and imprecise stage. Under a rule of alternatives,⁷⁴ the ultimate outcome could have actually been more consistently explained: since the defendant’s hotel services were offered in Copenhagen, alternatives to a stay at the defendant’s hotel were to be found in Copenhagen and the vicinity. The marketplace at issue was coterminous with the city. Application of German law, thus, would have been an issue of non-confusion-based goodwill invasion only.⁷⁵

What is further remarkable in the court’s arguments is that only private-party concerns and interests were deemed essential. This also was the analysis in Sterling Drug.⁷⁶ Since the economic or commercial effects of the defendant’s advertising on the plaintiff’s business in Germany were deemed insignificant (unwesentlich), the balancing of interests could not provide for a prevalence of the plaintiff’s concerns.⁷⁷ Similar to the collision-of-rights theories in legal scholarship, the solution of international trademark conflicts seems to lie in a doctrine of international-individual equity.⁷⁸ Another facet of this perspective actually comes to the fore in a more recent opinion by the Bundesgerichtshof. In a cross-border dispute over the use of the trademark “Oscar,” the owner of a German trademark registration (and organizer of the annual Academy Awards ceremony) asserted infringement through the defendant’s use of the word “Oscar” in the titles of several Italian television broadcasts. The defendant (an Italian television company) had broadcast its programs via satellite to Germany and a number of other countries. As the court explained:

What is required is … a comprehensive balancing of the parties’ interests that, in addition to considering the weight of effects of the symbol’s use at issue on the trademark owner’s domestic economic interests, must also give regard to the question of how far it was possible and reasonable for the defendant to avoid an inland infringement.⁷⁹

Since satellite transmission cannot be precisely separated along national boundaries, the court assumed, it will inevitably find receivers in several countries. An overly strict enforcement of trademark rights might hence make European satellite transmission impossible. Necessarily, therefore, the court concluded, the spillover of trademark-infringing effects—if and to the extent that these effects are unavoidable—must be acknowledged and cannot be infringed by a domestic right owner.⁸⁰

In this light, the status quo of de minimis analysis can be summarized by a few basic rules: confronted with cross-border trademark infringement, courts will not limit their consideration of interests to those of the plaintiff; equally important are the other side’s concerns. In addition, as long as effects within a certain jurisdiction cannot be reasonably “avoided,” application of the respective regime is problematic. In particular, if economic or commercial activity (e.g., satellite or internet communication) would be hindered by an overly strict enforcement of trademark rights, the tendency is to abstain from judicial intervention. However, there is no structured and precise qualitative standard for determining the significance or sufficiency of effects. Instead, courts apply an ad hoc rule of international-individual equity.

2 Reconceptualization

Of course, the current practice of de minimis testing provides for acceptable results in many cases. As I have already alluded to, there often is no “costless” solution for international trademark and unfair competition disputes—many cases will not allow for a complete avoidance of right and policy conflicts. Hence, a balancing of individual interests seems to be most pragmatic and reasonable. After all, from the court’s perspective, the litigant parties’ concrete dispute is the only thing that must be resolved. In addition, these are the interests that a judge can practically ascertain with acceptable effort. Yet distortion looms beyond the picture of interparty relations. If courts limit their analysis accordingly—especially by neglecting long-term effects with respect to the policies involved—they lose sight of the structure of conflicts resolution and choice of law “under the surface.”⁸¹

a Structural Underpinning and Relevant Interests

First, it is necessary to challenge the governing technique of ad hoc decision making, which claims to focus on rather vague concepts of interest balancing and proportionality in general. Even though “interests” are indeed what must be evaluated and balanced, the conceptual self-limitation of current doctrine overlooks a critical aspect: the evaluation and balancing of interests in international trademark and unfair competition conflicts is not an issue of general “fairness” or “equity,” nor is the judge left to her own devices. The area of trademark and unfair competition conflicts is founded on dense structures of international agreements—in particular, however, on a transnational convergence of substantive law policies. My analysis of substantive trademark and unfair competition law has illustrated this infrastructural underpinning, which also predetermines choice of law. Any interest, in order to be eligible for consideration, must thus be founded on or be correlated with the regulation of market information. The lack of a qualitative functional analysis aside, at this point, it is questionable whether the HOTEL MARITIME court actually used the correct numbers—that is, the costs ensuing from the defendant’s impact on market information, not the actual turnover numbers or other costs—for its calculation and balancing. Unfortunately, the court’s reasons do not explain the metric that was used to compute the “insignificance” of commercial effects.⁸²

Second, following from this qualification, it is important to distinguish private-party concerns from state interests. Under current doctrine, interest balancing is based on the private parties’ “legitimate interests”⁸³ or “equities.”⁸⁴ While this may seem fair to the parties, it neglects the overall impact of conflicts resolution in the field. This defect stems from the field’s doctrinal history of private rights protection. And it is somewhat reflective of the courts’ limited focus on private-individual parties and their interests—not on the overall consequences of legal doctrine in the field.⁸⁵ Yet the analysis must not be limited to individual or private-party interests; it always requires reconciling the state interests involved.⁸⁶ All issues in the core area of trademark and unfair competition law are oriented toward market information infrastructure and its protection.⁸⁷ For conflicts law and choice of law, the conclusion is inevitable: the conflict is between legal regimes, not private rights.⁸⁸ It is thus not a rule of international “equity” but one of international “comity” that must be applied.

In this light, international trademark and unfair competition conflicts usually entail at least two states’ divergent interests⁸⁹—namely, divergent interests regarding the freedom or the limitations that are necessary to establish or maintain the optimum status of market information. We must be aware of the fact that what fosters one state’s interest often undermines the other’s. Accordingly, conflicts resolution will seldom be costless in the sense of allowing for a comprehensive avoidance or reconciliation of all concerns involved. Let us return to the HOTEL MARITIME scenario for illustration, with slightly modified facts: if a market actor uses a certain trademark in online advertising targeted at her seat jurisdiction, but if this advertising can also be accessed abroad where the identical symbol is already in use as a trademark by a competitor (for identical products), the conflict is not limited to the individual parties. While the first actor’s seat jurisdiction has an interest in the trademark’s domestic functions (notably search cost reduction among local customers), accessibility of the advertising in the other competitor’s jurisdiction may cause consumer confusion there—and, accordingly, higher search costs among the other jurisdiction’s consumers. If a binary technique of conflicts law requires an all-or-nothing approach under one single chosen law, it is either the defendant’s domestic customer base that is divested of valuable market information or the foreign-based consumers who are being confused.⁹⁰ At an abstract level, both jurisdictions involved may follow widely corresponding policies fostering an optimal level of market information within their respective territories. But the policies in action—and each jurisdiction’s concrete interest in the outcome of the case—are in open conflict.

But the analysis is not complete with a look at the immediate costs and benefits alone. In addition, the long-term consequences of the court’s decision must be taken into account. This brings us back to my analysis of international economic law and the doctrine of international comity: extending domestic law beyond national borders, as is commonly argued, risks invading foreign jurisdictions’ sovereignty with respect to regulating their internal information infrastructure. But this is not the only problem. Although US practice under the McBee fallacy still contends otherwise,⁹¹ it is usually impossible to regulate foreign markets through the cross-border extension of national law. What may then ensue is an anticompetitive discriminatory application of different regulatory standards—mostly to the detriment of domestic parties.⁹² And even if extraterritorial regulation should be effective, an ultimate distortion of international transacting may yet result from retaliation. In other words, the risk that other jurisdictions could apply the same overextensive rules would ultimately paralyze many sectors of communication.⁹³

b Practical Rules and Presumptions

Against this backdrop, a more detailed practical guideline of de minimis analysis can be suggested. Even though the reconceptualization of jurisdictional self-restraint cannot provide for a one-size-fits-all rule or an exact demarcation between admissible and inadmissible extension of national rights and policies, it does provide for a more solid and comprehensive test.

(i) Starting Point: Fact-Based Crafting of Remedies

In essence, the decision maker is referred to a multistep analysis. Since there is no one-size-fits-all solution if the defendant’s conduct covers more than one kind of marketing activity, the court must undertake a separate analysis for all single instances of alleged infringement.⁹⁴ For each single instance, then, she must determine the overall consequences of both the alleged infringer’s or violator’s activity and a court-crafted remedy (usually an injunction). This fact-based analysis provides the fundament for a flexible tailoring of redress.

In detail, what is to be considered with respect to market information requires comparing all market parameters, particularly the parties’ sales numbers, the geographical span of their markets, sales channels, and consumer sophistication. Sterling Drug provides for an illustration of such a salient fact-based analysis. Under the rubric of “Background,”⁹⁵ both the district and circuit court summed up the factual background and extensively analyzed the economic setting of the conflict. On this basis, then, with respect to the actual redress, a court should compare all available options of court-crafted remedies. Quite often, the judge has a continuum of measures of corrective court invasion in her hands. An example of such a comparison can be found, at least rudimentarily, in Judge Newman’s instructions to the district court: inter alia, he required the lower instance to consider adequate restrictions to the defendant’s international marketing activity by means of a categorization of the relevant print media.⁹⁶ In this regard, he explained, the “placing [of] a full-page ‘Bayer’ advertisement in the U.S. edition of a foreign magazine or newspaper” should not be considered admissible.⁹⁷ An injunction in Sterling Drug’s favor would, accordingly, cover a prohibition on this kind of marketing activity. However, he went on to explain that “it might be inappropriate [for the district court] to leave the injunction so broad as to ban the announcement of new medical research in Lancet, or an employment notice in Handelsblatt [a leading German business newspaper].”⁹⁸ This differentiation reflects the fact that the dispute primarily concerned a consumer product (pharmaceuticals) and its marketing vis-à-vis consumers. Necessarily, therefore, the instruments of market communication would cover popular national media but not scientific journals or foreign newspapers. In essence, the court undertook an individualized and probability-based infringement analysis: the higher the probability of infringing by a certain communication instrument’s circulation in the United States, the more stringent the court’s remedy.

One aspect is important under a methodological perspective: this approach can provide for only a reasonable approximation of the cost-benefit ratio of cross-border regulation and extraterritoriality of rights. Yet by spelling out the factual basis of its verdict, the court helps avoid a defect in future decision making. An expressly fact-based finding makes clear that the court’s holding is founded on the concrete and individual case. It thereby prevents a mis- and overinterpretation of the holding in the sense of an all-encompassing and generally valid rule of law.⁹⁹ In other words, the more detailed a court’s factual analysis, the clearer the instruction to later courts to question their cases’ factual basis and, if necessary, distinguish their case from the precedent. The message should be clear: each case requires its own fact-specific analysis.

In this regard, apart from the lack of qualification of effects, the HOTEL MARITIME holding can be shown as imprecise, if not incorrectly decided, in at least one more respect. The court’s analysis suffered from the traditional individual-party focus. By looking just at the litigants, the court found that the defendant’s use of the mark caused only negligible and insignificant injury. After all, the small competitor-defendant did not seem to pose a real threat to the trademark owner’s large hotel chain. The issue would have become more complex, however, had the court also taken into account potential later-comers’ use of the plaintiff’s trademark—hence, if it had also considered the long-term effects of their holding. In essence, under the court’s doctrine, hardly any kind of use of a competitor’s trademark—if only the alleged infringer is small enough—can be characterized as exerting sufficient effects abroad. By this means, the court created a risk that the number of small competitors’ use of an identical or similar trademark may multiply. In the end, large-company trademark owners have been factually outlawed with respect to attacks by small foreign-based competitors.

(ii) Prima Facie “Effects Sufficiency”: Defendant’s Intent

We have already seen that if intent on the side of the defendant can be proven, the foreseeability of effects should be acknowledged: an attempt to reach across the border will be undertaken only if economic success looms.¹⁰⁰ A similar rule can be formulated when a limitation in light of international comity is at issue. Here as well, the parties’ cost-benefit analysis has a higher probability of being correct than the judge’s evaluation. If that is true, however, a long-term perspective suggests that the expectation of a positive output will invite imitation by others. It is then no longer the individual and concrete interparty cost-benefit analysis that matters. Over time, the sum of effects must be expected to increase. In light of the long-run perspective, therefore, effects are not negligible and must be presumed to be above the de minimis threshold. Here again, we must be aware that a presumption can provide only for a rule of approximation. Of course, the defendant’s intent is a test factor that focuses primarily on the concrete case and the individual parties. In the long run, however, it provides a guideline for private-party marketplace activity in general and thereby functions as a regulatory corrective. It will ultimately bring out the next-best result to a precise cost-benefit calculation.

Once more, we can explain the HOTEL MARITIME holding as partly imprecise. As already mentioned, with respect to the choice-of-law question, the court failed to account for the defendant’s active targeting of a German customer base.¹⁰¹ In the case, both the defendant’s website and its mail advertising were directed at a German public. They offered extensive information in German, and it was even possible to make online reservations in German. There was no explanation for the use of the German language or for the soliciting of German consumers other than that it was a lucrative kind of marketing from the defendant’s point of view. Accordingly, the court should not have neglected this kind of marketing as having only insignificant commercial effects. Quite differently, for the sake of illustration, the English case Euromarket Designs Inc. v. Peters and Crate & Barrel Ltd.¹⁰² presented a case of apparently de minimis effects in a similar scenario: the defendants ran a single retail store for household goods and furniture in Dublin, Ireland, and operated a website on which the plaintiff’s UK and European Community trademark “Crate & Barrel” was used. Judge Jacob, denying an infringement, emphasized two facts. First, the mere accessibility of the website was not enough. As he put it, “[T]he website owner should [not] be regarded as putting a tentacle onto the user’s screen.” Second, the defendants had not actively gone out to solicit customers in the UK.¹⁰³

(iii) Caveat: “Effects Unavoidability”

Finally, case law and commentary have discussed cases where effects within a jurisdiction are deemed de minimis if a defendant’s actions include a reasonable effort to avoid or minimize an infringement. A common-sense approach seems to be that infringing use of a trademark (or unfair competition) should not be deemed to occur if the defendant takes reasonable steps to avoid interference with the domestic market to the best possible degree.¹⁰⁴ While analysis in these cases has often been unduly curtailed, asking for the avoidability of effects makes sense as a practical proxy for determining the sufficiency of effects.

Of course, if effects within a jurisdiction are truly avoidable, there does not exist any conflict of policies or interests. This is the case, for instance, where confusion can effectively be excluded by a disclaimer.¹⁰⁵ In many disputes, however, the alleged infringer or violator cannot prevent her conduct from having effects in more than one jurisdiction—no matter what she does. One example is the use of a word mark that is not well known in the defendant’s seat jurisdiction as part of a domain name under this jurisdiction’s top-level domain. If the website can be accessed in other jurisdictions where the symbol is well known for a competitor’s products, the conflict—in the sense of effects on the trademark’s reputation and prestige—can hardly be avoided (at least not by a disclaimer).¹⁰⁶ Asking for the avoidability of effects in such a case implies that the defendant’s conduct—and, accordingly, this conduct’s effect—is legitimate. Otherwise, a genuine rule of avoidability would actually require completely ceasing the activity at issue. This assumption of legitimacy can best be explained by reference to the collision-of-rights perspective.

As we have seen, under a lens centered on individual parties and rights, it seems as if the conflicts panacea can be found in a rule of international-individual equity.¹⁰⁷ Looking at the same dispute in light of the policies and state interests involved, however, indicates the need for a balancing rule beyond the individual parties’ concerns. In certain cases, the result of the interest balancing may actually differ from the individual parties’ avoidability scheme. Back to our example: even if the defendant’s online use of a symbol that is a famous trademark abroad should be limited to a “reasonable minimum” or to the “best possible degree,” she may still not be ordered to cease using the disputed domain name. Hence, one might have to find the injury resulting from its ongoing use to be “unavoidable.” Yet the overall perspective on all short-term and long-term interests in regulating market information may indicate that completely enjoining the defendant’s use would be less costly in terms of the public interests involved in all jurisdictions. A probability-based infringement analysis and cost balancing may bring out different results than the individual equity balancing. Particularly if the plaintiff’s market information capital is large, even an absolute minimum use of the symbol—the defendant’s bad faith aside—may so significantly distort the market information infrastructure (e.g., through misguided online searches) that it will ultimately result in an overall negative cost-benefit account in both jurisdictions.

Nonetheless, the rule of avoidability has practical value. Often, the analysis of conflicting interests and the cost-benefit computation is difficult. A court will then shy away from making harsh all-or-nothing decisions and will tend to find the equitable “compromise.” And this need not be unreasonable in terms of procedural efficiency. As long as it is unclear whether an alternative structuring of the transnational information infrastructure—by means of a court-crafted remedy for the dispute at bar—is less costly and more beneficial under an overall and long-term perspective, the court should follow a rule of avoidability. As a rule of practical approximation, it brings substantive-policy analysis and procedural efficiency to conformity.

IV Summary

Looking at trademark and unfair competition conflicts in light of the underlying policies indicates a uniform approach. Whenever a conflict involves conduct that has an effect on consumer decision making and transacting, the point of attachment must be found at the place (or places) where alternative transactions exist. As this convergence implies, there is no difference between the fields with respect to the quality of effects required for conflicts determination. In light of this qualitative assessment of effects, the necessary quantity or intensity of actual or potential effects will be determined by testing for objective foreseeability. What judges must undertake is a market analysis concerning the product and marketing communication at issue. Finally, conflicts resolution calls for a separate testing of international comity aspects in addition to the analysis of effects foreseeability. In essence, courts must undertake a fact-based analysis and craft their remedies accordingly. Both defendant intent and effects avoidability may serve as a practical proxy.

Section 2 The Reinterpretation of Steele and Rome II

This consolidated conflicts resolution structure calls for a new interpretation of existing trademark conflicts and unfair competition choice-of-law rules. Both US conflicts law and European choice of law can be reconceptualized through moderate modifications to the Bulova test and its variants, as well as by reinterpreting the Rome II Regulation.

I US Lanham Act Subject-Matter Jurisdiction

As discussed earlier, some have suggested extending the unilateral Bulova test in order to establish a multilateral rule of trademark conflicts law.¹⁰⁸ This would result in the application of foreign laws in US federal fora. Indeed, this option is not too exotic. For example, international copyright infringements are deemed to bring into existence a so-called transitory cause of action when foreign copyright laws are applied by US courts.¹⁰⁹ Also, international tort conflicts may require that foreign laws be applied.¹¹⁰ And finally, the Judicial Improvements Act of 1990 allows district courts in civil actions, if they have original jurisdiction over the action, to exercise supplemental jurisdiction “over all other claims that are so related to claims in the action within such original jurisdiction that they form part of the same case or controversy.”¹¹¹ At least in cases where the effects at issue trigger the application of the Lanham Act and, concurrently, of other countries’ trademark and unfair competition laws, a multilateral conflicts resolution would not be anathema to procedural doctrine.¹¹² Indeed, the concurrent application of different trademark or unfair competition regimes may provide for a reasonable resolution of conflicts, particularly with regard to economic and regulatory concerns.

Nonetheless, it is questionable whether conceptions of such a groundbreaking reformulation of US conflicts doctrine have much prospect of success. A more realistic suggestion should focus on the most oft-debated factor in both multilateral and unilateral conflicts resolution—the restriction of the domestic regime’s scope of application. My focus will thus be on current doctrine. As I will demonstrate, shifting the focus of the Bulova effects prong toward a more functional analysis can be consistently adapted to and implemented in the analysis of Lanham Act subject-matter jurisdiction. This means that it will not be “some,” “substantial,” or “significant” effects that determine the finding of a relevant impact on US commerce but a modified test factor of foreseeable minimum effects—on consumer decision making. In addition, modification of the “nationality” and “conflicts with foreign law” factors will bring US doctrine into conformity with a comity-based requirement of jurisdictional self-restraint.

A Modification: A Qualitative Reformulation of “Effects on US Commerce”

The chapter on international comity explored the detrimental effects of trademark extraterritoriality on international competition.¹¹³ In fact, this is the most crucial problem with Lanham Act subject-matter jurisdiction, albeit one that is widely ignored. It is due, among other things, to the indeterminate terminology and application of the Bulova effects prong in different circuits.¹¹⁴

The divergence can be illustrated, for instance, by a comparison between the Fifth Circuit’s American Rice “some effects” factor and the Second Circuit’s “substantial effects” requirement. Interestingly, neither the Supreme Court majority in Steele nor the dissenting justices gave an express definition of effects as “substantial,” “significant,” or otherwise.¹¹⁵ This was different for the appellate decision at the lower level. The Fifth Circuit majority opinion in Bulova Watch Co. v. Steele¹¹⁶ was the first to make use of the term “substantial economic effects.”¹¹⁷ Although the opinion has not received significant attention in case law or scholarship, it apparently had some genuinely “terminological” influence on the Second Circuit in Vanity Fair. There, Judge Waterman, making reference to the Fifth Circuit’s decision, adopted a requirement of “substantial effects” without further elaboration.¹¹⁸ In the end, Vanity Fair was not based on a finding of effects. Nonetheless, it has proven to be remarkably successful. Even today, this requirement remains an element of the Second Circuit’s test, and it has influenced the tests used by a majority of federal circuits.¹¹⁹ Outside New York, the Nevada district court in Wells Fargo was the first court to pick up the Vanity Fair standard of “substantial effects.”¹²⁰ The Ninth Circuit appellate court, however, corrected the district court’s holding in 1977, stating that Steele contains no substantiality requirement.¹²¹ Shortly after, the Fifth Circuit in American Rice adopted the Ninth Circuit’s standard and held that “some effects” would be sufficient.¹²² Over time, these different standards have spread throughout the circuits. The Eleventh Circuit, for example, still interprets Steele as requiring “substantial effects.”¹²³ Other circuits either follow one specific circuit’s approach or apply a combined test.¹²⁴ The Fourth Circuit, for instance, has established a “significant effects” standard derived from the three-pronged Vanity Fair test.¹²⁵

As illustrated by the variety of tests, particularly the terminological noise and confusion in the debate on effects determination, there is no truly authoritative standard of qualification. Ultimately, therefore, as a practical consequence, almost any effect might be sufficient to trigger the application of the Lanham Act. This brings out the specter of overextension by effects indeterminacy. One case in the wake of Steele lucidly illustrates this problem. In the 1983 American Rice case, both parties were American agricultural cooperatives acting in the United States and abroad. Their dispute arose out of a trademark resembling the plaintiff’s US registration, which the defendant used for selling rice in Saudi Arabia.¹²⁶ Even though the defendant’s sales occurred solely in Saudi Arabia and “none of [the] products found their way back into the United States,”¹²⁷ the Fifth Circuit held that the plaintiff’s US trademark had been infringed on. Sufficient effects were found on the basis that processing, packaging, transporting, and distributing US-produced rice constituted activities “within commerce.”¹²⁸ It was not reported whether other competitors in the Saudi Arabian market made use of allegedly infringing symbols. Very likely, however, most of them could not be haled into a US court for want of personal jurisdiction, and the prospects of successfully litigating before Saudi Arabian courts were likely also dismal (at least if compared with litigation in US federal courts). Consequently, deciding on the dispute between domestic competitors, the American Rice court—even though formally extending protection for the owner of a domestic right—factually burdened a national competitor. Ultimately, the discriminatory application of US law resulted in an uneven burden to other domestic parties competing with the right owner abroad.¹²⁹

What is most striking in American Rice is that virtually completely unqualified effects—fully detached from the actual marketplace—sufficed to trigger the application of the Lanham Act, a statute that is specifically designed to regulate market communication and information.¹³⁰ More generally, this tendency of rights extension can actually be seen in a large portion of the Steele progeny between 1952 and 2014. As my bird’s-eye view in chapter 2 has brought up, both the idiosyncrasies of an unqualified effects test and the common law pedigree of transnational goodwill and trademark rights acquisition and protection have contributed to a wide extension of national rights into foreign-based marketplaces.¹³¹ A first corrective is thus necessary: the void of qualitative guidance must be filled by looking at the core policy of trademark protection and unfair competition prevention.¹³² In essence, this requires saying farewell to the debate on merely terminological nuances; it does not matter whether effects are “substantial” or “significant,” or whether “some” effects will do. Instead, the basis of effects testing must be a policy-oriented analysis of the impact that defendant activities have on market information infrastructure. Only if the infrastructure is affected, will relevant “effects on US commerce” be found. On this basis, for the American Rice scenario, the correct outcome is then easy to find: selling rice under an allegedly infringing trademark in Saudi Arabia did not affect the market infrastructure—nor consumer decision making—in the United States. There were no relevant domestic effects for the regulatory policies at issue. Accordingly, there should have also been none for the triggering of the Bulova effects prong.

B Reinterpretation: Dusting Off “Nationality” and “Conflicts with Foreign Law”

The picture is no less complicated regarding the other two Bulova test factors: “nationality” and “conflicts with foreign law.” A constant corrosion has been going on behind the scenes, in which both of these factors have been increasingly invalidated. In this regard as well, however, a few modifications help correct existing aberrations.

1 Nationality, Citizenship, and What Else—or Nothing at All?

Testing the nationality or citizenship of a defendant in an international trademark or unfair competition dispute seems a simple task. In fact, the nationality principle has always been acknowledged as a legitimate instrument of choice of law.¹³³ Practically, it is a convenient test factor. The preconditions set by Steele, Vanity Fair, and subsequent case law are straightforward. In the same vein, the First Circuit’s McBee test recently established the defendant’s nationality as the primary test factor by setting different standards for US and foreign infringers. For US citizens, the court explained, jurisdiction is a matter of domestic law “that raises no serious international concerns, even when the citizen is located abroad.”¹³⁴ By contrast, for foreign defendants, a court’s subject-matter jurisdiction would have to be based on the conduct’s effects on US commerce and, therefore, on a different constitutional power.¹³⁵ Even though virtually all courts adhere to this seemingly easy and unbiased test, the results can differ significantly. In fact, as a closer look reveals, the nationality factor has developed into an empty shell.

Steele already treated the nationality factor casually. The majority phrased it simply:

The issue is whether a US District Court has jurisdiction to award relief to an American corporation against acts of trade-mark infringement and unfair competition consummated in a foreign country by a citizen and resident of the United States.¹³⁶

In a footnote, the court stated, “Joined as parties defendant were S. Steele y Cia, S.A., a Mexican corporation to whose rights Steele had succeeded, and Steele’s wife Sofia who possessed a community interest under Texas law.”¹³⁷ In spite of this actually quite heterogeneous “citizenship” status, the court continued to refer to Sidney Steele as the sole petitioner. Accordingly, the majority based its constitutional analysis of jurisdiction on Steele’s US citizenship.¹³⁸ Certainly, the existence of a Mexican corporation would not (and should not) have been enough to alter the outcome. But the majority’s treatment of the issue may be understood as having set the stage for the lower courts’ ultimately broad construction.

In fact, opinions after Steele have gone far beyond that of the Supreme Court. As demonstrated in my bird’s-eye view on the Steele progeny between 1952 and 2014, the nationality prong has been a weak corrective for the effects factor’s dominance. At best, a defendant’s foreign nationality may present an obstacle to subject-matter jurisdiction when both nationality and conflicts with foreign law point toward nonapplication of the Lanham Act.¹³⁹ And the nationality prong also has an enforcing impact when it bends in the other direction. Among 58 opinions where the defendant’s nationality (or allegiance) was found to point toward application of the Lanham Act, the courts applied US law 50 times (86.21%).¹⁴⁰

In addition, another aspect becomes visible upon a closer look at the microstructure of nationality testing: among the opinions that substantially discussed the defendant’s nationality, citizenship, or allegiance, the definition of “US nationality” has significantly varied and has ultimately been extended. First, after 1977, in the Ninth Circuit, the Timberlane comity test expressly allowed for a substitution of citizenship by a finding on “the nationality or allegiance of the parties and the locations or principal places of business of corporations,” which may include a party’s residence.¹⁴¹ This gave courts wide discretion to neglect their defendants’ foreign nationalities. In other circuits, nationality neglect may have evolved under a surface of formalities, but it was nonetheless drastic: quite often, where a judge had to find that the defendant was not a US national or entity, or that the group of defendants contained at least one foreign national or entity, many courts extended their definition of “US citizenship” to include foreign nationals with US residence and corporate responsibility for a US entity, or some other responsibility for the alleged infringements.¹⁴² Typically, this was expressed by finding a foreign defendant to be the “controlling force”¹⁴³ behind a US company. In addition, many opinions relied on more unspecified findings of close corporate relationships between American and foreign defendants, the commission of allegedly infringing acts by corporate subsidiaries in the United States, or the nationality prong already being satisfied through at least one defendant’s US citizenship.¹⁴⁴ Two opinions even based their finding of the defendant’s American “citizenship” on a prior agreement between the parties that submitted certain issues to US law and jurisdiction.¹⁴⁵ Courts in the Second Circuit in particular have coined this extensive understanding of nationality as “constructive citizenship.”¹⁴⁶ In terms of numbers, among all 140 opinions in the Steele progeny, a total of 28 (20%) regarded foreign defendant parties as US nationals for reasons of “constructive citizenship” or for a similar connex to the United States, notably based on their corporate function, their residence in the United States, or other significant contacts (e.g., choice-of-law or choice-of-court agreements). The extraterritoriality rate among these opinions (i.e., the rate of Lanham Act application) was 26 out of 28 (92.86%, compared to 59.29% overall).¹⁴⁷ It seems that whenever necessary, the nationality prong has been handled with remarkable discretion—usually in favor of applying US law.

2 Conflicts with Foreign Law: Another Shell of Formalities

The nationality test is not the only one to have been twisted and turned. The conflicts test—that is, courts’ determination of whether a “conflict which might afford [the defendant] a pretext that … relief would impugn foreign law”¹⁴⁸ exists—also paints a complicated picture.

I will refer once again to decisions issued between 1952 and 2014. Apart from considering whether the defendant actually owned a foreign trademark or had applied for registration in a foreign jurisdiction (32 opinions, or 22.86%),¹⁴⁹ some courts looked at the overall legality of a defendant’s activities abroad, including by comparing the plaintiff’s claims with the defendant’s defense under foreign laws (15 opinions, or 10.71%).¹⁵⁰ In this regard, some courts even expressly found a conflict to exist if a foreign jurisdiction’s interest in freedom of competition afforded the defendant immunity against the extension of US law.¹⁵¹ These cases account for a relatively internationalist approach, acknowledging both foreign rights and policies to warrant consideration. This also applied to another group of decisions that defined conflicts more formally, based on the stage of actual or potential litigation. Courts there asked, inter alia, whether litigation outcomes in the United States and abroad might be inconsistent (8 opinions, or 5.71%).¹⁵²

A large group of opinions, however, established a less egalitarian perspective on foreign laws. Courts within this group refused to find a conflict to exist as long as no foreign court had actually ruled that the defendant had a legal right to use a trademark (6 opinions, or 4.29%). This approach governs especially in the Fifth and Ninth Circuits.¹⁵³ In the same vein, some courts found the burden to be on the defendant “to show that [she] has a superior right in a foreign country to prevent the imposition of an injunction.”¹⁵⁴ Evidently, therefore, both substantive law doctrine and procedural law can prevent a court from placing too much emphasis on the existence of conflicts with foreign law.¹⁵⁵ Like the nationality test, the conflicts test thus—even though formally tailored to give regard to international consensus and convenience of transacting—tends to neglect comity.

3 A New Paradigm

Against this backdrop, it is clear that modernization of the Bulova test factors for “nationality” and “conflicts with foreign law” requires a dual reorientation. The test for a defendant’s nationality is not only obsolete but economically misconceived; parties’ nationalities should be disregarded. And the test for conflicts with foreign law—though not as problematic as the nationality test—requires at least some restructuring.

a The Neutralization of Nationality and Citizenship

As seen earlier, the nationality factor’s application has resulted in circumvention and invalidation, suggesting that the test factor is outdated. In addition to practical failure, aspects of conflicts theory and economic reason counsel jettisoning nationality as one of the pillars of Lanham Act subject-matter jurisdiction.

Let us start with a deontological argument: I have already alluded to Lea Brilmayer’s political rights theory requiring that conflicts determination and application of a certain national law must have a rights-based justification. As Brilmayer explains, in conflicts scenarios, a party’s nationality or citizenship is one of the relevant aspects that may justify a state’s coercion exerted through choice of this state’s law and its application.¹⁵⁶ But this is only one factor. In addition, she also specifies more flexible connections that can justify choice and application of a state’s law. One example is individual consent, notably in the form of a party’s residence or traveling in a state’s territory; another example is the benefit that a party receives upon initiating purposeful contact with a state.¹⁵⁷ The rise of transnational communication and transacting requires acknowledging a shift in the relative importance of the different connecting factors. In the old days, of course, nationality and citizenship constituted a rather precise presumption for a connex between the sedentary actors’ activities, these activities’ consequences, and the local regulator. Since individual mobility was low, and the effects of most activities were local, nationality was a valid basis for the application of a state’s law in most cases. But modernity no longer allows for such an automatic conclusion. First of all, in today’s world, the connection between conduct and the local law’s regulatory purpose is no longer guaranteed. Before the rise of international transactions and communication, conduct and effects were often conflated within a single place. Effects occurred in the vicinity of conduct and thus served as a handy proxy for effects.¹⁵⁸ Today, however, this connection is no longer guaranteed. This does not necessarily mean, as is sometimes contended, that the traditional “link between law and land” has been broken.¹⁵⁹ There is actually no change in the subject matter of what is regulated: effects have been and will be the ultimate determinant. Accordingly, particularly in the area of economic regulation, it has become more and more acknowledged that “territorial” jurisdiction may be based either on local conduct or on the local occurrence of effects.¹⁶⁰ In the same vein, the national affiliation of individuals and corporate entities has increasingly lost its relevance for where they act or where effects of their activities will ultimately occur. In a world of highly mobile individuals and corporate actors, where everybody can virtually cause effects everywhere, national affiliation and citizenship no longer provide for a significant connex.

Finally, a closely related economic aspect must be considered. As we saw in chapter 5, giving regard to the parties’ nationality in trademark and unfair competition conflicts law may result in an anticompetitive burden for domestic competitors in foreign markets.¹⁶¹ By the 1930s, Arthur Nussbaum’s explanation of the lex domicilii communis in unfair competition choice of law had been criticized for this structural deficit.¹⁶² The same problem exists with the nationality prong in Steele: domestic competitors in foreign markets are the primary actors subjected to US courts’ personal jurisdiction; in addition, with respect to subject-matter jurisdiction, a nationality-based Bulova test further extends the risk of constraining competition. After all, as my bird’s-eye view has shown, it will multiply chances that stricter rules of US trademark protection apply if the defendant is a US national or corporate entity—but not if a foreign actor is on the defendant’s bench.

b The Deformalization and Depropertization of “Conflicts with Foreign Law”

With respect to the third test factor, the Second Circuit’s decision in Sterling Drug is again remarkable. While the court’s analysis of effects and its flexible tailoring of remedies were certainly innovative, its testing of “conflicts with foreign law” was anything but. Chief Judge Newman’s analysis of the foreign-compulsion doctrine in Hartford Fire is revealing:

In the context of [Hartford Fire], the Court found no “conflict” warranting a declination of jurisdiction because there was no claim that conformity with the requirements of United States law required the defendants to do any act in violation of British law. … [W]e think [this approach to the comity issue] is not automatically transferable to the trademark context, especially where the contending parties both hold rights in the same mark under the respective laws of their countries. It is one thing for the British reinsurers in Hartford Fire to be barred under United States law from boycotting activity that they might be free to engage in without violating British law. But it is quite a different thing for the holder of rights in a mark under German law to be ordered by a United States court to refrain from uses of that mark protected by German law.¹⁶³

Both parties, Sterling Drug and Bayer, owned trademarks in their respective home jurisdictions. Consequently, the court expected to be on solid ground when it distinguished Hartford Fire on the basis that the defendant could refer to a valid right in Germany. Yet, by this means, the existing policy conflict and its resolution were hidden behind a veil of formal rights. This is due to, inter alia, the fact that the reasoning in Sterling Drug, like US doctrine in general, was founded on a remainder of the act-of-state doctrine:¹⁶⁴ the validity of foreign trademark rights is generally not questioned since these rights are conceived of as foreign political acts. This deceptively clear-cut situation changes, however, if the conflict involves not two private “rights” but one formal entitlement and a nonformal nation-state policy, or two divergent nonformal nation-state policies. While one jurisdiction may favor the trademark owner’s interest in protecting her right, the other jurisdiction may foster freedom of competition by granting a more liberal domain of market communication. From a court’s perspective, it is already virtually impossible to determine which of the conflicting policies involved—the domestic one or the foreign one—should prevail.¹⁶⁵ How much more should it then be feasible to determine the “weight” of domestic and foreign policies on the basis of their formal implementation? In other words and more concretely, are common law use-based rights “weaker” or “less valid” than civil law registered trademarks?

To further illustrate this point, let us modify the Sterling Drug facts: if the parties’ dispute had not been on the use of both parties’ trademark “Bayer” but on an allegedly improper claim of advertising, the outcome would not have been different—even though in this variation of the facts, there was no conflict of “rights.” Let us assume that Sterling Drug (this time as the defendant) had expressly advertised its products as “imitations” of the European Bayer company’s drugs and that there had been significant spillover of the advertising to Europe. Then, a similar issue of potential “conflict with foreign law” would have existed. This is due to the fact that in cases of “imitation” or “replication” claims, European doctrine, under the Misleading and Comparative Advertising Directive,¹⁶⁶ not only finds both trademark and unfair competition law affected but also disallows comparative advertising stating (whether explicitly or implicitly) that the product at issue is an imitation or replica of a product bearing a well-known trademark.¹⁶⁷ US doctrine, by contrast, is grounded on the conviction that if a seller has a legal right to copy a product, she must also have the right to inform the public accordingly.¹⁶⁸ Imitation and replication claims are thus not improper per se. In an international conflict like in the modified Sterling Drug scenario, the issue—if phrased individually—would be to resolve a dispute between one party’s “right” (i.e., trademark) and the other party’s economic freedom (i.e., liberty to correctly describe her product). In essence, however, as under the unmodified facts, the dispute requires a reconciliation of divergent levels of economic freedom. The existence or nonexistence of formal entitlements does not make a difference. It therefore cannot determine the conflicts issue.

II European Trademark and Unfair Competition Choice of Law

As with US doctrine, European trademark and unfair competition choice of law can be reformulated without much turbulence by several modest but effective modifications of the existing rules. This requires acknowledging that the common core of trademark and unfair competition policies implies a unified approach at the conflicts level. In addition, a qualitative effects test helps reconceptualize the regulatory function of choice of law in the field. Finally, effects testing must be accompanied by a comity-based rule of self-restraint.

A Clarification: Characterization of Trademark and Unfair Competition Conflicts

While terminology varies, characterization is widely understood as the classification, qualification, or interpretation of laws that may apply to an international dispute. The judge, therefore, before addressing the actual choice of law must first determine the nature and character of the dispute before her. She will thereby have to find out which conflicts norm(s) to use to identify the applicable substantive law.¹⁶⁹ The public international law system of trademark protection and unfair competition prevention does not provide a detailed guideline for characterization. There are no common normative standards that universally define “unfairness” or “dishonesty” in competition.¹⁷⁰ Nor are there express rules on the differentiation between trademark protection and unfair competition prevention. The question whether a litigated claim is to be qualified as a cause of “trademark infringement,” “unfair competition,” or “torts” is therefore still an issue of forum law.¹⁷¹

Accordingly, the unified European choice of law under the Rome II Regulation is based on the idea of autonomous characterization.¹⁷² This requires the application of a single supranational standard, notably in light of recital 21 of the Rome II Regulation, which provides for a triple purpose of protection:

In matters of unfair competition, the conflict-of-law rule should protect competitors, consumers and the general public and ensure that the market economy functions properly. The connection to the law of the country where competitive relations or the collective interests of consumers are, or are likely to be, affected generally satisfies these objectives.

In light of member states’ diverging substantive laws, however, such a uniform approach used to be difficult to conceive.¹⁷³ For a long time, the supranational stage of substantive law harmonization was not much more advanced than the international level; at least with respect to a larger sector of unfair competition prevention, though, this seems to have changed more recently in the course of an increasing harmonization through European secondary legislation. As Peter Mankowski explains, a modest convergence concerning the definition of what constitutes unfair competitive conduct can be found in European substantive law, particularly in the Unfair Commercial Practices Directive’s¹⁷⁴ definition of “commercial practices” in article 2(d), which provides that

“business-to-consumer commercial practices” … means any act, omission, course of conduct or representation, commercial communication including advertising and marketing, by a trader, directly connected with the promotion, sale or supply of a product to consumers.¹⁷⁵

As Mankowski concludes, within this core area of substantive law convergence, a common and uniform characterization of “unfair competition” is both possible and required.¹⁷⁶ One could also refer to the Directive on Misleading and Comparative Advertising as an additional part of the relevant European acquis.¹⁷⁷ Yet despite the incipiencies of substantive law consolidation, these directives provide for a rudiment at best. Both instruments have a limited scope.¹⁷⁸ Notably, the Unfair Commercial Practices Directive provides for “unfair business-to-consumer commercial practices” but “neither covers nor affects the national laws on unfair commercial practices which harm only competitors’ economic interests or which relate to a transaction between traders.”¹⁷⁹ Not surprisingly, therefore, scholarly suggestions try to amend the arsenal of references and sources by including not only article 10^bis of the 1883 Paris Convention and the TRIPS Agreement but also, for instance, soft-law standards like the 1983 Conflict-of-Laws Rules on Unfair Competition by the Institut de Droit International¹⁸⁰ and the WIPO Model Provisions on Protection against Unfair Competition.¹⁸¹ In addition, agreement exists that the European member states’ national law concepts must be given regard to in the promulgation of an autonomous concept of “unfair competition.”¹⁸²

Contrast this with US conflicts law. If a scenario does not concern federal law issues, notably in cases where common law or state statutory unfair competition claims are litigated (e.g., trade secret misappropriation), courts treat unfair competition claims as genuine torts. They then apply the forum state’s choice-of-law rules, regularly referring to an interstate consensus expressed in common law rules on when and how to characterize a tortious act as unfair competition.¹⁸³ The technique is different with respect to federal law trademark and unfair competition claims. Under the Lanham Act, international trademark infringements and violations of federal unfair competition law are a question of the federal courts’ subject-matter jurisdiction. The scope of Lanham Act subject-matter jurisdiction is determined under Steele and the circuits’ different test variants.¹⁸⁴ Moreover, absent concurrent causes of action under US law, courts tend to dismiss claims based on foreign law under the doctrine of forum non conveniens.¹⁸⁵ Hence, if the case at bar does not contain a sufficient connection to the United States, it will be dismissed. Foreign law is not applied. Factually, this is a unilateral conflicts rule. Even though characterization appears redundant, this analysis implicitly still determines whether the case at bar falls within the domain of “trademarks” or “unfair competition.” Hence, it contains at least a rudiment of characterization on the basis of the Lanham Act’s substantive law policies. Essentially, however, the analysis of jurisdiction includes and replaces an actual choice-of-law decision.¹⁸⁶

But characterization need not be an issue of public international law conventions, supranational legal instruments, and national or state law alone. These approaches may be useful for producing adequate results in inter-US or intra-European conflicts. But they cannot provide for uniformity beyond the respective federal or supranational entity. What is required instead is a transnational standard. This must be founded on a broader consensus, which brings us back to the analysis of a functional core of policies in trademark and unfair competition law.¹⁸⁷ While a genuinely comparative characterization, as suggested by Ernst Rabel,¹⁸⁸ may still lack a solid foundation with regard to practical feasibility, characterization based on universal structures of substantive law policy provides for a different situation. As my analysis has revealed, the core function of both trademark and unfair competition law is the protection of consumer decision making—this is the “whole Law and the Prophets on the subject.”¹⁸⁹ Protection of market information infrastructure and unmanipulated consumer decision making are the pillars of a transnational architecture of competition fairness. This foundation guarantees universality on the basis of economic theory. And since such a functionally structured approach provides a uniform basis for characterization in the core areas of both trademark and unfair competition law, a rule of lex specialis differentiation between the two sectors is not required.¹⁹⁰

Beyond this functional core of policies, of course, there is still no harmonized concept. We have seen that the area extends much wider, particularly with respect to the prevention of unfairly competitive torts at the horizontal level. Another area beyond the core sector is international antitrust conflicts.¹⁹¹ In these cases, whether a specific instance of competitive conduct falls into the formal category of “unfair competition” is not an issue of its immediate impact on consumer decision making. For want of a uniform functional basis, characterization will then remain an issue of the forum’s (supra)national law.¹⁹²

B Foundation: Marketplace Effects Rule and the Lex Loci Protectionis

Putting consumer decision making at the center of conflicts analysis does not require a reformulation of statutory choice-of-law rules. For Rome II, the specification of the place “where competitive relations or the collective interests of consumers are, or are likely to be, affected” in article 6(1) can be reinterpreted. The regulation’s recitals explain that the collision-of-interests approach or the marketplace rule is “not an exception to the general rule in Article 4(1) but rather a clarification of it.”¹⁹³ Formally, therefore, the country where competitive relations or the collective interests of consumers are affected is the place “in which the damage occurs irrespective of the [place] in which the event giving rise to the damage occurred and irrespective of the [places] in which the indirect consequences of that event occur” (art. 4(1)).¹⁹⁴ While this clarification has overcome conduct- and damage-centrism, it needs further specification regarding the element of “competition” or “marketplace” that must be affected in order to establish “damage.” In this regard, as we have seen, reference to concepts of economic theory, notably the “marketplace” or the place of “competition,” as well as the proxy of collective consumer interests, can be problematic.¹⁹⁵ Problems can be avoided, though, by looking at the model of the market mechanism in more detail. Such a perspective clarifies the issue of what kind of marketplace effects should be seen as relevant damage. It is not, as we have seen, effects on a competitor’s position or her market share; rather, it is effects on consumer decision making and transacting that determine the place where the damage occurs and where competitive relations or consumer interests are affected.

While this makes clear that no structural reinterpretation of the marketplace rule in article 6 of Rome II is required, reconstruction—at least upon first sight—appears more complicated for the lex loci protectionis rule in article 8 of Rome II. The law applicable to a trademark infringement is “the law of the country for which protection is claimed.” Prima facie, no effects rule is implemented. The lex loci protectionis, however, is a multilateral and a quasi statutist rule of conflicts determination.¹⁹⁶ This means that each regime determines its own scope of application. The substantive law provides for the relevant aspects required to find an infringement of domestic rights. On this basis, no national law is obliged to implement a strictly conduct-based concept of territoriality. On the contrary, applying domestic law to effects within the state’s territory—particularly to effects on consumer decision making—will not extend national trademark rights illegitimately.¹⁹⁷ In sum, therefore, under the Rome II Regulation, both for trademark and unfair competition conflicts, a qualified effects approach can be implemented by reinterpretation of the lex lata.

C Application: Marketplace Effects and the Gran Canaria Conundrum

Before we develop a concluding typology of trademark and unfair competition conflicts,¹⁹⁸ let us test this concept on one of the most contested scenarios in European doctrine: the German Bundesgerichtshof’s 1990 Kauf im Ausland decision, also known as the Gran Canaria case. This case is a result of what can be characterized as a merger of once separate national markets into a single multijurisdictional marketplace. I have already explained how globalization has perforated national borders. And we have seen that in order for regulation to keep up with this development, adherence to a paradigm of conduct-based choice-of-law theory must be overcome. The place of conduct or impact no longer provides for a sound attachment.¹⁹⁹ This perspective also requires a change of directions in cases of the Gran Canaria kind.

1 Recapitulation: The Gran Canaria Scenario

For many marketing activities, the place of conduct and the place of transacting still coincide. Transactions occur at the point of (or in close proximity to) the competitor’s preceding marketing conduct. The place of conduct, then, is also the place of interest collision or marketplace effects. This is where the consumer has an interest in true and honest information, where competitors’ interest in fair competition arises, and where the public is interested in the free and unhindered functioning of competition and the market mechanism.²⁰⁰ Conduct in, impact on, and effects within the marketplace coincide.²⁰¹ This is not the case, however, if the place of competitor conduct and the place of consumer transacting are different—for example, when advertising takes place in one state’s territory for a product that is available only in another’s,²⁰² or when advertising takes place abroad but targets domestic consumers in order to lead to an inland transaction. These cases, though most common in an online environment, can also occur in the offline world. The Bundesgerichtshof’s decision in the Kauf im Ausland case addressed such a scenario.²⁰³

The facts are well known:²⁰⁴ German tourists in Gran Canaria, Spain, were solicited through German advertising materials. The contract was in German, and delivery of the products (merino wool duvets and pillows) was to occur in Germany upon the tourists’ return. According to its terms, Spanish law was to apply to the contract. Even though the core issue may have appeared to be whether the choice of Spanish contract law was valid, the case was initiated on claims that the contract, in fact, violated German unfair competition law. The plaintiff, a German consumer association, contended that the defendant’s Spanish sales associate had not properly informed buyers of their right to rescind the contract under German consumer protection law. The information requirement in German law was due to European secondary law²⁰⁵ and was considered an issue of both consumer protection and unfair competition law. Accordingly, the consumer association as plaintiff claimed that actual noninformation constituted a case of statutory breach under German unfair competition law.²⁰⁶ As the Bundesgerichtshof explained in its decision—which has been widely approved by courts and scholars—the place of the advertising market (Werbemarkt) generally serves as the point of attachment if advertising conduct and effects occur in different territories.²⁰⁷ Critics are in the minority when they contend that the advertising market need not necessarily be the place where effects materialize, and that consumer and competitor interests may also exist in the sales market (Absatzmarkt), constituting the relevant effects for conflicts determination.²⁰⁸

2 Problem: Economic Concepts and Legal Terminology

The case and its doctrinal handling reveal a general deficit of conflicts theory and practice. As alluded to earlier, this deficit is due to, among other things, the incongruity between economic concepts and legal terminology.²⁰⁹ Lawyers tend to directly “translate” economic concepts into legal terms. However, in doing so, they neglect to integrate these concepts into proper doctrinal structures. Legal arguments are often founded on the localization of a “market” without proper guidance as to how this market’s geographical scope should be determined. As we have also seen, modern statutory language has further replicated this unfortunate tendency in unfair competition conflicts. Under article 6(1) of Rome II, effects on the “competitive relations” or the “collective interests of consumers” are supposed to help determine the relevant point of attachment. But this proxy is a poor substitute. It is of particularly low utility in a transnational market setting, where consumer and competitor interests are nearly ubiquitous and highly elusive.²¹⁰

In order to avoid this untested and sweeping translation of economic concepts, a more detailed perspective is required. We have seen that consumer decision making is central to market functioning; and it is market information that provides the basis for the consumer’s decision making.²¹¹ If trademark and unfair competition law are considered to regulate the market’s information infrastructure, it is this infrastructure that must determine choice of the applicable law. This change of perspective brings a new understanding that the geographical extension of the market must be determined by the consumer’s transaction alternatives.²¹² As long as two options present themselves to the consumer as true alternatives, they are part of the same market.²¹³ In this light, the Gran Canaria scenario featured only one single marketplace—albeit multijurisdictionally extended. Formerly separate markets had “merged” in the course of increasing consumer mobility.

3 Analysis: The Chronology of Consumer Decision Making

The Kauf im Ausland decision and scholarly commentary in its wake have established a dichotomous understanding of what constitutes the “marketplace.” Only two relevant stages of market transacting appear to exist—accordingly, two separate market segments must be distinguished. One stage of transacting is the competitor’s “conduct.” This is deemed to determine the place where the consumers are affected—in other words, the locale of the advertising market (Werbemarkt). The other stage of transacting is performance or delivery. This is described as the sales or performance market (Absatzmarkt). But this model of market transacting is unduly curtailed. The sequence of marketing activities and transacting conduct is more complex. Different stages must be distinguished.

The Kauf im Ausland reasons start with a reference to the tort foundation of unfair competition law and traditional conflicts determination under the locus delicti rule.²¹⁴ On the basis of the Kindersaugflaschen doctrine,²¹⁵ the court explained that the place where the competitors’ interests collided would serve as the point of attachment.²¹⁶ For advertising activity, this place would have to be located within the marketplace in which an impact on the customers’ decision was intended.²¹⁷ Other factors, such as the customers’ nationality or place of residence, were deemed irrelevant. The court also explained that places where preparatory activities are undertaken or where damage to the victim-competitor occurs should be irrelevant.²¹⁸ Therefore, the applicable law would be based on the place where the specific activity at issue was intended to affect the customer, regardless of where a later transaction might occur.²¹⁹ Scholarly commentary has widely followed, looking at the impact on the consumer or on the other side of the market (Einwirkungsprinzip).²²⁰ Leading scholars have expressly described the “place of conduct” as determinative.²²¹ Even the most sophisticated analyses, looking at the consumer’s referee function, have limited the scope to the place where the conduct at issue actually affects the consumer.²²²

This understanding risks unnecessarily limiting the perspective on Kauf im Ausland scenarios, and it also partly misconceives the structure of unfair competition conflicts. First, it is important to note that the Bundesgerichtshof did not require the defendant’s conduct to serve as the point of attachment. Rather, the place where marketing or advertising activity was intended to affect the customer—and not the place where the defendant would actually act—was defined as the place where interests collided and, accordingly, served as the point of attachment.²²³ Actual conduct in marketing activity is thus only the starting point. As we have seen, it precedes the first stage of consumer decision making.²²⁴ Therefore, the place of “impact” on consumer decision making may well differ from the place of “conduct.” Internet advertising provides an evident example of such a divergence: there is uploading on one side (“conduct”) and accessing or downloading on the other (“impact”). Another example where this correlation is apparent can be found in cases where the consumer relocates after having contact with the marketing conduct at issue—but prior to transacting. This can be found, for instance, in advertising vis-à-vis commuters crossing a state border on their way to work and back home. The consumer-commuter is “impacted” at her workplace abroad but “transacts” at home.

And one more aspect is important. The Kauf im Ausland court seemed unperturbed when declaring that the sales market (Absatzmarkt) should be deemed irrelevant as a point of attachment if it is understood only as a place where the results of improper conduct come into existence.²²⁵ Here as well, more precision is required. Of course, the sales market is commonly understood as the place where the contract is performed through the delivery of goods or the performance of services. In laymen’s terms, this is usually the successful final stage of the contract. But the consumer’s decision to transact and the localization of its implementation into the marketplace must be distinguished;²²⁶ neither product delivery nor performance of services constitutes the completion of the transaction in economic terms. Those acts may complete the performance scheme under the parties’ contract. Yet with respect to the market mechanism, it is generally the conclusion of the contract that determines the final act in the consumer’s decision-making process. When and where performance will later actually take place is of secondary relevance at best.

4 Implementation: Alternative Transactions and the Merger of Markets

Of course, resolving cross-border unfair competition conflicts in light of the consumer decision-making model does not necessarily mean that the place of competitor conduct is irrelevant. The place of acting will provide for at least one possible point of attachment if the transaction is also concluded there. In Kauf im Ausland, therefore, under the assumption that actual or potential competition existed in Spain—in other words, that the German tourists could have also transacted with the defendant’s competitors in Gran Canaria—application of Spanish law would be part of a consistent solution.²²⁷ But the issue is not resolved with this conclusion.

Dominant scholarly commentary still widely agrees with the Bundesgerichtshof’s exclusive choice of the Spanish unfair competition regime. Josef Drexl, for instance, has defended the Kauf im Ausland holding under a perspective of regulatory sovereignty and with respect to the aim of upholding a level playing field of international competition.²²⁸ As he posits, the Spanish state’s regulatory interest is attached to the specific conduct within its national territory; neither consumer nationality nor residence will matter. In addition, he explains, unfair competition law aims to establish a par conditio concurrentium in the marketplace. Any extraterritorial extension of other states’ laws (in this case, German law) would therefore not only affect the Spanish state’s interest in an autonomous regulation of its own markets but also distort competition. In other words, if German law had been applied, German competitors would have been disadvantaged, for they would have had to comply with their stricter home regime. Spanish enterprises, by contrast, could act under the more liberal conditions of Spanish law, thereby bearing lower costs.²²⁹

I have already analyzed the underlying economics of a discriminatory application of stricter forum laws.²³⁰ Yet there is an additional aspect that must not be overlooked. The Kauf im Ausland case presented a truly multijurisdictional conflict. Of course, all lawmakers are interested in their local order of market communication—this necessarily requires a certain degree of regulation of local conduct. But this does not mean that any other state’s interest must be rejected ab initio. Even if we assume that there was competition in Spain, this does not invalidate German lawmakers’ concurrent interest in regulating their own local order of market communication and transacting. Exclusive application of Spanish law cannot be argued on the grounds that the tourists were targeted in Spanish territory only and that they had no option of transacting with competitors in Germany.²³¹ This perspective loses sight of the correlation between market extension and product properties.²³² It thus necessarily overlooks the fact that actual effects on consumer decision making and on the process of market transacting also existed in Germany. Among scholarly commentary, Jochen Glöckner provides one example of this approach. As he argues, it would be far-fetched to imply that the German tourists would have waited until their return to Germany to buy the products.²³³ This argument may be convincing for products that have a short consumption period (e.g., restaurant or entertainment services). A Gran Canarian bar will thus generally not compete with a Spanish restaurant in Munich. The situation changes, however, if the product at issue is not to be used or consumed within a short time. Duvets and pillows, as textiles in general, have a far longer life and consumption span. Potential buyers thus might well consider deferring a purchase for more than a few days and ultimately buying the product back home. And products of the kind offered by the defendant were actually also available in Germany. Hence, the tourists did have “German alternatives” to their purchase in Gran Canaria. Technically, the tourists as peripatetic consumers had extended the scope of their activities geographically, thereby bringing the market with them—literally in their luggage.

Moreover, it is also not a valid argument that competition in the “Spanish market” could be distorted by the application of German law to the defendant’s conduct. Of course, as we have seen, extraterritoriality of stricter national policies bears a risk of anticompetitive overregulation.²³⁴ But this problem must not lead to an ab initio rejection of the application of laws other than the local regime. On the contrary, if qualified and sufficiently intensive effects exist within another lawmaker’s territory, application of the respective regime is principally justified. As we have also seen, effects, not conduct, are what trigger the interest in regulating market communication and transacting. Per se, the mere occurrence of conduct in one territory does not provide for a prevalence of the local regulatory interest vis-à-vis other regimes’ concerns—even though it may be territorial effects only that can be found there.²³⁵ In sum, it is determinative that the marketplace in Kauf im Ausland extended across both jurisdictions. Accordingly, both Spanish and German unfair competition law should have been applied concurrently.²³⁶

A different question is whether the application of a certain regime—here, German law—should yield to a comity-based rule of jurisdictional self-restraint. As illustrated earlier, this is the last prong in choice-of-law analysis.²³⁷ In this regard, Drexl’s proposal to avoid anticompetitive distortion bears some value. Nonetheless, this argument does not require excluding the application of German law. On the contrary, the Kauf im Ausland facts called for an application of German law. Even though actual transaction numbers may have been small, the defendant found it worthwhile to conceive of a scheme of “market relocation.” In other words, it decided to selectively affect German consumers during their holidays in Spain. The aim was to circumvent stricter German laws on consumer protection. In this light, no legitimate interest on the side of Spanish lawmakers could be found to provide for “more leeway of commercial transacting” on their territory.²³⁸ Unlike Germany, Spain at that time was in default with its implementation of the European directive that provided for the relevant provision on consumer information and protection. Under the intent-based proxy rule of de minimis effects explained above,²³⁹ this suffices to overcome the threshold that is required in the interest of jurisdictional self-restraint.

5 Conclusion

The consumer’s decision making and the implementation of her decision by transacting constitute the cynosure of the market mechanism. The core of trademark and unfair competition policies aims at the regulation of market information infrastructure as the basis of the consumer’s decision-making process. Accordingly, the process of the consumer’s decision making must determine the place of conflicts attachment. Looking at the consumer’s alternatives to her transaction (or nontransaction) helps clarify and interpret concepts of the “marketplace” and the place of “interest collision” that are often too casually referred to in statutory provisions, case law, and commentary. At the same time, the concept of transaction alternatives also helps explain that there is no reasonable differentiation between an “advertising market” and a “sales market.” In Gran Canaria scenarios, the consumer has left her place of residence, but she has not transacted in a truly “foreign” market. She has extended the market’s geographical scope by her own mobility. Accordingly, more than one national regime applies.

Section 3 The Typology of Trademark and Unfair Competition Conflicts

I now have arrived at the final part of my inquiry—the presentation of a practical typology of trademark and unfair competition choice of law. Autonomy in consumer decision making and transacting is the ultimate subject matter of protection in trademark and unfair competition law. Choice of law requires an accordant conceptualization. This shifts the perspective away from traditional conduct-based and formality-founded methods, thus yielding, in many cases, results contrary to those proclaimed by conventional wisdom.

I Consumer Decision Making: Protecting the Market Information Infrastructure

Both in trademark conflicts and in unfair competition choice of law, the analysis will require giving regard to the effects on the market mechanism. With respect to the core policy of protecting consumer decision making and transacting, a “rule of alternatives” must be used to determine the applicable law.²⁴⁰

A The Common Core of Trademark and Unfair Competition Policies

1 Advertising Communication: A General “Rule of Alternatives”

Example (similar scenario to the Court of Justice’s L’Oréal v. Bellure case²⁴¹): Acme GmbH, incorporated in Germany, manufactures “smell alike” perfumes for sale on the British market. Acme GmbH’s products replicate the fragrances of numerous brand-name perfumes, including those of Blammo Perfumes SARL, a company incorporated in France that sells its “original” fragrances worldwide. Acme GmbH’s products are packaged in a way to avoid confusion with Blammo Perfumes SARL’s trademarks. Yet Acme GmbH advertises its products as “imitations” and “replicas” of the original brands.

The category of unfair competitive conduct that concerns market information and its transmission comprises, inter alia, deceptive and confusing advertising, including comparative advertising and hidden advertising. In terms of current US doctrine, section 1 of the Restatement of Unfair Competition (Third) explains these scenarios as practices relating to “deceptive marketing” and “infringement of trademarks and other indicia of identification.”²⁴² As shown in the analysis of the Gran Canaria scenario, an alleged violator-competitor’s conduct in the advertising market (Werbemarkt) does not necessarily serve as the point of attachment.²⁴³ What matters is the place of alternatives to the consumer’s transaction or nontransaction.²⁴⁴ These alternatives may be located in a territory other than the place of the violator-competitor’s conduct and the place of impact on the consumer.

Solution: In the example, the places of conduct (advertising), impact (perception of the advertising by consumers), and transacting (point of sale) are situated in the UK. The UK also seems to be the only place where alternatives to consumers’ transactions with Acme GmbH exist. Thus, British law applies.

At this point, it is important to note that this rule of alternatives under article 6(1) of Rome II will also apply to most scenarios within the specific categories of common law torts of passing off, as well as malicious falsehood and defamation in a business context.²⁴⁵ English scholarly commentary partly contends otherwise; yet the bulk of cases regarding passing off concern issues of actual or potential manipulation of consumer decision making (“misrepresentation”).²⁴⁶ Accordingly, in passing-off scenarios, the market mechanism is usually directly affected and there is no bilaterality under article 6(2) of Rome II.²⁴⁷

Finally, the category of unfair advertising communication may also include communication vis-à-vis the customer after concluding a contract whenever such communication is intended to affect decision making with respect to the existing contract and the consumer’s potential alternatives. Contrary to dominant scholarly commentary, it is also a rule of alternatives—not an attachment to the place of the consumer’s residence (where she sees, reads, or otherwise perceives the communication at issue)²⁴⁸—that must govern.

2 No Exceptions: Trademarks, Trade Names, Geographical Indications, and Designations of Origin

Example (modified scenario of Grupo Gigante SA De CV v. Dallo & Co., Inc.²⁴⁹): Acme GmbH, incorporated in Germany, is a grocery chain that intends to open an online shop under the name Titan Marché. The website (which uses, among others, the address “titan-marche.fr”) will offer online grocery sales and delivery services in many German cities, including those in the German-French border region. The brand name Titan Marché is well known for grocery services in Algeria. Blammo Groceries SARL owns rights to the service mark for Algeria but not for the European Union, Germany, or France, where, so far, no registration exists. A significant number of French citizens, however, are familiar with the Algerian grocery chain.

Existing doctrine on trademark conflicts and the rule of the lex loci protectionis need not be changed with respect to its territorial foundation. However, it should be amended by jettisoning the conduct/formality dichotomy. Under a reconceptualized effects principle for trademark conflicts, a plaintiff (i.e., right owner) will have to specify the applicable regime under which she believes her rights are infringed on.²⁵⁰ The court will then determine the existence and foreseeability of effects and will undertake a comity-based analysis of jurisdictional self-restraint in order to establish the admissible territorial scope of the relevant regime.

Solution: In order for Blammo Groceries SARL to enjoin Acme GmbH from using the service mark Titan Marché, it must find a cause of action under either German or French trademark or unfair competition law. Grocery services are local; hence, alternatives to consumers’ transactions with Acme GmbH—notably the potential use of the service mark by Blammo Groceries SARL—must be found in either Germany or France. Since the mark is not registered in either of these places, it is essential for Blammo Groceries SARL to prove the existence of an unregistered service mark, or—if no such right exists—to resort to enjoining Acme GmbH by means of an unfair competition claim. Considering that the potential customer base is located in France, the case will have to be decided under French law.

If a state’s lawmakers have implemented additional policies to prevent, for instance, preparatory activities, the approach may vary. One example of such concurrent policies is article 9(2) of the European Community’s trademark regulation, which prohibits conduct at the premarket level (e.g., affixing trademarks to goods or packaging).²⁵¹ In these scenarios, lawmakers have expressly detached infringing conduct from effects on consumer decision making. Accordingly, the place of the alleged infringer’s activity—regardless of its ultimate effects on the market mechanism—determines the applicable law.

Beyond the core of trademark rights protection, finally, there is some dispute concerning the treatment of trade names, geographical indications, and designations of origin. While the protection of trade names and work titles in Germany, for instance, has traditionally been an issue of unfair competition doctrine,²⁵² modern statutory law has implemented protection into express provisions of the Trademark Act.²⁵³ Not surprisingly, therefore, despite wide agreement that trade-name protection will not amount to full “rights” status, conflicts resolution has been based on the lex loci protectionis.²⁵⁴ Under Rome II, this requires application of the conflicts provision for intellectual property rights in article 8.²⁵⁵ Similarly, application of article 8 of Rome II has been suggested for the infringement of geographical indications.²⁵⁶ This approach has been supported by a number of arguments, among them the assertion that geographical indications are akin to intellectual property rights.²⁵⁷ As we have seen, however, these arguments are based on a misguided concept of the formality of rights.²⁵⁸ The trademark-as-property (or geographical-indication-as-property) paradigm must be rejected. Accordingly, with respect to conflicts resolution, a homogeneous treatment under the modernized marketplace effects principle is indicated. Categories of “conduct” and “property” are irrelevant. Instead, effects on consumer decision making and transacting are what determine the applicable law. Only if, however, the policy at issue goes beyond the protection of navigation goodwill (e.g., for famous geographical indications) will the conflicts rule have to accommodate different requirements for the protection of surplus goodwill. I will address this exception in more detail in an instant.²⁵⁹

B Implementation of Decision-Making Results: Transacting

1 The Core Policy

As my look at trademark and unfair competition policies has shown, the domain of unfair competition prevention goes beyond the rationale of trademark protection.²⁶⁰ In contrast to the two fields’ common core policy, the prevention of other acts of unfair competition protects the consumer’s decision-making process subsequent to the transmission of information. This notably concerns cases of unfair conduct that force the consumer into a transaction that she would not have made if she had been free to decide. The goal there, too, however, is to prevent unfree and presumptively unreasonable transacting. Among the numerous scenarios of such impact on stages following the transmission of marketplace information, undue psychological pressure may be the most common example.²⁶¹

Example (compare with no. 30 of annex I of the Unfair Commercial Practices Directive²⁶²): Acme Books, an English bookseller, sells postcards to German customers. These postcards are marketed as “handpainted” by the handicapped. In its advertising materials, Acme Books accentuates that if these cards are not sold successfully within the weeks to come, the painters’ livelihoods will be in jeopardy.

With regard to conflicts attachment, this case is no different from those where the transmission of market information is affected. The information transmitted here need not be incorrect; there may be no element of confusion involved. Ultimately, however, both scenarios concern improper impacts on consumer transacting. Unless the specific policy is combined with a concern for protecting the consumer as an individual—notably with respect to her individual rights—the place of alternatives will constitute the point of attachment. In the example case, therefore, German law will apply.

Finally, this rule also governs fact patterns where other policies intended to regulate commercial communication are at issue. One example (implying neither an improper transmission of information nor an undue influence on the decision-making process) is the prohibition against the bundling of commercial offers with a lottery. Such marketing methods are popular, notably in the form of sales that include an option to participate in a lottery by means of returning a part of the product packaging (e.g., the label). Not long ago, some European civil law regimes used to qualify such kind of advertising as improper manipulation of the consumer’s decision making.²⁶³ The underlying policies of such a prohibition may be deemed obsolete in light of the modern European consumer paradigm. Nonetheless, such a rule against bundling is intended to prevent “irrational” consumer decision making. Accordingly, choice-of-law determination will—contrary to what dominant commentary argues—be based on the place where the consumer’s alternatives to the underlying transaction exist.²⁶⁴ The place where potential or actual lottery prizes will be granted and transferred is also irrelevant for conflicts determination.

2 Policies Beyond

Under a strict separation of protection policies, the point of attachment will vary if a marketing method is deemed to constitute an additional wrong beyond the mere manipulation of consumer decision making. The subject matter of protection, then, is different: it is not the consumer-as-referee but the consumer-as-citizen.

Example: Acme Co., incorporated in England, is a marketing company that employs telephone operators and offers “advertising and marketing services” for special opportunities. For instance, one can hire Acme Co.’s personnel to call potential customers almost anywhere in Europe, including Germany.²⁶⁵

If a legislature prohibits cold calling or spamming for the sake of protecting consumers beyond their capacity as referees in competition, conflicts determination will differ from the alternative-transaction model. If marketing conduct is prohibited only to protect consumer privacy, traditional tort choice-of-law rules will govern. This would then be a case not of article 6 of Rome II but of the national regime’s autonomous choice of law.²⁶⁶ The practical problem is determining which policy ultimately prevails. Usually, a joint concern for protecting the consumer’s privacy and her role as referee in competition will drive the implementation of the relevant unfair competition norm. Two (or even more) different jurisdictions’ laws may then be eligible for choice of law: the one (or more) where the market mechanism is affected and the one where the consumer’s privacy is invaded. Choice-of-law attachment will have to function accordingly—that is, with a potential multitude of applicable regimes.

Solution: Depending on what the plaintiff asserts, a European court will either apply article 6 of the Rome II Regulation or resort to the national choice-of-law regime. If the claim is invasion of privacy in violation of article 13 of the ePrivacy Directive²⁶⁷ only, and there is no claim of effects on consumer decision making, the national forum’s choice of law will determine the applicable tort regime. If, however, the plaintiff also asserts a manipulation of the consumer’s decision-making process through cold calling, the court will also have to consider the regime where alternative transactions can be found.

II Theories of Misappropriation and Other Impact on Competition

Apart from their aim to ensure unmanipulated consumer decision making, trademark and unfair competition law are sometimes also founded on an alternative concept of preventing the misappropriation of a competitor’s position in the market. The thrust of such protection is quite different from the functional core of trademark and unfair competition policies. Rather than fostering information correctness and unhindered consumer decision making, misappropriation doctrine is aimed at preventing a moral wrong.²⁶⁸ I have already discussed the scholarly criticism. In the United States, antidilution, initial-interest, and postsale confusion theories, as well as the recognition of merchandising rights, have been cited as examples of such aberrations.²⁶⁹ Similarly, civil law concepts of unfair competition law preventing the misappropriation of a competitor’s goodwill (often discussed in the context of product imitation) have been characterized as improper outgrowths and overprotection. Finally, beyond both these domains of anticonfusion and antimisappropriation theories, additional policies of unfair competition prevention exist. Some are unique to European civil law. Most prominently debated are cases of antitrust and unfair competition concurrence and the so-called breach of statutory duties. As a closer look at these fact patterns reveals, conflicts resolution must be categorized in accordance with whether the substantive law policy at issue is one of protection of consumer decision making or whether it has a broader aim.

A Modern Extensions of Trademark-Infringement Theory

Modern extensions of trademark protection and the traditional theory of point-of-sale confusion prevention have one thing in common: they are founded on a general policy aimed at preventing improper consumer instrumentalization. Unlike in cases of confusion prevention, however, the instrumentalization at issue in extended protection theories is not necessarily achieved through the transmission of incorrect information. It is a different form of usurping the consumer’s mind.²⁷⁰ Accordingly, choice of law must be conceived of differently.

Example (variation of the Gran Canaria case):²⁷¹ Acme SL, incorporated in Spain, runs a supermarket chain on the Canary Islands. The symbol that it utilizes as a product designation for its own food products is similar to a well-known trademark registered and used in Germany by Best Decoration GmbH. Best Decoration GmbH, however, has registered the trademark, which it uses exclusively for jewelry; it has no registration in Spain or the European Union. Acme SARL makes intentional use of the designation in its advertising vis-à-vis German tourists on holiday in Gran Canaria.

Under the rule of the lex loci protectionis, the law of the protecting country decides on its own scope of protection and the extension of domestic trademark rights. While the traditional rule provided for a requirement of conduct within the country of protection,²⁷² modifications are indicated on the basis of marketplace effects comparable to the rule in unfair competition choice of law. However, the subject matter of protection varies. This particularly concerns the area of non-confusion-based infringement theories in trademark law. In a number of fact patterns governed by theories beyond confusion prevention, no competition exists. Use of the trademark by a second-comer will then not immediately affect the market mechanism with regard to the original brand’s product. Cases of trademark dilution provide one example where relevant effects will generally be found not in a collision of competitive interests but in the actual or potential deterioration of market information capital (goodwill).²⁷³ Accordingly, the point of attachment must be the place where the senior trademark owner’s market capitalization is in danger of being diminished. In most cases, this is the place of residence of the relevant consumer group for the original brand’s product.²⁷⁴ But it will not necessarily coincide with the place of the alleged infringer’s conduct.

Solution: In the example, conduct on Spanish territory can be found. However, the case must be seen primarily as a scenario of trademark-impairing effects to be found with regard to the right owner’s goodwill in Germany. The target group—as in the Gran Canaria case—is German consumers. Accordingly, at issue is the concern that tourist-consumers’ minds could be affected and that (in the long run) a deterioration of trademark distinctiveness with regard to the German public could take place. In principle, therefore, German trademark law is to be applied to the foreign-based dilutive conduct.²⁷⁵

Similarly, theories of preventing initial-interest confusion or bait-and-switch schemes require a differentiated perspective. Two variants must be distinguished.²⁷⁶ If lawmakers have established a concept of preventing initial-interest confusion that provides for protection without regard to the impact on the consumer’s decision making, a theory of genuine misappropriation prevention governs. We then apply the conflicts rule described above for antidilution prevention.

Example (modification of Brookfield Communications, Inc. v. West Coast Entertainment Corp.²⁷⁷): Plaintiff Brookfield Communications is an entertainment-industry information provider incorporated in the United States. It owns the US trademark registration MOVIEBUFF. Defendant West Coast GmbH is a German-based company and provides internet services, including downloads and streaming of movies and music. West Coast has included “moviebuff” and similar terms in the metatags of its movie-search websites, which are offered in both English and German.

Solution: Use of the MOVIEBUFF trademark in metatags may not create actual confusion among the visitors of West Coast’s websites, as long as these sites prominently display West Coast’s own trademarks.²⁷⁸ Under German law, however, unauthorized use of a competitor’s trademark in metatags may constitute an infringement without regard to actual confusion. Even such hidden use of the trademark may suffice if it is used to improperly redirect internet searches to the company’s own website.²⁷⁹ Yet if the plaintiff has no trademark rights in Germany, no claim exists. Under the rule of the lex loci protectionis, however, the decision maker must also look at other jurisdictions that could be affected (at least if indicated by the plaintiff). Following the rules on initial-interest confusion in US doctrine, use of the term “moviebuff” in metatags will divert potential customers to West Coast’s website. West Coast thereby “improperly benefits from the goodwill that Brookfield developed in its mark.”²⁸⁰ Considering this specific goodwill-centered policy in American trademark law—and disregarding the lack of actual confusion²⁸¹—relevant effects may be found in the United States. Accordingly, at least with respect to the requirement of “effects on US commerce,” Lanham Act subject-matter jurisdiction exists.²⁸²

On the contrary, if it is impact on the consumer’s decision-making process that is conceived of as the only element of impropriety (notably in terms of increasing search costs), a structurally different conflicts rule applies. This is usually the case when so-called reorientation or switching costs for the consumer are large.²⁸³ These scenarios must be treated under the general conflicts rule for the manipulation of market information.²⁸⁴

Example²⁸⁵: Acme GmbH, incorporated in Germany, runs food restaurants A-Burgers along the German and French highways. Along the French side of a highway, it erects a billboard that reads, “A-Burgers—next exit, 10 kilometers (across the French border)—everything 50% off.” When automobilists leave the highway and drive into the German countryside, however, they learn that the discount period is over. “The billboard,” as they are told by an employee, “has been outdated since last year—but it still boosts our sales. Isn’t that great!” Most of the misdirected customers, although frustrated by the sham, shy away from a new search for a burger place and ultimately eat at A-Burgers (and pay full prices).

In these cases, choice of law must take into account that the consumers’ decision making has been immediately affected with the initial confusion. The manipulation—that is, misinformation—may not have endured until the point of actual transaction. Yet the decision-making process has still been distorted by means of raising the costs of alternative transacting (i.e., the additional effort of searching anew). In this light, it is clear that the applicable law will have to be determined in accordance with the place(s) where alternative transactions existed.

Solution: The detour forced upon the consumer can lead to a transaction that is different from what the consumer would have decided without manipulation. The consumer may have been aware of the scam at the time of transacting, but reorientation or switching costs were too large to make it reasonable to start a new search. Depending on where the misdirected consumers had alternatives, German and/or French law applies.

Finally, for the category of postsale confusion, choice of law also requires a differentiation. In all subcategories of postsale confusion, the trademark owner’s goodwill as market information capital must be protected against deterioration.²⁸⁶ The place where a trademark owner’s market information capital exists and will (or may) be damaged determines the applicable law. Again, the place of conduct and the place of relevant effects can diverge. Even though the sale of an improperly branded product may occur abroad, its trademark-impairing effects can still occur, after the point of sale, in domestic territory.

Example (Steele v. Bulova Watch Co.²⁸⁷): The defendant manufactures and sells wristwatches in Mexico, where he also stamps the watches with the plaintiff’s trademark, Bulova. These watches are mostly purchased by American tourists and ultimately “filter” into Texas and other US states. There, the plaintiff’s contract dealers are confronted with customers’ complaints of low quality.

Solution: The defendant’s conduct—even though carried out exclusively on Mexican territory—creates the risk of deteriorating the plaintiff’s US-based goodwill. Hence, effects on trademark goodwill occur mainly in the United States. Accordingly, US trademark law applies.²⁸⁸

B Product Imitation

As we have seen, the consumer’s referee function does not necessarily stand at the center of protection with regard to the prevention of improper product imitation.²⁸⁹ Before I address the implications of substantive law policies, however, I must clarify one aspect. The scholarly debate sometimes centers on the question whether to apply an unfair competition conflicts rule or to follow the lex loci protectionis principle. While the practical results are largely the same, a difference exists at the doctrinal level. The prevention of unfair product imitation may be related to intellectual property rights protection, particularly to the protection of trademarks and design rights. Accordingly, despite the fact that no formally state-granted rights exist, it may not be too far-fetched to suggest that the lex loci protectionis should apply.²⁹⁰ The contrary position follows a characterization as unfair competition and accordingly contends that the law at the place of the sales market should govern.²⁹¹ This is the place where the original product and its imitation can be found to be in competition.²⁹² An effects-based approach helps avoid both approaches’ inherent defect of conduct foundation. The analysis differs with respect to the substantive law policy at issue. Two cases must be distinguished.

Example (variation of BGH Rillenkoffer case²⁹³): R-Bag GmbH manufactures exquisite luggage and travel accessories. Among its products is a pilot case made from aluminum. This case is well known for its outer appearance, notably the characteristic striation pattern. R-Bag GmbH sells the case in countries across Europe, through a network of exclusive dealers. It does not sell in Switzerland, though. I-Bag GmbH, a start-up incorporated in Switzerland, manufactures medium-priced luggage, which it sells in its only store, at the Zurich airport. For customers who cannot take the purchase with them, the store offers delivery abroad. When I-Bag GmbH starts to sell an almost identical replica of the original aluminum striation case, R-Bag GmbH sues in Germany.

With respect to cases of consumer deception and confusion, regardless of whether protecting consumer decision making is to be seen as the primary subject matter of protection, the plaintiff’s quasi IP right will be invaded by manipulation of the consumer’s mind. Here as well, in terms of the English passing-off doctrine, misrepresentation is the most significant element.²⁹⁴ Despite what is argued by dominant opinion, however, it is not the sales market of the imitation that matters.²⁹⁵ Instead, the place of alternatives—in other words, where the original has been or is being offered—will serve as the point of attachment. In the example case, even though the imitation is marketed in Switzerland, the court should also consider applying German or other European countries’ unfair competition laws since the sale of the imitation to travelers from abroad also affects the original’s European consumer base.

The issue appears similarly straightforward for cases of improper exploitation and impairment of a product’s reputation or goodwill.²⁹⁶ Scholarly commentary sweepingly characterizes cases of exploitation as an invasion of the competitor’s interests and, accordingly, calls for the law of the sales market of the imitation (Absatzmarkt) to be applied. This is deemed to be the place of confusion, reputational exploitation, and impact on the original producer’s sales.²⁹⁷ Even though, in practice, the “sales market” may often be where choice of law has to be attached, doctrinal intricacies of what is being protected make a more precise determination necessary. Most importantly, this scenario will often not entail an element of “competition” among the parties. As illustrated by, for instance, the German Rolex case and the US Ferrari doctrine, a junior party may be trying to benefit from utilizing the senior party’s brand or other indication of source. The junior party’s products, however, will not be seen as an alternative for the consumer.

Example (variation of BGH Rolex case²⁹⁸): A German supermarket chain sells wristwatches that imitate the design of a famous Swiss timepiece. The sales price of the “imitation” is only 1% of the Swiss original’s price. Since Swiss customers are increasingly frequenting the chain’s German supermarkets in the South of Germany, the original’s manufacturer is concerned about the potential spill of cheap imitations into Switzerland and the deterioration of its timepieces’ air of exclusivity and prestige.

For want of actual competition, this scenario implies a different determination of the connecting factor. The issue is not consumer confusion or deception; accordingly, there is no attachment to the place of alternative transactions. Instead, it is the situs of the plaintiff’s goodwill that matters for choice-of-law determination. This is the place where the original product is marketed—and while this can be the place where the imitation is offered, it not need be. As Rolex (similar to Ferrari in the United States) implies, the factor that prevails in cases where the original’s reputation and scarcity are to be protected is not transacting or transaction-related confusion; it is damage to market information capital in the sense of surplus goodwill. In the example case, therefore, Swiss law will apply.²⁹⁹

C The Antitrust Concurrence

Example (variation of BGH 20 Minuten Köln case³⁰⁰): Acme SARL, a large French publisher, offers a free weekly German-language magazine at all kiosks, petrol stations, and other sales locations in the border region with Germany. B-News GmbH, a local newspaper publisher in the city of Freiburg (near the French border), complains that the free offer by Acme SARL will eventually deteriorate its sales and ultimately push its newspaper “out of the market.”

For certain situations of competitive conduct, a concurrent application of unfair competition and antitrust law is debated. Examples include boycotting a competitor, selling below the cost of production, and market disruption and disturbance (Marktbehinderung and Marktstörung).³⁰¹ All of these scenarios bear a general risk of distorting competition.³⁰² While cases of boycotting may be understood to also target an individual competitor, in the other cases, no obvious individualized focus on competitive unfairness exists.³⁰³ In this light, application of a multilateral antitrust effects principle—as implemented in article 6(3) of Rome II—seems to suggest itself. According to scholarly commentary, under such a rule, national law should apply to effects within the national market, while foreign law should apply if a foreign market has been affected.³⁰⁴ Yet Peter Mankowski has criticized this approach. As he posits, a genuine marketplace effects rule will provide the same results. The specific policy of protecting fair competition as an institution (as implemented in antitrust law) should not disguise the fact that market disruption and distortion—when effectuated through unfair competitive conduct—would also (and in particular) affect competitors’ interests, notably through an intentional attack on their market positions.³⁰⁵ While the practical results under both approaches may not differ too much, structural differences exist. Again, looking at the core of unfair competition policies helps resolve the issue. What matters is the impact on the consumer’s decision-making process.

Solution: As the example of selling below production costs (or giving away for free) illustrates, such conduct will not lead to an unduly influenced consumer decision. After all, the consumer will execute an economically rational transaction, buying the cheapest or accepting a costless product. It is the subsequent effect on market conditions that lawmakers may intend to prevent. This, however, is a genuine antitrust concern. As such, it must also be treated under a genuine antitrust conflicts rule.

D Breach of Statutory Duties as Unfair Competition

Example (taken from BGH Rotpreis-Revolution³⁰⁶): Acme GmbH runs a retail store in Luxembourg, near the border with Germany. It advertises a “special sales deal” in the daily newspaper of Trier, a nearby German city where many of its customers come from. This sales deal will take place on a Sunday. Unlike in Luxembourg, in Germany stores are generally prohibited from opening on Sundays and public holidays. A breach of such shopping-hours regulations is qualified as unfair competition under section 3a of the German Unfair Competition Act.

As explained in my analysis of substantive law policies, the category of statutory breach is structurally foreign to the field of trademark and unfair competition law. It does not directly concern the market information infrastructure.³⁰⁷ Nonetheless, a market-oriented conflicts approach has been suggested. In scholarly literature, for instance, it is contended that cases of breach should be characterized in accordance with the respective interests affected. Hence, the market effects rule would apply whenever the interests of consumers or the public have been affected by a breach of a statutory duty.³⁰⁸ Similarly, the market where a competitor attains a competitive advantage (by the breach) has been deemed the relevant point of attachment.³⁰⁹

On the basis of a reconceptualized approach, a different focus is indicated. As demonstrated earlier, the prerequisite for unfair competition within this category is that a statutory norm be breached.³¹⁰ The breach per se, however, neither implies nor requires an impact on market information and consumer decision making; take, for instance, the bribery of a competitor’s employees. It is not effects on the market mechanism that determine whether a breach is found to be unfair; it is the breach as such that constitutes the requirement for unfairness. Consequently, the applicable regime will have to be determined by the plaintiff’s indication. She must plead that a cause of action exists under a certain law. Within this framework, the judge must first inquire whether the applicable law acknowledges a claim for breach of statutory duties at all, and—if this is the case—resolve the issue of an actual statutory breach.³¹¹ Depending on the norm that is claimed to have been breached, this may or may not require territorial conduct on behalf of the defendant.

Solution: The breach of a shopping-hours regulation may constitute the basis for an unfair competition claim under German law. A German-based competitor may thus claim unfair competitive conduct and indicate the application of German law. Yet Acme GmbH’s conduct in Luxembourg cannot breach the local regulation in Germany. Nor can the nonexistent breach trigger the application of an unfair competition cause of action under the German Unfair Competition Act.

III Competitor-Related and Bilateral Commercial Torts

If unfair competitive conduct directly targets a specific competitor, conflicts determination may be more complicated. In this case, individual rights—in addition to or instead of third-party consumer interests or the public—will be affected. This special category of unfair competition has been defined (and the European Commission has acknowledged its categorization for Rome II) as including defamation, bribery and corruption, theft and use of competitors’ trade secrets (industrial espionage), the improper approaching or enticing away of foreign staff, and the inducing of a breach of contract.³¹² Since conduct in these cases is targeted toward a specific competitor, the purpose of individual rights protection seems to suggest the application of the lex loci delicti commissi. Both the place of conduct and the seat of the victim-competitor have accordingly been identified as possible points of attachment.³¹³ Recourse to tort conflicts principles is also what the Rome II Regulation has implemented in its article 6(2): “Where an act of unfair competition affects exclusively the interests of a specific competitor, Article 4 shall apply.” As we have seen, however, practical problems exist with the terminology and the concept of bilateral situations in competition in general.³¹⁴ While some instances of competitive conduct may be easily categorized as either market related or primarily competitor related, certain cases require more extensive debate. To summarize: at the center of all problems is the fact that almost all conduct classified as “bilateral” can be said to concurrently also affect consumer interests and the public’s interest in unhindered competition. There is seldom such a thing as competitive conduct that has no effects on the market at all. Indeed, the notion of “competitive conduct” implies that there must be at least some contact with the market, as well as some effect—at least sooner or later. Concepts of competitor relatedness, market impact, or effects directness, therefore, are not helpful. What is required, again, is a look at the triangular structure of the market mechanism. Competition unravels between individuals on the basis of consumer decision making and transacting (or nontransacting). This is the domain of the marketplace effects rule. Bilateral torts, by contrast, do not target the consumer directly; there is no attempt to influence decision making or its implementation. A different sphere is thus being invaded—the victim-competitor’s assets. Of course, such an impact may also immediately affect the victim-competitor’s market activities—for example, by impinging on her capacity to compete (especially in cases of actual hindrance). The invasion, however, will not be undertaken by means of the market mechanism. Thus, the justification for a different attachment for bilateral situations is not the closer proximity to general torts or delicts, or the stronger focus on individual rights protection. What is amiss in these situations is an involvement of the market mechanism by means of a direct targeting of consumer decision making.

In this light, the examples cited in the European Commission’s explanation reveal a number of scenarios that require clarification:

Example: While on a business trip in India, Alice, an English producer of technical equipment, writes a letter to one of her customers in Germany. The letter is concerned primarily with a sales offer to the customer. However, it also contains a paragraph on Alice’s strongest competitor, the French producer Claire. Alice “discloses” to her customer that Claire is close to being declared bankrupt (which is actually not true).

Defamation, the spreading of malicious falsehoods, and calls for boycotting a market participant will usually directly affect market information transmission.³¹⁵ It is not only the defamed competitor’s or boycotted participant’s individual interest at stake but also the public’s interest in unhindered competition by availability of unmanipulated information.³¹⁶ These torts, therefore, are not isolated from the consumer’s decision making. Consequently, at the conflicts level, the marketplace principle governs.³¹⁷ Under a rule of alternatives, however, this must not correspond to the consumer’s residence or (more generally) the location of the other side of the market (Marktgegenseite).³¹⁸

Similarly, the enticing away of a competitor’s personnel or agents will immediately affect the market—at least with respect to the market for employees and employers. This is the only direct effect, but it suffices to preclude application of article 6(2) of Rome II.³¹⁹

Example (continued): On the same business trip, Alice visits a job fair at Bangalore University, where she has several job talks with local graduates. She learns that they have all signed contracts with her English competitor Best Ltd. to work in its Indian branch in Mumbai. Alice successfully offers several of these employees a “financial incentive” to breach their contract and start working for her Indian subsidiary instead.

Of course, the improper poaching of foreign employees will seldom directly affect markets at the next level of production.³²⁰ There may be reverberations on the product market over time, due to the improper enticing. Yet these are indirect effects. Effects on the labor market, by contrast, are direct. Whether poaching is accompanied by deception, bribery, or other forms of undue influence does not matter. In each case, the triangular structure of competitor-employee relations parallels the competitor-consumer relationship. In the example case, therefore, Indian unfair competition law, not the law of the victim-competitor’s seat, will apply. On the same basis, enticing a competitor’s customers to breach their contracts will affect the market.³²¹ It is thus the marketplace rule modified as a rule of alternatives that applies.

Example (continued): On the same business trip, Alice learns that a Chinese competitor is close to filing for patent protection with respect to a certain apparatus. Alice has tried to gain patent protection for the same mechanism for some time but has been unsuccessful. By accident, she gets to know a former employee of the Chinese competitor, who discloses the relevant technical information. Back in England, Alice immediately adjusts her production methods to the more cost-efficient innovation. She is thereby able to lower her prices.

The analysis differs with respect to theft of trade secrets. The theft as such may provide a competitive advantage to the thief since competitive superiority usually depends on innovation and know-how.³²² Over time, this advantage will affect the market: cheaper production due to savings in research and development may ultimately increase the thief’s market shares. However, this effect is not immediate—in particular, it does not affect consumer decision making.³²³ It is a tort on the premarket stage. None of the protective purposes related to market information and consumer decision making are affected. Consequently, traditional tort choice of law rules will determine the applicable regime.³²⁴