Melbourne Journal of International Law
In BBS Kraftfahrzeugtechnik AG v Racimex Japan KK; Jap Auto Products KK the Supreme Court of Japan had the opportunity to decide an important question concerning the parallel importation of patent-protected goods, which it did in an interesting way. It held that the rights of Japanese patent holders are not infringed by the importation into Japan of patent-protected goods placed on the international market by the holder of an equivalent patent in another country. This is provided that the holder(s) of the patents in the different markets concerned are the same legal entity or licensee(s) of the same entity. The case is of significance as it directly addresses the conflict between an intellectual property owner’s desire to maximise its profits by engineering variations in price for the same goods in different markets, and the desire of entrepreneurs to exploit those price differentials for their own gain. The Supreme Court considered the situation in which a patent holder sells goods in one country (in this case, Germany), and then seeks to block their importation into another country (Japan), on the basis of an implied contractual restriction between itself and the first buyer as to where the goods could be resold. The issue before the Court became one of the legal effect of the absence of knowledge of the geographical restriction on the ability of subsequent purchasers to resell patent-protected goods. This is an issue that, in one way or another, will eventually have to be addressed by the European Court of Justice (‘ECJ’), the most likely occasion being the long-awaited Davidoff case. In this regard, the conclusions reached by the Supreme Court of Japan are useful in illuminating the ‘Gordian knot’ of competing legal influences in the parallel importation problem.
BBS Kraftfahrzeugtechnik AG (‘BBS’), a German company, held patents in both Europe and Japan for a particular type of car wheel. It sold these goods in Germany and Japan through its subsidiaries Nihon BBS KK and Washi Mayer KK. There was a sufficient price differential between the same goods in Germany and Japan to enable an arbitrageur, Jap Auto Products KK (‘Jap Auto’), to buy the goods in Germany, export them to Japan, and resell them to Racimex Japan KK (‘Racimex’).
BBS brought an action in Japan against both Jap Auto and Racimex, claiming infringement of its Japanese patent. The premise of the claim was that, because BBS had not consented to the sale of those particular goods in Japan, their importation and sale was akin to that of pirated goods, even though the goods had been legitimately manufactured and sold in Germany under the relevant European patent. Jap Auto and Racimex responded by claiming that the voluntary first sale of the goods in Germany resulted in the international lapse of the effect of the patent. Accordingly, the sale by BBS in Germany of goods sold around the world under different patents in different countries should be considered to have exhausted BBS’s patent rights to those goods in every country in which BBS held an applicable patent. Therefore the respondents were not violating BBS’s patent rights in Japan and were free to export or import the goods which BBS had put on the German market.
After the initial trial of the infringement issue, the Tokyo District Court held that Japanese law recognises the legal principle of internal market exhaustion: that a voluntary sale of patent-protected goods in Japan exhausts the patent owner’s control over those goods in Japan. However, it also held that a foreign sale of the goods does not have the same effect. The District Court reasoned that BBS’s sale of the goods in Germany did not affect its patent-based control over them in Japan because BBS never consented to their export to, and sale in, Japan. The Court awarded BBS a royalty of seven per cent of Racimex’s sales of the patented products, an amount equal to approximately US$38 589.
The Tokyo High Court allowed an appeal by the defendants, holding that they did not infringe BBS’s Japanese patent. It reasoned that:
[I]t is beyond dispute that the plaintiff owns a patent in the Federal Republic of Germany for an invention which is identical to the invention in question and has lawfully put the products in circulation. Moreover, the plaintiff has already had the opportunity of securing compensation for the disclosure of the invention. Since there is no evidence in the case under review that a legal restraint exists concerning the opportunity of securing compensation, it is the understanding of this court that the patent right in question is exhausted by the aforementioned lawful putting into circulation of the patented products within the Federal Republic of Germany.
The High Court then rejected BBS’s claim for damages and injunctive relief against future, non-consensual importation and sale of its goods. It found, as a matter of policy, that the patent owner’s monopolistic right to control the resale of its goods must have been exhausted by the voluntary first sale of the goods. Otherwise, the High Court held, the development of industry and the expectations of commerce would be damaged. The High Court added one proviso to its adoption of a principle of international exhaustion, holding that if a patent owner’s ability to receive adequate compensation for the exploitation of its patent in the original market was in some way legally constrained (such as through government imposed price controls or compulsory licenses), then the sale of the goods under those conditions would not be a voluntary first sale for the purposes of the international exhaustion principle. BBS subsequently appealed to the Supreme Court of Japan.
The Supreme Court’s first concern was whether the Convention for the Protection of Industrial Property (‘Paris Convention’) dictates the adoption of an inflexible rule as to the international exhaustion of patent holders’ rights. Accordingly, it focused on the first two sections of art 4bis of the Paris Convention, which provide:
1 Patents applied for in the various countries of the Union by persons entitled to the benefits of [nationals of countries of] the Union shall be independent of patents obtained for the same invention in other countries, whether members of the Union or not.
2 This provision is to be understood in an unrestricted sense, in particular, in the sense that patents applied for during the period of priority are independent, both as regards the grounds for invalidation and for forfeiture, and as regards their normal duration.
From these sections of the Paris Convention the Court concluded:
The provision denies the interdependency of patents and affirms that patent rights as regards their formation, amendments and forfeiture are interdependent of each other in each country. The existence of a patent right does not depend on patent rights being declared null and void in another country, or on the duration of the patent right in such country. It is understood that all this does not reciprocally affect patent rights. Having established this, the above provision [of the Paris Convention] does not concern the exercise of patent rights by the patentee.
The Court summarised its interpretation of the Paris Convention as follows:
[T]he principle of territoriality with regard to patents means that patent rights in each country are subject to domestic legal provisions insofar as establishment, transfer, and validity are concerned. It means that the patent rights only have effect within the geographical limits of such country.
The Supreme Court then held that it was not, in the present case, constrained by the Paris Convention:
It is only to be determined under the domestic patent law as to what constitutes a use of the patent right, and in how far exercise of a patent right can also relate to goods the patentee has circulated outside Japan. This question bears no relation to the principle of territoriality under the Paris Convention, for which reason a violation of the provision of art 4(2) Paris Convention cannot be successfully claimed in the present case.
The Supreme Court then went on to deal directly with the issue of enforcement of a patent holder’s rights after a voluntary first sale of the goods which embody the patented invention. The Court noted that s 68 of the Patent Act 1957 (Japan) creates the sole right to work or license a patent, and ‘“working” means the act of manufacturing, using, assigning, leasing, importing, or offering for assignment or lease of the product.’
The Supreme Court considered the situation where a patent holder or its licensee puts the patented goods on the market. The Court noted that a subsequent transfer to, or commercial use of, the goods by a third party in Japan, without the permission of the patent holder, would appear to violate the Patent Act 1957 (Japan). However, the Court concluded that
if patented products are sold domestically, either by the patentee or with his consent, the patent is deemed exhausted because it has fulfilled its purpose. The patent does not give rights to subsequent use of the patented product by acts of transfer or lease.
In essence, the Supreme Court was pointing out the generally accepted principle relating to the internal market exhaustion of goods protected by intellectual property rights: if the intellectual property owner agrees to sell the protected subject matter, the owner’s control over those goods ceases to exist. This has the effect of exhausting the controls based upon intellectual property rights, making them unenforceable within the territory which gave the intellectual property protection in the first place. The Court justified upholding this principle as follows:
First, patent law has to be understood as balancing the interests of invention protection, and the public benefit of society at large. Next, if a tangible object is transferred, the rights are obtained by the transferee, and the transferee obtains those rights that were originally vested in the transferor. Also insofar as patented products are distributed on the market, the transferee obtains an object from the patentee whose exercise of the right suggests that the right in further acts of re-sale has been transferred as well. If with respect to any acts of marketing patented products, the patentee’s consent were necessary each time a transfer occurs, the free movement of goods will be seriously impeded, the smooth distribution of patented products would be hampered and as a result, the interests of the patentee himself would suffer. This would run counter to the purpose of the Patent Act ‘to encourage inventions by promoting their protection and utilisation so as to contribute to the development of industry’ … Finally, by making the invention available to the public, the patentee will have the opportunity to obtain the reward from selling the product or granting a licence for the use of the patent and thereby obtain a licensing fee. In order to protect the financial interests of the patentee who has made his invention public, it would not seem necessary to give the patentee or the licensee rights beyond the first act of marketing, as the patentee would then obtain an unnecessary double reward in the course of further distribution.
Here the Supreme Court of Japan sought to reconcile three basic and often conflicting aspects of patent law. In doing so, the Court began by stating that the first aspect, the ‘benefit of society at large’, must be considered when trying to reconcile the second aspect, a patent owner’s desire to control the marketing of its patent-protected goods. These must be considered with the third, namely the needs of both the subsequent purchaser and the overall market. The Supreme Court recognised that both a subsequent purchaser and the national market must have some assurance that the patent-protected goods are readily transferable once the goods have been placed voluntarily on that market by the patent holder or its licensee.
The issue is essentially one of privity of contract. The owner of intellectual property can sell the goods to whomever it chooses subject to whatever terms it chooses. Often one of those sale terms is a restriction on where the goods may be resold. In an effort to limit the marketability of the goods to a particular area, the intellectual property owner might even condition the first sale of the protected goods on the requirement that its first buyer make the same territorial restriction applicable to any subsequent sale of the goods. In doing so, it seeks to maintain an advantageous price differential.
Problems arise when, through neglect, misunderstanding or deception, the goods are sold on to subsequent purchasers who have no knowledge of the original resale restriction. The subsequent purchaser could acquire the goods thinking that there are no geographical restrictions on where it could resell them, (apart from those imposed by the applicable customs laws), and, finding a more profitable market for the goods, sell them in a location that was prohibited under the initial agreement between the intellectual property owner and the first buyer. In this scenario, the second buyer did not have a contract with the intellectual property owner and did not know of any restriction imposed by the intellectual property owner as to whether the distribution of the goods in question could be resold.
This represents a clear conflict between the intellectual property owner’s contractual expectancy of a geographical limit to the distribution of the goods and the subsequent purchaser’s expectations of being free to transfer the goods wherever and to whomever it likes. The Supreme Court of Japan resolved this issue in the following manner:
[I]f the patentee reserves his rights at the date of transfer with respect to the use in Japan when selling the patented products abroad, at the time of the transfer the patentee has agreed with the transferee that sale or use of the product should not be allowed in Japan. If clearly indicated on the products, such restriction is also valid against subsequent purchasers of the patented product along the distribution chain even with a number of intermediaries. Here it is understood that the above products have been sold under certain restrictive conditions, and any purchaser is free if he wants to obtain products bearing such a limitation or not.
The Court then went on to hold that BBS could not enforce its Japanese patent against the parallel importer because it could not show that the respondents had knowledge of a geographical restriction on where they could resell the goods:
The product was sold by the plaintiff and patentee in the Federal Republic of Germany. When the plaintiff sold the products, no limitation was imposed on the purchaser about any restrictions of sale or use, or even if so, the plaintiff has offered no proof of this product. For this reason, with regard to the products, the patentee is not entitled to claim an injunction or damages under his patent right.
The obvious way to define the market is to base the market’s boundaries on the extent of the relevant intellectual property protection where the goods are sold. A patent owner’s control over goods protected by a German patent is exhausted in Germany when the goods are voluntarily put on the German market by the patent holder or its licensee. Under this definition, the Japanese patent owner’s ability to exclude patented goods from Japan that were originally sold in Germany would not be affected, because Japan was not the market to which the exhaustion principle would apply. The German patent owner could have given its consent for the goods to be resold anywhere, or only in geographically limited areas. This would be a case of expressed consent for the resale of the goods. However, to require the express consent of the intellectual property owner is a disguised restriction on parallel importation and the free movement of goods within the international market.
An even more interesting question, and one that for the purposes of the efficiency of international trade needs answering, is when will the consent to import and resell goods protected by intellectual property be implied. The Supreme Court of Japan seems to say that such consent is to be implied unless there is clear evidence that the consent was expressly withheld. There are obvious questions arising from this position. To whom must the expressed withholding of consent be clear? Is it to those who have a contractual relationship with the intellectual property owner? Or, alternatively, is it to subsequent purchasers who are knowledgeable in the type of goods being traded, to whom knowledge of the likelihood of a lack of consent might be imputed (a ‘reasonable merchant’ standard)? Or does it require knowledge on the part of the public at large (a ‘reasonable person’ standard), whereby the average consumer would be expected to know that consent for a resale of the goods had either been given or withheld?
Each construction has numerous problems ranging from the procedural to the practical. If the standard is based on contractual relationships, then a procedural problem arises for the intellectual property owner and subsequent purchasers: how do they enforce or discover contractual terms that they may not have control over? Consequently, the problem arises of how to seek damages or indemnity from successors or the ‘predecessors in ownership’ of the underlying goods.
Similarly, the ‘reasonable merchant’ standard complicates every transaction by requiring the merchant to question the marketability of the goods. They would have to inquire of the intellectual property owner, or seek assurance from its seller, whether the goods are freely resellable in the markets in question. The ‘reasonable person’ standard is the least workable of the three as it reduces every parallel importation case to a subjective finding of fact: what would a reasonable person have known about any limitations placed on the marketability of the particular goods involved in the transaction being considered?
By way of comparison, in Roussel Uclaf v Hockley a company received a licence to work a process patent in the People’s Republic of China for the formulation of insecticide. Roussel Uclaf SA (‘Roussel’), the patent holder, supplied the Chinese company with the chemical resulting from the patented process. The Chinese company resold the chemical on the international market. When the defendant, Hockley International Ltd (‘Hockley’), came into possession of a batch of that chemical in the United Kingdom, Roussel brought an infringement suit. It claimed that it had only given permission for the goods in question to be used in China and that it had expressly forbidden their re-export.
On Roussel’s motion for summary judgment, Hockley responded by saying that it had no notice of any such geographical restriction, either from the Chinese company or from Roussel. It stated that it was in possession of drums of a chemical manufactured by Roussel that did not bear any label indicating the claimed resale restriction.
In denying Roussel’s motion, the English Patents Court held:
It is the law that where the patentee supplies his product and at the time of the supply informs the person supplied (normally via the contract) that there are limitations as to what may be done with the product supplied then, provided those terms are brought home first to the person originally supplied and, second, to subsequent dealers in the product, no licence to carry out or do any act outside the terms of the licence runs with the goods.
The Court took this further and stated:
If no limited licence is imposed on them at the time of the first supply no amount of notice thereafter either to the original supplyee … or persons who derive title from him can turn the general licence into a limited licence.
Because Roussel could not prove that it had limited the licence granted to the Chinese, the Patents Court found that the resale restrictions claimed by the patent holder did not apply. Furthermore, even if Roussel had only granted the Chinese a limited licence, the limitation was not communicated to subsequent purchasers such as Hockley:
Limited licences require that notice be brought to the attention of every person down the chain. That is normally done by putting notice of the licence on the goods, but it can of course … be done in other ways. Once the goods are sold without a limited licence then the purchaser buys them free of any patent restriction.
The Court noted that while notice is usually given by affixing a statement of any applicable resale restriction directly to the product, ‘[i]t is of course not necessary that the label be on the product provided the message is brought home somehow.’
A second definition of ‘the market’ for the purposes of the exhaustion principle (one that would be favoured by BBS and most other intellectual property owners) is that the market in which the goods may freely circulate is the area in which the intellectual property owner has expressly given consent for the goods to be resold. This definition would allow patent holders or their licensees to sell goods to buyers in several countries that were initially protected under one set of national patent laws. Alternatively, they could authorise the resale of the goods in several countries without jeopardising their patent-based control over those same goods in all other countries, giving patent holders an implicit veto over the free circulation of the goods.
The Supreme Court of Japan used another definition, which was less obvious but more economically realistic:
[S]ome thought should be given to the relationship between the free movement of goods in international trade and the interest of the patentee. In our present day society, international trade and economy affect us very broadly and permit conditions of rapid development. Even in the case that goods are purchased abroad, imported into Japan and put into circulation in the domestic market, there is the need to create conditions for the free distribution of goods, including their importation. Even if economic transactions have been made abroad, as a general principle, the transferee obtains not only the object as such, but also the rights vested therein. In other words, the transferor transfers his rights vested therein. To enable such transactions and to set the conditions for international trade in modern society, it is assumed that the patentee who has transferred the ownership of patented goods abroad has also endowed the transferee or any subsequent purchaser with the right to undertake further transactions with third parties, including the importation to Japan, use in Japan, and transfer of ownership on our domestic market.
This definition of the market considers the realities of international trade. It allows for the exhaustion of a patent holder’s rights in one country (in this case, Japan), when the underlying goods have been put on the international market by the patent holder or with its consent through a sale in another country. The Court noted that international traders should be able to expect, unless notified to the contrary, that goods are free to move between countries unfettered by patent holders. In addition, patent holders should foresee that their goods will move around the world once put into circulation.
One point which is unclear from the wording of the decision is whether the principle formulated by the Supreme Court is limited to cases where the Japanese patent holder is itself selling the same goods in Japan. In the present case, BBS, through its subsidiaries, was selling its German manufactured goods in Japan. It may be that the Court intended its holding also apply to a Japanese patent holder who is not trading the goods in Japan.
A variation on this issue was addressed by the ECJ in Sebago Inc v G-B Unic SA. G-B Unic obtained a quantity of Sebago’s trade mark protected shoes, which had been parallel imported into the European Economic Area (‘EEA’) from North America. At trial, G-B Unic was unable to prove where all of the shoes in its possession came from. It argued that Sebago sold identical shoes in the EEA, and that if the brand owner consented to some of the shoes being sold there, then G-B Unic was reasonable in assuming that it had consented to all legitimate copies of its shoes being sold in the market. The ECJ held that consent by a trade mark owner to the marketing of some goods inside of the EEA, and some goods outside of the EEA, does not exhaust the trade mark owner’s right to object to the importation of goods which were originally sold outside of the market. This means, in effect, that a retailer in the EEA must be prepared to show that it has the consent of the intellectual property owner concerned for the sale in the EEA of each example of a product protected by intellectual property rights.
Another important question is how closely the Japanese patent holder and the foreign patent holder have to be related. In this case, BBS held both the Japanese and the German patents for the same invention. The Supreme Court did state that it would not make a distinction between the Japanese and foreign patent holders based on a parent–subsidiary relationship:
In the case where the marketing activities abroad have been undertaken by an affiliated company, a subsidiary or a person with the same standing as the patentee, the case should be treated as if the patentee himself had marketed the patented products.
The same effect would be reached where the patent holder in Japan and the patent holder abroad have a licensor–licensee relationship.
Though not clearly stated in the Court’s judgment, the non-infringement rule employed in the present case may not be applied where the Japanese patent holder and the foreign patent holder are entirely independent. The closest the Court came to acknowledging this was to note that the considerations detailed above in Part III(B)
are not applicable if the owner of a Japanese patent markets the patented products abroad. For one thing, in the country where the products have been marketed, the patentee may not even hold a parallel patent that corresponds to the one in Japan. But even so, the domestic patent right and the parallel patent right in the country of marketing have to be regarded as separate rights. It cannot be argued that although selling the products under a parallel patent would give double reward, it may qualify as an act of use under the domestic patent right.
The Supreme Court went on to address the practice of intellectual property owners using national patents so as to segregate the international market. The Court perceived this practice as giving intellectual property owners the ability to create price differences in order to achieve increased profits by controlling the flow of their goods after the first sale. This was the case where the difference in prices charged by the plaintiff for its aluminium wheels in Germany and Japan was enough to make it worthwhile for the defendants to buy the wheels in Germany, ship them to Japan, and resell them. There is an element of realism in this practice. Intellectual property owners recognise that not all markets are the same. Consumers in one market are often more willing, or more able, to pay a higher price than those in another market. While the profit margin may be more lucrative for the intellectual property owner in the higher-priced market, there is still a profit to be made in the lower-priced market. Otherwise, the intellectual property owner would not work its patent or exploit its other intellectual property rights there.
The main issue here is the legal effect to be given to assignments of intellectual property rights that isolate markets from one another. In the European Union, the ECJ addressed this question in the context of eliminating intellectual property-based internal market divisions by formulating the ‘common origin’ doctrine in the case of Van Zuylen Frères v Hag AG. According to this principle, when independent owners of equivalent intellectual property rights in different countries could trace the origins of those rights back to a common predecessor, the parallel importation of goods protected under the laws of one country would be allowed into another country in competition with goods protected there by that country’s equivalent intellectual property laws.
The common origin doctrine was later modified by the ECJ in SA CNL-Sucal NV v Hag GF AG, and eventually rejected in IHT Internationale Heiztechnik GmbH v Ideal-Standard GmbH. The ECJ held that a where a trade mark owned in two or more countries by the one company is voluntarily transferred, such that the common market is divided, a parallel importer from the country of the assignee is not entitled to compete against the assignor or its successors in other countries. The Court’s decision was based upon the function of trade marks as indicators of the origin of the underlying goods.
The BBS v Racimex decision is unclear as to the extent to which the Court would allow this same manipulation, as described above, to be carried out through agreements between independent patent holders in different countries, so as to prevent the resale of their goods in each other’s country. Such agreements enable each to charge a higher price in their home markets through the lack of competition from the other. The most likely instance of this is through an assignment or licence of one national patent by the holder of that patent and another equivalent patent in a different country. Here, the inventor would be able to retain its most profitable market and exploit the second market through its assignee or licensee. The decision in BBS v Racimex considers this situation; the Supreme Court acknowledged that a patent licensor receives compensation for its inventive effort. Therefore, in the absence of an expressed intention, the licensor is not entitled to additional compensation in the form of inter-market control over subsequent sales of the patented goods.
The same situation as described above could occur in other ways, such as where the patent in one country is given as security for a loan that enabled the original patent holder (through a subsidiary or licensee) to exploit a secondary market. Upon default, the lien holder could sell the patent right in the secondary market as partial liquidation of the debt. In doing so, they transfer the patent rights in the secondary market on to an entity that has no direct connection with the original patent holder. Under such circumstances, the Supreme Court’s decision does not give guidance as to whether the original, primary market patent holder would be able to object to the parallel importation of patented goods by the lien beneficiary or its successor in interest from the secondary market.
Similarly, it is possible for a patent holder or a subsidiary in one country to file for bankruptcy protection, with its national patents in that country becoming part of its bankruptcy estate. Upon liquidation, those patents could be sold to a party that is unrelated to the patent holder. This would result in the patent buyer competing with the patent holder in another market. If the estate were to be put under the control of a receiver, then, in seeking to get the best possible price for the bankrupt’s goods, the bankruptcy receiver would be compelled to compete with the patent holder in the most lucrative markets, parallel importing goods from the market where the patent holder was bankrupt into markets where the patent holder was still solvent. The most interesting problem, from a parallel importation point of view, arises where the patent holder in several national markets files for bankruptcy. This would place the bankruptcy trustee in the difficult position of allowing parallel importation from one part of the bankrupt enterprise in competition with other parts in other markets. This would maximise the first patent’s value while working to the detriment of the other patents in the more lucrative markets, or prohibiting the same parallel importation for the opposite purpose.
At the far end of this spectrum of factually distinct circumstances is the case where separate patents arise simultaneously for the same invention in different markets. This is a very rare occurrence, due to the priority rules that apply to patents, but not an impossible one, especially in highly competitive fields of research. There are also other, even more unusual possibilities brought about through events beyond the patent holder’s control.
Also absent from the decision is any discussion of the reimportation into Japan of goods originally sold there and exported under an agreement between the Japanese patent holder and the first buyer that the goods were not to be sold in Japan. The question remains whether the decision in this case would allow an arbitrageur to purchase the goods abroad and bring them back into Japan for resale.
In BBS v Racimex the Supreme Court of Japan found itself free under provisions of the Paris Convention to lay down a rule of non-infringement of patents by the importation of goods originally sold outside of the country, where there is no notice given to subsequent purchasers of any import restriction. This stops short of imposing a rule of international exhaustion of patent rights under Japanese law. The Court held that the Japanese patent could still be relied upon as a bar to imports of patented goods if adequate notice of any applicable import ban is given to subsequent purchasers, although what constitutes adequate notice remains in question. The Court decided this in response to the perceived need in international trade for the free movement of patent-protected goods. Moreover, international traders have a need for assurance that the importation into Japan of patent-protected goods would not be blocked by the Japanese patent holder who has placed the goods on a foreign market under a foreign patent. The Court recognised that patent holders can agree with their first buyers that the goods are not to be sold in Japan. However, it also noted that it would be practically impossible for subsequent purchasers of the goods to know of such a restriction short of a notice of the geographical limitation on resale being affixed directly to the goods themselves. The Court held that once BBS had voluntarily put the goods onto the market, it could not enforce its Japanese patent against an importer.
Many of these same issues are currently being explored elsewhere in the world, particularly in Europe in Davidoff. The questions before the ECJ in that case will require it to determine, like the Supreme Court of Japan did here, whether an intellectual property owner that puts protected goods into circulation on the world market may object to their further commercialisation in the intellectual property owner’s home country. Furthermore, Davidoff will have to address the issue of what sort of notice will be effective in restricting the actions of subsequent purchasers of the goods.
Additional questions are left unaddressed by the decision of the Supreme Court of Japan. The Court did not expand upon the various combinations of possible relationships between a Japanese patent holder and its foreign counterparts. Furthermore, it did not address the situation where goods protected by patents are first put on the market in Japan, exported, and then reimported over the objection of the Japanese patent holder. The future resolution of these issues will require a careful balancing of the interests of intellectual property owners, antitrust laws, and the drive towards free trade at a global level.
[*] Case No H-7 (O) 1988, dated 1 July 1997.
 Case No H-7 (O) 1988, dated 1 July 1997. An English translation of this judgment has been published at (1998) 29 International Review of Industrial Property and Copyright Law 331 (‘BBS v Racimex Translation’).
 See Nanao Naoko, Koyama Takahiro and Sudo Hiromi, ‘Decisions of Parallel Imports of Patented Goods’ (1996) 36 IDEA: The Journal of Law and Technology 567.
 This conflict causes recurring legal battles between intellectual property owners and arbitrageurs all over the world: see, eg, Quality King Distributors, Inc v L’anza Research International, Inc,  USSC 25; 523 US 135 (1998) (‘L’anza’) and the cases cited therein for the various attempts at resolving the conflict (or avoiding it) in North America. The European version is emerging in Zino Davidoff SA v A & G Imports Ltd  Ch 127 (‘Davidoff’), currently on referral to the European Court of Justice from the English High Court of Justice. For a general discussion of the importance and political sensitivity of the parallel importation problem, see W R Cornish, Intellectual Property: Patents, Copyright, Trademarks and Allied Rights (4th ed, 1999) 47–9; Warwick Rothnie, Parallel Imports (1993).
 See, eg, Silhouette International Schmied GmbH & Co KG v Hartlauer Handelsgesellschaft mbH  ECR I-4799 (‘Silhouette’). Here, the ECJ enforced trade mark protection against a parallel importer who had no knowledge of the importation restriction agreed to by the trade mark owner and its first buyer. For a discussion of the implications of Silhouette, see W R Cornish, ‘Trade Marks: Portcullis for the EEA?’ (1998) 20 European Intellectual Property Review 172; Thomas Hays and Peter Hansen, ‘Silhouette is Not the Proper Case upon Which to Decide the Parallel Importation Question’ (1998) 20 European Intellectual Property Review 277.
 BBS v Racimex Translation, above n 1, 331.
 The implication of this argument is that a voluntary sale in a national market of goods protected by a national patent activates the emerging principle of international law that the freedom of international trade takes precedence over national intellectual property protection. It also states that goods protected by intellectual property rights, once freely put on a national market, can be considered as having been freely put on the international market, regardless of how many different patents the patent holder may have for the same invention in different countries: see Frederick Abbott, ‘First Report (Final) to the Committee on International Trade Law of the International Law Association on the Subject of Parallel Importation’ (1998) 1 Journal of International Economic Law 607.
 BBS Kraftfahrzeugtechnik AG v Racimex Japan KK; Jap Auto Products KK, Case No Hei 4 (wa) 16565 (Tokyo District Court, 22 July 1994). An English translation of this trial judgment has been published at (1996) 27 International Review of Industrial Property and Copyright Law 546. For a thorough discussion of this case at the trial and intermediate appellate levels, and its historical background in Japanese law, see Shusaku Yamamoto, ‘A Reversal of Fortune for Patentees and Parallel Importers in Japan: The Aluminium Wheel Case’ (1995) 17 European Intellectual Property Review 341.
 BBS Kraftfahrzeugtechnik AG v Racimex Japan KK; Jap Auto Products KK (1996) 27 International Review of Industrial Property and Copyright Law 546, 547.
 Ibid 548.
 Based on an exchange rate of US$1 = ¥121.45, current as at 1 May 2001.
 Racimex Japan KK; Jap Auto Products KK v BBS Kraftfahrzeugtechnik AG, Case No Hei 6 (ne) 3272 (Tokyo High Court, 23 March 1995). An English translation of this intermediate appellate judgment has been published at (1996) 27 International Review of Industrial Property and Copyright Law 550.
 Racimex Japan KK; Jap Auto Products KK v BBS Kraftfahrzeugtechnik AG (1996) 27 International Review of Industrial Property and Copyright Law 550, 554.
 Ibid 552–3.
 Ibid 552.
 Ibid 554.
 Opened for signature 20 March 1883,  ATS 6 (entered into force 6 July 1884). As at 15 April 2001, there were 162 states parties.
 BBS v Racimex Translation, above n 1, 333.
 Ibid 332.
 Paris Convention, above n 17, art 4bis(1), (2).
 BBS v Racimex Translation, above n 1, 333.
 Ibid. The ‘principle of territoriality’ is an important element of the Paris Convention. It stands for the proposition that national laws on the creation, existence and lapse or expiration of a patent only have a legal effect within the territory of the particular nation involved, and do not affect an equivalent patent elsewhere: see Cornish, Intellectual Property, above n 3, 247. Thus, the costs of international patent litigation are astronomically high due to the fact that multinational infringement cases — particularly where the validity of the patent is challenged by a defendant — have to be re-litigated separately in each country: see Friedrich-Karl Beier, ‘Territoriality of Trademark Law and International Trade’ (1970) 1 International Review of Industrial Property and Copyright Law 48.
 BBS v Racimex Translation, above n 1, 333.
 Ibid, citing the Patent Act 1957 (Japan) s 2(3)(i).
 BBS v Racimex Translation, above n 1, 333. BBS argued that the unauthorised resale in Japan of its European patent protected products was a violation of its Japanese patent.
 See generally Cornish, Intellectual Property, above n 3, 253–5, and the citations given therein.
 BBS v Racimex, above n 1, 333–4.
 Ibid 333.
 This issue is addressed in detail in Thomas Hays, ‘The Burden of Proof in Parallel Importation Cases’ (2000) 22 European Intellectual Property Review 353.
 For a discussion of this problem in a European context, see Thomas Hays, ‘The Silhouette Case: The European Union Moves to the Highest Common Denominator on the Gray Market Question’ (1998) 88 The Trademark Reporter 234, 247–8.
 There are two limitations on this generality, namely the free market economic pressures that would require the intellectual property owner to limit its demands on its first buyer in order to receive a better initial price, and the relevant competition or antitrust laws in force in jurisdictions affected by the transaction.
 See, eg, the facts of Silhouette  ECR I-4799.
 This is the intellectual property equivalent of the Uniform Commercial Code’s definition of a bona fide purchaser for value without knowledge: American Law Institute and the National Conference of Commissioners on Uniform State Laws, Uniform Commercial Code (1952), arts 8-302, 9-309 <http://www.law.cornell.edu:80/ucc/ucc.table.html> at 11 May 2001 (copy on file with author).
 BBS v Racimex Translation, above n 1, 334. The Supreme Court of Japan recognised that the patent owner could have imprinted ‘Not for sale in Japan’ on its products. This solution to the parallel importation problem has been suggested elsewhere, particularly in the context of trade mark protected goods: see Hays, ‘The Burden of Proof in Parallel Importation Cases’, above n 30, 357. In that article it was argued that the simplest and surest way of giving notice to subsequent purchasers of any territorial resale restriction, is to print the restrictions onto the goods in the same manner and in the same medium as the relevant trade mark. Thus if a brand owner did not want a bottle of perfume sold in a particular market, it should affix a notice on the label of the bottle saying, for example, ‘Not for Sale in the European Economic Area’. Commending this argument is the fact that it would give consumers knowledge of any restrictions affecting their purchase. However, the scheme has the disadvantage of leading consumers to think that the resale restriction implies that the goods, because they are being excluded from another market, are somehow inferior.
 BBS v Racimex Translation, above n 1, 335.
 See below n 39.
 BBS v Racimex Translation, above n 1, 334.
 In a typical case, the intellectual property owner sells goods to its first buyer with a geographical limitation as a condition of the sale. Another condition of the sale is that the first buyer extend the same conditions to all secondary purchasers and that they similarly extend the conditions to all subsequent purchasers. At some point in the chain of ownership it can be expected that the limitation will not be carried forward and that the expressed limitation will be broken. Can it then be said that an end-purchaser should have known of resale restrictions of which it had no direct knowledge? See Roussel Uclaf SA v Hockley International Ltd  RPC 441, 444–5 (‘Roussel Uclaf v Hockley’); Christopher Heath, ‘Continuity and Development in Japanese Intellectual Property Rights’ (1995) 26 International Review of Industrial Property and Copyright Law 900, 902–3.
 Such a construction might create a new form of business insurance. Rather than take the time to determine whether intellectual property protected goods were freely tradeable, a purchaser might prefer to take out a parallel importation assurance policy, which would protect them from loss in the event that they were sued by the intellectual property owner or a subsequent purchaser.
 The defence in every parallel importation case might become: ‘I reasonably did not know of any withheld consent’. Presumably the burden of proof would be on the intellectual property owner to prove that the reseller had acted unreasonably. This would create, in effect, a negligence standard for intellectual property infringement.
  RPC 441, 443.
 Ibid 444.
 Ibid 443.
 Ibid 444.
 Ibid 445.
 Ibid 444.
 This was, to some degree, the definition adopted by Advocate-General Jacobs in Silhouette  ECR I-4799, I-4810. Silhouette had authorised the export of its trade marked eyeglass frames to Bulgaria and the states of the former Soviet Union. Silhouette successfully argued (although the ECJ’s final decision was based on more technical issues of European Community law) that because it did not give permission for the reimportation of the frames into the European Economic Area, its trade mark rights were not exhausted there. The ECJ held that a principle of the international exhaustion of a trade mark proprietor’s rights, as was recognised under Austrian law, was inconsistent with the free movement of goods within the common European market. It so held because the overriding goal of integrating the market required that all states parties to the European Economic Area act together to prevent the formation of internal barriers to the free movement of goods based on national intellectual property laws. A similar argument in the case of copyright protected goods failed in L’anza USSC 25; , 523 US 135, 152 (1998), where goods that were intended for resale in Malta made their way back into the United States. The Supreme Court of the United States held in L’anza that, even where a territorial restriction was one of the terms of sale, the voluntary first sale of the goods exhausted the copyright owner’s control over those goods.
 BBS v Racimex Translation, above n 1, 334. For a discussion of and arguments for this definition of ‘the market’ for intellectual property law purposes, see Hays and Hansen, ‘Silhouette is Not the Proper Case’, above n 4, 281–6.
 This definition of the market, for the purposes of exhaustion of intellectual property rights, is based upon what could be called the ‘John Donne Principle’ (‘No man is an island, entire of itself’: John Donne, Devotions upon Emergent Occasions (first published 1624, Anthony Raspa (ed) (1975)) Meditation 17, 87. International trade has become so prevalent that individual national markets have become interdependent to the point that a sale of goods in one market affects the others.
 BBS v Racimex, above n 1, 333–4.
 But see below n 71 and accompanying text.
  2 CMLR 1317.
 Ibid 1321.
 Ibid 1330–1.
 Ibid. This issue has recently arisen in the United Kingdom in Davidoff  Ch 127,
134–5. A & G Imports argued that Davidoff had not sufficiently proven (or at least not for the purposes of Davidoff’s motion for summary judgment) that the bottles of scent that A & G Imports was selling had come from outside of the EEA. Laddie J disposed of the issue by noting that Davidoff was alleging infringement and that it was overwhelmingly likely that at least some of the bottles had come from outside the market. He then ruled that the case could go forward in order to determine whether there had been an infringement of Davidoff’s trade mark in a qualitative sense, leaving the quantity of any infringement for a later stage in the proceeding.
 BBS v Racimex Translation, above n 1, 335.
 Ibid 333–4.
 Ibid 334.
 This point is discussed by the High Court in Davidoff  Ch 127, 132.
 It must be said that this motivation on the part of intellectual property owners is not the only reason that parallel importation occurs. Quite often there are circumstances beyond the control of the intellectual property owner that give rise to international arbitrage in its goods. These circumstances include currency fluctuations and favourable tariff conditions. However, the General Agreement on Tariffs and Trade, opened for signature 30 October 1947, 55 UNTS 187 (entered into force 1 January 1948) (‘GATT’) and its successor agreements have done much to eliminate the latter. The GATT has reduced the average world tariff on manufactured goods from 40 per cent in 1947 to a present level of five per cent, making parallel importing more profitable on average, but eliminating the tariff-driven incentive: Michael Trebilcock and Robert Howse, The Regulation of International Trade (2nd ed, 1999) 21.
  ECR 731 (‘Hag I’).
 Ibid. See also BBS v Racimex Translation, above n 1, 334.
  EUECJ C-10/89;  ECR I-3711 (‘Hag II’). The Hag I and Hag II coffee cases were factually unique. Though the ‘Hag’ trade mark in different countries had a common origin, the splitting of the ownership of the mark was involuntary, the Benelux version having been sequestered by the Allied Nations after World War II: see Hag I  ECR 731.
  EUECJ C-9/93;  ECR I-2789 (‘Ideal-Standard’).
 Ibid. The effect of Ideal-Standard — allowing trade marks to segregate the common market along national lines — is nullified in most instances by the free movement of goods provision of the Treaty Establishing the European Community, opened for signature 25 March 1957, 298 UNTS 11, 37 ILM 56, art 30 (entered into force 1 January 1958); as amended by the Treaty of Amsterdam Amending the Treaty on European Union, the Treaties Establishing the European Communities and Certain Related Acts, opened for signature 2 October 1997, (1997) OJ C340 (entered into force 1 May 1999).
 BBS v Racimex Translation, above n 1, 333–4.
 This country could be called the secondary market of the patent holder, in so far as it is not a market that the patent holder intends to exploit directly. It is one in which it has patent rights (or other intellectual property rights) and chooses to sell or license those rights to another. The holder thereby receives the benefits of the sale or licensing agreement, in exchange for its consent to the forfeiture of control over the exploitation of the national patent(s) in the market concerned. Such agreements are sometimes accompanied by ‘non-compete clauses’. Under such clauses, the patent holder in the primary market promises not to allow the export of goods from the primary market that would compete with the same goods being sold by the licensee or assignee, in return for the compensation paid by the licensee or assignee in the secondary market. Often these agreements contain provisions whereby the assignee or licensee promises not to allow the export of the goods protected by the patent from the secondary market into the original patent holder’s primary market. Whether such agreements are enforceable against subsequent purchasers of the underlying goods, be they patent, trade mark or copyright protected, is one of the key questions posed by international intellectual property law.
 Many pharmaceutical and computer software firms, in particular, have limited assets apart from their intellectual property portfolios. These portfolios, being a large part of their corporate value, are often subject to the liens imposed by lenders.
 This is not a particularly unrealistic scenario, considering the present financial situation in Asia, and the terms imposed by the International Monetary Fund upon some Asian governments and upon banks in those countries, many of which have loaned millions of dollars to companies with significant assets tied up in intellectual property rights. It is therefore entirely plausible that Asian patents will eventually come into the hands of someone other than the original patent holders: see, eg, Christopher Heath, ‘All Her Troubles Seemed So Far Away: EU v Japan before the WTO’ (1996) 12 European Intellectual Property Review 677. Whether, under the particular national laws involved, intellectual property rights are subject to valid security interests is another question entirely: see, eg, Uniform Commercial Code, above n 34, art 9. Based on the language used by the Supreme Court of Japan (BBS v Racimex Translation, above n 1, 334), the argument could be made that the sale of the secondary market patent by the lien holder results in compensation to the original patent holder/debtor-in-default. Such competition results by way of the reduction of the original patent holder’s debt to its creditor and, therefore, the original patent holder should not be allowed to object to the parallel importation of patented goods originating in the secondary market.
 As a general rule, a bankruptcy trustee or receiver of a licensee takes the assets of the bankruptcy estate subject to whatever contractual terms were in place before the bankruptcy filing. There are, however, provisions under many national bankruptcy codes that enable a trustee or receiver to void contractual terms that significantly diminish the value of the estate and the ability of creditors to recover from its assets. This would be particularly true where the bankrupt was a subsidiary of the patent holder who imposed restrictions on the export of goods protected by patents.
 This was the problem faced by the ECJ in the Hag I and Hag II coffee cases, where the ‘Hag’ trade mark in Belgium was confiscated at the end of the World War II by the Allied Nations and was assigned by them to an entity that had no relationship to the owner of the identical, German trade mark: Hag I  ECR 731; Hag II  EUECJ C-10/89;  ECR I-3711; Warwick Rothnie, ‘Hag II: Putting the Common Origin Doctrine to Sleep’ (1991) 13 European Intellectual Property Review 24. While these cases were based on the ‘Hag’ trade mark, they could just as easily have been based on a sequestered patent.
 Like Silhouette before it, Davidoff is a reimportation case, where the goods originated within the EEA, and were exported subject to an agreement that the goods were to be sold abroad and, by implication, not to be reimported. Like the Supreme Court of Japan, the ECJ has not addressed the finer differences in parallel importation fact patterns. Such differences include those between reimportation cases and direct parallel importation cases, where the goods are not reimported but begin outside of the relevant market. See also John Tessensohn, ‘Japan Speeds Up Patent Reform’ (1999) 114 Patent World 25.
 These were the facts in Silhouette  ECR I-4799, and L’anza USSC 25; , 523 US 135 (1998).
 It is an old question. In Betts v Willmott (1871) 6 LR Ch App 239, 245, Lord Hatherley LC held that where the British patent holder places goods on the international market, it could block their importation into Britain only if it could prove that the importation of the goods into Britain was unauthorised. Different rules applied where the patented goods were imported into Britain from abroad by a foreign licensee of the British patent holder: see, eg, SA des Manufactures de Glaces v Tilghman’s Patent Sand Blast Co (1884) 25 Ch D 1, 9. It is interesting that the Supreme Court of the United States’ decision in L’anza USSC 25; , 523 US 135 (1998), covers similar ground to the British cases without citing them.
[†] BA, JD (Arkansas), LLM (Exeter), PhD candidate (Cantab); Lecturer, School of Law, University of Aberdeen.