A rights holder can control the initial marketing in the European Economic Area (EEA) of goods bearing its trademark. Consequently, it can prohibit the use of its trademark in relation to goods bearing that mark which have been placed on the market in the EEA without its consent.
In such unauthorised parallel import cases, the European Court of Justice (ECJ) judgment in Van Doren v Lifestyle(1) is an important reference for determining the burden of proof. The ECJ held that a rule of evidence according to which exhaustion of the trademark right constitutes a plea in defence for a third party against which the rights holder brings an action – so that the existence of the conditions for such exhaustion must as a rule be proved by the third party that relies on it – is consistent with EU law. In other words, the third party will bear the burden of proving that the particular goods bearing the trademark have been put on the market in the EEA by or with the consent of the rights holder. However, the principle of the free movement of goods can reverse the burden of proof – for example, if the third party can establish that there is a real risk of partitioning of national markets by the rights holder. In that instance, the rights holder must establish that the goods were initially placed on the market outside the EEA by it or with its consent. If such evidence is produced, the third party must prove that the rights holder consented to subsequent marketing of the products in the EEA.
On September 22 2015 the 's-Hertogenbosch Court of Appeal delivered an interlocutory judgment on assessing the real risk of market partitioning, in an unauthorised parallel import case between Converse Inc and Dutch company Aspo concerning the sale of Converse-branded shoes.
In an earlier interlocutory judgment, the 's-Hertogenbosch Court of Appeal had agreed to hear Aspo's claim that there was a real risk of market partitioning by Converse if it could provide evidence regarding the source of authorised parallel imports. Before weighing the evidence provided by Aspo, the court considered whether the assessment on the risk of partitioning should take place ex nunc (ie, not retroactively) or ex tunc (ie, retroactively).
The court agreed with Converse's argument that the rationale for reversing the burden of proof implies that the assessment must take place ex nunc. It held that the reversal is not a punishment for past events; nor is it intended to protect the sources of goods. The reversal is appropriate only if it is necessary to prevent the rights holder – by using information that it would otherwise not acquire – from partitioning the markets and risking price differences between member states. The court, pointing out that the ECJ had considered in Van Doren whether providing evidence could create a risk of market partitioning, held that – by the very nature of the case – this was a risk assessment that should take place ex nunc.
However, the court acknowledged that, in practice, it will be difficult to prove that the rights holder has been acting in a manner that created a risk of market partitioning at the time that the evidence is provided. It therefore accepted that indications implying that the rights holder had acted in such a manner in recent years are a sufficient indication of the current risk of market partitioning, especially if the behaviour fits a pattern or has a certain degree of continuity to it.
Thus, the court ruled that Aspo must provide evidence that Converse had – in the past five years (approximately) – organised or applied its distribution system in such a manner that created a real risk of market partitioning. The court commented that as it is difficult to obtain solid proof of market partitioning, an estimation must be made – based on the indications and circumstances of the case – as to whether a real risk of partitioning exists.
During the assessment of evidence brought forward by Aspo, the court stressed that the fundamental question is whether the rights holder has maintained or applied a distribution system that creates a real risk of market partitioning. According to the court, its decision must therefore be based on decisions and statements of the rights holder itself. However, the court stated that "it cannot be ruled out that under circumstances" evidence could also be derived from statements of licensees.
The court concluded that Aspo had not successfully provided evidence of a real risk of market partitioning. The various evidence put forward by Aspo included a witness statement, emails from licensees implying that they had refused to deliver outside their territories and French appellate court judgments which considered that there was a real risk that Converse had partitioned the market. The court also considered counter-evidence provided by Converse. The court determined that the emails from licensees to third parties in which licensees refused to ship converse goods outside their territory were not sufficient evidence because it was unclear whether the refusals were instigated by Converse as the licensor.
The judgment is an interesting insight into what the 's-Hertogenbosch Court of Appeal deems important when assessing a real risk of market partitioning by a rights holder. In light of the ECJ's reasoning in Van Doren, that the third party must provide evidence, the real risk of partitioning should arguably be assessed ex tunc (ie, at the time the evidence is provided). Surprisingly, the court held that the purpose of a reversal of the burden of proof is not to protect the sources of goods; whereas the ECJ in Van Doren explicitly mentioned obstruction of sources by the rights holder as one of the risks associated with market partitioning.
Acknowledgement of the difficulties in providing such evidence prompted the court to consider that such evidence may also relate to behaviour in recent years. What qualifies as 'recent years' remains unknown; in this case, the court specified a period of approximately five years. However, this could result in rights holders deliberately delaying or prolonging proceedings to ensure that certain events fall outside the scope of 'recent years'. The period taken into account should date back to when the goods in question were obtained by the third party or the rights holder initiated proceedings. After all, in that period, the third party may have already had issues obtaining a paper trail to the source of the goods because of reluctance in the supply chain to share information. Further, if at the time proceedings were initiated the rights holder then partitioned the markets, it could be argued that it cannot invoke its trademark rights (ie, abuse of right).
The court set a high standard for what evidence constitutes proof of the risk of market partitioning. Essentially, the court held that the evidence must be based on acts or communications of the rights holder itself. In this case, the court seemed to attribute little weight to statements from licensees, stating "that it cannot be ruled out that under circumstances evidence could also be derived from statements of licensees". Evidence that is derived from statements or acts of licensees must evidently be sufficiently connected to acts or statements of the rights holder.
The court's judgment suggests that evidence from licensees is not considered sufficient evidence. In practice, it will be hard to collect evidence directly from the rights holder. One could reason that the acts or policies of the rights holder will often be visible only through the acts of its licensees. If there is a real risk of market partitioning, the licensees may be reluctant to provide information implying market partitioning because of their dependency on the rights holder. If and when a licensee does provide information, there will be strong arguments that such evidence should qualify as sufficient evidence that is significant to the overall assessment.
Rights holders will welcome the 's-Hertogenbosch Court of Appeal's judgment because of the high standard it sets for what evidence qualifies as proof of a real risk of market partitioning. Conversely, third parties (ie, parallel importers) are faced with a burden of proof that was already difficult, but will be even more difficult if evidence regarding licensees is not considered sufficient evidence.
It seems that the court lost sight of the fact that the free movement of goods and prohibition of quantitative restrictions are cornerstones of the Treaty on the Functioning of the European Union.(2) While Article 36 of the treaty allows derogations from the fundamental principle of the free movement of goods in order to protect industrial and commercial property, such derogations are – according to established ECJ case law – permitted only to the extent to which they are justified by the fact that they safeguard the rights which constitute the specific subject matter of that property. One of the permitted derogations is the exclusive right of the rights holder to control the initial marketing in the EEA of goods bearing its trademark. Article 36 of the treaty also prescribes that such prohibitions or restrictions may not constitute a disguised restriction on trade between member states.
By setting high standards for the evidence required from a parallel importer, the court appears to perceive the exercise of IP rights as a starting point on which the principle of free movement of goods allows derogation, not vice versa. In light of the free movement of goods being the fundamental principle – and the exercise of IP rights being an allowable derogation to the minimum extent – it would be reasonable to give the third party the benefit of the doubt when there are serious indications of market partitioning. If the third party successfully provides evidence that there was a real risk of market partitioning in the past, which was caused by acts of the rights holder, it would be reasonable to shift the burden of proof so that the rights holder must prove that the goods were initially placed on the market outside the EEA by it or with its consent (as was ruled in Van Doren). Moreover, the rights holder would have to bear the burden of proof that it did not partition the markets or risk losing its ability to invoke its IP rights relating to the goods at stake.
For further information on this topic please contact Peter Claassen or Bram Woltering at AKD by telephone (+31 88 253 50 00) or email (firstname.lastname@example.org or email@example.com). The AKD website can be accessed at www.akd.nl.
This article was first published by the International Law Office, a premium online legal update service for major companies and law firms worldwide. Register for a free subscription.