Case C-217/13, Oberbank – seeing trade marked red in financial affairs [judgment 2014, ECLI:EU:C:2014:2012]

Red is in the trade marks of Coca-Cola, McDonalds, and Marlboro. In 2007, a German bank registered a particular type of red as its trade mark. This registration was challenged by two other banks new to the German financial services market. Austria’s Oberbank and Spain’s Bank Santander both use the colour red in their corporate identity. So can the German bank keep its mark by claiming that through ‘use’ its red has acquired distinctiveness? And if so, how can it prove this?

Facts
In 2007, the German patent and trade mark office registered an abstract colour mark for a type of red known as ‘HKS 13′. The mark was registered for Class 36 (financial services). The right holder was the German bank ‘Deutscher Sparkassen- und Giroverband e.V’.

In 2008, the Austrian bank ‘Oberbank AG’ challenged the mark’s registration and it asked the German patent and trade mark office to annul the registration. In support of its request, it raised three main arguments. First, it was simply illogical to allow an abstract contourless mark to serve as an indication of origin – services were intangible and they were not linked to any particular colour.

Second, in the financial services sector red had no intrinsically distinctive characteristic. Red was used by a number of companies. Further, and as a matter of principle, colours ought to belong to the public domain. Red was the advertiser’s colour of choice. In any event, the strict legal conditions governing whether a colour was capable of indicating origin were not satisfied in this instance – there was neither a ‘very specific market’, nor were there just a few competitors recognisable by ‘their colour’.

Third, and most importantly, the Austrian bank anticipated the German bank’s ‘defence’ of ‘red might not itself be distinctive but this one had acquired a distinctive character through use’. No, said the Austrian bank, there was a real need to keep the colour red free for use. And in any event, the Austrian bank remarked, the German bank’s studies had failed to show a sufficient link between the colour and the German company. For whereas the German bank’s studies had shown there was a 68% ‘degree of association’, the Austrian bank submitted that there were errors in the study. Moreover, given the importance of what was at stake the Austrian bank submitted that the law should really require almost a 100% ‘degree of association’.

The Austrian bank’s submissions were refuted by the German bank. In the latter’s view, abstract colour marks were recognised in case law. Red HKS 13 possessed the necessary distinctive character in the retail banking sector. Competitors in this sector of the market were identifiable by colour – for example, there was a ‘Dresdner bank’ green, and a ‘Deutschen bank’ blue. Through use by the Deutscher Sparkassen- und Giroverband, this particular shade of red had acquired a distinctive character.

The German bank added that its consumer surveys were fair and established that a distinctive character had been acquired by use of this red. The German bank had used Red HKS 13 as part of its corporate house style from the 1960s and 1970s – for example, the colour was used on its customers’ savings books. The colour had been depicted in the bank’s advertising as far back as 1940. And the German bank had spent a considerable sum of money on the use of this red in its advertising. Indeed, it had spent a far greater proportion of its money on advertising when compared to its competitors.

On hearing the submissions of the Austrian and German banks, the German patent and trade mark office decided in April 2012 that the German bank’s mark lacked distinctiveness. However, it accepted the reports and documents supplied by the German bank. It also took no objection to the way in which the questions had been asked. And it held that the percentage rates relating to the degree of association should be regarded as sufficient.

The decision was appealed up to the German Federal Patent Court. The Austrian bank again repeated its submission that a degree of association far in excess of 50% ought to be required. Red was an effective colour for advertising purposes.

At the German Federal Patent Court
To determine whether the particular colour of red had acquired a distinctive character through use, the German court repeated the guidance of the ECJ expressed in Case C-108/97, Windsurfing Chiemsee Produktions- und Vertriebs GmbH (WSC). The factors which a court had to take into account included not only the market share of the mark but also the intensity, geographic distribution and how long-standing the use of the mark has been, and the scale of expenditure devoted to advertising this mark. Market research could also be taken into account.

German case law and commentary on consumer surveys made the point that whether a mark has acquired a distinctive character through use depended on how many interviewees recognised the sign as an indication of origin (see for example, Pflüger, GRUR [2004] at 652). For those interviewees who did recognise the sign, the key question was then what is the name of the company referred to by the sign? According to the majority of writers, the answer to that question provided the requisite ‘degree of association’ to apply Article 8(3) of the German Law on Trade Marks [Markengesetz].

Applying those factors here, it did not seem particularly likely to the German Federal Patent Court that an abstract colour mark could be understood as an indication of origin for financial services. These services were not as a rule linked with particular objects recognisable by consumers. Financial services by their nature were complex and offered with the necessary supplementary oral and written information. A consumer was not going to decide to invest in a product or take out a loan without sufficient information – and their decision would not depend on colour alone. Therefore, it was highly likely that the consumer in this area would be particularly concerned by the written information rather than understand an abstract colour as an indication of origin. The colour may be understood to be an element of design, which can be used by any supplier.

Equally, the amount of money spent on advertising using the HKS 13 colour red was irrelevant. The investment does not relate to the use of the colour red in isolation but was almost always used in combination with the word ‘Sparkasse’, or ‘Sparkassen S’ – that is to say, together with another sign that refers to those banks. The amount invested provided no answer to the question of whether the savings banks concerned had been successful in establishing the degree of association between the colour alone and their services.

Nevertheless, the results of the consumer study were of decisive importance given that those studies only concerned the colour HKS 13 and not the combination of the colour with another sign. In so far as the consumer surveys themselves were concerned, the various studies generated divergent results. The differences arose because of the questions asked. The CJEU in its ‘chocolate mouse’ judgment of Case C-96/11, August Storck KG v OHIM had indicated that the mark had to make an immediate and certain identification of origin with the claimants’ products or services. To the Federal German Patent Court, consumer surveys did not seem to be the appropriate way of ascertaining this – not too many questions could be asked in an interview, and interviews could not take too long. Rather what was required was an analysis of those surveys which have already been conducted. In this instance, although the surveys contained useful information when they were taken as a whole, the surveys needed to be understood as meaning that even if more than 60% of consumers recognised the colour as an indication of a specific company, then doubts still remained as to whether there was the requisite immediate and certain indication rather than just a vague indication.

Such doubts were not be removed even where the majority of consumers recognised the colour as an indication of the origin of the specific services of a particular company, or rather they recognised the colour was only an indication of a company and thereby necessarily a package of services offered by that company. In the latter situation, although the colour of the mark may well constitute a company sign, the colour itself was most probably not a mark because the function of a mark is to indicate the origin of specific, separate services. A fresh survey would be unlikely to remove these doubts. Accordingly, a degree of association in excess of 60% was probably insufficient to alleviate these concerns and a much higher degree of association was most likely required.

In conclusion, extremely convincing evidence was required to overcome these doubts and arrive at a finding that the contested contourless abstract colour mark had achieved a degree of association for consumers sufficient to indicate the origin of the financial services. A degree of association of 50% was insufficient – it should be at least two-thirds, namely at least 70%.

The German Federal Patent Court then turned to the issue of evidence. This matter, it believed, was not just a matter of national law. Admittedly, the CJEU had held that whether a sign had a distinctive character could be determined by consumer research conducted under the rules of national law. Yet, in this case, the question was not confined to just the nature and way in which consumer research should be conducted and evaluated. The only guaranteed way of ensuring a uniform standard for annulling marks was if it fell within the scope of the Directive. And on that point, there was no case law from the CJEU as to the requisite degree of association for allowing an abstract colour to be said to have acquired a distinctive character. Equally, there was no case law from the CJEU surrounding how high these percentages ought to be and which arguments were decisive.

The German Federal Patent Court also wondered about nullity depending on whether the sign had acquired distinctive character through use at the time of application. Thus, if the relevant time was taken to be the date of registration (July 2007), then the degree of association had not achieved the rate of 70%. However, if the relevant time was taken to be the application filing date (February 2002), then in this case there was a problem – there was a gap of almost three and a half years from the moment when the mark was applied for and the moment when the consumer studies were conducted. In that period, considerable changes could also have occurred particularly since the banking sector had been liberalised. There were also new entrants to the German market which used a similar colour red. Consumer perceptions could have changed. As such, it could not be concluded that at the time when application was made that the mark had not acquired the necessary degree of association.

In that context, the German Federal Patent Court noted that the EU CTM regulation set down a procedure for annulling a trade mark. However, according to the case law of the ECJ the time of the decision to register does not determine the matter but rather attention is directed to when the application was filed and whether there were grounds to refuse the mark. That approach might be applied by analogy to the EU’s trade mark directive.

The final issue raised in this reference of the German Federal Patent Court concerned the burden of proof. Again, the German court believed that the CJEU enjoyed competence in this area for what was at stake was interpreting the Directive in light of and together with Treaty law. For whereas those rules relating to the risk of confusion and the concomitant rules on the burden of proof may well be a matter of national civil procedure, this did not mean that the rules relating to the burden of proof were exclusively a matter for the member states.

The rules of evidence do not belong to the rules of civil procedure governed by the Directive in so far as they concern the nature and adducing of evidence. National rules of civil procedure can indicate which party bears the burden of proof, which types of evidence are admissible, the form of the evidence, and whether the judge can stipulate the point in time up until which evidence may be adduced. These types of rules related to the nature of procedure and were not governed by the Directive.

However, this is quite separate to the question as to which party was responsible for ensuring that a specific factual matrix was clarified. The portent of this question extended beyond the mere modalities of the procedure and should be answered in light of both the significance and the aim of the directive. Consequently, this question was inextricably linked to that of the substantive law of the Directive. In that context it was pertinent to point out that in the Vienna Sales Convention, and the Spanish, French and German rules of international private law and civil procedure, the burden of proof was seen as an aspect of substantive law.

For these reasons, it would be in keeping with the aim of the Directive for the rule relating to the burden of proof to have a uniform content. The CJEU had already declared questions relating to the burden of proof to be admissible. Furthermore, if trade marks were incorrectly registered they were capable of restricting market access and constituting a barrier to the freedom to provide services so the CJEU ought to have the competence to answer the questions.

Questions Referred
According to the Curia website, the German Federal Patent Court has asked:

1. Does Article 3(1) and (3) of the directive preclude an interpretation of national law according to which, for an abstract colour mark (in this case: red HKS 13) which is claimed for services in the financial affairs sector, a consumer survey must indicate an adjusted degree of association of at least 70% in order to form a basis for the assumption that the trade mark has acquired a distinctive character following the use which has been made of it?

2. Is the first sentence of Article 3(3) of the directive to be interpreted to the effect that the time at which the application for the trade mark was filed – and not the time at which it was registered – is also relevant in the case where the trade mark proprietor claims, in his defence against an application for a declaration invalidating the trade mark, that the trade mark acquired a distinctive character, following the use made of it, in any event more than three years after the application, but prior to registration?

In the event that, under the abovementioned conditions, the time at which the application was filed is also relevant:

3. Is the trade mark to be declared invalid if it is not clarified, and can no longer be clarified, whether it had acquired a distinctive character, following the use made of it, at the time when the application was filed? Or does the declaration of invalidity require the applicant seeking that declaration to prove that the trade mark had not acquired a distinctive character, following the use made of it, at the time when the application was filed?