Competition soft law in French and German courts

Abstract

Ten years after the decentralization of EU competition law enforcement, this paper aims to ascertain through empirical means whether Commission-issued competition guidelines and notices (soft law) actually help to enhance enforcement consistency at the national level. This is what the European Commission itself maintained in White Paper on Modernization in 1999. The main premise of the White Paper is that if soft law is to help enhance consistency (a goal central to the current regime), it needs to be treated consistently by national judiciaries within and across EU jurisdictions – a difficult task, however, given its lack of binding force. Therefore, to ascertain the extent to which and how national courts in two select jurisdictions – France and Germany – engage or refuse to engage with supranational soft law, a theoretical framework of ‘judicial recognition’ of soft law is superimposed on a sample of 84 national judgments. While not discouraging, the results show some significant discrepancies that require further attention.

1. Introduction

More than 15 years ago, in its White Paper on Modernization, ¹ the European Commission set out its plans for the decentralization and ‘modernization’ of EU competition law, while emphasizing the importance of preserving the consistency of enforcement in the newly designed multi-stakeholder system. The consistency goal was subsequently sealed into legislation – Regulation 1/2003 and Articles 3, 11 and 16 thereof. ² Those provisions penetrate into national (judicial) autonomy by, among other things, requiring national courts not to render judgments that are contrary to existing or even envisioned Commission decisions (Article 16(1) of Regulation 1/2003). Therefore, save for the ability of national courts to challenge the Commission’s choices under Article 234 TFEU, consistent decentralized enforcement is achieved through a focus on the Commission’s output as an enforcer that functions as a ‘first among equals’.

More specifically, in its White Paper the Commission explained that, in addition to its Block Exemption Regulations, enforcement decisions, and the case law of the supranational courts, the decentralized enforcement system would benefit from added legal certainty and consistency by means of notices and guidelines issued ‘to explain (…) policy and provide guidance for the application of the Community competition rules by national authorities’. ³ Given their non-binding nature, the latter instruments were deemed particularly suitable for usage as interpretative aids to hard law, thereby making ‘a valuable contribution to the consistent application of Community law’. ⁴ This enhanced consistency-securing function of guidelines also increased their relative importance as enforcement instruments. ⁵

Ten years down the road of modernization, however, the first indications appear that challenge the originally envisioned, consistency-enhancing role that Commission-issued guidelines were supposed to play in the decentralized competition enforcement system. ⁶ Discrepancies are observable in recent judgments of French and German courts, which have adopted opposing stances on a specific rule introduced by the Commission in its 2010 Vertical Restraint Guidelines. ⁷

The rule in question spells out the Commission’s policy on the issue of online platform sales bans. It seems as if this rule has not resonated with the majority of (French and German) courts that have adjudicated on the matter. However, at the same time, there are also several German courts that have endorsed the Commission’s approach. While this article acknowledges that the observed divergences can partly be explained with regard to the peculiarities of national (judicial) decision-making, it claims that they are also related to the uncertain legal effects of the soft law instruments in national courts.

In that context, this article asks whether – as claimed in the White Paper – Commission-issued competition guidelines (soft law) actually contribute to the achievement of consistency in national judicial practice. ⁸ The main premise is that if soft law is to help enhance consistency, it needs to be treated similarly by national judiciaries within and across EU jurisdictions. An obstacle in that respect is the lack of binding force of the instruments. However, supranational courts have (sparingly) acknowledged the ability of soft law to produce legal effects and, ⁹ by such means, to anchor itself in judicial reasoning. ¹⁰ The question, therefore, is whether and how national courts acknowledge the legal effects of soft law, which directly affects the latter’s ability to contribute to the achievement of consistent judicial outcomes. ¹¹

This question will be approached in two stages. Firstly, the article will examine the case study of online sales bans to show that consistent national judicial outcomes cannot be secured if the relevant soft law rules are novel and/or unclear. Consistency is also undermined, as the Pierre Fabre case ¹² exemplifies, when supranational courts miss an opportunity to elaborate on the legal effects of soft law. Secondly, the work undertakes a broader empirical examination into how other Commission-issued competition soft instruments, similar to the Vertical Restraint Guidelines, ¹³ fare in national judicial practice with regard to their consistency-enhancing function.

The question posed by this article will be tackled through a comparative study that focuses on France and Germany, as these are jurisdictions likely to offer fruitful ground for analysis due to their differing treatment of online sales bans. These two jurisdictions are also interesting to compare because they have the longest competition enforcement traditions in the EU and are thus equally experienced, but also equally diverse in their attitudes towards certain competition issues, as will be demonstrated in the following sections.

The article is divided in four parts: in Section 2, the case of differential judicial treatment of online sales bans is explored in more detail. Section 3 tackles the broader question of national judicial attitudes to Commission-issued competition soft law by presenting further empirical data. Section 4 then provides an analysis of the empirical results, while Section 5 contains the concluding observations.

2. The peculiar case of online sales bans in France and Germany

Currently, e-commerce and its regulation are hotly debated topics on the Commission’s (competition) agenda. ¹⁴ However, this was not the case 10 years ago ¹⁵ when only a limited number of legal sources, in particular the Vertical Restraint Guidelines, provided guidance on the competitive assessment of online sales.

In this context, several examples pertaining to the operation of the Vertical Restraint Guidelines in situations concerning uncertainty in the law will be put forward. What transpires is that, under such conditions, while (national) competition enforcement bodies are explicit about their reliance on soft law rules, courts hesitate even to involve those instruments in their discourse. As will be explained in Section 3 below, this attitude is likely to have its roots in the judicial apprehension over soft law rules being perceived as forming the ratio decideni of a judgment, which would essentially convert them into hard law. ¹⁶ Nonetheless, by simply excluding relevant soft law from their deliberations, courts hinder the ability of these instruments to foster certainty and consistency in the law.

To illustrate, in 2008, the French Competition Authority (Autorité de la concurrence, FCA) issued a decision ¹⁷ containing the novel finding that a manufacturer of high-end cosmetic products (Pierre Fabre) had infringed Article 101 TFEU and the equivalent national provision ¹⁸ by imposing, in its selective distribution agreements, a total ban on the online sale of cosmetics and personal care products to end-users. The infringement was held to constitute a ‘by object’ restriction because it denied a marketing channel that would otherwise have been available to Pierre Fabre’s distributors. The FCA’s reasoning relied heavily on paragraphs 51, 53 and 54 of the old Vertical Restraint Guidelines, read together with the Vertical Block Exemption Regulation (VBER) ¹⁹ and Article 4(c) thereof. Most significantly, the FCA broadened the ambit of Article 4(c) to include online sales bans by relying on paragraph 53 of the 2001 Vertical Restraint Guidelines, which stated that ‘in a selective distribution system the dealer should be free to advertise and sell with the help of the internet’. ²⁰ Pierre Fabre objected to this interpretation on the basis that Commission guidelines were of no legal value in the dispute. ²¹ The FCA replied that – although not binding on national judicial and administrative bodies – those soft law instruments were a useful ‘analytical guide’ for the interpretation of hard legal provisions to which they pertained, thus guaranteeing – to a certain extent – the uniform application of EU law by National Competition Authorities. ²² However, on appeal, ²³ the Paris Court of Appeal was not entirely convinced, noting that it was uncertain of the law in the area and that neither the Commission guidelines nor the subsequent amicus brief that it received from the Commission were legally binding. ²⁴ Given these circumstances, a request for a preliminary ruling on the legality of online sales bans was submitted to the Court of Justice of the European Union (CJEU).

The CJEU clarified the law in the matter at hand (agreeing with the FCA that online sales bans constituted ‘by object’ restrictions), but at no point did it discuss the relevance of the Vertical Restraint Guidelines, which were clearly used as an important guidance tool by the FCA. The only brisk reference to the soft law instrument in the CJEU’s judgment came in the statement of facts, whereby the CJEU noted that the Paris Court of Appeal’s position on soft law was ‘that neither the Commission’s guidelines nor its observations were binding on the national courts’. ²⁵

While undoubtedly correct, this statement says nothing about the CJEU’s views on the matter; in this sense, the CJEU missed an opportunity to clarify the legal effects of the Vertical Restraint Guidelines for national courts.

It is therefore not surprising that, in a later case very similar to Pierre Fabre, ²⁶ the Paris Court of Appeal reduced a fine initially imposed on the plaintiff by the FCA because of uncertainty in the law relating to online selling. The French court disagreed with the FCA’s argument that pursuant to the provisions of its Vertical Restraint Guidelines, the plaintiff should have been aware of the ‘by object’ nature of their conduct. The law, as it stood in the guidelines, was thus seen as being uncertain.

With the Pierre Fabre developments in mind, and from the perspective of consistency in the judicial treatment of soft law across the EU, it is of particular interest to this paper to track national cases where courts were challenged – similarly to the CJEU – to rule on the related issue of banning sales via online third-party platforms such as Amazon and Ebay. Importantly, and similarly to the situation in the Pierre Fabre case, there is no supranational case law or legislation spelling out the rules applicable to the latter situation and the only guidance on the matter can be found in paragraph 54 of the Vertical Restraint Guidelines. Paragraph 54, which was introduced in 2010, spells out the following rule (hereinafter, the ‘paragraph 54 rule’):

a supplier may require that its distributors use third party platforms to distribute the contract products only in accordance with the standards and conditions agreed between the supplier and its distributors for the distributors’ use of the internet. For instance, where the distributor’s website is hosted by a third party platform, the supplier may require that customers do not visit the distributor’s website through a site carrying the name or logo of the third party platform.

The rather unclear second sentence of this passage permits a supplier to prevent its distributors from placing the contract goods/services on online platforms such as marketplaces or price comparison sites. ²⁷ As will be demonstrated below, despite the similar circumstances, reactions of national judicial bodies to the paragraph 54 rule did not quite resemble that of the CJEU in Pierre Fabre – in particular, and unlike the CJEU, national judges did not steer away from a direct discussion of the paragraph 54 rule. On the contrary, they clashed in their interpretations.

It is hereby maintained that the lack of an authoritative CJEU judgment on the legal effects of the Vertical Restraint Guidelines, combined with paragraph 54’s uncertain meaning and the type of instrument it is contained in (that is, soft law), contributed to the divergent judicial reactions to the issue in both France and Germany. This observation will be backed by evidence in the following sections.

A. Online sales bans and the German judiciary

The paragraph 54 rule was criticized by several German courts in the context of private enforcement claims ²⁸ and also seems to have been rejected by the German Federal Cartel Office (Bundeskartellamt, BKartA) in its 2013 Background Paper on Vertical Restraints in the Internet Economy. ²⁹ In this document, the German Federal Cartel Office suggested that the CJEU’s Pierre Fabre judgment clashed with the aforementioned rule and, by implication, superseded it.

In this sense, both judicial instances in the Casio case ³⁰ maintained that the defendant – a manufacturer of digital cameras – had included a ‘by object’ restriction in its distribution contracts by prohibiting its distributors from selling on so-called ‘internet auction platforms’ and ‘internet marketplaces’ (for example, Ebay and Amazon Marketplace). The courts arrived at that conclusion by arguing that the sales prohibition considerably hindered the distributors’ access to certain customer groups (marketplace and platform users). This reasoning echoes Article 4(b) of the VBER, which provides that an agreement is restrictive ‘by object’ if it restricts the territory into which, or the customers to whom a buyer could sell the contract goods. The court further elaborated on this argument by relying on passages of the Vertical Restraint Guidelines that clarify the content of Article 4(b) of the VBER. ³¹

Importantly, with regard to paragraph 54 of the guidelines that precludes an interpretation of total online platform sales bans as ‘by object’ restrictions, the lower court reasoned by inferring the Commission’s intent. The judge maintained that with this paragraph, the Commission did not support a total online platform sales ban because such a ban would surely be anti-competitive ‘by object’. ³² On appeal, this reasoning was upheld and it was further argued that had the Commission wished to allow a total ban on online platform sales, it would have formulated this in simple terms in paragraph 54 (which, accordingly, it did not). ³³ Both courts were therefore of the opinion that by adopting paragraph 54, the Commission simply wanted to establish the rule that the manufacturer could introduce certain justifiable quality requirements (such as those in brick-and-mortar shops) to its distribution system.

A similar reasoning was exhibited by the Frankfurt District Court in its Coty and Deuter cases. ³⁴ Both cases dealt with bans on online platform sales within the selective distribution systems of the perfume manufacturer Coty Germany and the rucksack manufacturer Deuter, respectively. The court decided both cases by viewing the bans in question as absolute restrictions completely excluding certain customer groups from the reach of distributors. The restrictions were thus found to be ‘by object’ restrictions. In contrast to the Casio case, however, the paragraph 54 rule was not subjected to re-interpretation, but was dismissed on the grounds that the paragraph’s content was not compatible with hard law provisions (namely, Article 101 TFEU and Article 4 of the VBER) and CJEU case law – the Pierre Fabre judgment. ³⁵ Additionally, in both cases the court held that paragraph 54 was superseded by the Pierre Fabre judgment. ³⁶

It is important to note, however, that this constitutes a very broad reading of the Pierre Fabre case, which essentially dealt with the fact that a manufacturer cannot – without legitimate reasons – restrict sales of its products only to physical stores; it must allow online placement too. This is a different issue from the discussion on whether or not to use online platforms, which are but one – amongst many – methods of placing products on the internet.

A variation of this argument is used by courts that disagree with the ‘by object’ characterization of online platform bans. ³⁷ The reasoning goes that one cannot see an object restriction excluding a particular customer group ³⁸ in a situation, such as sales via platforms, when such a group cannot be clearly demarcated. As the Higher Regional Court of Munich reasoned in its Sports Articles decision: ‘[s]ince internet platforms are destined for all internet users, the customers of those platforms can also be reached through other means’. ³⁹

Yet, a different interpretation was given by the Higher Regional Court of Frankfurt on appeal to the aforementioned Deuter case. ⁴⁰ The reasoning of the lower instance court was modified and the ban on sales via online platforms was re-framed as a restriction that could be legitimized where it was applied in a non-discriminatory fashion. In particular, the appellate court stated that a brand manufacturer has a legitimate interest that its branded products are perceived as high quality products accompanied with high quality sales advice. Consequently, the manufacturer is, in principle, free to decide under which conditions its products are sold, provided that these conditions are necessary to meet the quality standards. In that sense, the appellate court endorsed the Commission’s reasoning contained in the guidelines.

The Higher Regional Court of Berlin, in its Scout satchels case, also took a view similar to the Higher Regional Court of Frankfurt in Deuter. ⁴¹ In Scout, the dispute centered around distribution agreements for the sale of luxury schoolbags, whereby the manufacturer had imposed a prohibition on Ebay sales. Although, at a first glance, the ultimate decision of the court to prohibit the Ebay selling restriction as a hardcore restriction seems to go against the provision of paragraph 54 of the guidelines, the case must be read on its facts. The court only prohibited the ‘Ebay clause’ because the manufacturer itself had sold some of its (allegedly) high-quality produce to discount retailers. In this sense, the Ebay clause was seen as discriminatory to other distributors and thus prohibited. The decision did not take issue with the Vertical Restraint Guidelines themselves.

Overall, the position of the German judiciary on online platform sales bans seems to depend on whether the court sees these practices as either:

restricting market access to a particular customer group, in which case they will be deemed as ‘by object’ restrictions; or

restricting a particular method of distribution, whereby they are unlikely to be seen as anti-competitive. ⁴²

For that determination, the judicial interpretation of the paragraph 54 rule (together with the VBER) is key and it is unfortunate that the judgments discussed above disagree not only on the meaning, but also on the relevance and applicability of the provision. As discussed above, this fact can certainly be attributed not only to the peculiarities of national (judicial) decision-making, but also to the unclear phrasing of the paragraph 54 rule, combined with the uncertain legal effects of the Vertical Restraint Guidelines and the lack of supranational judicial guidance/precedent on the latter.

B. Online sales bans and the French judiciary

The French outcry on online platform sales bans is more recent than the German one; hence, fewer cases exist and – for the most part – relevant cases are still pending. The most recent case had its roots in a conflict between the companies Concurrence and Samsung and was subject not only to litigation in the French courts (that was ultimately referred to the CJEU), but also to a public investigation by the FCA. Unfortunately, due to a request for interim measures by the complainant Concurrence, ⁴³ the FCA has not yet pronounced itself on the substance of the dispute – namely the legality of the manufacturer’s (Samsung) ban on its former distributor’s (Concurrence) online marketplace sales. While the factual setup of the case embodies the classical scenario ‘supplier prohibits authorized distributor from selling via certain online channels’, there are certain important details which must be taken into consideration. In particular, Samsung allowed sales by means of online marketplaces to other distributors taking part in its distribution network. In that sense, Concurrence also alleged discriminatory treatment, which makes the facts similar to those of the German Scout case discussed above. Thus, the future decision of the FCA might also have to be read on its facts and, in that sense, will not lead the interpretation of paragraph 54 of the Vertical Restraint Guidelines.

An important judgment that could prove to be leading on the interpretation of the paragraph 54 rule though is the Paris Court of Appeal’s recent pronouncement in the Caudalie case. ⁴⁴ In this judgment, the court prohibited manufacturers using selective distribution networks from completely banning marketplace retailers from offering their products. In other words, the scope of the Pierre Fabre rule is broadened to encompass bans on specific types of online sales.

In that case, the cosmetics company Caudalie applied for an injunction before the Paris Commercial Court against a marketplace platform operated by the company eNOVA, which was selling Caudalie brand products online without being an approved distributor. Even authorized Caudalie distributors were only to sell online via their own websites, as opposed to offering the products on an online marketplace. This selective distribution system was defended by eNOVA on the basis that it had been approved by the FCA on a prior occasion (in 2007) and therefore was perfectly legal. Nevertheless, Caudalie was granted an order for interim measures, ⁴⁵ which stated that eNOVA should cease all marketing of the products in question. Consequently, eNOVA filed an appeal against this order, citing among other things recent decisions of the French and German Competition Authorities favoring completely unrestricted sales via online marketplaces. ⁴⁶ The lower court’s injunction was then overturned by the Paris Court of Appeal, which instead ruled that a blanket prohibition of selling online via a marketplace platform may well constitute an impermissible restriction of competition.

Although, in ruling so, the court did not expressly mention the paragraph 54 rule, the sources on which it based the final judgment – decisions of the French and German Competition Authorities – are decisions that directly affect the interpretation of this provision. In particular, they all contradict the position of the Commission expressed in the said rule. In this sense, a rule initially proposed in supranational soft law has been picked up, re-interpreted in the decisions of national administrative authorities, and – in this latter form – informed a French judicial decision. This development can be seen as the reverse image of what the Commission initially hoped to achieve with its guidelines, namely that

They (guidelines) might not be binding on national authorities, but they would make a valuable contribution to the consistent application of Community law, because in its decisions in individual cases the Commission would confirm the approach they set out. Provided those individual decisions were upheld by the Court of Justice, then, notices and guidelines would come to form part of the rules that must be applied by national authorities. ⁴⁷

On the contrary, in the Caudalie case, national authorities not bound by Commission soft law resisted the original idea behind paragraph 54 and were then followed by a French appellate court, which disregarded the views of some of its German colleagues. ⁴⁸ In the absence of a Commission decision or a CJEU precedent on the matter, there is certainly no obligation on the French judiciary to follow the approach taken by German courts, but the fact remains that the observed development can hardly be seen as certainty and consistency-enhancing. From this perspective, now that the issue of online platform sales bans has reached the CJEU by means of two preliminary references, ⁴⁹ the CJEU’s answer on the legality of these practices will be extremely important. Should the court adhere to the majority position that online platform sales bans are restrictive ‘by object’ contrary to what the Commission believes, this would hint at an ability of national enforcers to formally set EU competition policy in a bottom-up fashion, thus interfering with the top-down (formal and informal) ⁵⁰ processes of EU competition policy-setting by the Commission.

Having examined the dynamics underlying national judicial treatment of the paragraph 54 rule and the implications for enforcement consistency, this article will now turn to the broader question of how other Commission-issued competition soft law instruments are seen by national courts and whether their legal effects are recognized or not.

3. Judicial reception of Commission-issued soft instruments in France and Germany

In order to analyse a broader empirical sample of judicial attitudes to supranational soft law, this paper adopts a classification developed in the previous works of this author ⁵¹ that draws from and builds on theoretical accounts proposed by several legal scholars ⁵² and political scientists. ⁵³ The generated categories of judicial attitudes are as follows: (1) national judicial recognition of supranational soft law (comprising explicit agreement or disagreement with the instruments’ contents, or, alternatively, implicit persuasion ⁵⁴ of their value); and (2) national judicial refusal for recognition of supranational soft law (comprising explicit rejection to engage with the instruments’ contents, or, alternatively, implicit neglect thereof). The recognition category presupposes the attribution of legal effects to soft law, while refusal for recognition suggests the opposite outcome.

Before fitting the generated empirical data within this framework, it is necessary to define the parameters of the empirical sample.

A. Aggregate presentation of empirical observations

The total empirical sample, including the cases discussed in Section 2 above, amounts to 84 judgments (44 for France and 40 for Germany) that mention Commission-issued competition soft law since the entry into force of Regulation 1/2003. ⁵⁵ The observations – comprising of both public and private enforcement judgments – were selected by means of a key-term search ⁵⁶ through national ⁵⁷ and EU ⁵⁸ case law databases. Finally, the sample aims to encompass all relevant judgments – both purely national and with an EU dimension – that were issued by French and German courts in the above-delineated period.

The final number of competition judgments dealing with supranational soft law forms a relatively small percentage of the totality of competition judgments delivered in the researched jurisdictions during the relevant reference period. Empirical studies point to consistently high levels of overall competition litigation activity in Germany – circa 200 cases per year in the period of 2000-2010. ⁵⁹ This figure seems realistic when juxtaposed with data on private enforcement cases, which, after excluding judgments that do not have competition matters as their core, constituted 608 judgments within the period of 2004-2009. ⁶⁰ With regard to public enforcement, in the period of May 2004 to May 2016, records of the German Federal Cartel Office show a total of about 200 administrative decisions taken, which formed the basis for subsequent judicial activity.

For France, the public enforcement figures available seem to be more reliable than the private enforcement figures. The latter figure is a mere 58 judgments in the period 2004-2012 by a report drafted by Chagny and Fourgoux. ⁶¹ However, the authors admitted that there was a lack of easily accessible information on French private competition judgments. The given number is thus likely to be underreported. As to public enforcement, Idot testified to the existence of 219 FCA decisions and 94 envisaged decisions for the period of 2004-2013, ⁶² which reflected (in rough terms) the numbers available on the public database of the FCA – namely, 464 decisions in the period between mid-2004 and late-2015.

Finally, the empirical data-gathering exercise showed that cases mentioning the Vertical Restraint Guidelines form a significant part of the total findings per country: 77% for France and 62% for Germany. This can be explained by the general trend identified by de Vries and others ⁶³ that nowadays ‘vertical’ cases before the Commission are very scarce, while they are on the rise in Member States. Additionally, the fact that both the French and German competition authorities have a historical proclivity for enforcing the competition rules against vertical agreements, sheds light on the observed statistics. ⁶⁴ For Germany, according to Rodger, vertical agreements are also often the subject matter of stand-alone civil litigation. ⁶⁵ Also, as seen above, the advent of e-commerce and the novel issues it poses for selective distribution in particular, gives further enforcement impetus to the authorities in question as testified by their sector inquiries ⁶⁶ and position papers ⁶⁷ on the matter. As to other types of possibly anti-competitive agreements, experts in both jurisdictions testify to the fact that because cartels and horizontal cooperation agreements are, in general, more difficult to prove, those cases are rarer than the ones concerning vertical relationships. ⁶⁸

In light of the significantly higher number of judgments related to the Vertical Restraint Guidelines, not all of them are going to be discussed in detail, thus giving way to a discussion of judgments dealing with other supranational competition soft instruments. The selection in that regard comprises of substantively significant Commission-issued soft instruments similar to the Vertical Restraint Guidelines, namely ⁶⁹ the Horizontal Agreements Guidelines, ⁷⁰ the Article 81(3) Guidelines, ⁷¹ the Article 82 Guidance Paper ⁷² and the Technology Transfer Guidelines. ⁷³ It is also worth mentioning that the instances where judicial recognition of soft law was observed significantly outnumbered the rest of the envisioned judicial attitudes. This is, at least at first glance, a positive development from a consistency perspective.

B. National judicial approaches to supranational competition soft law

Recognition – explicit agreement or disagreement

Judicial recognition of supranational soft law happens when courts explicitly acknowledge the relevance of soft law instruments to a dispute, be it because of a soft law invocation by the parties or on the courts’ own initiative. In that respect, the conclusion that surfaces from an in-depth examination of the empirical sample is that supranational soft law is judicially recognized when used together with relevant hard law (case law or legislation) that it complements and/or clarifies. A soft law instrument can also be judicially invoked and recognized on its own (stand-alone invocation) when the court sees it as a ‘shorthand reference’ to well-known principles and rules established in hard law. To illustrate those two instances of judicial recognition, a few examples will now be discussed.

In the German HRS case, an appeal to a 2013 German Federal Cartel Office decision, ⁷⁴ the Higher Regional Court of Düsseldorf invoked the Article 81(3) Guidelines in order to assess the plaintiff’s efficiencies claim. ⁷⁵ The case concerned the so-called retail ‘most favoured nation agreements’ (MFNs) – vertical arrangements by means of which sellers through an internet retail platform agree not to sell at a lower price elsewhere, including through other retail platforms. ⁷⁶ This practice was sanctioned by the German Federal Cartel Office as ‘a significant restraint of competition’, which was subsequently appealed because the alleged infringer (HRS) believed the market definition and therefore its market share estimation to be incorrect, which led to an inaccurate disapplication of the VBER.

The court dismissed HRS’s arguments and proceeded to examine a possible individual exemption under Article 101(3) TFEU. In the context of the balancing test to be performed under the said provision, ⁷⁷ the judge used the Article 81(3) Guidelines and paragraph 51 thereof to assess whether the first of the four cumulative conditions under Article 101(3) was fulfilled. In this context, the relevant information contained in the soft law instrument, which listed the conditions for substantiation of efficiency claims, was invoked on the court’s own initiative and on a stand-alone basis. It is important to observe, however, that the information contained in paragraph 51 of the guidelines is well established in hard law (more specifically, in prior Commission decisions). ⁷⁸ Therefore, in this case, the 81(3) Guidelines and paragraph 51 thereof were used as a ‘shorthand reference’ to rules already established in hard law. ⁷⁹ The same phenomenon with regard to the 81(3) Guidelines is also detectable in the decisions of the French courts. ⁸⁰

A French ‘shorthand’ use of the 81(3) Guidelines is exemplified by the Crédit Lyonnais judgment – an appeal to an FCA decision which held that an agreement between major French banks to reform the national interbank check-processing system amounted to a cartel-like, ‘by object’ violation of competition law. The Paris Court of Appeal criticized the authority for being too eager to place the agreement in the ‘object’ box. In that context, an extensive discussion of the difference between the ‘object’ and ‘effect’ categories in EU competition law followed, an illustration of which was given (among others) by means of a stand-alone reference to paragraph 20 of the 81(3) Guidelines that essentially summarized relevant case law on the matter.

The Horizontal Guidelines ⁸¹ received similar treatment by the German judiciary in the Press Wholesaler Association (PWA) case – a cartel-like arrangement that was eventually legitimized under Article 106(2) TFEU. ⁸² The case concerned the practices of the PWA, which was in the business of negotiating uniform contract conditions with publishers (suppliers) on behalf of its members (press wholesalers). An independent publisher negatively affected by this practice complained to the Cologne District court, which consequently denounced the practice as anti-competitive. One of PWA’s defending arguments in that regard was that its members were not competing undertakings and thus competition could not be restricted. To counter this reasoning, among other things, the court cited the content of a footnote to the 2001 Horizontal Guidelines that defined the concept ‘potential competitor’ under EU law. This stand-alone citation was only possible because the concept had already been discussed in a preceding Commission decision and in other hard law instruments. ⁸³ The citation to the guidelines was thus used as a shorthand reference to those sources.

In turn, on appeal, the Vertical Restraint Guidelines were cited by the Dusseldorf Court of Appeal. However, they were not used as a shorthand reference to hard law, but rather in a supporting function to legislation – namely, the VBER. ⁸⁴ In particular, PWA argued that its press distribution model was to be seen as ‘exclusive distribution’ under the VBER and thus exempted under the latter’s provisions. However, a precondition for exemption of exclusive distribution under the VBER ⁸⁵ is the non-preclusion of passive sales into delineated exclusive territories. Thus, the court had to determine what constituted ‘passive sales’ and consulted the provisions of the Vertical Restraint Guidelines, which further complemented and clarified the VBER on that point. This judicial engagement instance is therefore one that reflects the interpretation of soft law together with pertinent hard law. ⁸⁶ Both of the described types of recognition testify to soft law’s ability to produce legal effects under the circumstances delineated above.

Refusal for recognition – neglect

The theoretical model that this article utilizes sees judicial neglect as an instance whereby an argument based on soft law is made by a party to the dispute, but is ignored by the court. Such cases are significantly fewer than the ones constituting explicit recognition; still, their occurrence is certainly not conducive to consistency in enforcement at national level. To illustrate, an example of judicial neglect by a French court will be given.

In the Crédit Lyonnais case discussed above, the 2001 Horizontal Guidelines were neglected by the judge when the plaintiff banks invoked paragraph 24 of the said instrument to argue that their check-processing agreement fell outside the remit of competition law altogether. In particular, this part of the Horizontal Guidelines provides that when undertakings agree to join forces in order to carry out an activity that they cannot single-handedly achieve (as was the case with the check-issuing agreements), that activity does not imply a coordination of the parties’ competitive behaviours on the market and thus falls outside the remit of competition law. The court was silent on this issue; it chose not to engage with it and ruled that the practice was not anti-competitive because of the applicability of a particular national regulatory framework to the hybrid (public and private) banking sector. ⁸⁷

In this sense, one can argue that the Horizontal Guidelines and paragraph 24 thereof were subject to judicial neglect. ⁸⁸ This will not come as a surprise if one considers that this same passage was also subjected to a similar judicial attitude (that is, rejection) by Justice Morgan of the English Court of Appeal. ⁸⁹ Such instances could be explained by a judicial hesitation to engage with soft law passages that could, in principle, be perceived as the relevant rule that solves the case (the so-called ratio decidendi). Since the ratio must be exclusively informed by hard law rules, ⁹⁰ judges would not risk interpreting soft law in cases where such interpretation could be seen as forming the ratio of a decision.

As a second and final point on judicial neglect, it is worth mentioning that, after a thorough examination of the empirical data, it transpires that neglect can also be the result of situations where courts simply do not need to involve soft law in the discussion, even if it is invoked by the parties to a dispute. In such cases, judicial economy considerations are a good reason for non-engagement with certain arguments. Such situations do not exhibit a particular judicial attitude to soft law and therefore have no bearing on the research question posed by this work. To illustrate, a party to the proceedings invokes Commission-issued guidelines, but the court decides the dispute on the basis of national competition law only. ⁹¹ Other cases were detected where the competition matter, although pled by the parties, was not decided due to lack of sufficient evidence ⁹² or because of insufficiently elaborate argumentation. ⁹³ Another relevant illustrative example concerns situations where national appellate courts have to judge on the legality of an administrative decision and, in this sense, can disregard party arguments (including such based on soft law) that do not contribute to the purpose of judicial review. ⁹⁴

3. Refusal for recognition – explicit rejection

With regard to the hypothesized explicit judicial rejection of soft law, no further cases beside those already identified in Section 2 above were found in French judicial practice. The cases detected in German courts, however, further confirm the skeptical attitude of some judges towards the rules on online selling in vertical distribution agreements enunciated in the Vertical Restraint Guidelines.

To illustrate, in 2013 the Cologne District Court had to rule on the anti-competitiveness of a refusal of membership to an applicant to the German Wholesaler’s Association ‘Haustechnik’. ⁹⁵ The applicant alleged discriminatory treatment that – unless objectively justified – was anti-competitive. The plaintiff also claimed that one of the reasons for its rejection – the fact that it handled its business predominantly over the internet – is protected by the European Commission in its Guidelines on Vertical Restraints. However, the judge did not even come to rule on the substance of this point, but rather denied the applicability of the Vertical Restraint Guidelines by explicitly stating they were not controlling in relations between private parties. ⁹⁶ Although the soft law instrument could have been interpreted together with the pertinent VBER, which both the French and German courts have done on other occasions, the German judiciary refused to interpret the guidelines in this case altogether. This could be due to the fact that the paragraph of the Vertical Restraint Guidelines relevant to the dispute (paragraph 52 regarding online distribution) was only inserted in the latest version of the guidelines, dating to 2010, and, similarly to paragraph 54 discussed above, was not yet ‘tested in battle’ before the CJEU. In that sense, the national judiciary was careful not to tread new ground and set in stone rules that have not yet surfaced as ‘hard principles’ at EU level. ⁹⁷

The reason why some German courts were less cautious when endorsing paragraph 54 of the Vertical Restraint Guidelines might lie in the fact that that passage had already been subject to several decisions of the German Federal Cartel Office, which – by interpreting soft instruments – signaled how the authority saw the particular issue. ⁹⁸ This, in turn, charts a possible line of reasoning that could be endorsed or rejected by courts as observed above.

Recognition – persuasion

Instances of judicial persuasion are generally hard to determine because in such cases courts never explicitly engage with the content of guidelines, but still reach a conclusion compatible with their content. This study only detected two such instances, both in French courts. ⁹⁹ On both occasions, the Vertical Restraint Guidelines were invoked by a party to the proceedings together with the pertinent VBER. In the grounds for the decision, on both occasions, the judges preferred merely to quote the provisions of the VBER as informing their conclusions, although in doing so they were very likely also agreeing with the content of the Vertical Restraint Guidelines previously invoked by the parties.

4. Discussion of results and their implications for enforcement consistency

In this Section, the above account on national judicial attitudes to supranational competition soft law will be summarized, analysed and represented graphically. ¹⁰⁰

A first observation that stands out is that unlike the Horizontal and 81(3) Guidelines, which are the subjects of relatively homogenous treatment (recognition) across jurisdictions, the Vertical Restraint Guidelines have engendered more varied judicial attitudes. Nevertheless, more than half of the observations dealing with the Vertical Restraint Guidelines in both jurisdictions exhibit a ‘recognition’ attitude by the courts. In fact, scholarly accounts on the Vertical Restraint Guidelines point out that, unlike other soft instruments (namely, the Guidance Paper ¹⁰¹ ), those guidelines (together with the Horizontal Guidelines) enunciate precise and clear rules that help businesses to correctly self-assess. ¹⁰² Therefore, it is only logical that the judiciary treats such unambiguous rules relatively consistently. Additionally, the majority of cases dealing with the Vertical Restraint Guidelines in which neglect or rejection by national courts can be observed are either cases exhibiting judicial economy concerns (as explained in Section 3) or cases dealing with rules newly introduced by the 2010 version of the Vertical Restraint Guidelines (discussed above). ¹⁰³

Table 1.

Germany

Table 2.

France

This observation thus confirms the intuition that courts are hesitant to interpret rules contained in soft law that have not yet been ‘tested in battle’ before the CJEU. ¹⁰⁴ Besides precedent established by the CJEU, what makes courts more willing to engage with soft law content and thus endow it with legal effect, is the existence of a legislative instrument that forms the basis for the existence of the soft law instrument. This is the case for the VBER that forms the backbone of the Vertical Restraint Guidelines and the Technology Transfer Block Exemption Regulation ¹⁰⁵ that plays the same role for the Technology Transfer Guidelines. By the same token, the Block Exemption Regulations (BERs) on (1) specialization and (2) research and development agreements are further elaborated on in the Horizontal Guidelines. The existence of these Block Exemption Regulations likely preconditions the largely positive national judicial attitude (recognition) to both the Technology Transfer and the Horizontal Guidelines. Although the latter guidelines were subject to judicial recognition only in Germany, the neglect observed in France in that regard is – as explained in Section 3 above – an aberration.

Finally, with regard to the 102 Guidance Paper, no relevant judgments were detected in Germany and only one – a judicial recognition instance – was found in France. This might well be due to the fact that German law on abuse of dominance is stricter than its supranational counterpart, which would imply that it is easier for claimants to establish abuse under national law where the Guidance Paper is of limited or no relevance. ¹⁰⁶ This fact would undoubtedly make claims under German law more attractive. Another, more general, explanation for the lack of judicial engagement with the Guidance Paper might lie in the fact that the principles enunciated in it are not entirely consistent with the CJEU’s case law. ¹⁰⁷ France, although it was initially very positive about the upcoming Guidance Paper ¹⁰⁸ and impatiently expected it ¹⁰⁹ does not yet have a significant track record of judicial pronouncements in that regard either. This could indeed be owing to the fact that the national judiciary is more likely to get its cues from CJEU case law, which is essentially contradicted by the Guidance Paper. ¹¹⁰

The same reasoning also holds true for Germany, with the added fact that – at the time of adoption of the Guidance Paper – there was already strong political resistance to its adoption. ¹¹¹ As to the French position, Prieto testifies that ‘[u]nlike the BKartA, the Conseil de la Concurrence was convinced that the Discussion Paper was a workable framework to improve the enforcement of Article 82 EC and the concept of consumer welfare in particular’. ¹¹² The empirical data hereby generated, however, does not confirm this statement. A final explanation to account for the very limited number of references to the Guidance Paper could be a scarce statistical basis – relative to Article 101 TFEU cases, dominance claims are in short supply both at EU and national levels according to both Peyer ¹¹³ and Rodger. ¹¹⁴

5. Concluding remarks

The empirical sample presented in this paper shows that national courts are hesitant to engage with competition guidelines and do not endow them with legal effects when those instruments spell out vague and/or novel rules that have not been previously endorsed by supranational hard law (case law or legislation). The outcome is the same when a certain soft-law enunciated rule can serve as the ratio decidendi of a judgment. In those instances, soft law does not produce legal effects and its consistency-enhancing function is undermined.

On the contrary, when the content of Commission guidelines does reflect and/or clarify rules previously established in supranational hard law, judges readily endow soft law with legal effects through interpreting it together with the said hard law; its consistency-enhancing function is thus secured.

The online sales bans case study introduces a variation to the above conclusions, showing that when soft law fails to secure consistent outcomes at national level, the classical road of the preliminary ruling procedure is always open. However, empirical studies show that not many national cases make it to that stage. ¹¹⁵ Thus, unless the intensity of preliminary references in that regard increases significantly, national courts do need a common approach to tackling the issue of the legal effects of soft law in the ambit of competition law. This common approach can still be spearheaded by the CJEU, which should not eschew opportunities to delineate the legal effects of Commission-issued competition soft law.