Dutch Antitrust Authority investigates Apple: does this support Spotify’s complaint?

Kluwer Competition Law Blog
April 25, 2019

Please refer to this post as: , ‘Dutch Antitrust Authority investigates Apple: does this support Spotify’s complaint?’, Kluwer Competition Law Blog, April 25 2019, http://competitionlawblog.kluwercompetitionlaw.com/2019/04/25/dutch-antitrust-authority-investigates-apple-does-this-support-spotifys-complaint/


After Google has been fined several times for breach of (EU) competition rules the last couple of years, it might now be Apple’s turn. Last month (March 2019), Spotify filed a complaint against Apple with the European Commission. The Dutch Authority for Consumers & Markets (“ACM”) immediately responded by stating that ACM is finalizing its market study into mobile app stores which it launched a year before. ACM added that it examines similar practices as Spotify is complaining about to the Commission. On 11 April 2019, ACM indeed published its market study into mobile app stores. Most interesting development is, however, perhaps that ACM simultaneously announced the launch of an investigation into abuse of dominance by Apple in its App Store. Does this mean that Spotify indeed has a case? And/or is ACM stepping in to assist the European Commission in applying competition rules to curtail the market power of online giants, like Apple?

Why did Spotify file a complaint at the European Commission in the first place?

 According to Spotify, Apple does not play fair. Amongst others, Spotify alleges that:

  1. Apple charges a discriminatory “tax”. Apple requires that certain apps pay a 30% fee for use of their in-app purchase system (“IAP”).[fn]That revenue share is 30 percent for the first year of an annual subscription but it drops to 15 percent in the years after.[/fn] According to Spotify, this fee is discriminatory because only digital goods and services that are delivered inside an app are charged. Thus apps like Uber or Deliveroo are not charged this fee. Also, Spotify alleges that Apple does not apply this fee to Apple Music giving its own app an unfair advantage;
  2. Apple does not let Spotify share deals – like 99c for three months of Spotify Premium – if Spotify does not use IAP, while Apple promotes its own offer for a free month of Apple Music via a push notification, Spotify argues;
  3. Apple would allegedly not allow customers to upgrade to Spotify Premium with ease;
  4. Apple rejects Spotify’s app enhancements, and
  5. Apple does not allow Spotify access to all devices, like HomePod, Apple Watch and a connection with Siri while – again – Apple Music does have access.[fn]See also the five “facts” that Spotify published and would shows that Apple doesn’t play fair: https://timetoplayfair.com/facts/[/fn]

As a result, according to Spotify, Apple abuses its dominant position as both the owner of the iOS platform and App Store and a competitor to services like Spotify.[fn]See for more information about the possible qualifications of the abuse (discrimination, tying, refusal to supply, margin squeeze etc.) an interesting blog of Friso Bostoen of 24 April 2019.[/fn]

Why is the ACM investigating Apple?

The concerns expressed by ACM in its press release reflect Spotify’s complaints about Apple’s App Store. According to the press release, ACM will investigate, amongst others, whether Apple gave its own apps a preferential treatment in violation of the prohibition to abuse a dominant position. ACM received indications that app providers do not always have a fair chance against Apple’s own apps. Providers of digital products and services are required to use Apple’s payment systems for in-app purchases, and they are also required to pay a 30% commission. Moreover, they are not always able to use all functionalities of an iPhone. And finally, they say they have difficulties when communicating with Apple about the application of their conditions. In other words, ACM clearly recognizes the issues that Spotify has complained about to the European Commission.

What evidence follows from ACM’s market study?

Although the objective of ACM’s market study was explicitly not to carry out a competition law analysis, ACM describes a number of restrictive practices by both Apple and Google in the findings of the market study that may be found at odds with competition rules.

For example, on the one hand the market study acknowledges that an app store with popular apps contributes to the attractiveness of the overall ecosystem, which means that Apple and Google both have an incentive to assist app providers. However, on the other hand, the market study also stresses the dual role of both parties since they also compete directly with a selection of third-party app providers. According to the market study, this could give them an incentive to favor their own apps over apps of their competitors. The important position that Apple and Google both hold with the app stores on their respective ecosystems might also give them the opportunity to act in such way.[fn]The market study reveals that the app stores from both Apple and Google form a bottleneck for app providers to reach consumers on both mobile OSs. The app stores are becoming a default gateway for consumers to reach online content. On iOS, moreover, no viable alternatives exist for the Apple Store, while on Android alternatives are only realistic if the app provider already has an established user base.[/fn]

Examples thereof are limitations faced by third-party app providers regarding interoperability with the mobile OSs and the requirement to use IAP for apps that directly compete with Apple and Google. Another issue addressed in the market study is the unequal treatment of apps in general.[fn]ACM therefore also suggests to further debate whether there is a need for ex-ante regulation. For example a regulation comparable to the Net Neutrality Regulation which requires Internet Service Providers to treat similar Internet traffic non-discriminatory. [/fn] Examples mentioned in the market study report are (i) the technical or interoperability restrictions such as access to APIs while other apps were apparently granted access to these APIs and access to the NFC-chip, (ii) limited access to certain data concerning payments and customer relations while app providers are required to use the app stores payment systems for IAP and (iii) the featuring of apps in the app store (rankings).[fn]The market study also revealed problems that app providers experience with regard to the lack of communication with Apple and Google. These issues are most apparent when it concerns a discussion about the interpretation of the terms and conditions or a removal (unjustified or otherwise) of an app. In the opinion of ACM, transparency and the ability to get in touch with Apple and Google are important requisites for app providers to do business. ACM does note that the upcoming EU Platform to Business regulation might be a step in the right direction and will not prioritize transparency issues at this moment insofar as these issues fall under the aforementioned regulation.[/fn]

Does this mean that Spotify has a case?

It goes beyond the scope of this blog to discuss all aspects that would provide an initial answer to this question. At first sight, however, the issues revealed in the market study report seem to underpin the complaints filed by Spotify against Apple at the Commission. Also, the fact that ACM commenced an investigation against Apple based on these issues could be seen by some as an indication that Spotify indeed has a case. However, whether these issues also qualify as an abuse of a dominant position remains to be seen.

Especially since Apple fiercely rebutted Spotify’s arguments and defended its App Store rules publicly. According to Apple, Spotify does, for example, have access to Apple Watch and has the same app development tools and resources that any other developer has. Besides that, Apple emphasizes that Spotify cannot expect to use all functionalities of the App Store ecosystem, paid by Apple, while also retaining 100 percent of the revenue. After all, Spotify wouldn’t be the business they are today without the App Store ecosystem. Moreover, Apple stressed towards ACM, amongst others, that it would not be rational for Apple to discriminate unfairly against third-party apps, because Apple earns the vast majority of its revenues from devices. And the more popular Apple’s App Store is, the more devices it sells. In other words, Apple seems to argue that the conduct either did not take place, was justified and/or was just an incident – which according to the ACM could well be the case since, in a market place as large as the app store, mistakes can be made.

Furthermore, ACM’s investigation will initially focus on Dutch apps for news media that offer their apps in Apple’s App Store; not on music apps. The reason for this is that ACM received many indications about such apps (and perhaps not so much about music apps).[fn]Nevertheless, there are some similarities between the news and music apps. Both kinds of apps offer digital goods. Moreover, Apple plays in both situations a dual role as Apple offers not only its own Apple Music app, but recently also started a news service offering readers access to a range of publications on a subscription basis, known as Apple News Plus. This might also explain the support by Schibsted, Scandinavia’s largest publisher, and European Publishers Council for Spotify’s complaint. These news media parties highlighted that the same challenge is faced by news publishers offering apps for media consumers. See for example: https://musically.com/2019/04/16/nordic-publisher-schibsted-backs-spotifys-apple-battle/ and http://epceurope.eu/news-media-development-on-mobile-stifled-by-anti-competitive-app-stores-european-publishers-council-supports-spotifys-eu-competition-case/. [/fn] It will be interesting to see whether ACM will expand its investigation from only news media apps into other kinds of apps, like music apps. Although Spotify filed its complaint at the Commission, ACM has given an open invitation to Spotify and any other app provider that experiences restrictions in Apple’s App Store to feed into the ACM’s investigation. If the European Commission takes up Spotify’s complaint for investigation – which seems to be the case as apparently questionnaires have been sent to the market[fn] See http://www.mlex.com/GlobalAntitrust/DetailView.aspx?cid=1084314&siteid=190&rdir=1[/fn], however, ACM’s market study might perhaps in any case aid the Commission’s case.

Finally, it will also be interesting to see whether comparable complaints will be filed and investigations will be commenced against Google.[fn] It appears that at least one complaint has already been filed at the European Commission by the app developer Disconnect Mobile. At the time of this writing, it is unclear how the Commission will deal with this claim. See, for example, https://www.politico.eu/article/why-margrethe-vestagers-android-case-is-googles-worst-nightmare/. The Dutch market study and investigation also seem to complement the Commission’s prohibition decision against Google’s Android mobile operating system.[/fn] The Dutch market study revealed that Google uses comparable terms & conditions as Apple. ACM has, therefore, called upon app providers to come forward if they experience any problems with not only Apple’s App Store, but also if they experience similar problems with Google’s Play Store.[fn]According to ACM, the investigation will currently focus on Apple because, at the moment, the most detailed reports have been received about Apple’s App Store. ACM believes that these reports may indicate conduct that is at odds with competition law.[/fn] Perhaps Apple is thus not the new target, but the first target of ACM.

 

This blog’s authors are: Pauline Kuipers and Mariska van de Sanden
Pauline.Kuipers@twobirds.com and Mariska.van.de.Sanden@twobirds.com 

The views and opinions set forth herein are the personal views or opinions of the authors; they do not necessarily reflect views or opinions of the law firm with which they are associated.This blog has been prepared for informational purposes only and does not constitute legal advice. Readers should not act upon this without seeking advice from professional advisers. 

 




Abuse of Stronger Bargaining Position in Bulgaria – Initial Trends in Judicial Review

Kluwer Competition Law Blog
April 25, 2019

Please refer to this post as: , ‘Dutch Antitrust Authority investigates Apple: does this support Spotify’s complaint?’, Kluwer Competition Law Blog, April 25 2019, http://competitionlawblog.kluwercompetitionlaw.com/2019/04/25/dutch-antitrust-authority-investigates-apple-does-this-support-spotifys-complaint/


On 24 July 2015 an amendment to the Bulgarian Protection of Competition Act (PCA) was promulgated, introducing a novel type of infringement – abuse of stronger bargaining position.

Article 37a of the PCA reads as follows:

Paragraph 1: “Any act or omission of an undertaking with a stronger bargaining position, which contravenes good-faith commercial practice and impairs or may impair the interests of the weaker contracting party and the interests of consumers, shall be prohibited. Bad-faith shall be those acts or omissions which have no objective economic justification, such as unjustified refusal to deliver or purchase goods and services, imposing unjustifiably burdensome or discriminatory terms or unjustified termination of commercial relations.”

Paragraph 2: “The existence of a stronger bargaining position shall be ascertained in view of the characteristics of the structure of the respective market and the specific legal relationship between the affected undertakings, while taking into account the degree of dependence between them, the nature of their activity and the difference in its scale, the probability of finding an alternative trade partner, including the existence of alternative sources of supply, distribution channels and/or clients.”

The Bulgarian Commission on Protection of Competition (“CPC”/”the Commission”) has since rendered a number of decisions under Art. 37a PCA, some of which were then appealed.

To date, there have been only four court judgments reviewing decisions of the CPC under Art. 37a PCA. Three of those judgments were rendered in the period between 20 February 2019 and 27 March 2019. In all these three cases the three-member panels (different in each case) of the Supreme Administrative Court (“SAC”), acting as first instance in judicial review, annulled the respective CPC decisions.

The purpose of the present post is to provide an overview of the respective court judgments since they arguably contain key takeaways as to how Art. 37a of the PCA must be applied in practice.

 

  1. Cable operators v. Nova TV

 1.1. The CPC decision

In their application to the CPC, three cable operators alleged that the TV broadcaster Nova TV abused its stronger bargaining position by imposing a guaranteed minimum number of subscribers and using unclear methods to determine this number. The CPC found that Nova TV had stronger bargaining position due to the high rating and audience shares of its channels. As to the conduct at issue, the Commission concluded that there was no bad-faith/unjustified conduct by Nova TV since the guaranteed minimum number of subscribers was renegotiable and the methods used to determine this number were clearly formulated in Nova’s general terms and conditions. Based on those considerations, the CPC found that no infringements were committed by Nova TV vis-à-vis the applicant cable operators.

1.2. The SAC judgment

In its judgment dated 20 February 2019 the SAC disagreed with the definition by the CPC of the relevant market as national and noted that the market is actually regional. The court reproached the Commission for not investigating the presence or absence of alternative sources of supply for the Nova TV channels and how that would influence consumers. According to the court panel, the CPC must have examined the various components of the remuneration due by the cable operators to Nova TV in order to assess the economic justification of the price increases. The SAC disagreed with the conclusion of the CPC that the price was renegotiable, as the actual negotiations led to no renegotiation. Further, the court noted that the CPC failed to analyze the impact on consumers of the conduct at issue.

Based on the above considerations, the SAC annulled the CPC decision and remanded the case to the CPC with mandatory instructions to collect evidence regarding: the relevant regional markets; the possibility for the cable operators to find alternative suppliers for the channels of Nova TV; the good-faith commercial practice for TV distribution contracts by way of comparative analysis of contracts of other TV broadcasters; the price components and changes therein throughout the relevant period (2017-2019); the impact of the price increases between the TV broadcaster and the cable operators on the prices payable by consumers.

 

  1. Cable operators v. BTV[1] and Nova TV

2.1. The CPC decision

In its decision the CPC found that the two respondent TV broadcasters – BTV and Nova TV, had stronger bargaining position vis-à-vis the four applicant cable operators because of the high rating and audience shares of their respective TV channels. The Commission, based on the wording of BTV’s General Terms and Conditions and the contracts with the applicant cable operators, found that BTV had applied unclear methods when determining the actual number of subscribers as basis for the remuneration due to BTV by three of the four cable operators. The CPC thus found three separate infringements by BTV and imposed a sanction of nearly EUR 500,000 for each infringement (i.e., a total of approx. EUR 1,500,000).

2.2. The SAC judgment

In its judgment dated 15 March 2019 the three-member court panel of the SAC criticized the CPC for using rating and audience share data as indicators for market power on the market of wholesale supply of TV channels, whereas those numbers are relevant for another – the TV advertising – market. The court further reproached the unclear finding of the Commission that the applicant cable operators were “small regional operators” despite one of them being even in the national top 8 in terms of number of subscribers. According to the SAC, the CPC failed to analyse the dependence between the TV broadcasters and the cable operators (the latter being the means for the TV broadcasters to reach the targeted audience). Further, the Commission incorrectly viewed the two respondent companies together instead of treating them as mutually interchangeable – and thus alternative – trade partners for the cable operators.

As to the conduct at issue in the proceedings, the court underlined that only negotiations/contracting which occurred after the entry into force of Art. 37a PCA fall within the temporal scope of the latter (as opposed to application/performance of general terms and conditions/contractual provisions which had already been adopted at the date of entry into force of the Art. 37a prohibition). When reviewing the actions of BTV undertaken after July 2015, the SAC found, based on various factual details, that there was no bad-faith/unjustified conduct on behalf of BTV.

For the above reasons, the SAC annulled the CPC decision in its parts whereby the CPC had found and sanctioned abuse of stronger bargaining position by BTV.

 

  1. Supplier of alcoholic drinks v Kaufland

3.1. The CPC decision

In its application to the CPC, a supplier of low-price alcoholic drinks complained that Kaufland had systematically requested additional bonuses, which had led to reduction of the net supply price and resulted in sales at a loss by the supplier. Further, Kaufland had ultimately terminated the commercial relations with the supplier without objective economic justification.

The CPC adopted a narrow market definition encompassing “the supply of alcoholic drinks in the low-price segment, distributed in the Kaufland retail chain”.  The Commission decided that Kaufland had stronger bargaining position vis-à-vis the supplier due to the former being part of an international economic group with many outlets and serious investment opportunities. The CPC took into account that a large portion of the supplier’s turnover was formed by supplies to Kaufland and stated that the relocation to another trade partner would entail additional costs for the supplier. According to the Commission, Kaufland had unilaterally imposed on the supplier reduction of the supply prices by charging bonuses/discounts with no economic justification and through unclear mechanism to determine the amount thereof. The termination of the commercial relations between the parties was found to be part of an overall bad-faith model of conduct by Kaufland. The CPC concluded that the reduction of supply prices not accompanied by a respective decrease in the prices to consumers harmed consumer interests. The latter were also impaired by the blocking of sales of certain products from the supplier’s portfolio, which limited the available product variety.

Based on the above considerations, the CPC found that Kaufland had abused its stronger bargaining position and imposed on Kaufland a sanction of BGN 157,981 (approx. EUR 80,000).

3.2. The SAC judgment

In its judgment dated 27 March 2019 the SAC criticized the Commission for its abstract declaration that switching to a different channel of distribution is a process entailing lots of additional costs and time. According to the court, although this stance is theoretically true, such finding must have been made in concreto for the applicant. The CPC however had not examined what risky long-term investments the supplier must make in order to switch to another retailer, how the production must be restructured and what impact this would have on the production costs of the applicant supplier. The court concluded that the analysis of the market should have also covered other retail chains apart from Kaufland when deciding on the probability of finding an alternative trade partner. The SAC further underlined that the Commission had not investigated the good-faith commercial practice on the market, whereas it must have established what the customary commercial practice of other retail chains was in order to reach the conclusion that Kaufland’s conduct indeed contravened good-faith commercial practice on the relevant market.

Based on the above considerations, the SAC annulled the CPC decision and remanded the case to the CPC with mandatory instructions to conduct an economic analysis of the market of low-price alcoholic drinks sold in retail chains by inquiring into the commercial policies of the retail chains, in order to properly examine the probability of finding an alternative trade partner by the supplier and to establish what exactly the good-faith commercial practice on the market is.

 

  1. Key takeaways

Although the above SAC judgments are subject to cassation appeal, they arguably indicate the following initial trends in the judicial review of CPC decisions regarding abuse of stronger bargaining position:

  • market definition should be prudently done so as not to define the relevant market too narrowly (as in the Kaufland case) or too broadly (as in the Nova TV case);
  • the CPC must apply the Art. 37a prohibition only to negotiations/contracting which have taken place after the entry into force of Art. 37a;
  • when examining the alleged contravention of the respective conduct with good-faith commercial practice, the CPC must inquire into the actual practices by other market players – competitors of the respondent undertaking;
  • the Commission must thoroughly examine the probability of finding an alternative trade partner based on empirical evidence instead of relying on general theoretical statements of its own;
  • conclusions regarding the impact on consumers must also be made on the basis of actual data and not by way of mere assumptions.

After all, the ultimate outcome of the above cases as well as future CPC decisions and court judgments in other proceedings regarding alleged abuse of stronger bargaining position will show whether the aforementioned points would establish themselves as permanent trends in the application of Art. 37a PCA.


[1] The author has been and is currently representing BTV in this case




Apple v. EBizcuss.com: Agreeing A Forum For Your Antitrust Disputes

Kluwer Competition Law Blog
April 25, 2019

Please refer to this post as: , ‘Dutch Antitrust Authority investigates Apple: does this support Spotify’s complaint?’, Kluwer Competition Law Blog, April 25 2019, http://competitionlawblog.kluwercompetitionlaw.com/2019/04/25/dutch-antitrust-authority-investigates-apple-does-this-support-spotifys-complaint/


In a recent judgment providing a preliminary ruling in the case, Apple Sales International et al. v. EBizcuss.com (C-595/17, October 24, 2018) (“EBizcuss.com”), the Court of Justice of the European Union (“CJEU”) affirmed that jurisdiction clauses subject to EU law may be enforced by Member State courts in the context of actions for damages for abuse of dominance based on Article 102 TFEU.

As part of its judgment, the CJEU rejected any requirement that the jurisdiction clause make explicit reference to disputes relating to liability incurred as a result of an infringement of competition law.  The CJEU thereby confined its earlier judgment in CDC v. Akzo Nobel et al. (C-352/13, May 21, 2015) (“CDC”), which had held that such an explicit reference was a prerequisite for applying a jurisdiction clause to cartel-related claims based on violations of Article 101 TFEU.  CDC turned on the CJEU’s view that cartel-based claims would not have been foreseen by the parties at the time of contracting and thus could not be found to have been included within the scope of the jurisdictional agreement, unless explicitly identified. [see here; R. Harms/J. Sanner/J. Schmidt, EuZW 2015, pp. 584-592].  The CJEU reasoned in EBizcuss.com that claims alleging abuse of dominance based on Article 102 TFEU, unlike cartel-damages claims based on Article 101 TFEU, may be foreseeable to parties at the time of contracting.

 

Case Law In The EU Member State Courts Interpreting CDC In Connection With Agreements To Arbitrate

The judgment in EBizcuss.com did not address agreements to arbitrate and as such has no formal significance for questions of arbitral jurisdiction.  Whereas the CJEU has jurisdiction to resolve questions of EU law related to the scope of jurisdictional agreements subject to the Brussels Regulation and its Recast, it is questionable whether the CJEU has jurisdiction to resolve questions related to the interpretation of agreements to arbitrate, which are subject to national law.

Nonetheless, following the CJEU’s ruling in CDC, national courts in various EU Member States have been asked to consider the relevance of the CJEU’s case law related to jurisdiction clauses when deciding whether agreements to arbitrate drafted in general terms should be enforced to refer cartel-based damages claims to arbitration.  These Member State courts have reached different conclusions, which may be relevant where a party is considering invoking or has invoked an agreement to arbitrate in relation to claims based on EU competition law before EU Member State courts.  The CJEU’s most recent decision will likely enter into discussion as part of this emerging case law.

The first decision to consider CDC in relation to the interpretation of the scope of an agreement to arbitrate involved an action seeking damages based on an infringement of Article 81 of the (old) EC Treaty (now Article 101 TFEU).  In a 2015 decision that does not contain detailed reasoning, the Court of Appeal of Amsterdam concluded that there was no “good reason” to depart from the rule in CDC when interpreting an agreement to arbitrate drafted in general terms.  [see Kemira Chemicals Oy v. CDC, Case No. 200.156.295/01 (July 21, 2015, Court of Appeal Amsterdam), para. 2.16.]  The Court of Appeal did not consider in its decision the distinctions that exist between arbitration law and EU law related to jurisdiction clauses.

A second decision emerged from the English High Court in February 2017.  In its judgment in Microsoft Mobile OY (Ltd) v Sony Europe Limited & Ors [2017] EWHC 374 (Ch), the High Court was called upon to determine, in the context of an action alleging liability pursuant to Article 101 TFEU, an objection based on EU law to the enforcement of an arbitration agreement.  Specifically, while the defendant contended that the claim for damages under Article 101 TFEU was precluded by the parties’ agreement to arbitrate in their underlying supply agreement, the claimant contended that that agreement did not reach the claim because it did not refer expressly to disputes concerning liability incurred as a result of a competition law infringement.

The High Court first found that the relevant agreement to arbitrate reached the competition damages claim, relying on the inclusion in the contract of a specific obligation to negotiate pricing in good faith.  The English court observed that this obligation would in principle have been breached if pricing was distorted by the existence of a price-fixing cartel.  As such, there was a contractual claim falling within the arbitration agreement and the parties could be expected to have intended any tortious claim relating to the same matters to be resolved in the same forum as the corresponding contractual claim.  The fact that the claimant asserted a purely statutory tort claim and did not rely upon a contractual breach as the basis for its claims did not alter the court’s analysis, since the claimant could have alleged contractual claims.  To allow otherwise would have allowed the claimant to circumvent the agreement to arbitrate simply by choosing a tortious rather than contractual cause of action.

The High Court then addressed the question of whether enforcement of the relevant agreement to arbitrate would be contrary to the requirements of EU law, which would have rendered the agreement inoperable and ineffective if found to be the case.  In concluding that this was not the case, the English court considered at length the Opinion of Advocate General Jääskinen in CDC, whose arguments in that case were cited by the claimant in opposing arbitration before the High Court.  Specifically, the court was referred to the Advocate General’s argument that allowing arbitration would undermine EU law, including by causing the fragmentation of the relevant litigation.  The High Court rejected this line of argumentation, noting that the opinion had not been followed by the CJEU, whose judgment did not in any event reach arbitration. Thus, the High Court held as a matter of English law that the arbitration agreement could reach the tortious claim for breach of Article 101 based on an alleged secret cartel.

Most recently, in a judgment dated September 13, 2017, the Regional Court of Dortmund agreed to apply agreements to arbitrate drafted in general terms to cartel-related damages claims based on an infringement decision issued by the Bundeskartellamt (Federal Cartel Office). [seeLandgericht Dortmund, Judgment of September 13, 2017 – 8 O 30/16 Kart].  The court observed that, as a matter of principle, arbitration agreements need to be interpreted broadly.  Since parties usually do not intend to split claims based on contractual obligations (falling with the scope of an arbitration agreement) and based on statutory tort claims (to be litigated in court unless deemed to fall within the scope of an arbitration agreement), the court found that the arbitration clauses should be fully applied to the relevant competition-based claim.

While the case involved two German parties and was thus not subject to the then-Brussels Regulation, the court explicitly declined to follow the CJEU’s position on foreseeability in relation to jurisdictional clauses, observing that claims of an unforeseeable nature (such as claims alleging fraud), which are unknown to one party at the time of contracting, may validly be submitted to arbitration under German law.  The court questioned the proposition that the CJEU case law relating to jurisdiction clauses subject to EU law could be applied automatically to the interpretation of agreements to arbitrate, and further questioned the competence of the CJEU to interpret agreement to arbitrate in light of the fact that arbitration is expressly excluded from the scope of both the Brussels Regulation and the Brussels Recast.  Finally, like the English High Court, the Regional Court of Dortmund rejected an argument that the principle of effectiveness of EU law would require a different conclusion.

 

Potential Relevance Of Apple Sales International et al. v. EBizcuss.com To Arbitration

The CJEU’s most recent judgment provides helpful guidance in clarifying that the CJEU’s earlier case law should not be construed as automatically precluding reliance on jurisdiction agreements worded in general terms whenever a dispute relates to infringements of EU competition law.  The recent case law should have a similar moderating effect in relation to debates over the applicability of agreements to arbitrate drafted in broad terms, at least in relation to many disputes based on infringements of Article 102 TFEU.

The CJEU’s decision to cabin its earlier case law in CDC was not a foregone conclusion.  Prior to referring this question to the CJEU, France’s Court of Cassation, in an earlier 2015 decision in the same matter, construed CDC  as precluding application of the relevant jurisdiction clause to the distributor’s claims based on an infringement of Article 102 TFEU.  [see Apple Sales International et al. v. EBizcuss.com (C-595/17, October 24, 2018), para. 15 (citing the October 7, 2015 decision of the Court of Cassation).]

While the EBizcuss.com judgment appears to allow for greater contractual flexibility overall, the CJEU’s apparent attachment to the decisional logic of the CDC case law, which turns on the foreseeability of specific types of claims, raises questions.  Just as contracting parties rarely contemplate that their contractual relationship will be affected by a secret cartel, contracting parties rarely anticipate distortions resulting from abusive behaviors by a dominant undertaking.  Parties often seek through the use of broad language to refer all disputes arising out of a commercial relationship to one forum for adjudication, and thereby avoid debates over what might or might not have been foreseen at the time of contracting.

The practical importance of the distinction that has been drawn by the CJEU will depend on parties’ willingness to invoke agreements to arbitrate in relation to different types of claims involving EU competition law.  Whereas disputes related to abuse of dominance claims often are bilateral in nature, and thus may be addressed effectively through a single arbitration, cartel-damages actions in practice usually involve multiple parties and multiple contractual relationships.  Multi-party actions may be more difficult to manage through arbitration clauses found in agreements that bind only a limited number of parties to the action.  This limitation poses particular challenges in the context of damages actions based on EU competition law.  On the other hand, certain features of international arbitration can be particularly appealing for complex disputes involving EU competition law.  For further discussion of related issues, see here; A. Goldsmith, “Arbitration and EU Antitrust Follow-On Damages Actions,” 34 ASA Bulletin 1 / 2016, pp. 10-40 (2016).]

Competition-based claims and defenses are asserted regularly in the context of complex commercial arbitrations in Europe.  Thus, parties considering the use of arbitration for disputes involving European business interests would be well advised to follow the case law discussed in this post and to consider its ramifications both at the time of contracting and when any dispute involving violations (confirmed or alleged) of EU competition law breaks out.

 

The views expressed herein are those of the authors and should not be construed as necessarily reflecting those of their firm or of any of its clients. 




The CMA Remicade decision: discount schemes and abuse of dominance – effects matter!

Kluwer Competition Law Blog
April 25, 2019

Please refer to this post as: , ‘Dutch Antitrust Authority investigates Apple: does this support Spotify’s complaint?’, Kluwer Competition Law Blog, April 25 2019, http://competitionlawblog.kluwercompetitionlaw.com/2019/04/25/dutch-antitrust-authority-investigates-apple-does-this-support-spotifys-complaint/


Summary

On 14 March 2019, the UK Competition and Markets Authority (CMA) decided to close its investigation into a discount scheme by Merck Sharp & Dohme Limited (MSD).[1] The CMA concluded that there were no grounds for it to take action, since MSD’s discount strategy was not likely to limit competition in anticipation of the market entry of competitive products. In other words, when analysing a discount, the CMA looked at the likelihood of it having exclusionary effects. Since it found no such likelihood, the investigation was closed.

 

Background

In December 2015, the CMA opened a formal investigation, alleging that MSD had abused its dominant position by offering a discount scheme for the sale of Remicade in the UK, contrary to section 18 of the Competition Act 1998 and Article 102 of the Treaty on the Functioning of the European Union (TFEU).

MSD’s Remicade (molecule: infliximab) was the only product used in the UK to treat autoimmune inflammatory disorders before March 2015. After its patent expired on 24 February 2015, a number of biosimilar infliximab products entered the UK market (and specifically Inflectra by Hospira, Resmina by Napp, and Flixabi by Samsung Biopis). Infliximab is a prescription-only medicine purchased by the National Health Service (NHS) Trusts in the UK through tenders. In principle, the NHS’s purchasing decisions are based on cost-effective principles assessing the pharmaceutical companies’ pricing and tender offers against the market dynamics.

The under review rebate scheme for the sale of Remicade designed by MSD was based on (i) a matrix that set out a series of bands, with each band being associated to a specific price and a specific purchase volume of Remicade; and (ii) a quarterly review mechanism where purchases of Remicade would be reviewed against the volume specified in the matrix mentioned above. In the investigation at issue, the CMA examined whether MSD’s discount scheme was likely to lead to market foreclosing effects for the period from 1 April 2015 to 31 December 2015.

 

The CMA Decision

Market definition and dominant position

Although a final position on the definition of the relevant market and on MSD’s alleged dominance was not necessary given the absence of grounds for action, the CMA worked on the assumption that the relevant market was the supply of infliximab products in England, and that MSD held a dominant position therein.

The relevant product market was defined as Remicade and infliximab biosimilars. The CMA considered defining it more widely based on the products’ therapeutic substitutability. This criterion would have led to a wider product market, encompassing other biological medicines and TNF alpha inhibitors. However, the CMA chose a narrower market, notably because of the way the products were administered. Remicade was (at that time) administered in a hospital by intravenous injection. Other TNF alpha inhibitors were administered in the home by subcutaneous injection (a pen) by the patient himself. The CMA considered this an important difference. The relevant geographic market was limited to England because of the different tendering procedures through which infliximab was marketed in different parts of the UK.

As regards MSD’s alleged dominant position, the CMA took into account the following elements: (i) the biosimilars’ barriers to enter the market and grow due to the NHS’s clinical caution, (ii) the evolution of MSD’s market shares by volume and value, and (iii) the absence of constraints on MSD’s conduct through countervailing buyer power.

Discount schemes by dominant undertakings are not per se abusive

In terms of abuse, and in line with the ECJ case law in Intel, [2] the CMA’s starting point was that “not all discounts granted by undertakings in a dominant position are” abusive.[3] Rather, a variety of factors need to be assessed in order to determine the existence of an abuse. These factors include the rules applicable on the grant of the discount as well as the discount’s tendency to bar competitors from accessing the market, to strengthen the dominant position of the undertaking concerned and to influence purchasing behaviour.

On that basis, the analysis of the CMA focused on the likelihood of MSD’s discount scheme to produce exclusionary effects. Thus, the CMA, in line with Intel, highlighted the importance of the effects-based approach for the assessment of the abusive nature of the applicable discounts.

The likelihood of exclusionary effects matters

The CMA examined the rules applicable to the discount scheme and considered that it was designed with the alleged intention of disincentivising the NHS to switch to biosimilar products. The idea was that biosimilar suppliers would have to charge low prices in order to compete with MSD, essentially for contestable new patients. The CMA also considered that, at the time of the introduction of the rebate scheme, the NHS believed that the scheme could lead to exclusionary effects.

Nonetheless, the turning point for the CMA’s reasoning was the objective assessment of the circumstances of the market when the discount scheme was introduced in March and April 2015. The CMA clarified that “the likely effect of a dominant undertaking’s conduct should be assessed by reference to the point at which the allegedly abusive conduct was implemented rather than at some point after the allegedly abusive conduct had been in place”.[4] After carrying out detailed research which included surveying NHS staff, the CMA concluded that the market reality at that time prevented the scheme from developing any likelihood of exclusionary effects, since it proved that MSD’s assumptions as to the NHS’s potential reaction were incorrect. In effect, the NHS showed less clinical caution and a much greater willingness to use infliximab biosimilars. This meant that the pricing scheme did not have the alleged anticipated effect, as the contestable part of the market was much greater.

Thus, the CMA concluded that even if the discount scheme may have allegedly intended to exclude biosimilars from the market, there was no abuse of any dominant position since it could not practically lead to such anticompetitive effects.

The role of the as-efficient competitor test

In reply to MSD’s arguments, the CMA justified the choice not to apply the as-efficient competitor test (AEC price/cost test) in the Statement of Objections, by stressing that – despite being informative and useful – it was not required here. (While this may not entirely reflect the framework of analysis set out in the CJEU’s judgment in the Intel case, the CMA cannot be criticised for not applying the AEC price/cost test in a case where it closed the investigation based on no effects.) In any event, the CMA focused on the importance of assessing all the relevant circumstances of each case as stipulated in Intel.

 

Conclusions

The CMA’s decision that there were no grounds for action in respect of Remicade is a development that endorses the effects-based approach in unilateral conduct cases. The CJEU spelled out that the European Commission should carry out an economic analysis of the effects of discount schemes in cases of dominant undertakings. That is exactly what the CMA has done in the Remicade case. Without over-relying on the alleged intention of the dominant undertaking in question, it conducted its own economic analysis and concluded that MSD’s discount scheme could not have produced any anticompetitive effects. In short, the CMA confirmed that effects matter.


[1]       See the CMA’s statement at: https://www.gov.uk/government/news/cma-warns-businesses-after-ending-remicade-investigation

[2]       In Intel, the Court of Justice broke with its more formalistic previous case law on rebates and stipulated that: “the case law must be further clarified where the undertaking concerned submits, during the administrative procedure, on the basis of supporting evidence, that its conduct was not capable of restricting competition and, in particular, of producing the alleged foreclosure effects.” (par. 138)  In essence, the EU court underscores that, if evidence is adduced that rebates are not capable of foreclosing access to the market at issue, then dominant undertakings cannot be accused of any antitrust infringement irrespective of the discount system’s design.

[3]       See page 42 of the CMA’s Remicade Decision at: https://assets.publishing.service.gov.uk/media/5c8a353bed915d5c071e1588/Remicade_No_Grounds_For_Action_decision_PDF_A.pdf

[4]       Ibid, page 63.




The first competition case in the Presidium of the Supreme Court of the Russian Federation – resolution of the dispute between the FAS and the largest stevedore

Kluwer Competition Law Blog
April 25, 2019

Please refer to this post as: , ‘Dutch Antitrust Authority investigates Apple: does this support Spotify’s complaint?’, Kluwer Competition Law Blog, April 25 2019, http://competitionlawblog.kluwercompetitionlaw.com/2019/04/25/dutch-antitrust-authority-investigates-apple-does-this-support-spotifys-complaint/


Background and main problems

Novorossiysk Commercial Sea Port PJSC (“NCSP”) is one of the Europe’s largest port operator in terms of cargo turnover and the leader in the Russian market.

The Federal Antimonopoly Service (the “FAS”) found that NCSP abused its dominant position, by fixing and maintaining monopolistically high prices in 2015 for transshipment services of ore, fertilizers, containers, ferrous and non-ferrous metals, oil and petrochemicals in Novorossiysk seaport.

Tariffs for the services were fixed in USD. It was alleged that RUB devaluation in 2014-2015 led to serious increase of tariffs by NCSP, while market conditions did not fundamentally change.

Generally, the antitrust investigation in respect of stevedores began in June 2016, when the FAS initiated cases, on antitrust law violation, in respect of nine companies that provided services for the loading, unloading and storage of cargo in seaports of the Russian Federation. However, the NCSP case could be deemed as the most notable one.

The case raised two main problems.

  • Appropriate type and size of a sanction. After a number of complicated hearings, the FAS issued its decision and the prescription demanding, inter alia, that NCSP is obliged to transfer its unlawfully-gained income to the federal budget (approx. RUB 9.7 billion – around EUR 130 million, one of the largest sanctions in history of the FAS) instead of imposing a much smaller administrative turnover fine.

The problem is that, according to the law, the FAS can impose an administrative fine of up to 15% of a company’s annual turnover in the market where the violation occurred. At the same time, the law empowers the FAS to issue a prescription, obliging the transfer of unlawfully-gained income, resulting from violations, to the federal budget of Russia. According to the law, the FAS cannot do both. However, the latter may be much higher in terms of amount of payments and the appropriate size of the sanction is of primarily importance for companies.

Earlier, on July 08, 2016, the FAS issued clarifications stating that the FAS shall impose an administrative turnover fine, if it can be calculated, rather than issue an alternative prescription. However, this case, as well as some other similar ones, raised this question once more.

  • Market analysis. Narrow geographical boundaries were established (within the particular port) despite of some evidence on the contrary (if the market had been defined within the boundaries of sea basin). The accusatory decision was based on the presumption that NCSP is formally a natural monopoly and prices for the services on comparable markets were not analyzed.

 

The courts’ decisions

NCSP judicially appealed the decision and the prescription of the antimonopoly authority. Courts of three instances confirmed the illegitimacy of the FAS acts.

Nevertheless, the Judicial Chamber on Economic Disputes of Russia’s Supreme Court decided to unwind all of the courts’ decisions and sent the case, for new consideration, to the court of the first instance.

The Supreme Court’s decision confirmed that the company is subject to the special natural monopoly regulation and abuses its dominant position in the market, by raising tariffs for its services. In particular, the point was that the essence of the stevedore service is access to the special technological equipment and infrastructure of the particular port. Therefore, geographical boundaries were determined by the antimonopoly authority correctly. It was also mentioned that being a natural monopoly presumes absence of competitive environment in this market.

Shortly afterwards, the decision of the Judicial Chamber on Economic Disputes was appealed to the Presidium of Russia’s Supreme Court (the highest supervising instance in Russia), which finally, on December 26, 2018, supported the NCSP.

In particular, the Supreme Court mentioned that the FAS improperly determined the sanction. The Supreme Court clarified the findings of the courts, which recognized that the FAS, by demanding that the company is obliged to transfer income to the federal budget instead of imposing a smaller administrative fine, “improperly determined the sanction for the administrative offense, without clearly defining its application in a regulatory document”. This is very important because incorrect determination of sanctions, which could lead to a several-times higher liability, may seriously affect activities of companies operating in the Russian market.

Separately, the Supreme Court confirmed mistakes in the market analysis in terms of using narrower definition of geographical boundaries. The court also confirmed that being a natural monopoly does not exclude the necessity to duly conduct market analysis in terms of analyzing prices in comparable markets, when it comes to monopolistically high prices allegations.

This above-mentioned case is the first competition case revised by the Presidium of Russia’s Supreme Court, since the merger of the Supreme Court and the High Arbitration Court in 2014 and is the first economic dispute reviewed by the Presidium of Russia’s Supreme Court over the past two years.

 

Conclusion

The case showed that an economic approach should prevail over the formal one while determining dominant position and its abuse. Also, it confirmed that type and size of a sanction should be reasonably chosen and calculated.

The case will standardise administrative and court practice and we see that it has already started affecting similar cases involving large stevedores by applying the same approaches. Moreover, more careful approaches to market analysis could be expected as a result.




Boost to Dawn Raids in India – Supreme Court Rules Power to Search Includes Seizure as well

Kluwer Competition Law Blog
April 25, 2019

Please refer to this post as: , ‘Dutch Antitrust Authority investigates Apple: does this support Spotify’s complaint?’, Kluwer Competition Law Blog, April 25 2019, http://competitionlawblog.kluwercompetitionlaw.com/2019/04/25/dutch-antitrust-authority-investigates-apple-does-this-support-spotifys-complaint/


Introduction

The Supreme Court of India in its order dated 15 January, 2019 ratified the power of the Office of Director General (DG), i.e., the investigating arm of the Competition Commission of India (CCI) to conduct ‘search and seizure’ operations. This order comes in light of a protracted dispute with JCB India Private Limited (JCB). The CCI in 2014 had directed the DG to carry out investigation into the alleged abuse of dominance by JCB pursuant to an order passed under Section 26(1) of the Competition Act, 2002 (the Act). In line with this, the DG obtained a search warrant from the Chief Metropolitan Magistrate and conducted a dawn raid at JCB’s premises. JCB challenging the investigation, sought to set aside the search and seizure operation by moving the Delhi High Court. The High Court perfunctorily accepted JCB’s contention and restrained the CCI from using the seized material on the ground that the search warrant merely authorized the DG to search the premises and that the same could not be used for purposes other than search (impugned order). CCI against this backdrop preferred an appeal before the Supreme Court of India assailing the impugned order. Providing great impetus to the investigative powers of the DG, the apex court vacated the impugned order and affirmed the validity of the seizure proceedings. In the following section, the authors present the rationale of the Supreme Court in reaching this conclusion and concurrently analyze the same.

 

The Ruling – An analysis

Section 41(3) of the Act states that the DG has powers equivalent to that of an Inspector as given under Section 240A of the Companies Act, 1956 (CA1956). Under Section 240A (2) of CA1956, a sanction by the Magistrate of First Class empowers the Inspector:

“(a) to enter, with such assistance, as may be required, the place or places where such books and papers are kept;

(b) to search that place or those places in the manner specified in the order; and

(c) to seize books and papers he considers necessary for the purposes of his investigation.”

JCB in its contentions, firstly, endorsed a disjunctive reading of the above produced section and argued that the warrant obtained, being a ‘search warrant’ authorized the DG to merely search the premises and not seize any material. In other words, JCB pointed that the power to enter, search and seize materials were separate from one another and thus, each act required a distinct warrant for its execution. Secondly, JCB placed reliance on Section 240A (4) which reads, “every search or seizure made under this section shall be carried out in accordance with the provisions of the Code of Criminal Procedure, 1898.” The use of word ‘or’ between search and seizure is indicative of the dissimilar meaning and the corresponding mutual exclusivity between the two terms. However, the Supreme Court took a contrasting view and remarked that the provisions of Section 240A relating to search extend to authorization of seizure as well, because a search by itself will not be sufficient for the purposes of investigation. This line of thought stems from the underlying purpose of Section 41(3) of the Act which requires both search and seizure to allow the DG to reach a conclusive finding (page 12 of CCI Advocacy Series 4).

In the case of H.N. Rishbud and Inder Singh v. The State of Delhi, the Supreme Court concluded that an investigation under the Code of Criminal Procedure generally includes the search of places as well as the seizure of things considered necessary for the investigation. The authors opine, the fact that the verdict in JCB is in line with its judicial practice gives more validity to the ruling. Furthermore, a perusal of the prayer before magistrate for grant of search warrant reveals that the basic agenda behind solicitation of the search warrant was for recovering incriminating documents and papers related with the case. Therefore, the Supreme Court ruling finds its credence in sound logical argumentation and any criticism of the same would be both unwarranted and unfounded.

 

Suggested amendments

Section 41(3) of the Act still places reliance on the CA1956 which has been substituted by the Companies Act, 2013 (CA2013). Further the CA1956 stands repealed from 30th January, 2019. Thus, there is not only a need for amendment in the Act for this reason but also due to the fact that the CA2013 simplifies the process of obtaining a warrant of seizure. This is because the CA1956 under Section 240A requires the authorization of Magistrate of First Class for a warrant of seizure but the CA2013 under Section 220 does away with this requirement and merely requires compliance with the provisions of Code of Criminal Procedure, an investigation under which has been interpreted to include search as well as seizure. This is suggestive of an intent to facilitate easier conduct of dawn raids on part of the legislators.

The move of the Supreme Court in JCB as well as the foregoing suggested amendment will ensure that position of law in India is in consonance with the internationally recognized and mature competition jurisdictions like UK and USA. Section 194 of the Enterprise Act, 2002 of UK clearly provides that a warrant authorizes an officer of the Office of Fair Trading to not only search premises but also to take possession of any relevant document. In a similar vein, a warrant under Rule 41 of U.S. Federal Rules of Criminal Procedure issued by magistrate empowers search and seizure of person or property located within the district. This establishes that a mere search by itself is not considered sufficient for the purposes of an investigation in foreign jurisdictions as well.

 

Conclusion

The Indian competition regime is a continually growing one and progressive rulings like that of JCB play an important role in furthering the scope and ambit of the Indian competition watchdog. Dawn raids play a crucial role in snaring clandestine anti-competitive activities, which if, not for dawn raids, would continue unperturbed. There already have been three instances of dawn raids in India – JCB, dry cell battery markets case and beer cartel case; and this number is bound to escalate in light of the Supreme Court ruling.

Interestingly, in the dry cell batteries market case, the regulator saw the opposite party come forward and file a leniency petition three days subsequent to its premises being subject to a dawn raid. The fact that a 100 percent penalty waiver was allowed to the opposite party in the said case evinces the character of dawn raids as effective tools of detecting cartels. The authors believe that the Supreme Court has done great service to the Indian competition regime by giving a legal character to antitrust search and seizure operations. Expansion of search to include seizure has ensured that the power preserved under Section 41(3) of the Act does not remain a mere lip service.




Bundeskartellamt hits „don’t like“-button on Facebook

Kluwer Competition Law Blog
April 25, 2019

Please refer to this post as: , ‘Dutch Antitrust Authority investigates Apple: does this support Spotify’s complaint?’, Kluwer Competition Law Blog, April 25 2019, http://competitionlawblog.kluwercompetitionlaw.com/2019/04/25/dutch-antitrust-authority-investigates-apple-does-this-support-spotifys-complaint/


On February 7, 2019, Germany’s Federal Cartel Office (“FCO”) issued its long-awaited decision in the Facebook case, see press release and background paper in English here.

It qualifies Facebook’s current practice of collecting and matching data of its users from third-party services/websites, including on What’sApp and Instagram, without explicit consent as an abuse of dominance.  Facebook is ordered to stop the conduct within 12 months: it must obtain explicit user consent and change the current system by offering an opt-out solution.  Facebook must propose adequate solutions to the FCO within four months.

  1. Dominance in social networks market in Germany

Market definition

The FCO defines a separate (digital) social networks market in Germany.  Professional networks (LinkedIn and Xing), messaging services (WhatsApp and Snapchat) or other social media (YouTube or Twitter) are not in the same market, because they largely satisfy complementary needs.  Even if they were included, the FCO says Facebook’s market share would still be indicative of dominance – which is the case under German law with a market share exceeding 40%.  The market is limited to Germany, because German users mainly use Facebook to keep in touch with users in Germany.

Dominance

Facebook has a market share of more than 90% (by user numbers).  The large size of its network leads to high entry barriers.  The FCO finds that switching to other social networks is difficult, so that users are locked in.  Indirect network effects impede market entry – it would be difficult for a new entrant to offer services financed by advertising revenues, which requires a critical number of users.  The FCO did not find multi-homing in the social networks market.  In addition, Facebook has superior access to (user) data and thus a competition parameter in the market, because user data are important for the product design and allow monetarizing the services through advertising, which may create further entry barriers.

  1. Theory of harm – abusive t&cs due to violating data protection laws

Concept

The theory of harm is that Facebook abuses its dominant position through imposing abusive, i.e., exploitative, user terms and conditions (“t&cs”).  The concept of abusive business terms is stipulated in Section 19(2) no. 2 ARC, and does not only protect companies, but generally the demand side of a market, which can include consumers like here.  The FCO primarily relies on German caselaw pursuant to which violating legal principles (including civil law) can render business terms abusive.

Data protection laws violation

The FCO finds the t&cs are abusive as they violate data protection laws: they make using the social network pre-conditional upon Facebook’s vast data collection and combination, i.e., matching data from other websites or services to Facebook user accounts.  Facebook collects the data from users surfing third-party websites or apps with an embedded Facebook API (application programming interface), e.g., the Facebook “Like-Button”.  It then matches the data with the users’ Facebook accounts for additional data processing activities.  Users do not have to click on the said button – surfing on such sites is sufficient.  The FCO says that users are often unaware of this practice and have not been requested for explicit consent, which violates EU data protection laws.

Balancing of interests

In the balancing of interests exercise, required under German law, the FCO concludes that the vast data collection and matching is inappropriate as it is not necessary for Facebook in order to offer its social network services. In addition, users have no choice to opt out – even if they are aware of the practices, they can only take it or leave it.  Given that they are locked in, leaving is no realistic option.  The FCO weighs the user interest to have control over its personal data and the fundamental right of so-called informational self-determination higher than Facebook’s interests to collect and match the data.

  1. Termination of abuse through change of t&cs and system

The FCO requests that Facebook change its t&cs, including seeking explicit user consent for its data collection and matching practices and providing for an opt-out possibility: if users reject consent, they must be able to continuously use Facebook and should only be subject to very restricted data collection and matching.  This also applies to the Facebook-owned services What’sApp and Instagram.  The FCO argues that only with an opt-out solution any user consent can be deemed truly voluntary.

Regarding the restricted data collection and processing, the FCO mentions as example restrictions on the amount of data, purpose of use, type of data processing, additional control options for users, anonymisation, processing only upon instruction by third party providers, limitations on data storage periods, etc.  The FCO obliges Facebook to offer solutions how to implement this within the next four months, so it obviously wants to have a say in how Facebook intends to implement the decision.

  1. Comments

P2C-side in focus

The decision is a big bang.  It is first time that abuse proceedings into digital platforms concern the P2C (i.e., platform to consumer)-side of a multi-sided platform market.  (The Commission’s Google cases concerned the platform’s practices vis-à-vis companies, i.e., the P2B side.)

Precedents stress need for causality

While the concept of abusive terms is not new, the reliance on violating data protection laws as the key (and maybe only) reason for finding abuse is.  The quoted precedents stipulate that there must be a causal link between the company imposing an illegal clause and dominance.  The currently available materials on the FCO’s decision do not really deal with this aspect, other than generally finding that consumers are locked in.  Hopefully the decision can shed more light on this crucial point.

Additional theory of harm?

The background materials also allude to possible negative effects of Facebook’s practice in the online advertising market, i.e., on competitors (foreclosure due to superior data access) and online advertising customers (increased dependence on Facebook).  While this could be a more common theory of harm, it is unclear to which extent the FCO effectively relies on it.

Prohibition covers intra-group exchange

It is remarkable that any data flow between Facebook-owned services (What’sApp and Instagram) and Facebook is subject to the same obligations as third-party websites:  generally, a group of companies is treated as one single economic entity in abuse proceedings, which means that internal information exchange/flow is normally permissible.  (And unlike in the ongoing Commission Amazon probe, the data exchanged here do not include sensitive data from customers who are also competitors.)

Far-reaching obligations: “internal unbundling”

In essence, the FCO requires Facebook to change its business model – not only in terms of consent requirements and IT firewalls.  If a significant number of users refuses consent (which is unknown), Facebook cannot use their data for targeted advertising or other services in the same way as today.  The FCO’s president Mr. Mundt has described this as “internal unbundling”.  Potentially, this concept might serve as a blueprint for other abuse cases in digital platform markets.  While it is less severe than the sometimes-voiced request to break up digital giants, it essentially serves as a functional separation and limit on conglomerate strategies.

In fact, regarding What’sApp, the decision almost seems like a partial unbundling of or imposing post-clearance remedies in the Facebook/What’sApp merger, which may cause controversy:  In its clearance decision, the Commission did not consider potential conglomerate practices between the two services as a concern, given sufficient competition in the online advertising market(s).  (It also noted that any potential data protection violation issues would be a matter of EU data protection and not competition law.)

Limited geographic scope

Of course, the decision is limited to Germany.  Irrespective of whether one agrees with the decision, it does not seem meaningful to pursue this type of case only on a national basis.  The FCO says it has closely coordinated the case with Brussels.  But it seems that the Commission would not bring the FCO’s theory of harm under EU abuse of dominance proceedings.




A new standard for abusive denigration? Danish NCA condemns covert media campaign

Kluwer Competition Law Blog
April 25, 2019

Please refer to this post as: , ‘Dutch Antitrust Authority investigates Apple: does this support Spotify’s complaint?’, Kluwer Competition Law Blog, April 25 2019, http://competitionlawblog.kluwercompetitionlaw.com/2019/04/25/dutch-antitrust-authority-investigates-apple-does-this-support-spotifys-complaint/


Antitrust enforcers are good at regularly reminding the competition law community that the various forms of abuse of dominance listed in Article 102 TFEU are not exhaustive. Indeed, the idea of what conduct falls outside “competition on the merits” is ever evolving. And this can make it difficult for practitioners to set clear lines on what constitutes abuse and what does not. In fairness though, that is exactly what makes being a competition lawyer so interesting.

The Danish Competition Council’s (DCC) decision of 30 January 2019 is one of these reminders. The DCC holds that following a failed tender bid for the provision of ambulance services, the company Falck devised a comprehensive internal and external communications’ strategy to make it difficult for the winning bidder, BIOS, to effectively take over the services it had won ahead of Falck. The purpose of this strategy, says the DCC, was to generate uncertainty and concern among all stakeholders about BIOS’ ability to carry out these critical services and to effectively prevent BIOS from recruiting the necessary number of paramedics that were a scarce resource.

The DCC bases its decision on the broad-stroke “competition on the merits” standard while expressly acknowledging that there is no case-law serving as clear precedent for condemning Falck’s conduct.

Despite the lack of clear precedent, the decision does contain elements similar to the decisional practice on abusive denigration developed particularly in France. There are also similarities with the CJEU’s Article 101 TFEU judgment last year in the Hoffman-La Roche/Novartis case and, to a lesser extent, the European Commission’s AstraZeneca case from a while back. The novelty in the Falck case is that the DCC places little emphasis – if any – on whether Falck’s statements & communication were wrong or misleading in substance. According to the DCC, it was sufficient to find that the communication strategy was systematic, covert and likely to make life difficult for the competitor, BIOS. And it was therefore abusive, even if it conveyed information that was correct in substance.

As Falck has announced that it will not be appealing the decision, the precedent stands, and dominant companies are well-advised to review their public relations strategies & measures, in particular if they make use of external PR agencies or other intermediaries.

Falck’s reaction to losing the tender for ambulance services to BIOS

Falck is the world’s biggest international ambulance service company and in Denmark more or less synonymous with emergency medical services and roadside assistance. In 2014, the Region of Southern Denmark tendered ambulance services for four areas in the region, three of which were won by BIOS, a new Dutch-owned entrant, and none won by Falck.

Upon losing the tender, the DCC finds that Falck embarked on a strategy targeting BIOS and consisting of secretly feeding the press with negative stories about them. Falck also sought to dissuade Falck’s paramedics from seeking employment with BIOS, making it more difficult for the latter to deliver on the contracts it had won.

In July 2016, BIOS (Denmark) went into bankruptcy, and from August 2016 the Region of Southern Denmark took over the provision of ambulances service in the region.

The “Masterplan”, the PR agencies and the “double arms’ length” communication

The DCC describes Falck’s PR strategy as two-fold in that it targeted external stakeholders by way of the general public, politicians, and media outlets in the region of southern Denmark as well as internal stakeholders by way of paramedics employed with Falck.

Falck’s communications strategy comprised feeding negative stories about BIOS to the press and mobilizing opinion ambassadors that would publicly make the case for Falck and against BIOS. To this end, Falck engaged a PR agency that drafted what it called a “Masterplan,” laying out the most effective way to pitch stories to the press and taking various other measures. Part of the strategy was to ensure that the stories about BIOS were not traceable back to Falck. This served to ensure that the stories were seen as objective and as credible as possible. The DCC summarizes this in the following:

“A large part of Falck’s exclusionary activities were carried out at “arms’ length” – and in some cases “double arms’ length” – from Falck through [Communications agency X], [independent communications consultant] and [Communications agency Y]. “Arms’ length” is to be understood as meaning that the specific activities could not be traced back to Falck as the real initiative taker and sender, and the activities were thus perceived as objective and credible by the paramedics in Falck and the general public.” (translated from Danish)

One measure that the DCC describes in particular detail is how Falck created a group on a social media platform that appeared to be managed by a particular paramedic:

“The purpose of the [social media group] was to create uncertainty and concern for the paramedics in the region about BIOS as an employer and to mobilise the paramedics, their families, friends, and other members of the public in protest over BIOS as the future provider of ambulance services. The goal was for paramedics to refrain from seeking employment with BIOS. […] At the instruction of Falck, [Communications agency X] followed the group’s activity and on a daily basis provided possible news topics that the [Paramedic in Region South] could post in the group. […] Moreover, Falck was in possession of log-in and password to [Paramedic in Region South’s] profile on the [social media platform]. Falck thus had direct access to the group and could post, approve, and reject posts and remove comments on  posts in the group.” (translated from Danish) 

An important aspect of the case is that BIOS had to recruit a large number of Falck’s paramedics in order to supply the services specified in the tender. With qualified paramedics being a scarce resource, Falck’s strategy, according to the DCC, was to prevent BIOS from recruiting these in sufficient numbers. This is why the communications strategy was to target not only public stakeholders but Falck’s own paramedics in particular. An attempt at input foreclosure of sorts.

Departing from the established standard for abusive denigration

The legal benchmark for considering commercial disparagement of a competitor to be abusive has generally been that the negative communication must be wrong or misleading on substance, or at least based on unverified assertions. The French case-law – e.g. the Sanofi decision in 2013 upheld by the Cour de Cassation in 2016 – seems to be a testament to this benchmark, as does the CJEU’s judgment in Hoffman-La Roche/Novartis last year, although that was an Article 101 TFEU case. In a similar vein, in AstraZeneca, the European Commission took issue with information provided to the patent office on the basis that the information was misleading although that was not a denigration case as such.

In the Falck decision, however, the DCC does not seem to consider it all that significant as to what extent Falck’s internal and external communication was misleading in substance:

“The Authority finds moreover that, even if some of the published stories are not factually inaccurate, the fact that (parts of) the implementation of Falck’s communications strategy was carried out through an “arms’ length’ principle” – and with respect to the [social media group], a “double arms’ length principle” – is in itself misleading and manipulative. Nor does this constitute “competition on the merits” according to the Authority.” (translated from Danish)

In other words, the DCC’s position is that abuse can be established – even if the disparaging statements are completely accurate on substance – in cases where the communication to the market has taken place in a covert manner to create an air of credibility and objectivity. For example, by making use of an intermediary.

A lawyer’s abuse case and no ne bis in idem

As abuse of dominance cases go, this decision does not contain much elaborate economic analysis by way of graphs, curves or equations and the like. And although the DCC does go into how the market works and how Falck’s behavior was to be seen in that context, the decision reads more as one of unfair commercial & marketing practices. In fact, Falck did argue that boundaries as to commercial communication should be dealt with under the rules on marketing practices and not under competition law / antitrust.

The DCC rejected this argument saying that the potential application of marketing practices rules to Falck’s conduct does not preclude enforcement under antitrust rules. This is essentially the same reasoning as the Bundeskartellamt has just used in its Facebook case regarding data protection law vs. competition law. Intervention under one legal regime does not prevent sanctioning of the same conduct under another and there is no (successful) double jeopardy or ne bis in idem argument to be made here it seems.

The DCC’s press release in the case is available in English here.




The biggest Bulgarian telecom operator A1, member of Austria Telekom Group, fined for abuse of stronger bargaining position

Kluwer Competition Law Blog
April 25, 2019

Please refer to this post as: , ‘Dutch Antitrust Authority investigates Apple: does this support Spotify’s complaint?’, Kluwer Competition Law Blog, April 25 2019, http://competitionlawblog.kluwercompetitionlaw.com/2019/04/25/dutch-antitrust-authority-investigates-apple-does-this-support-spotifys-complaint/


A fine of slightly above EUR 400 000 was imposed to the biggest Bulgarian telecom operator – A1 Bulgaria, member of Telekom Austria Group (“A1”) for abuse of stronger bargaining position in its contractual relationship with its former sales representative – Handy Bulgaria (“Handy”).

The relationship between the two companies started back in 2005, when Handy started rolling a large store network across the entire country to distribute A1 products, mainly mobile telephone subscription plans and mobile phones. In 2017 the telecom operator decided to terminate the sales representation agreement. Handy, being almost fully dependent by the legal and economic relation with A1, saw this act as abuse of stronger bargaining position in the meaning of Art. 37a of the Bulgarian Competition Protection Act (“CPA”). In addition to the claim for the termination of the agreement, Handy arose allegations concerning the imposition of unjustified conditions under the agreements, i.e. unrealistic targets combined with coercion of Handy to divest part of its points of sales to A1 and termination of marketing activities of A1 in Handy’s shops.

The breach under Art. 37a CPA might be seen as specific for Bulgarian competition law. It has been introduced in the CPA in 2015 and is not yet much developed in the case-law of the Bulgarian Commission for Protection of Competition (“CPC”). The key procedures are these for A1, BTV (one of the private national TVs in Bulgaria), the pharma company AMGEN and Siemens. The breach basically comprises abuse of a stronger bargaining position under a specific relationship, where the economically weaker party may be or is harmed. The CPC sets a test for analysis when a company is considered in a stronger bargaining position., i.e.:

  • The characteristic and the structure of the specific market (defined for the purposes of this breach);
  • The characteristic and the structure of the relationship between the two companies;
  • The characteristics of the activity of the companies and its scale;
  • The possibility for alternative commercial partner, including alternative distribution channels, supply/delivery channels, etc.

First, the CPC had to deal with the period of the breach. As mentioned, Art. 37a was effective as from 24 July 2015, however the relationship between A1 and Handy dated back in 2005. The CPC decided that the relevant period starts in 2015 and finishes on the date of its decision (22 November 2018) and although the agreement between Handy and A1 was signed in 2005, it has effect also after 25 July 2015 and will be analyzed for the purposes of Art. 37a CPA.

Second, the CPC had to deal with the presence of A1’s stronger bargaining position that harms or may harm Handy. In this part, commission’s analysis is predominantly economic and comprised:

  • Definition of the product and geographic market (market of distribution of telecommunication services on the territory of the Republic of Bulgaria);
  • Analysis of the structure of the market (CPC took the net turnover of A1, its position compared with the remaining two telecom companies (Telenor and BTK), the services its offers, etc.);
  • The position of A1 and Handy in their contractual relationship (CPC found that Handy was almost fully dependent by A1 as its sales representative only to A1 and along with the presence of a clause prohibiting Handy’s cooperation with A1’s competitors, the former is seen as economic weaker party);
  • Analysis of Handy’s possibility to find a partner alternative to A1 (having in mind the prohibition to cooperate with A1’s competitors and the limited number of telecom operators in Bulgaria, the CPC found Handy’s possibilities to find alternatives to A1 as almost fully limited).

In light of the above criteria, CPC decided that A1 has a “stronger bargaining position” under the sales representation agreement with Handy.

Concerning the harm that is caused or may be cause out of the actions of A1, CPC decided as follows:

  • The imposition of unjustified conditions under the agreement, i.e. increase of targets combined with coercion for divestment of point of sales from Handy to A1 and termination of marketing campaigns of A1 in Handy’s shops cannot be seen as abuse actions by means of Art. 37a, since Handy agreed to the new targets (marked as confidential information in the decision of the commission), which the CPC does not found as abusive and simultaneously failed to prove in the course of the proceedings before the CPC that A1 coerced it to divest its point of sales to the latter.
  • The termination of the sales representation agreement, however, is considered as breach of Art. 37a CPA, since the grounds for termination of the agreement with Handy (marked as confidential information in the decision) are abusive and cause harm to Handy, being the economically weaker party under the agreement with A1. The CPC found that the A1’s reference to non-compliance with the targets under the agreement are wrong, since based on the excerpts form the systems of both A1 and Handy, the latter complied systematically with the set targets for performance for 2015, 2016 and 2017 (except for the last trimester of 2017). In addition, the CPC ruled that A1 groundlessly reduces Hady’s performance by deduction form its results of the new activations that are terminated or unfinished.

In light of the above factual, legal and economic background and taking into consideration the termination of the business of Handy as result of the terminated agreement with A1, the CPC imposed a sanction to A1 in the amount of BGN 804 340 (approximately EUR 400 000). The decision is subject to appeal before the Supreme Administrative Court within 14 days as from party’s notification and as from publication on the webpage of the CPC for any third interested party (who also make seek private damages by means of Art. 105 et seq CPA).

The decision was widely reflected by the Bulgarian media and discussed among competition law professionals. The particular interest derives from the fact that A1, being the biggest telecom company in Bulgaria, is involved, from the fact that termination of a business of sale representative of a telecom company in Bulgaria due to agreement termination by the telecom is not a precedent and from the fact that the decision concerns Art. 37a, which as mentioned, is relatively new and poorly developed in our caselaw. It worth mentioning that this time the CPC conducted relatively structured and detailed factual, legal and economic analysis. The commission also provides a helpful overview on what the test for application of Art. 37a CPA, elaborated by the authority is.

 




Competition Disputes: Will a Jurisdiction Clause Govern Where They are Brought?

Kluwer Competition Law Blog
April 25, 2019

Please refer to this post as: , ‘Dutch Antitrust Authority investigates Apple: does this support Spotify’s complaint?’, Kluwer Competition Law Blog, April 25 2019, http://competitionlawblog.kluwercompetitionlaw.com/2019/04/25/dutch-antitrust-authority-investigates-apple-does-this-support-spotifys-complaint/


The European Court of Justice (CJEU) held recently in Apple Sales International v MJA acting as liquidator of eBizcuss.com[1] that claims alleging abuse of a dominant position could come within the terms of a jurisdiction clause even where the clause did not expressly refer to claims based on competition law.

 

Relevant Rules

The rules of jurisdiction within the EU are designed to be transparent and predictable. Brussels I Recast, which subject to some exceptions, governs civil and commercial claims, provides as a default position that persons domiciled in a Member State shall, whatever their nationality, be sued in the courts of that Member State.

Apart from this rule of general jurisdiction, there are certain rules of special jurisdiction.  Article 25 of Brussels I Recast, the successor of Article 23 of the Brussels I Regulation[2], provides that if parties, regardless of their domicile, agree that the courts of a particular Member State are to have jurisdiction to settle any disputes between them in connection with a particular legal relationship, those courts shall have jurisdiction, unless the agreement is null and void under the law of that Member State. This jurisdiction is exclusive unless the parties have agreed otherwise.

 

Background

eBizcuss was an Apple distributor in France and sued Apple in France for abuse of a dominant position alleging that Apple was favouring its own distributor network.   However, a clause in the distribution agreement between the parties provided that the agreement and the corresponding relationship between them was to be governed by Irish law, with the parties also submitting to the jurisdiction of the Irish courts.   Notably the clause was framed in general terms and contained no reference to disputes concerning liability incurred as a result of an infringement of competition law.  The French courts had to consider the effect of this provision on their jurisdiction to hear the matter and in that context questions were referred to the CJEU for a preliminary ruling on the interpretation of Article 23.

Of relevance to the issues arising was the CJEU’s decision in CDC Hydrogen Peroxide[3] which held that a jurisdiction clause which abstractly referred to disputes arising from contractual relationships did not extend to a dispute relating to the alleged liability in tort of one party to that contract following its participation in an unlawful cartel.  This was on the basis that the other party to the contract which suffered the loss could not reasonably have foreseen such litigation when it agreed to the jurisdiction clause.  Where it had no knowledge of the unlawful cartel involving the other party, the CJEU found that the litigation could not be regarded as stemming from the contractual relationship.  Therefore, in the context of those proceedings arising from the activities of a cartel, the CJEU’s view was that Article 23 permitted a jurisdiction clause in a contract for the supply of goods to only be taken into account where the clause expressly referred to disputes concerning liability incurred as a result of an infringement of competition law.

 

The Latest CJEU Decision 

The CJEU held in the Apple case that, in an action for damages brought by a distributor against its supplier for abuse of a dominant position, Article 23 should be interpreted as meaning that the application of a jurisdiction clause within the contract binding the parties was not excluded solely because the clause did not expressly refer to disputes relating to liability incurred as a result of an infringement of competition law.

The CJEU noted that a jurisdiction clause can only apply to disputes arising in connection with a particular legal relationship. This avoids a party being taken by surprise by the assignment of jurisdiction to a certain forum for all disputes which may arise out of its relationship generally with the other party but which may stem from a relationship other than that in connection with which the jurisdiction clause was agreed.

In reaching its decision the CJEU distinguished two types of anti-competitive behaviour:

  • anti‑competitive conduct in the form of an unlawful cartel and arising under Article 101 TFEU, which is in principle not directly linked to the contractual relationship between a member of a cartel and a third party affected by that cartel; and
  • abuse of a dominant position arising under Article 102 TFEU, which can materialise in contractual relations involving an undertaking in a dominant position.

In respect of the latter, the CJEU noted that it could not be regarded as surprising to one of the parties that a court would take account of a jurisdiction clause that referred to a contract and “the corresponding relationship” arising from it.

 

Comment

The Apple judgment reiterates the need for drafters of jurisdiction clauses to focus on the question of where disputes concerning liability incurred as a result of an infringement of competition law are to be dealt with and for antitrust litigators alike to take cognisance of the terms of such clauses prior to initiating proceedings.


[1] Apple Sales International  v MJA acting as liquidator of eBizcuss.com, Case C-595/17.

[2] While there are some differences between the two provisions, those aspects of Article 23 considered by the court are reproduced in Article 25 of the new instrument.

[3] CDC Hydrogen Peroxide, Case C‑352/13.