Europe: private practice perspective

Directorate General for Competition is the enforcement authority for Europe. Read their profile

The following summary highlights key developments in EU competition law between July 2020 and September 2021. In addition to key European Commission (Commission) decisions, it covers selected judgments of the EU courts (the General Court and the Court of Justice) as well as significant policy initiatives.

Mergers

Although the Commission did not block any merger during the reporting period, there have been several highly relevant developments in the area of merger control. These developments concern both the substantive assessment of mergers in the digital economy as well as procedural developments which have been largely motivated by the debate around large digital platforms, but have consequences across all economic sectors.

Commission approves Google’s acquisition of Fitbit, requiring remedies to address alleged competition concerns but dismissing concerns not related to competition law

On 17 December 2020, the Commission conditionally approved Google’s acquisition of Fitbit, the maker of well-known wearable fitness devices such as smartwatches and fitness trackers. Reflecting the ongoing and intense debate about the role of large digital platforms and the application of competition law in the digital economy, the transaction had been opposed by various interest groups which raised a host of concerns over issues ranging from data privacy to health data and informational advantages that would allow the exploitation of users and employees. Google/Fitbit therefore served a test case to see to what extent the Commission might respond to pressure to include ‘non-conventional’ concerns in the analysis of a merger involving a prominent digital player. In the end, the Commission was not willing to take these wider concerns into account.

The Commission, however, did find a number of competition law concerns related to Google’s use of data collected through Fitbit devices and the interoperability between Fitbit and Google’s Android devices, which Google was required to address by way of commitments.

First, the Commission considered that the merger could harm competition in online advertising, in which Google was found to have a dominant position, by giving Google access to Fitbit’s user health and fitness database. This would, in the Commission’s view, increase the amount of data that Google could use to generate personalised advertising, making it more difficult for rivals to enter or compete in the ‘ad-tech’ ecosystem. The Commission’s theory of harm is remarkable because it is reminiscent of an ‘efficiency offence’, as any competitive advantage Google would derive from the Fitbit data would be based on the improvement of its search advertising services, thus benefiting consumers. Nevertheless, Google committed to maintain Fitbit health and wellness user data separate, not to use such data to generate advertising through Google Ads, and to allow users to choose whether such data are shared with other Google applications.

Second, the Commission feared that Google might restrict competitors from accessing the Web Application Programming Interface (Web API) that provides services to Fitbit users and allows them to upload their health data to Fitbit or third-party devices. This might, in the Commission’s view, stifle the emergence of start-ups in the growing European digital healthcare space. The Commission reached this conclusion even though it may be questionable whether Google would have any incentive to degrade the compatibility of third-party devices. To address the Commission’s concerns, Google committed to maintaining third-party access to user data through the Fitbit Web API, and to license to all Android OEMs the public APIs covering the functionalities needed for Android wrist-worn devices to interoperate with Android smartphones.

Finally, the Commission noted that the merger might incentivise Google to degrade the Android smartphone interoperability of rival wrist-worn devices that compete with Fitbit. Google agreed to maintain access to users' health and fitness data to software applications through the Fitbit Web API.

The commitments entered into by Google will last for 10 years, but the Commission may extend the duration of the advertising commitments by an additional 10 years if it can justify an ongoing need to do so.

The Commission’s Guidance on referrals under article 22 EUMR increases uncertainty for many transactions

The new guidance on article 22 referrals

On 23 March 2021, the Commission published Guidance outlining its new policy for referrals under article 22 of the EU Merger Regulation (EUMR). In a sharp deviation from prior practice, whereby the Commission considered article 22 referrals appropriate only for transactions that are reviewable under member state merger control laws, the Guidance encourages referrals even for certain transactions that fail to meet the notification thresholds under both the EUMR and national merger control laws. This shift in the Commission’s longstanding policy could signal a new era of expansive Commission powers and of increased uncertainty for merging parties.

The Guidance reflects the Commission’s attempt to address, without changing the EUMR’s notification thresholds, so-called ‘killer acquisitions’, whereby an incumbent acquires a nascent rival – that has a very limited market presence and therefore very limited revenues – before it can develop into a competitive threat. The term ‘killer acquisitions’ has become a widely – and uncritically – used label especially in discussions of the digital economy, although there is very little to no evidence that this ‘acquire to kill’ strategy has commonly motivated acquisitions in the digital space.

The Guidance seeks to provide some predictability by explaining that ‘normally’ the new policy would cover transactions where the turnover of one party does not reflect its actual or future competitive potential (eg, because it is a start-up, an important innovator, an actual or potential important competitive force, has access to competitively significant assets, or provides products or services that are key components to downstream industries). This broad list is, however, purely ‘illustrative’. Nothing would prevent the Commission from accepting any case that it considers worth reviewing, provided the low statutory conditions are met, that is, that a transaction affects trade between member states and that the referring national competition authority has demonstrated that the transaction could prima facie raise significant concerns.

The Guidance also notes that the Commission ‘generally’ does not intend to examine cases more than six months after closing. Nevertheless, the Guidance firmly asserts that the Commission retains the right to examine (and potentially undo) any deal, no matter how long after closing, based on its assessment of the magnitude of the competition concerns involved or the potential effect on consumers.

The Guidance purports to provide ‘transparency, predictability and legal certainty’ to merging parties. A senior Commission official even asserted that the Guidance would increase legal certainty for market participants, compared to the previous article 22 practice. This assessment, however, is certainly not shared by most market participants. Interestingly, even some member states, most notably Germany, have taken the position that they would not refer cases to the Commission under article 22 unless a transaction meets the national notification thresholds.

Illumina/Grail – testing the Commission’s powers under the new referral policy

Shortly after issuing the Guidance, the Commission accepted its first article 22 referral under the new policy on 20 April 2021, asserting jurisdiction over the lllumina/Grail transaction involving the acquisition of US cancer detection test start-up Grail by the US genomics firm Illumina. The Commission asserted jurisdiction although the acquisition was not notifiable at EU or member state level, and Grail has no activities at all in the EU.

Although Illumina was required to notify the transaction to the Commission, it has appealed the Commission’s decision to assert jurisdiction under article 22 before the General Court, thus testing the legality of the new referral policy. And after the Commission decided to open a Phase II investigation, Illumina decided not to wait for the conclusion of a lengthy review of a deal over which it believed the Commission should never have asserted jurisdiction in the first place. Instead, it announced that it would close the acquisition, but would hold Grail separate until the Commission’s review was complete. This prompted the Commission to pursue a gun-jumping investigation, and it is also expected to issue interim measures with a hold separate obligation. It will likely conclude that the standstill provision was violated and one might expect a significant fine. However, the Commission’s ability to impose a fine will ultimately turn on whether it has jurisdiction at all, as determined by the case pending before the General Court.

The Commission seized on Illumina/Grail as a first chance to flex its jurisdictional muscles under the new article 22 EUMR referral policy. Illumina’s willingness to fight back at every turn has created a significant test case to determine the limits to the Commission’s jurisdictional reach.

Abuse of dominance

Two judgments, presenting different factual backgrounds, allowed the European Courts to clarify the boundaries of the demanding refusal to deal test established in Bronner, and on both occasions the Courts rejected attempts to extend Bronner beyond outright refusals to supply.

Slovak Telekom – Bronner strict refusal to supply test not applicable when dominant firms offer access on unfair contract terms

In 2014, the Commission had imposed fines on Slovak Telekom and its parent company, Deutsche Telekom, for abusing its dominant position on the Slovak market for broadband internet services. Slovak Telekom, the incumbent telecom operator, provided unbundled access to its fixed-line telecommunications network (‘local loop’) to allow other operators to compete in accordance with regulatory obligations, but the Commission found that Slovak Telekom had infringed article 102 because it failed to provide alternative operators with fair terms in granting this access to its local loop network.

On appeal from the General Court, which had upheld this part of the Commission’s analysis, the Court of Justice (ECJ) rejected Slovak Telekom’s argument that the unfair contract terms should have been considered an implicit refusal to grant access to the infrastructure and that, therefore, the Commission was required to establish that the strict conditions identified by the ECJ in Bronner were met, including the requirement to prove that the dominant firm’s infrastructure was indispensable for competitors. The ECJ clarified that the Bronner requirements did not apply because Slovak Telekom did not refuse access to its network – indeed, it was required by EU telecommunications regulations to grant access – but merely made access more difficult through the imposition of certain terms and conditions.

The ECJ therefore declined to expand the scope of the strict test set down in Bronner to situations where the conduct does not rise to the level of an outright refusal to supply. In this case, the applicable regulatory framework meant that a refusal to supply strategy was not possible for Slovak Telekom. As it was under a regulatory duty to supply, the only conduct that could possibly fall under article 102 was its use of allegedly unfair access terms. But the ECJ’s language goes beyond the specific facts of the case and appears to establish a broad, generally applicable principle.

Somewhat counterintuitively, the ECJ’s ruling means that it will be easier to challenge conduct that is alleged to make access unreasonably difficult, even if an outright refusal to grant access altogether would not be abusive. In making it harder for dominant firms to defend strategies that are considered to make access more difficult for competitors without denying them access altogether, the ruling would seem to create incentives for dominant firms to refuse to supply in the first place in order to avoid the (greater) risk that the terms at which they would be willing to supply could be found abusive.

Lithuanian Railways – dominant firm’s destruction of assets not subject to Bronner’s strict refusal to supply test

In its 18 November 2020 Lithuanian Railways judgment, the General Court confirmed that the Commission had applied the correct legal test when finding that Lithuanian Railways had abused its dominant position on the national rail freight market by dismantling infrastructure that a rival rail freight operator was planning to use to compete with Lithuanian Railways.

Lithuanian Railways, which was both the railway infrastructure manager and a provider of rail freight transport services, had agreed with Orlen to transport Orlen’s oil refinery products to the Lithuanian seaport of Klaipėda. When Orlen sought to shift its export business to alternative seaports in Latvia, and to engage the services of the Latvian national railway company LDZ which would transport its products via a short railway line connecting the two countries, Lithuanian Railways, acting in its capacity as rail infrastructure manager, dismantled the railway line in question, citing damage to the tracks. As a result, Orlen was forced to abandon its plans to engage the services of LDZ.

On appeal from the Commission decision finding that Lithuanian Railways had infringed article 102, Lithuanian Railways argued that its failure to ensure access to the railway track should have been assessed in light of the case law on refusal to provide access to essential facilities, and that the Commission had failed to establish that exceptional circumstances identified in Bronner existed in this case.

The GC rejected this argument, reasoning that the stricter Bronner test, which must be met before a dominant company can be compelled to provide competitors with access to its infrastructure, are designed to protect the incentive to invest in such infrastructure in the first place. This incentive to invest was not at stake in this case as Lithuanian Railways’ dominant position derived from a former state monopoly and the infrastructure in question had been built using public funds. In addition, the General Court reasoned that the Bronner standards are meant to protect the infrastructure owner’s exclusive right to exploit its investments. However, in this case, Lithuanian Railways was already under a regulatory obligation to grant access to public railway infrastructure and to ensure that the infrastructure was in good technical condition. Therefore, it was not necessary to apply the Bronner test in a competition law analysis, as the necessary balancing of the economic incentives had already been carried out by the legislature.

Vertical restraints

Territorial restraints featured prominently in a Commission decision as well as an ECJ judgment, both confirming a continued strict approach to territorial restraints under article 101 TFEU.

PC gaming platform operator Valve and five video game publishers infringed article 101 by using geo-blocked activation keys

On 20 January 2021, the Commission announced that it had found that the US company Valve, owner of the popular online PC gaming platform ‘Steam’, and five publishers of video games had infringed article 101 by implementing cross-border restrictions on the sale and use of video games in the EEA.

Steam is one of the world’s largest online PC gaming platforms. In addition to directly downloading PC video games from the Steam platform, users can also activate a game on the platform after purchasing a physical or digital version elsewhere. In that case, players can use so-called ‘Steam activation keys’ (keys), which are supplied by Valve to publishers, who in turn provide them to distributors, and are included in the physical or digital version of games.

The Commission found that Valve and the publishers restricted cross-border purchases of certain video games by means of geo-blocked Steam activation keys, which prevented activation of the games by users outside certain territories in Central and Eastern Europe. For example, the territory control function could be used to prevent a game (physically or digitally) bought in the Czech Republic being activated by users located in France. This prevented passive sales of the games to parts of the EEA where they could not be activated. In addition, certain of the publishers included cross border passive sales restrictions in agreements concluded with distributors. These practices were found to partition the EEA market, denying consumers the benefits of the Digital Single Market, and thereby violated article 101 TFEU and article 53 of the EEA Agreement.

The case is a further example of the recently reinvigorated enforcement practice of the Commission against cross-border sales restrictions, in particular restrictions of passive sales. The Commission has taken the robust view in recent cases that contractual restrictions on cross-border sales may infringe article 101 even where those sales would infringe intellectual property rights.

A novel feature of this case appears to be the application of article 101 to the use by the publishers of activation keys to achieve geographic restrictions on the use of their products. Geo-blocking practices which do not involve contractual restrictions on distributors would normally amount to unilateral conduct and thereby fall outside of the scope article 101. However, in this case the Commission brought the practice within the scope of article 101 by virtue of the upstream contractual relations that existed between the publishers and Valve, which supplied the keys. The full decision is not yet published at the time of this writing, and it will be interesting to analyse the full facts and precise legal reasoning applied by the Commission, as, for example, a supplier of inputs which permit a customer to engage in geo-blocking in respect of the latter’s own products may not, at least generally speaking, assume that it could be liable for that conduct.

Canal + – Court of Justice confirms that absolute territorial protection in licensing agreements can be a by object infringement but annuls Commission decision for failure to consider adverse effects on third-party interests

On 9 December 2020, on appeal by Canal +, the European Court of Justice (ECJ) annulled the General Court’s (GC) judgment that had confirmed the Commission’s commitments decision in the Paramount case on the ground that that the Commission had failed to take the adverse effects on third-party interests into account when accepting Paramount’s commitments. On substance, however, the ECJ confirmed the GC’s finding that licensing agreements which confer absolute territorial protection on licensees likely could be considered a restriction of competition by object.

After opening an investigation into licensing agreements by US film studios for pay television services, the Commission raised concerns regarding: (i) territorial exclusivity clauses by which a studio would grant an exclusive territorial licence to a broadcaster, while at the same time committing not to grant to any other third party any licensing rights on the territory concerned; and (ii) clauses which prevented broadcasters from responding to any unsolicited service requests from customers located in a member state different from that of the broadcaster. The Commission took the preliminary view that these clauses had the object of eliminating cross-border competition between broadcasters in breach of article 101 TFEU.

To address these competition concerns, the US film studio Paramount offered a commitment, made binding by a 2016 Commission decision, whereby it would not implement the contested provisions over a five-year period. Paramount notified broadcasters – including French broadcaster Canal + – that it no longer intended to ensure compliance with the absolute territorial exclusivity rights granted to them. However, Canal + took the view that the contractual clauses in question did not restrict competition and, more generally, geographic exclusivity favoured cultural diversity and supported creative activities by enabling right holders to receive adequate remuneration. Therefore, it sought the annulment of the Commission decision before the GC. On 12 December 2018, however, the GC dismissed Canal +’s appeal.

Ruling on Canal +’s appeal against the GC’s judgment, the ECJ confirmed the GC assessment that the contested provisions aimed at creating absolute territorial protection for broadcasters and likely could be considered a restriction of competition by object, without prejudice to any decision definitively finding the existence or absence of an infringement of article 101 TFEU. Along the same lines, the ECJ found that the GC correctly held that the Commission was not obliged to analyse potential competition effects in each geographical market, since the provisions were intended to partition national markets, a practice which constitutes a restriction of competition by object on the EEA-wide market as a whole.

Nevertheless, the ECJ set aside the GC’s judgment on the ground that (i) the Commission’s decision violated the principle of proportionality by interfering with the contractual freedom of a third-party contracting partner, Canal +, and (ii) the GC incorrectly assumed that Canal + could enforce its contractual rights before national courts even if this conflicted with the commitments in the Commission decision. The ECJ ruled that national courts could not adequately remedy the Commission’s failure to consider the proportionality of its commitment decision in relation to third parties’ contractual rights because, were a national court to uphold Canal +’s contractual rights, this would infringe article 16 of Regulation 1/2003, which prohibits courts from adopting decisions contradicting a Commission decision or a decision the Commission may contemplate in the future.

Policy initiatives

Important policy initiatives continued during the reporting period, including the ongoing review of existing block exemptions and guidelines and proposals for two new regulations that would reshape the boundaries of competition law enforcement.

Commission publishes draft new Vertical Agreements Block Exemption and Vertical Guidelines

On 9 July 2021, the Commission published the draft revised Vertical Block Exemption Regulation (Draft VBER) and draft revised guidelines on vertical restraints (Draft VGL), a key step in the context of the revision of the current regime governing vertical agreements. The Draft VBER and Draft VGL would bring about a number of important changes to the application of the EU competition rules to a wide range of vertical agreements and would affect vertical agreements in all sectors of the economy.

In particular, the draft instruments envisage more lenient rules on online sales restrictions and active sales restrictions, although the increased flexibility this would give to suppliers would also come with increased legal uncertainty and risk as the line between permissible restraints and those considered to be a hardcore restriction is not clearly delineated. There would be greater legal certainty for most MFNs, which would continue to benefit from the VBER, as only so-called wide MFNs in favour of providers of online intermediation services would be excluded from the scope of the VBER and subject to self-assessment.

Widespread concerns have been raised about the proposal to exclude information exchanges in dual distribution systems from the benefit of the VBER as soon as combined retail market shares of the parties exceed 10 per cent, and to consider information exchanges that result in an ‘by object’ infringement as a hard-core restriction that would prevent the application of the VBER to the entire agreement. The proposed rules would materially increase uncertainty and costs in dual distribution systems, for no apparent reason. Not only does it remain unclear what policy concerns could justify stricter rules for information exchanges in dual distribution system, but it also appears that the proposed rules would not result in more effective competition law enforcement against collusion at the retail level, which is already unlawful under the current rules.

Proposals for new instruments and enforcement powers

Commission unveils proposed regulation on foreign subsidies

On 5 May 2021, the Commission adopted a proposal for a Regulation on foreign subsidies distorting the internal market (Proposed Regulation). The Proposed Regulation would empower the Commission to investigate financial contributions granted by non-EU governments to companies active in the EU to determine whether they distort the internal market and require redress. Broadly speaking, through the Proposed Regulation, the Commission aims to ‘level the playing field’ between EU companies and non-EU companies doing business in the EU that have received third country financial support.

For this purpose, the Commission would be armed with three new tools, comprising (i) a notification tool to investigate certain concentrations (ie, mergers, acquisitions and full-function joint ventures as defined in the Merger Regulation); (ii) a notification tool to investigate certain public procurement bids; and (iii) a general market investigation tool allowing the Commission to investigate ex-officio ‘all other market situations’ involving financial contributions by non-EU governments to companies active in the EU.

The Proposed Regulation is neither industry- nor country-specific. The nature of a ‘financial contribution’ is broadly defined to cover the conferral of a ‘benefit’ on an undertaking engaging in an economic activity in the EU where the benefit is limited to that company or to a group of particular companies (reflecting the principle of ‘selectivity’ applicable under the EU state aid rules). As a result, the concept covers a wide array of benefits a company might receive beyond direct payments, such as tax-advantaged treatment, favourable financing, or preferential loan arrangements.

Under the Proposed Regulation, the Commission would hold exclusive responsibility to enforce the three tools. Notably, the two notification-based tools would require notification to the Commission of any concentration or public procurement bid meeting the above-mentioned thresholds and impose a standstill obligation on such concentrations pending the conclusion of the Commission’s review (and, in a similar way, prevent the award of public procurement contracts pending such review). If a company fails to notify such a concentration, the Commission is empowered to impose fines and to assess the transaction as if it had been notified.

Under all three tools the Commission would have the power to balance any distortive effect on competition in the internal market resulting from the subsidy against any potential positive effects on ‘the development of the relevant economic activity’ to determine whether to impose corrective measures or accept commitments from the parties concerned. The Commission also has the power to prohibit notified transactions entirely.

The Proposed Regulation undoubtedly provides the Commission with a powerful and far-ranging arsenal of tools to address a relatively specific gap in its enforcement. But the question remains whether the tool is appropriately tailored to the task, or whether it risks creating more concerns than it seeks to solve, considering the breadth of the Commission’s powers, the lack of legal certainly for merging parties, the lack of guidance for a substantive assessment, and the high degree of discretion when it comes to remedies. Some of these concerns may be alleviated as implementing instruments are developed to complement the Proposed Regulation and to provide further detail and guidance.

Commission proposes Digital Markets Act and Digital Services Act

On 15 December 2020, the Commission presented a legislative package consisting of a proposed Regulation on a Single Market for Digital Services (Digital Services Act or DSA), which creates new rules applicable to providers of online intermediary services, as well as a proposed Regulation on Contestable and Fair Markets in the Digital Sector (Digital Markets Act or DMA), which is designed to regulate the conduct of large online platforms that can be classified as ‘gatekeepers’. The two proposed Regulations would have far-reaching implications for digital markets in the EU, and the Commission believes that the DMA rules, as well as enforcement powers, will improve competition in digital markets in Europe.

From a competition law perspective, the proposed DMA is of particular interest, despite statements by the Competition Commissioner that the DMA is not related to competition law enforcement. With the DMA, the Commission seeks to introduce a regulatory framework for large digital platforms or ‘core platform services’, which can include marketplaces, app stores, search engines, social networking services, operating systems, cloud services and advertising services.

Under the DMA, a platform provider designated by the Commission as a gatekeeper would have to comply with a detailed list of specific obligations that should ensure a fair and open online environment for consumers and businesses. Practices from which gatekeepers would have to refrain include (i) favouring their own services over those of their business users, (ii) combining data collected through different service offerings, (iii) restricting possibilities for business users to make offers and conclude contracts outside the platform, or (iv) preventing users from uninstalling pre-installed software or applications. Additionally, gatekeepers would have to abide by burdensome obligations, such as ensuring the interoperability of the platform with third-party apps, sharing the data provided or generated by business users and their customers in their use of the platform, and providing rival search engine providers with access on FRAND terms to certain search data.

The list of regulated practices is extraordinarily long, although it appears primarily to reflect complaints and experiences in individual competition cases, rather than a systematic and coherent attempt to identify practices that clearly harm market outcomes for consumers. The Commission could even impose additional obligations if it identifies more practices it perceives as unfair or undermining contestability, although it could also suspend certain obligations for a gatekeeper upon request.

Moreover, gatekeepers would have to inform the Commission of any intended transaction that is considered a concentration within the meaning of the EU Merger Regulation (EUMR) involving any other provider of a core platform service or of any other services provided in the digital sector, regardless of whether the transaction is notifiable under the EUMR or national merger control rules.

The DSA and DMA are currently under review by the European Parliament and the Council, with ongoing debates about the specific obligations imposed on gatekeepers and the allocation of enforcement powers between the Commission and member states. Whatever the final result, it will remain to be seen whether such cumbersome regulatory intrusion in highly dynamic industries will improve market outcomes for consumers and lead to more innovation.