July 3, 2022

Volume XII, Number 184


July 01, 2022

Subscribe to Latest Legal News and Analysis

June 30, 2022

Subscribe to Latest Legal News and Analysis

EU Competition Law Review- 2021



AT.40413, Focus Home – Video Games; AT.40414, Koch Media – Video Games; AT.40420, Zenimax – Video Games; AT.40422, Bandai Namco – Video Games; AT.40424, Capco – Video Games, 20 January 2021

On 20 January 2021, the EC fined Valve Corporation and five publishers, Bandai Namco, Capcom, Focus Home, Koch Media and ZeniMax, a total of EUR 7.9 million for breaching antitrust rules by partitioning the EEA market.  


Valve, together with the five publishers, restricted cross-border sales of certain PC video games based on the geographic location of the users, within the EEA (otherwise known as geo-blocking).

Valve operates Steam, one of the world’s largest online PC video gaming platforms. It allows users to download and activate PC video games. Valve provides publishers the opportunity to use Steam to ensure authenticated copies of their video games are played by customers. Alongside this service, Valve offered publishers the ability to control the country where a customer could activate the video game. In bilateral agreements, a publisher and Valve agreed that authentication codes provided to a customer would geo-block other Member States. These agreements would last between one and five years and were implemented between 2010 and 2015. The publisher in turn would provide its distributors with the restricted authentication code, and the distributors in turn would agree to restrict cross-border sales for between three and 11 years.

The EC Decision

The EC considered the practices described above as geo-blocking, for the simple reason that a customer who purchased a video game in Poland, for example, would not be able to activate it in Germany. The EC found that by bilaterally agreeing to geo-block certain PC video games from outside a specific territory, Valve and each publisher partitioned the EEA market in violation of Article 101(1) TFEU. The EC therefore sanctioned this conduct by imposing fines. It reduced the fines for each of the five publishers by between 10% and 15% because of their cooperation with the investigation, and their individual net fines ranged from EUR 340,000 to EUR 2.888 million.

Valve chose not to cooperate and was fined EUR 1.624 million.

Executive Vice-President Margrethe Vestager, in charge of competition policy, said: 

More than 50% of all Europeans play video games. The video game industry in Europe is thriving and it is now worth over EUR 17 billion. Today’s sanctions against the ‘geo-blocking’ practices of Valve and five PC video game publishers serve as a reminder that under EU competition law, companies are prohibited from contractually restricting cross-border sales. Such practices deprive European consumers of the benefits of the EU Digital Single Market and of the opportunity to shop around for the most suitable offer in the EU.


The EC’s investigation was initiated in February 2017 as one of the outcomes of its sector inquiry into the e-commerce sector in 2015/16, the final report of which was published on 10 May 2017 under reference COM (2017) 229 final. This report identified the practice of using geo-blocking to prevent consumers from making cross-border purchases online. The report observed that:

“. . . geo-blocking measures based on agreements or concerted practices between distinct undertakings may be caught under Article 101 TFEU. The European Courts have on several occasions held that agreements or concerted practices which are aimed at partitioning markets according to national borders or which make the interpenetration of national markets more difficult, in particular those which are aimed at preventing or restricting parallel exports, have as their object the restriction of competition pursuant to Article 101(1) TFEU.”

Since this practice amounts to a restriction of competition “by object”, it cannot generally benefit from an exemption under Article 101(3) TFEU. The EC’s proposed revisions to its Guidelines on Vertical Restraints, discussed on page 54, also address other restrictions aimed at preventing cross-border trade within the EU. The current draft expressly states that exemption under Article 101(3) will not be available for any requirement that the distributor prevent customers located in another territory from viewing its website through the automatic re-routing of customers to the manufacturer’s or other distributors’ websites, or the termination of online transactions when consumers’ credit card data reveal that they are located outside the distributor’s territory.

Geo-blocking is not just a matter of concern from the point of view of EU competition law. It also undermines the internal market as an area without internal frontiers in which the free movement of, inter alia, goods and services is ensured. It is for this reason that Regulation (EU) 2018/302 was adopted by the European Parliament and the Council, on 28 February 2018, addressing unjustified geo-blocking and other forms of discrimination based on customers’ nationality, place of residence or place of establishment within the internal market. Member States are obliged to lay down effective, proportionate and dissuasive measures to enforce compliance with the provisions of this Regulation. Thus a person that engages in geo-blocking runs the risk not only of sanctions for infringement of Article

101(1) TFEU but also sanctions for infringement of Regulation (EU) 2018/302.


AT.40330, Rail Cargo, 20 April 2021

On 20 April 2021, the EC issued its settlement decision by which it fined Deutsche Bahn (DB) and Société Nationale des Chemins de fer belges/Nationale Maatschappij der Belgische Spoorwegen (SNCB) a total of approximately EUR 48 million for participating in a cross-border customer allocation cartel for rail cargo transport in the EU (Decision C (2021) 2521 final in Case AT.40330, Rail Cargo).


Following an application for immunity from fines made by the Austrian railways company ÖBB on 24 April 2015, the EC carried out unannounced inspections at the premises of DB in Germany in September 2015. 

DB applied for leniency in October 2015, and SNCB in September 2016. 

The EC found that the three companies were participating in a customer allocation cartel in the EU market for cross-border rail cargo transport services on blocktrains. Blocktrains are cargo trains that ship goods from one site, such as the production site of the vendor of the transported goods, to another site, such as a warehouse, without being split up or stopped on the way. Freight sharing models are common practice in this industry. They allow railway companies to provide customers with a single overall price for the service required under a single multilateral contract. 

Cooperation by railway undertakings on the joint provision of cross-border rail cargo services, including joint pricing in the framework of the freight sharing model, falls outside the scope of Article 101(1) TFEU by virtue of a block exemption contained in Council Regulation (EC) No 169/2009.

The EC’s complaint in the present case, however, was that conduct took place that went beyond what was required to carry out joint cross-border rail cargo transport services, and so fell outside the scope of the block exemption.

The EC found that the cartel concerned trilateral transport carried out by DB, ÖBB and SNCB, and bilateral transport carried out by DB and ÖBB.

The lead carrier under the freight sharing model is the railway company, which acts as the main party dealing with the customer. The role of lead carrier can have important advantages, notably in building and maintaining customer relationships, which potentially provide further and/or future business opportunities. The three companies protected each other’s position as lead carrier for so-called “existing business”. They developed a wide understanding of the notion of “existing business”, covering different situations in which a customer relationship had been established. They mainly considered as “existing business” nearly all transport relations for which a (joint) freight sharing contract with a customer already existed, and in respect of which one of them therefore acted already as lead carrier. 

The mutual understanding between the three companies was that the lead carrier position for “existing business” should be protected for the railway company which held that position, and that any switching of the lead carrier position by the customer should be avoided. To protect the role of lead carrier, each of the companies that was not the lead carrier abstained from making offers to potential customers for “existing business”, or made “cover quotes” (i.e., quotes that contained a mark-up on top of the price of the “existing business”). 

The collusive scheme applied to cross-border rail cargo transport services on routes starting in, ending in or passing through Germany or Austria and carried out by DB and ÖBB. Cross-border rail cargo transports carried out by DB and ÖBB in this way extended also to Hungary (where ÖBB had taken over the incumbent railway undertaking) and to the Netherlands (where DB had taken over the incumbent railway undertaking). The collusive scheme also applied to transports starting or ending in Belgium carried out together with SNCB. SNCB participated in the infringement only to the extent that such trilateral transports were concerned.

The EC found that there was a consistent pattern of collusive contacts between DB and ÖBB relating to the lead carrier role in rail cargo transport services on blocktrains carried out under the freight sharing model since 8 December 2008. The collusive contacts became trilateral when SNCB started participating on 15 November 2011. The last collusive contact between DB, ÖBB and SNCB took place on 30 April 2014. 

The EC noted that cross-border trade between Member States is significant for high-volume cargo transports in the EU. The routes concerned are key West-East and North-South rail corridors connecting essential industrial areas in the EU. The collusive conduct was therefore capable of having an appreciable effect on trade between Member States.

On 4 April 2019, the EC initiated formal proceedings against the three companies with a view to engaging in settlement discussions with them. The three companies ultimately admitted their involvement in the customer allocation cartel, and settlement meetings took place with the EC. A Statement of Objections was adopted on 4 December 2020, and the parties confirmed that it reflected their settlement submissions.

The EC Decision

By its decision, the EC fined DB and SNCB a total of EUR 48 million for their participation in the customer allocation cartel. 

ÖBB received full immunity for being the first undertaking to provide information and evidence on the infringement which enabled the EC to carry out a targeted investigation in relation to the alleged cartel.

The fines imposed on the other two companies were calculated in accordance with the 2006 Fining Guidelines. Considering the nature of the infringement and the geographic scope, the percentage of the variable amount of the fines as well as the additional amount (the entry fee) were set at 15% of the value of “relevant” sales. “Relevant sales” for this purpose were each undertaking’s sales of conventional cross-border rail cargo transport services (except in the automotive sector) provided on blocktrains under the freight sharing model and carried out in cooperation:

  • By the three railway undertakings DB, ÖBB and SNCB and starting or ending in or passing through Austria or Hungary, Germany or the Netherlands and Belgium; and 

  • By DB and ÖBB and starting or ending in or passing through Austria or Hungary and Germany or the Netherlands.

DB’s fine thus calculated was increased by 50% for recidivism since it had previously been held liable for another cartel in the cargo transport sector in March 2012.  

DB and SNCB benefitted from a double reduction in fine. First, a leniency reduction was granted for providing information of significant added value to the EC. The amounts of the leniency reductions were 45% for DB and 30% for SNCB. Second, a standard 10% reduction was granted for acknowledging liability in the context of the settlement with the EC. 

In all, while ÖBB escaped a EUR 37 million fine, DB and SNCB were fined respectively EUR 48.3 million and EUR 270,000.


On the day of the EC’s decision, Executive VicePresident Margrethe Vestager, in charge of competition policy, said: 

Rail transport of cargo is vital for a sustainable economy model. Fair competition is important to provide customers with the best offer when using sustainable transport. A cartel between key operators offering rail cargo services on essential rail corridors across the EU goes fundamentally against this objective. Today’s decision sends a clear signal that this type of collusive behaviour is not acceptable.” 

The decision also illustrates the value of the EC’s leniency programme, which allows applicants to benefit from substantial reductions in fines if they disclose their participation in the infringement and bring additional value compared to the facts disclosed by the immunity applicant in order to facilitate the uncovering of cartels by the EC. This particular case was also considered by the EC as suitable for the settlement procedure. The three companies concerned agreed to settle, thus obtaining a further 10% reduction in fines, while enabling the EC to economise time in bringing the investigation to a conclusion.


AT.40324, European Government Bonds, 20 May 2021

By decision of 20 May 2021, the EC found that Bank of America, Natixis, Nomura, RBS (now NatWest), UBS, UniCredit and WestLB (now Portigon) breached EU antitrust rules through the participation of a group of traders in a cartel on the primary and secondary markets for European Government Bonds (EGBs).


EGBs are debt securities issued in euros by the central governments of the Eurozone Member States in order to raise funds in international financial markets. An EGB represents a borrowing of an amount in euros for a fixed term at a predefined interest rate. The bondholder receives interest payments periodically, and the principal amount borrowed is paid to the bondholder upon expiry of the fixed term.  

EGBs are first issued on the primary market where a limited number of investment banks, the “primary dealers”, can bid for them or sometimes acquire them via syndication. The primary dealers then place and trade the EGBs with other investors on the secondary market. These other investors can include other banks, asset managers, pension funds, hedge funds and major companies. They can hold the EGBs as investments or trade them via brokers in the same way as any other financial instrument.

The EC started investigating this matter as a result of an immunity application submitted by NatWest (then named RBS). NatWest applied for a marker on 29 July 2015, followed by a full immunity application, and was granted conditional immunity on 27 January 2016. On 31 January 2019, the EC initiated formal proceedings and issued a Statement of Objections to the investment banks involved.

The EC Decision

By its decision of 20 May 2021, the EC found that the traders of the investment banks exchanged commercially sensitive information, mainly in multilateral chatrooms on Bloomberg terminals between 2007 and 2011. They informed each other on a regular basis of their prices and volumes offered in the run-up to the auctions and the prices shown to their customers or to the market in general. They discussed and provided each other with recurring updates on their bidding strategies on the primary market, and on trading parameters on the secondary market.

Relevant information exchanged included information on prices, volumes and trading positions: mid-prices, yield curves and spreads of bonds recently traded or being offered on the secondary market, volumes envisaged to be purchased at the auctions, information on the bids, the level of overbidding and overbidding strategies at the auctions, etc.

Access to the multilateral chatrooms was based on the expectation that participants would disclose commercially sensitive information with other competitors within a trusted group of traders. The aim was to reduce the uncertainties regarding the issuing and trading of EGBs. Exchanges within the chatrooms enabled participants to identify and pursue opportunities for coordination with each other in appropriate constellations. This was illustrated by the exclusive character of the chatrooms and the working names given to them.

The main evidence in this case consisted of records of communications between traders in chatrooms. The accused parties contested the authenticity of these exchanges and argued that the use of jargon, abbreviation and informal language made the evidence less credible. However, the EC rejected this argumentation because it was familiar with such language—this was not its first decision in the financial sector—and had been assisted in the interpretation of the evidence by three banks not fully involved in the conduct.

In fixing the level of the fines, the EC took into consideration the very serious nature of the cartel infringements, their duration and geographic scope, and also the fact that the conduct permeated the whole EGB industry on the primary and secondary markets. The EC also took into consideration the fact that EGBs are used for raising public funding and that the conduct took place over a period that included the very serious 2008 financial crisis. 

The duration of each bank’s participation in the cartel was as follows:

  • Bank of America from 29 January 2007 until 6 November 2008

  • Natixis from 26 February 2008 until 6 August


  • Portigon from 19 October 2009 until 3 June


  • NatWest from 4 January 2007 until 28 November 2011

  • UBS from 4 January 2007 until 28 November


  • Nomura from 18 January 2011 until 28 November 2011

  • UniCredit from 9 September 2011 until 28 November 2011

The first two of these could not be fined because they left the cartel more than five years before the EC started its investigation and therefore fell outside the limitation period set by Article 25 of Regulation (EC) No 1/2003.

The third bank, Portigon, which was winding down its activities, had negative net turnover in the last business year prior to the EC’s decision and so its fine of EUR 4.888 million was reduced to zero (in application of the rule that the fine cannot exceed 10% of a company’s worldwide group turnover in the last business year prior to the decision imposing fines).

The fourth bank, NatWest (formerly RBS), was the first applicant for immunity or leniency, and received full immunity from fines under the Leniency Notice.  

The fifth bank, UBS, was the second applicant for immunity or leniency (having applied on 29 June 2016), and benefitted from a 45% fine reduction for its cooperation in the EC’s investigation, the resulting net fine being EUR 172 million.

The sixth bank, Nomura, was fined EUR 129 million, and the seventh bank, UniCredit, EUR 69 million. 

Since this case was initiated before the end of the transition period under the EU-UK Withdrawal Agreement, the EC will collect the fine once it has become definitive and will reimburse the United Kingdom for its share.


This case falls in line with the previous decisions of the EC in the financial sector. The EC sanctioned multiple anticompetitive practices that took place during the financial crisis of 2008: cartels in relation to yen interest rate derivatives (2013), euro interest rate derivatives (2013) and Swiss franc interest rate derivatives (2014).  

Executive Vice-President of the EC Margrethe

Vestager, in charge of competition policy, said:

It is unacceptable that, in the middle of the financial crisis, when many financial institutions had to be rescued by public funding, these investment banks colluded in this market at the expense of EU Member States.”


C-586/16 P, Sun Pharmaceutical Industries and Ranbaxy (UK) v European Commission; C-588/16 P, Generics (UK) v European Commission; C-591/16 P, Lundbeck v European Commission; C-601/16 P, Arrow Group and Arrow Generics v European Commission; C-611/16 P, Xellia Pharmaceuticals and Alpharma v European Commission; C-614/16 P, Merck v European Commission, 25 March 2021

On 25 March 2021, the CJEU dismissed all the appeals brought by Danish pharmaceutical company H. Lundbeck A/S and five generic manufacturers against the judgments of the GCEU, thus upholding a decision of the EC on patent settlement agreements between Lundbeck and five generic manufacturers (Case C-586/16 P, Sun Pharmaceutical Industries and Ranbaxy (UK) v Commission, EU:C:2021:241; Case C-588/16 P, Generics (UK) v Commission, EU:C:2021:242; Case C-591/16 P, Lundbeck v

Commission, EU:C:2020:428; Case C-601/16 P,

Arrow Group and Arrow Generics v Commission,

EU:C:2021:244; Case C-611/16 P, Xellia

Pharmaceuticals and Alpharma v Commission,

EU:C:2021:245; and Case C-614/16 P, Merck v

Commission, EU:C:2021:246). The judgment is significant as regards the interpretation and application of fundamental notions of EU competition law, such as “by object” restrictions and “potential competition”. It also provides important guidance on when companies need to keep documents on file.


Between 1977 and 1985, Lundbeck developed and patented an antidepressant medicine containing the active substance citalopram and two processes for the manufacture thereof (the original patents). The original patents were issued in Denmark and in a number of western European countries. Over time, Lundbeck developed additional processes for the manufacture of citalopram for which it obtained patents in several countries, such as Denmark, the United Kingdom and the Netherlands (the process patents).

Between 1994 and 2003, the original patents expired. However, Lundbeck still held the process patents covering possible ways to manufacture citalopram. As generic versions of citalopram were about to enter the market, a number of patent disputes arose between Lundbeck and generic producers. These disputes pertained to the potential infringement or, indeed, validity of the process patents. To settle the disputes, Lundbeck, in 2002, concluded six agreements with four groups of generic manufacturers (collectively, the generics). Pursuant to these agreements, Lundbeck provided significant financial compensation to the generics. In return, each of the generics committed to delay its entry on the market with the generic version of citalopram (agreements referred to collectively as the patent settlement agreements).

In October 2003, the EC was informed of the patent settlement agreements by the Danish competition authority. Between 2003 and 2006, the EC conducted inspections at the premises of Lundbeck, and in January 2008 the EC launched a sector inquiry into the European pharmaceutical sector. Following an investigation into the patent settlement agreements, in 2013 the EC concluded that they restricted competition “by object”, i.e., by their very nature.

The EC fined Lundbeck EUR 93.7 million and the generics a total of EUR 52.2 million (Commission Decision C (2013) 3803 final in Case AT.39226, Lundbeck). In setting the fines, the EC considered that the amounts paid by Lundbeck to the generics corresponded approximately to the profits that they would have made had they entered the market. The parties to the patent settlement agreements appealed against the EC’s decision to the GCEU, which, in September 2016, upheld the EC’s decision in full. Lundbeck and the generics subsequently brought appeals before the CJEU, which dismissed them in their entirety.

The CJEU Judgment

Potential Competition

The CJEU ruled that, at the time of the patent settlement agreements, Lundbeck and the generics were potential competitors. In line with its reasoning in Case C-307/18, Generics UK and others v Competition and Markets Authority (Paroxetine) (EU:C:2020:52), the CJEU held that, to assess whether a company is a potential competitor, there must be “real and concrete possibilities” of it joining the market and competing with those companies already present thereon (Case C-591/16 P, Lundbeck, cited above, paragraph 54). The test consists of two steps:

  • Step 1: It is necessary to assess whether “the generic manufacturer had taken sufficient preparatory steps to enable it to enter the market concerned [in this case citalopram] within such a period of time as would impose competitive pressure on the manufacturer of originator medicines” (Case C-591/16 P, Lundbeck, cited above, paragraph 57).

Such preparatory steps must be such that they demonstrate that “the manufacturer of generic medicines has in fact a firm intention and an inherent ability to enter the market.” These include, inter alia, actions taken by the generic manufacturer to obtain a marketing authorisation or equivalent authorisation necessary for the marketing of its generic medicine, or legal actions to challenge the validity of the patent held by the originator (Case C591/16 P, Lundbeck, cited above, paragraphs 56, 59 and 86). Meeting these criteria is sufficient, and it is not necessary to demonstrate with certainty that the generic manufacturers will in fact enter the market and successfully remain thereon.

  • Step 2: It must be determined “that the market entry of a generic manufacturer does not meet barriers to entry that are insurmountable”. Specifically, “the existence of a patent which protects the manufacturing process of an active ingredient that is in the public domain cannot, as such, be regarded as an insurmountable barrier, regardless of the presumption of validity attached to that patent” (Case C-591/16 P, Lundbeck, cited above, paragraph 58).  

The CJEU, in line with precedent, including Paroxetine, also ruled that “[a] finding of potential competition between a manufacturer of generic medicines and a manufacturer of originator medicines can be confirmed by additional factors, such as the conclusion of an agreement between them at a time when the former was not present on the market concerned” (Case C-591/16 P, Lundbeck, cited above, paragraph 57, and Case C-307/18, Paroxetine, cited above, paragraphs 54-56). A further indication of potential competition is the transfer of value by a manufacturer of originator medicines to a manufacturer of generic medicines in exchange for the postponement of the latter’s market entry, even though the former claims that the latter is infringing one or more of its process patents. The greater the transfer of value, the stronger the indication (Case C-307/18, Paroxetine, cited above, paragraph 56).

Drawing upon previous rulings, the CJEU therefore laid down a significant criterion of potential competition: The conclusion of an agreement between companies operating at the same level in the production chain, especially at a point when one of the parties does not yet have a presence in the market, is a strong indication of potential competition between the parties (Case C-307/18, Paroxetine, cited above, paragraph 55).

Restrictions of Competition “By Object”

The CJEU reaffirmed its conclusions in previous rulings, including Paroxetine, that:

“The concept of restriction of competition ‘by object’ must be interpreted strictly and can be applied only to some agreements between undertakings which reveal, in themselves and having regard to the content of their provisions, their objectives, and the economic and legal context of which they form part, a sufficient degree of harm to competition for the view to be taken that it is not necessary to assess their effects” (Case C-591/16 P, Lundbeck, cited above, paragraph 112).

In the view of the CJEU, a finding of “restriction by object” is appropriate when it is clear that a transfer of money from an originator to a generic company only serves the parties’ common commercial interest not to compete on the merits. Further, the assessment of whether the net gain via the value transfers was sufficiently significant to act as an incentive for the manufacturer of generic medicines to refrain from entering the market concerned and not to compete on the merits with the originator must be conducted on a case-by-case basis. Such an incentive—as induced by significant payments— was demonstrated in the cases at hand. Lundbeck had not identified any pro-competitive effects of the patent settlement agreements that could offset their harm to competition.

The CJEU ruled also that a settlement agreement between the holder of a process patent and a party allegedly infringing that patent constitutes the legitimate expression of the intellectual property right of the holder of the process patent insofar as that patent is protected. However, such agreement must not go beyond the specific subject matter of the protection of the intellectual property rights and infringe competition law. The CJEU found that the patent settlement agreements went beyond the specific subject matter of Lundbeck’s patents; indeed, while the right to challenge infringements does fall within the subject matter of the patent, the right to enter into agreements by which actual or potential competitors are paid for not competing does not fall within the subject matter of the patent.

The Undertakings’ Obligation of Diligence in the Context of a Sector Inquiry

Another significant point made by the CJEU in its judgment in the appeal of Xellia Pharmaceuticals and Alpharma (formerly Zoetis) (Case C-611/16 P, Xellia, cited above) concerns a company’s obligation of diligence to the extent that the generics were required to maintain any document relevant for their defence in the context of a future administrative procedure.

Xellia and Alpharma had argued that the EC infringed their rights of defence by failing to inform them in a timely manner of the existence of an investigation concerning the patent settlement agreements in question. As a result, they did not retain any exculpatory evidence. Xellia and Alpharma argued that the EC’s investigation started in 2003, but they were informed about it only in 2010 and 2011.

The CJEU held that the GCEU had erred in law by imposing an obligation of diligence that was applicable from the time of the initiation of the administrative procedure by the EC against an individual party. In respect of Xellia and Alpharma, the period of diligence could only apply from 2010 and 2011, when the procedure was opened against them.

Contrary to the GCEU, the CJEU found that the point from which a company is bound by a specific duty to retain all necessary evidence coincides with the opening of a general EC inquiry into a particular sector, i.e., a sector inquiry (as opposed to the opening of an administrative procedure against an individual party). Thus: “when the Commission initiates sector inquiries, undertakings belonging to the sector concerned and, in particular, those which have concluded agreements expressly referred to in the decision initiating the inquiry, as was the case with Zoetis and Xellia, must expect that individual procedures may possibly be initiated against them in the future” (Case C-611/16 P, Xellia, cited above, paragraph 154).

The opening of a sector inquiry “constitutes a factor which should lead them to take precautions against the loss, due to the passage of time, of evidence that might prove to be useful to them in the context of subsequent administrative procedures or judicial proceedings” (Case C-611/16 P, Xellia, cited above, paragraph 152).


Companies in the pharma sector in particular should bear in mind the following:

  • Patent settlement agreements between the holder of a (process) patent and an alleged infringer are not per se illegal. However, patent settlement agreements may be caught by the EU competition rules if they are concluded between potential competitors, such as an originator and a manufacturer of generics, and involve payments by the former to the latter in exchange for the latter’s delayed market entry (pay-fordelay).

  • An originator and a generic manufacturer may be found to be potential competitors if there is a real and concrete possibility—absent the agreement—of the generic manufacturer entering the market and competing with the originator. An agreement between an originator and generic manufacturers not yet present on the market may provide a strong indication that the originator and the generic manufacturers are potential competitors.

  • A patent settlement agreement will often be found restrictive of competition by object, i.e., by its very nature, if the transfer of value from the originator to the generic company is clearly sufficiently significant to incentivise the latter to refrain from entering the market. The greater the transfer of value, the stronger the indication.

  • If the payment corresponds to the generic manufacturer’s anticipated profits post-entry or to the damages that could have been paid if the generic manufacturer had succeeded in litigation against the originator, the agreement may be found restrictive of competition by its very nature.

  • If no pro-competitive effects of the agreement can be demonstrated, and it is plain that the agreement is inherently restrictive of competition, then the EC will likely not assume any need to look at the competitive effects of the agreement.

  • Companies active in a sector covered by an EC sector inquiry must be aware that an investigation could potentially be initiated against them. The launch of a sector inquiry by the EC implies a need to retain any and all exculpatory documents going forward.


C-440/19 P, Pometon SpA v European Commission, 18

MARCH 2021

On 18 March 2021, the CJEU ruled on the appeal by Pometon SpA against the GCEU’s judgment in the steel abrasives cartel case. The CJEU ruled that the GCEU had breached the principle of equal treatment when recalculating the EC’s fine imposed on Pometon, the only non-settling party in this case. The CJEU made its own reassessment of the fine, reducing it from EUR 6.2 million to EUR 2.6 million.


Following the opening of an EC cartel investigation, five producers of steel abrasives requested the benefit of the cartel settlement procedure. Italian producer Pometon participated initially in the settlement discussions but later withdrew from this procedure. As a result, the EC found itself engaged in a “hybrid” cartel proceeding in which a settlement procedure and a standard procedure were conducted in parallel. This hybrid proceeding resulted in two decisions: one in April 2014, whereby the EC fined the four participants in the settlement procedure a total of EUR 30.7 million, and another in May 2016, whereby it fined Pometon EUR 6.2 million.

During the settlement procedure, the EC discovered that calculating the fine in accordance with its 2006 Fining Guidelines would have taken the amount above the statutory limit of 10% of turnover for most of the companies involved. In these particular circumstances, the EC considered that application of the 10% of turnover cap would result in fines that did not reflect the respective gravity of each individual company’s participation in the cartel.

The EC therefore applied “exceptional” reductions having first calculated an amount of the fine in accordance with the 2006 Fining Guidelines. In so doing, the EC relied on paragraph 37 of its 2006 Fining Guidelines, which provides: “Although these Guidelines present the general methodology for the setting of fines, the particularities of a given case or the need to achieve deterrence in a particular case may justify departing from such methodology.” The exceptional reductions applied were 75% for each of

Ervin and Winoa, 90% for MTS and 67% for Würth. In the same way, an exceptional reduction of 60% was applied in Pometon’s case.

In August 2016, Pometon challenged the EC’s May 2016 decision before the GCEU, alleging, inter alia, infringement of the principle of the presumption of innocence and inadequate statement of reasons explaining the calculation of Pometon’s fine.

The GCEU gave judgment on 28 March 2019. 

The GCEU found that the EC did not breach the principle of the presumption of innocence. It took all the necessary precautions in drafting the settlement decision and expressly excluded any finding in relation to Pometon, mentioning that the examination of this company’s case was to take place in a subsequent adversarial procedure. The GCEU concluded that the references to Pometon in the settlement decision could not be regarded as an indication of the EC’s lack of impartiality, nor as a lack of respect for the principle of the presumption of innocence.

However, the GCEU found that the EC failed to provide an adequate statement of reasons for the calculation of Pometon’s fine. The GCEU compared Pometon’s individual liability and situation with that of the addressees of the settlement decision and noted that other steel abrasives producers that had played a limited role in the cartel, like Pometon, had received substantially higher reductions in fine. Accordingly, in exercise of its jurisdiction to review the amount of a fine in competition matters, the GCEU reduced Pometon’s fine to EUR 3.8 million, corresponding to an extraordinary reduction of 75%.  

CJEU Judgment

The CJEU gave judgment on 18 March 2021, holding that the GCEU breached the principle of equal treatment when it recalculated Pometon’s fine. The CJEU rejected Pometon’s other grounds of appeal, including its claim that the GCEU erred in law by finding that the EC had not breached the principle of the presumption of innocence (Case C-440/19 P, Pometon SpA v Commission, EU:C:2021:214).

Respect of the Principle of the Presumption of Innocence

The CJEU observed that the GCEU examined, first, whether the EC took sufficient drafting precautions in the settlement decision in order to avoid a premature judgment as to Pometon’s participation in the cartel and, second, whether the references to Pometon in the settlement decision were necessary. 

As to the drafting precautions, the CJEU observed that the GCEU found that the EC expressly excluded Pometon’s guilt at that stage of the proceedings, emphasising that the file concerning Pometon would be dealt with subsequently, in a separate and adversarial procedure. Moreover, the reasons for the settlement decision did not contain any legal classification of the facts relating to Pometon. The CJEU formed the view that Pometon’s arguments were not capable of establishing that these findings of the GCEU were vitiated by an error of law.

As to the necessity of the references made to Pometon in the settlement decision, the CJEU observed that the EC had to avoid disclosing more information about Pometon’s involvement in the cartel than was necessary for the characterisation of the role of the other participants in that cartel (see, by analogy, Case C-377/18, AH and Others (Presumption of innocence), EU:C:2019:670, paragraph 44).

The CJEU observed that the GCEU had respected this principle in its review of the EC’s decision. The GCEU had concluded that the references to certain conduct of Pometon in the settlement decision could “be objectively relevant to the description of the origin of the cartel as a whole.” The GCEU observed, moreover, that those references did not set out any legal characterisation of Pometon’s conduct. The CJEU considered that the arguments put forward by

Pometon in the appeal were not such as to call the GCEU’s findings into question.

Recalculation of the Fine

Pometon claimed that the GCEU treated two different situations identically when recalculating its fine, without objective justifications, thus breaching the principle of equal treatment.

The CJEU agreed, finding that the GCEU had failed to state why it applied the same extraordinary reduction rate to Pometon as to another cartel participant, Winoa, even though Pometon’s infringement was less serious than Winoa’s, as the GCEU itself had concluded. The CJEU pointed out, at paragraph 151 of its judgment, that “it was for the [GCEU] to set out the reasons why, despite the difference in situation, it was consistent with the principle of equal treatment to grant Pometon a rate of reduction identical to that granted to Winoa”, especially when the calculation of the fine departed from the EC’s general fining methodology.

The CJEU set aside this aspect of the GCEU’s judgment and reduced Pometon’s fine to EUR 2.6 million, corresponding to an extraordinary reduction of 83%. Pometon’s situation was similar to that of settling party MTS, which received an extraordinary reduction of 90%. These two undertakings played a relatively limited role in the cartel, and their overall weight in the cartel was proportionally low having regard to the value of their specific sales in the EEA. Although Pometon’s total turnover was higher than that of MTS, the CJEU considered that disproportionate importance should not be attributed to this in comparison with other relevant factors.


Although Pometon succeeded in obtaining a further reduction in its fine, this case is of much wider importance for its discussion of the protection of the rights of the defence in a so-called “hybrid” proceeding, namely a proceeding that results when a company withdraws from a settlement procedure and so becomes subject to an adversarial procedure.

As both the GCEU and the CJEU observed, a participant in a settlement procedure has the right to withdraw from such procedure at any time and be examined under an adversarial procedure, in accordance with the universal principle of no penalty without trial. However, the conduct of a settlement procedure and an adversarial procedure in parallel, or in quick succession, creates difficulties for the protection of the principle of the presumption of innocence. Two principles must be respected. First, if facts concerning third parties which will be tried later must be introduced, the relevant court should avoid giving more information about the third parties than necessary for the assessment of the legal responsibility of those persons who are accused in the trial before it. Second, decisions must be worded in such a way as to avoid prejudging the guilt of the third parties concerned and thereby jeopardising the fair examination of the charges that will be brought against them in the later proceedings.

Drawing the line at what is necessary for the assessment of the liability of the accused parties in the settlement procedure is not an easy task. The question whether the EC disregarded the presumption of innocence will depend on the specific facts of each individual case. Pometon was a case where the limit of what was “necessary” was not exceeded. On the other hand, the case of Icap (Case T-180/15, Icap and Others v Commission, EU:T:2017:795) provides an example where this limit was exceeded.  

As a final point, it should be observed that one way of safeguarding the presumption of innocence is to adopt the settlement and the adversarial decisions at the same time, as the EC did in the investigation that was challenged in Case T-456/10, Timab Industries and CFPR v Commission (EU:T:2015:296).


C-694/19P, Italmobiliare v European Commission, 15 April


On 24 June 2015, the EC fined 10 manufacturers of retail food packaging trays a total of EUR 115.8 million for having participated in a single and continuous infringement consisting of five separate cartels aimed at fixing prices, allocating customers and markets, bid-rigging and exchanging commercially sensitive information. The cartels took place between 2000 and 2008 in five different geographical regions within the EEA, namely Italy, South-West Europe (covering Spain and Portugal),

France, Central and Eastern Europe (covering Poland,

Slovakia, Czech Republic and Hungary), and NorthWest Europe (covering Belgium, Denmark, Finland, Germany, Luxembourg, the Netherlands, Norway and Sweden).

The EC carried out unannounced inspections on 4 and 6 June 2008 following an immunity application filed by Linpac on 18 March 2008 pursuant to the Leniency Notice. After the inspections, several cartel participants applied for immunity, or failing this, for a reduction in fine pursuant to the Leniency Notice. The present case before the CJEU concerns the application made on 1 July 2008 by Sirap-Gema and its parent company, Italmobiliare. In the EC’s decision imposing fines, Linpac was granted immunity while Sirap-Gema and Italmobiliare qualified for a reduction in the 30% to 50% leniency band in respect of the Central and Eastern Europe cartel (its application in respect of this cartel being next in line after Linpac) and a reduction in the 20% to 30% leniency band in respect of the separate cartels in Italy and France (its application in respect of these two cartels being next but one in line after Linpac). 

Before the GCEU, Italmobiliare challenged the fact that the EC held Italmobiliare jointly and severally liable for the infringement of its subsidiary, SirapGema, and also challenged the amount of the fine. By judgment of 11 July 2019, the GCEU rejected Italmobiliare’s challenges. 

Italmobiliare appealed to the CJEU, requesting that the GCEU’s judgment be set aside in whole and that the fine be cancelled or reduced.

The CJEU Judgment 

The CJEU gave judgment on 15 April 2021, rejecting all of Italmobiliare’s arguments (Case C-694/19 P, Italmobiliare v Commission, EU:C:2021:286).

Presumption of Decisive Influence of a Parent Company on Its Subsidiary Which Breached the EU Competition Law

Before the CJEU, Italmobiliare challenged the application of the presumption that the parent company actually exercises decisive influence over its subsidiaries. Italmobiliare argued that application of the presumption violates the principles of legal certainty, the legality of penalties and the presumption of innocence referred to in Article 6(2) and Article 7 of the European Convention on Human

Rights, and Articles 48 and 49 of the Charter of Fundamental Rights of the EU.

Italmobiliare also argued that, because a financial holding cannot qualify as an undertaking, the presumption of decisive influence exercised by a 100% parent company should not apply to it. The

CJEU rejected this argument in light of the case law

(Case C-440/11 P, Commission v Stichting Administratiekantoor Portielje, EU:C:2013:514) and declared that:

. . . even if the first appellant [Italmobiliare] were a holding company not qualifying as an undertaking, this fact could not affect the application to it of the presumption of decisive influence and parent company liability” (unofficial translation from the French).

The CJEU rejected Italmobiliare’s argument that the EC did not check that the parent company actually exercised decisive influence over its subsidiary. The CJEU recalled that, where a parent company holds 100% (or virtually all) of the shares in the subsidiary, there is a rebuttable presumption of decisive influence and it is then for the parent company to demonstrate that its subsidiary independently determined its own course of action in the market. 

Last of all, the CJEU observed that holding the parent company of a group jointly and severally liable for infringements of EU competition law committed by its subsidiaries does not constitute a breach of the principle of personal liability. On the contrary, it is an expression of that principle because the joint liability of both the parent company and its subsidiary is based on the fact that those companies were both part of the economic entity which committed the infringements.

Application of the Leniency Notice

The first leniency applicant, Linpac, benefitted from full immunity from fines. Italmobiliare considered that Linpac should not have benefited from immunity because it did not meet the condition laid down in the Leniency Notice, point 12 (b), according to which an undertaking must cease its anticompetitive conduct without unnecessary delay after the filing of its application. 

In that regard, Italmobiliare argued that it should have qualified for first place in applying for leniency in respect of the Central and Eastern Europe cartel and so should have been granted full immunity from fines in respect of that cartel. 

The CJEU did not agree with this argument. It pointed out that Linpac remained the first applicant to have supplied information and evidence to the EC which enabled the latter to carry out a targeted inspection and, as a consequence, remained the only undertaking able to claim full immunity under the Leniency Notice. As the CJEU observed, at paragraphs 77 and 78 of its judgment:

As follows from paragraph 8 of the Leniency Notice, an undertaking can only benefit from immunity from fines if it was ‘the first to submit information and evidence’ to the [EC] which enabled the latter to carry out a targeted inspection in connection with the alleged cartel or to make a finding of infringement of Article 101 TFEU in connection with this cartel” and

“. . . to the extent that Linpac was the first undertaking to provide the [EC] with information and elements of proof that permitted the latter to carry out these inspections, neither the undertaking to which the appellants belonged, nor any other undertaking, could claim immunity from fines pursuant to paragraph 8 of the Leniency Notice” (unofficial translation from the French).



This case provides another example of the wellestablished principle that, where a parent company holds 100% of the shares in a subsidiary, whether directly or indirectly, that parent company is presumed to exercise a decisive influence over the subsidiary. As a result, the parent company will be held jointly and severally liable for any infringement of competition law committed by the subsidiary unless the parent company rebuts the presumption by adducing sufficient proof that it did not exercise a decisive influence over its subsidiary. To date, there has been no case in which a 100% parent company has succeeded in rebutting this presumption. The case also shows that the presumption is not avoided just because the parent company is a financial holding.

It is also important to note that immunity is available only to the first leniency applicant to submit information and evidence which, in the EC’s view, will enable it to carry out a targeted inspection or to find an infringement of Article 101 TFEU in connection with the alleged cartel. In addition, this first leniency applicant must cooperate fully with the EC and, of course, must cease all further involvement in the cartel immediately following its application, except for what would, in the EC’s view, be reasonably necessary to preserve the integrity of the inspections. If this first leniency applicant forfeits its immunity because, for example, it does not cooperate fully or it delays unnecessarily in ceasing its involvement in the cartel, it is not possible for another leniency applicant to qualify for immunity in the place of the first applicant. By definition, such other applicant would not be the first to submit information and evidence to the EC. The CJEU restated this principle a few months later in its judgment in Case

C-563/19 P, Recylex v Commission,

(EU:C:2021:428), also reported on page 19.


C-563/19 P, Recylex S.A. v European Commission, 3 June


On 3 June 2021, the CJEU issued a judgment rejecting Recylex’s appeal against the GCEU’s dismissal of Recylex’s challenge to the EC’s infringement decision in the car battery recycling cartel. By this decision the EC rejected Recylex’s request for an increase in leniency reduction beyond the level already granted of 30%. The EC also rejected Recylex’s argument that it had been the first to submit certain new evidence that expanded the duration or gravity of the infringement and so should have been granted partial immunity in respect of the additional scope of the infringement revealed by this new evidence.


In 2017, the EC imposed fines totalling more than

EUR 26 million on four companies (JCI, Recylex, Campine and Eco-Bat) active in the production of recycled lead and other products, for having participated in the car battery recycling cartel. The undertakings participated from 23 September 2009 to 26 September 2012 in a single and continuous infringement consisting of agreements and/or concerted practices aimed at coordinating prices in the sector for the purchase of scrap car batteries. JCI’s application for immunity from fines, submitted on 22 June 2012, enabled the EC to initiate its investigation. The three other companies made leniency submissions in the context of that investigation. 

Pursuant to point 8(a) of the 2006 Leniency Notice, JCI was granted full immunity from fines, being the first to self-incriminate and report the cartel to the EC. Pursuant to point 26 of the 2006 Leniency Notice, Eco-Bat and Recylex were granted the maximum reductions possible, respectively 50% and 30%, for being the first and second undertakings to provide evidence of significant added value. Campine was not granted any reduction in fine because it did not provide any added value to the investigation.  

In the EC’s administrative proceedings, Recylex argued that the first leniency applicant, Eco-Bat, had failed to fulfil its duty of cooperation because it provided incomplete and misleading information. As a consequence, Recylex argued that it should have been granted a leniency reduction in the range applicable to a first leniency applicant, namely 30% to 50%. The GCEU rejected Recylex’s challenge on this point, and so Recylex appealed to the CJEU.

In the administrative proceedings, Recylex also claimed that it had been the first to provide the EC with evidence of a certain meeting and evidence that the geographical scope of the cartel extended to France. Consequently, Recylex argued that, under the 2006 Fining Guidelines, this information should not have been taken into account in determining Recylex’s fine (so-called partial immunity). In other words, Recylex argued that it should have benefited from a reduction in the overall fine in addition to the leniency reduction. The GCEU rejected Recylex’s challenge on this point, considering that the elements provided by Recylex were already in the hands of the EC. This point was also the subject of Recylex’s appeal to the CJEU.

The CJEU Judgment

By its judgment on 3 June 2021, the CJEU rejected Recylex’s appeal on both points: the argument that its ranking should have been raised from second to first leniency applicant, and the argument that it should have been granted partial immunity (Case C-563/19 P, Recylex S.A. v Commission, EU:C:2021:428). 

The CJEU recalled, first of all, that the conditions of qualification for partial immunity under the 2002 Leniency Notice did not change with the 2006 Leniency Notice, notwithstanding a difference in wording. Partial immunity is only granted when the applicant provides new evidence, previously unknown to the EC, having the effect of increasing the gravity or duration of the infringement. Therefore, the CJEU rejected Recylex’s application for partial immunity, based on the plea of additional evidence provided, because the EC was already aware of the relevant cartel meeting and of the fact that the territorial scope extended to France. 

The CJEU also established that the degree of cooperation of the other leniency applicants was irrelevant to the determination of Recylex’s ranking. Therefore, even if Eco-Bat did not fulfil its cooperation duties during the investigation, this could not result in Recylex taking up Eco-Bat’s ranking as first leniency applicant and thereby obtaining a higher percentage reduction in fine.


In this case, the CJEU clarified several important points concerning leniency:

  • Partial immunity can be granted only to a leniency applicant which provides the EC with new evidence that expands the duration or gravity of the infringement.

  • The provision of evidence that strengthens the value of the information that the EC has already is relevant to determining the amount of the leniency reduction to be granted within the applicable range (see next point). By definition, such evidence is not new evidence that expands the duration or gravity of the infringement and so cannot benefit from partial immunity. 

  • The 2006 Notice provides for a leniency ranking based on the date of submission of the leniency application. This ranking determines the range of reduction in fine possible: 30% to 50% for the first leniency applicant; 20% to 30% for the second leniency applicant; up to 20% for all subsequent leniency applicants. A leniency applicant cannot become eligible for a position of higher rank even if its cooperation in the investigation was better than that of a previous leniency applicant.


C-882/19, Sumal, S.L. v Mercedes Benz Trucks España, S.L., 6 October 2021

On 6 October 2021, the CJEU issued its judgment in Case C-882/19, Sumal, S.L. v Mercedes Benz Trucks España, S.L. (EU:C:1987:418), and confirmed that follow-on damages actions can be brought against subsidiaries of companies found to have infringed EU competition law. The victim of an infringement of EU competition law that was committed by a parent company and sanctioned by the EC may seek compensation from a subsidiary of the parent, even if the subsidiary was not referred to in the EC decision. To that end, the victim must prove that the parent and subsidiary companies constituted a single economic unit at the time of the infringement.

The single economic unit principle has been long established by the case law and is applied regularly in the context of private damages actions for breach of competition law. For example, in Case C-724/17, Vantaan kaupunki v Skanska Industrial Solutions Oy and Others (EU:C:2019:204), the CJEU applied the single economic unit doctrine and found that parent companies can be held liable for damages for competition infringements committed by their subsidiaries. By its judgment in Sumal, the CJEU goes one step further, placing liability of parent companies and subsidiaries on equal footing in follow-on damages claims.  


Between 1997 and 1999, Sumal SL acquired two trucks from the Spanish subsidiary of Daimler AG, Mercedes Benz Trucks España (MBTE). By a decision of July 2016, the EC found that between 1997 and 2011, Daimler AG had participated in the

European trucks cartel by concluding arrangements with 14 other European truck producers on pricing and gross price increases for trucks in the EEA. Following that decision, Sumal brought an action for damages against MBTE, seeking approximately EUR 22,000 for loss resulting from that cartel. Sumal’s action was rejected by the Commercial Court of Barcelona on the ground that MBTE was not an addressee of the EC’s decision.

Sumal appealed against this decision before the Spanish Provincial Court, which stayed the proceedings and referred to the CJEU the following questions:

  • “Does the doctrine of the single economic unit developed by the [Court] itself provide grounds for extending liability from the parent company to the subsidiary, or does the doctrine apply solely in order to extend liability from subsidiaries to the parent company?”

  • “In the context of intra-group relationships, should the concept of single economic unit be extended solely on the basis of issues of control, or can it also be extended on the basis of other criteria, including the possibility that the subsidiary may have benefited from the infringing acts?”

  • “If it is possible to extend liability from the parent company to the subsidiary, what would be required in order for it to be possible?”

The CJEU Judgment

The CJEU began by referring to the Skanska case and other relevant case law: 

At the outset, it should be recalled that Article 101(1) TFEU produces direct legal effects in relations between individuals and directly creates rights for individuals which national courts must protect.

The CJEU deduced from this that:

Any person is thus entitled to claim compensation for the harm suffered where there is a causal relationship between that harm and an agreement or practice prohibited under Article 101 TFEU . . . , it being understood that the determination of the entity which is required to provide compensation for damage caused by an infringement of Article 101 TFEU is directly governed by EU law.”

The CJEU next addressed the question of the entity required to provide compensation. The CJEU observed that actions for damages for infringement of EU competition rules are an integral part of the system for enforcement of these rules, which aim at sanctioning anticompetitive behaviour of undertakings and deterring them from engaging in such conduct. It would therefore be incorrect to apply the concept of an undertaking and the single economic unit principle differently, depending on whether it concerned fines imposed by the EC, or private actions for damages.

The CJEU observed that, according to wellestablished case law, the concept of an “undertaking”, within the meaning of Article 101 TFEU:

. . . covers any entity engaged in an economic activity, irrespective of the legal status of that entity and the way in which it is financed, and thus defines an economic unit even if in law that economic unit consists of several persons, natural or legal . . . . That economic unit consists of a unitary organisation of personal, tangible and intangible elements, which pursues a specific economic aim on a long-term basis and can contribute to the commission of an infringement of the kind in Article 101(1) TFEU . . .”

Then the CJEU observed that, according to the case law:

. . . the concept of an ‘undertaking’ and, through it, that of ‘economic unit’ automatically entail the application of joint and several liability amongst the entities of which the economic unit is made up at the time that the infringement was committed.”

The CJEU recognised, however, that, in conglomerate undertakings, the parent company could be part of several economic units and that it would be wrong to hold a subsidiary liable for infringements committed in the context of economic activities entirely unconnected to its own activity.

The CJEU therefore decided that:

. . . the possibility for the victim of an anticompetitive practice of invoking, in the context of an action for damages, the liability of a subsidiary company rather than that of the parent company cannot automatically be available against every subsidiary of a parent company targeted in a decision of the [EC sanctioning] conduct that amounts to an infringement.”

Drawing inspiration from the Advocate General’s opinion in the case, the CJEU stated that:

. . . the concept of an ‘undertaking’ used in Article 101 TFEU is a functional concept, in that the economic unit of which it is constituted must be identified having regard to the subject matter of the agreement at issue.”

The CJEU concluded that:

. . . there is nothing to prevent, in principle, a victim of an anticompetitive practice from bringing an action for damages against one of the legal entities which make up an economic unit and thus the undertaking which, by infringing Article 101(1) TFEU, caused the harm suffered by that victim.”

The CJEU drew a distinction, however, between cases where the infringement of Article 101(1) TFEU was found by the EC and other cases. As to the first, the CJEU simply applied Article 16 of Regulation No 1/2003:

However, as regards the situation in which an action for damages relies on a finding by the [EC] of an infringement of Article 101(1) TFEU in a decision addressed to the parent company of the defendant subsidiary company, the latter cannot challenge, before the national court, the existence of an infringement thus found by the [EC]. Article 16 (1) of Regulation No 1/2003 provides, inter alia, that where national courts rule on agreements, decisions or practices under Article 101 TFEU, which are already the subject of [an EC] decision, they cannot take decisions running counter to the decision adopted by the EC.”

And as a corollary:

. . . in a situation where the [EC] has not made a finding of conduct amounting to an infringement in a decision adopted under Article 101 TFEU, the subsidiary of a parent company which is accused of an infringement is naturally entitled to dispute not only that it belongs to the same undertaking as the parent company, but also the existence of the infringement alleged.” 

All of this is synthesised by the CJEU in the formal operative part of its “preliminary ruling” as follows:

. . . the victim of an anticompetitive practice by an undertaking may bring an action for damages, without distinction, either against a parent company who has been [sanctioned by the EC] for that practice in a decision or against a subsidiary of that company which is not referred to in that decision, where those companies together constitute a single economic unit. The subsidiary company concerned must be able effectively to rely on its rights of the defence in order to show that it does not belong to that undertaking and, where no decision has been adopted by the [EC] under Article 101 TFEU, it is also entitled to dispute the very existence of the conduct alleged to amount to an infringement.”


The practical consequence of the CJEU’s judgment is that a corporate group must think in terms of the different single economic units of which it is composed. An Article 101 TFEU infringement committed by one company in the corporate group will expose other companies in the group to possible fines and follow-on damages actions if those other companies are part of the same single economic unit as the first company. Liability can flow upwards from a subsidiary to the parent, and downwards from the parent to the subsidiary. Victims seeking large sums in damages will no doubt bring their claims against all companies in the relevant single economic unit in order to maximise the “solvability” available to fund an award in damages. 

There is no reason why the same principles should not also apply to infringements of competition law found by an NCA, except that it would be for national law to determine whether the NCA’s decision would be binding on the national courts (in the same way that an EC decision is binding pursuant to Article 16

(1) of Regulation No 1/2003).


C-450/19, Kilpailu-Ja Kuluttajavirasto, 14 January 2021

On 14 January 2021, the CJEU delivered a preliminary ruling in which it clarified when a bidrigging cartel offence comes to an end. This issue is particularly relevant with regard to the imposition of fines within the limitation period.


In April 2007, Fingrid Oyj, the company responsible for the development of the high-voltage electricity transmission network in Finland, published a call for tenders for the construction of a transmission line. The company Eltel submitted a bid on 4 June 2007 and subsequently won the tender. On 19 June 2007, it signed a contract with Fingrid. Eltel completed the works on 12 November 2009, and Fingrid paid the last instalment of the price on 7 January 2010.

In January 2013, the company Empower Oy submitted a leniency application to the Finnish Competition and Consumer Authority for bid-rigging practices implemented with Eltel. After investigation, in October 2014, the Competition and Consumer Authority submitted an application to the Finnish

Market Court to impose a fine of EUR 35 million on

Eltel for having reached an agreement with Empower Oy on prices, margins and market sharing for the design and construction of electricity transmission lines in Finland. 

In March 2016, the Finnish Market Court dismissed the application for the fine. It took the view that Eltel ceased to participate in the restriction of competition before 31 October 2009 and that therefore the infringement was time-barred. Under Finnish Law, a fine cannot be imposed if the application is not submitted to the Market Court within five years from the point at which the restraint of competition ended or the Competition and Consumer Authority became aware of the restriction of competition. According to the Market Court, the cartel covered the design work preceding the construction works (completed in January 2007) but did not cover the construction works themselves.

Subsequently, the Competition and Consumer

Authority appealed to the Supreme Administrative

Court of Finland against the decision of the Market

Court. According to the Competition and Consumer Authority, the date on which the last instalment was paid, 7 January 2010, had to be taken into account. Alternatively, the restriction of competition ended at the earliest on 12 November 2009, when the construction works were completed. 

According to Eltel, the limitation period began to run from the date of submission of the tender, 4 June 2007. Alternatively, Eltel stated that in cases in which the price can still be negotiated after the tender has been submitted, the limitation period began to run when the definitive contract for the project was signed, 19 June 2007.

The Supreme Administrative Court of Finland submitted a request for a preliminary ruling to the CJEU, to ascertain at what point in time the alleged participation of an undertaking in an infringement of Article 101 (1) TFEU can be regarded as having ended.

The CJEU Judgment

The CJEU held that the duration of Eltel’s participation in the alleged infringement covered the period during which it implemented the anticompetitive agreement concluded with competitors. This included the period during which Eltel’s price offer was in force or could have been converted into a final contract between Eltel and Fingrid. 

According to the CJEU, the restrictive effects of bidrigging on competition disappear “at the latest at the time when the essential characteristics of the contract, and in particular the overall price to be paid for the goods, works or services which are the subject of the contract, have been definitively determined”. At the moment of the conclusion of a contract between the successful tenderer and the contracting authority, the latter is definitively deprived of the opportunity to obtain the goods, works or services under normal market conditions.

The CJEU did not agree with the Competition and Consumer Authority that the harmful economic effects of the cartel lasted until the payment of the last instalment, and that downstream this could have resulted in higher electricity distribution tariffs.

According to the CJEU, a distinction must be drawn between, on the one hand, the restrictive effects that the cartel has on competition by depriving the contracting authority of normal market conditions and, on the other, the resulting wider adverse economic effects on other market players, on the basis of which such players can seek redress before national courts. Moreover, the question regarding the limitation period for an action for damages is distinct from the question as to when a competition law infringement ended.

The CJEU concluded by ruling formally that:

…the infringement period corresponds to the period up to the date of signature of the contract concluded between the undertaking and the contracting authority on the basis of the concerted bid submitted by that undertaking. It is for the national court to ascertain the date on which the essential characteristics of the relevant contract and, in particular, the total price to be paid for the work, have been definitively determined”.  (Case C-450/19, Kilpailu-ja kuluttajavirasto, EU:C:2021:10)


The CJEU judgment provides clarity as to the end date of a bid-rigging offence. In this respect, the determining factor is when the essential characteristics of the contract, and most importantly the price, were definitively specified. As such, it is irrelevant that the payment of the total price may have been staggered over time. 

Furthermore, the CJEU clearly distinguished the limitation period for the imposition of penalties from the limitation period for an action for damages. Therefore, the CJEU’s judgment has no bearing on the possibility for third parties to claim damages before the national courts.



AT.40394, Aspen, 10 February 2021

On 10 February 2021, the EC concluded its first excessive pricing investigation in the healthcare sector by accepting commitments from Aspen Pharmacare Holdings Ltd and Aspen Pharma Ireland

Limited (collectively, Aspen) (Decision C (2021) 724 final in Case AT.40394, Aspen). The decision highlights competition enforcers’ efforts to improve patients’ access to affordable, essential medicines.


On 15 May 2017, the EC launched its first excessive pricing investigation in the pharmaceutical industry, against Aspen, a South Africa-based company. The EC suspected that Aspen had been implementing significant price increases for six off-patent cancer medicines throughout the EEA, after acquiring them from another company. 

The Italian Competition Authority had already imposed a fine of EUR 5 million on Aspen in 2016 for implementing unfair price increases of up to 1500%, so the EC did not include Italy in its investigation. 

The EC’s Preliminary Assessment

First, the EC found that Aspen had very high market shares of more than 70%, and that the relevant markets were characterised by barriers to entry and expansion.

Second, the EC assessed Aspen’s pricing practices in light of the framework of analysis developed by the CJEU in the United Brands case. The EC applied a two-limb analysis: (1) whether the profits of Aspen were excessive, and (2) whether the prices charged by Aspen were unfair.

With regard to the first limb, the EC compared Aspen’s profitability with the profitability of comparable undertakings, i.e., a sample of other undertakings that sold similar products and had a profile similar to Aspen’s. Furthermore, the EC compared Aspen’s actual prices for the products to the products’ total cost increased by a “plus” element, i.e., a proxy for a reasonable profit margin under competitive conditions, based on the median measure of comparable undertakings in terms of profitability. In this respect, only a significant excess over that cost-plus level can be deemed excessive. As such, the EC found that Aspen generated profits exceeding the cost-plus level by almost 300% from 2013 to 2019. Therefore, the EC raised concerns that the prices charged by Aspen were excessive.

With regard to the second limb, the EC analysed whether there were legitimate reasons underlying the level of Aspen’s prices (e.g., superior efficiencies). The EC found that there were no such legitimate reasons. The medicines concerned had been offpatent for 50 years, so research and development costs should already have been recouped. Moreover, Aspen’s prices and profits did not reflect any commercial risk-taking, innovation, investment or any material improvement regarding the products. The absence of legitimate reasons suggested an unfairness of the prices. Furthermore, the EC’s concerns about unfairness were supported by its finding that Aspen’s price increases of the products were disproportionate to the limited increases in its costs of production. 

According to the EC, Aspen was able to increase its prices because there were mostly no available alternatives to its medicines. When national authorities tried to oppose Aspen’s requests for price increases, Aspen threatened to withdraw the medicines from the national list of reimbursable medicines, or even to withdraw normal supply. 

In the light of all these considerations, the EC formally adopted its Preliminary Assessment on 19 June 2020. In that assessment, the EC expressed concerns that on all or most of the relevant markets for the six cancer medicines, Aspen may have held a dominant position within the meaning of Article 102

TFEU during at least parts of the period from 1 July

2012 until 30 June 2019 and might continue to do so. In addition, the EC found provisionally that Aspen may have abused its dominant position by imposing excessive prices which were “unfair” within the meaning of Article 102 (a) TFEU.


In light of the EC’s findings, Aspen offered commitments without acknowledging that it had breached competition laws. By its 10 February 2021 decision, the EC made the following commitments legally binding on Aspen.

Price Commitments 

Aspen offered to reduce its prices for the medicines concerned by 73% on average throughout the EEA. These reduced prices are set per country and are, overall, below the prices charged in 2012, when Aspen started to increase its prices. 

Aspen committed to implement these reduced prices for a period of 10 years as of the commitment decision date. In addition to the 10-year price commitment, Aspen committed to apply the reduced net prices with retroactive effect from 1 October 2019, the date when Aspen first approached the EC with a concrete commitments proposal. Aspen would reimburse the amounts charged in excess of the reduced net prices as from 1 October 2019.

Supply Commitments

Aspen agreed to ensure appropriate and continued supplies of the medicines concerned for a period of five years. During the five years thereafter, Aspen will continue commercialising the drugs, but may also sell its market authorisation to a third party.


Even though Aspen was not actually fined, the EC’s decision sends an important message to pharmaceutical companies. Executive Vice-President Margrethe Vestager, in charge of competition policy, noted that it gives “a strong signal to other dominant pharmaceutical companies not to engage in abusive pricing practices to exploit our health systems.”

Next to the EC, NCAs have also stepped up enforcement against excessive pricing in the healthcare sector. For example, in July 2021, the Dutch Authority for Consumers & Markets fined Leadiant for charging an excessive price for prescription drug chenodeoxycholic acid. In 2018, the Danish Competition and Consumer Authority also found that CD Pharma abused its dominant position through a 2000% price increase for the drug Syntocinon.

The decision should be considered as crucial in view of the EC’s and NCAs’ overall competition enforcement action in the pharmaceutical industry. The EC considers that overall competition enforcement ensures access to cheaper medicines and to a broader choice. In addition, it stimulates innovation. 

Although the focus on excessive pricing cases has increased, it should be noted that the EC and NCAs are also tackling practices such as exclusionary conduct aimed at delaying the entry of generics onto the market, and pay-for-delay agreements, as part of their efforts to achieve fair prices for pharmaceutical products.



T-612/17, Google and Alphabet v European Commission

(Google Shopping), 10 November 2021

Google’s flagship product is the Google search engine, which provides search results to consumers, who pay for the service with their data. Almost 90% of Google’s revenues stem from adverts, such as those it shows consumers in response to a search query. In 2004, Google entered the separate market of comparison shopping in Europe with a product called Google Shopping. This product allows consumers to compare products and prices online and find deals from online retailers of all types.

In order to be competitive, comparison shopping services rely to a large extent on traffic. More traffic leads to more clicks, more revenue and more retailers wanting to list their products with a comparisonshopping service. Given Google’s dominance in general internet search, its search engine is an important source of traffic for comparison shopping services.

The EC found that from 2008, Google implemented in EU markets an anticompetitive strategy to promote its comparison-shopping service:

  • Google had systematically been giving prominent and more eye-catching placement in its search results to its own comparisonshopping service (in dedicated “boxes”).
  • Google had been demoting rival comparisonshopping services by placing them further down as simple generic results (blue links) with the use of adjustment algorithms.

By decision of 27 June 2017, the EC found that Google had abused its dominant position on the market for online general search services by giving its own comparison-shopping service an illegal advantage over competing comparison-shopping services.

The EC imposed a penalty of just over EUR 2.4 billion on Google. Just over EUR 0.5 billion of this amount was imposed jointly and severally with Alphabet, Google’s parent company. On 10 November 2021, the GCEU issued a judgment largely dismissing Google’s and Alphabet’s action for annulment of the EC’s decision.

The GCEU’s Judgment 

The GCEU recognised the anticompetitive nature of the practice at issue, agreeing with the EC that Google had used its dominant position on the market for general search services to succeed in the separate market for specialised search services for comparison-shopping. (Case T-612/17, Google and

Alphabet v Commission (Google Shopping),


The GCEU began its analysis by recalling that an undertaking’s dominant position alone, even one on the scale of Google’s, is not anticompetitive as such, even if it is to be expanded into a neighbouring market.

However, by favouring its own comparison-shopping service on its general results pages while demoting the results from competing comparison services, Google “departed from competition on the merits”, “intentionally, not negligently.” The GCEU considered in particular the importance for comparison-shopping services of the traffic generated by Google’s general search engine and the behaviour of consumers, who typically concentrate on the first few results. The GCEU concluded that Google’s conduct was liable to weaken competition in respect of the market for comparison-shopping services. 

The GCEU also ruled that Google’s practice of favouring its own comparison-shopping service over competing services, rather than a better result over another result, involved a certain form of abnormality. First, the GCEU inferred, from case law, a general principle of “equal treatment”, which, in the context of Article 102 TFEU, would require that like situations be treated alike unless objectively justified. Second, the GCEU explained that Google’s conduct was inconsistent with the “role and value” of a general search engine, which: 

. . . lie in its capacity to be open to results from external (third-party) sources and to display these multiple and diverse sources on its general results pages, sources which enrich and enhance the credibility of the search engine as far as the general public is concerned, and enable it to benefit from the network effects and economies of scale that are essential for its development and its subsistence”.

Furthermore, the GCEU described the distinction between a refusal to supply, as in the Bronner case (CJEU, Case C-7/97, Oscar Bronner GmbH & Co.

KG v Mediaprint Zeitungs- und Zeitschriftenverlag GmbH & Co. KG and others, EU:C:1998:569), and differential treatment, as in Google Shopping.

According to the GCEU, the EC had not erred in concluding that the “essential facilities” doctrine in the case law and its conditions were not applicable in this case. Google’s general results page has characteristics “akin to those of an essential facility, as there is no current genuine available substitute that would enable it to be replaced in an economically viable manner”, and is therefore considered indispensable for competing comparison shopping services. However, not every practice relating to access to such a facility must necessarily be assessed in the light of the conditions applicable to a refusal to supply, as set out in Bronner, on which Google relied in support of its arguments. The GCEU considered that the practice at issue was based not on a refusal to supply but on differential treatment by Google for the sole benefit of its own comparison service. 

The GCEU recalled that, under Article 102 TFEU, an abuse of a dominant position exists where the dominant undertaking hinders the maintenance or growth of competition in the market through the use of methods different from those adopted under normal competitive circumstances. Such abuse may be established merely by demonstrating that the conduct is capable of restricting competition. The GCEU argued that “in the case of an abuse of a dominant position, it is sufficient to establish that there are potential effects (rather than actual effects).” Therefore, “in order to find that Google had abused its dominant position, the Commission had to demonstrate the—at least potential—effects attributable to the impugned conduct of restricting or eliminating competition on the relevant markets, taking into account all the relevant circumstances.”

According to the GCEU, the EC was correct to find that the potential outcome of Google’s conduct would be the disappearance of comparison-shopping services, less innovation on the market for such services, and less choice for consumers. All these are characteristic features of a weakening of competition.

The GCEU considered two separate markets in the case at hand and confirmed that the concept of selfpreferencing was indeed met. In particular, Google had used its dominant position in the market for general search services to help it succeed in another market, namely the one for specialised search services for comparison-shopping. Although the GCEU confirmed the EC’s analysis in respect of the market for comparison-shopping, it considered that the EC did not establish that Google’s conduct had had—even potential—anticompetitive effects on the market for general search services, and therefore annulled the EC’s finding of an infringement in respect of that market alone. 

The GCEU stated that “the use of an as efficient competitor (‘AEC’) test is warranted in the case of pricing practices, but not in non-pricing practices.” The use of such a test, which involves comparing prices and costs, did not therefore make sense in the present case, since the competition issue identified was not one of pricing.

As regards objective justification of a company’s anticompetitive behaviour under Article 102 TFEU, the GCEU recalled that a company may argue either that its conduct is objectively necessary from a technical or commercial point of view, or that the exclusionary effect produced may be counterbalanced by efficiencies that also benefit consumers. Regarding efficiency gains in particular, the GCEU recalled that the burden of proof is: 

. . . on the dominant undertaking [which must] show that the efficiency gains likely to result from the conduct under consideration counteract any likely negative effects on competition and consumer welfare in the affected markets, that those gains have been, or are likely to be, brought about as a result of that conduct, that such conduct is necessary for the achievement of those gains in efficiency and that it does not eliminate effective competition by removing all or most existing sources of actual or potential competition.

The GCEU ruled out any objective justification for Google’s conduct. It found that while the algorithms for the ranking of generic results or the criteria for the positioning and display of Google’s specialised product results may, as such, represent procompetitive service improvements, that does not justify the unequal treatment of results from Google’s comparison shopping service and results from competing comparison shopping services. Google had not demonstrated efficiency gains linked to that practice that would counteract the negative effects on competition.

Last of all, the GCEU carried out its own assessment of the facts with a view to determining the level of the penalty. The GCEU found, first, that even though it annulled that part of the EC’s decision that concerned the market for general search services, this should not have any impact on the EC’s determination of the amount of the fine. The EC had not taken into account the value of Google’s sales on that market when evaluating the amount of the fine. Furthermore, the GCEU stressed the serious nature of the infringement and the fact that Google acted intentionally. For these different reasons, the GCEU maintained the amount of the fine (approximately EUR 2.42 billion). 

Practical Considerations

After the GCEU’s judgment was issued, a member of the European Parliament commented:

The Court’s Google shopping ruling highlights urgency of need to tackle monopoly position of ‘Big Tech’ companies. Legislators have to protect citizens’ choice and access to essential infrastructure to ensure that digital markets remain open and fair.

The GCEU’s ruling in Google Shopping is in line with the EC’s proposal for the Digital Markets Act (DMA), which will prohibit digital “gatekeepers” from engaging in any of a long list of practices that are considered to limit “contestability” or to be simply “unfair” (see article on page 57). A large online platform would qualify as a “gatekeeper” for the purposes of the DMA if (1) it had a significant impact on the internal market, (2) it operated a core platform service which served as an important gateway for business users to reach end users, and (3) it enjoyed an entrenched and durable position in its operations or it was foreseeable that it would enjoy such a position in the near future. 

Self-preferencing of the kind engaged in by Google in the Google Shopping case would be just one of the practices prohibited specifically by the DMA. The significance of the DMA is that it would list specific prohibited practices in far more detail than the broad definition of abuse of a dominant position contained in Article 102 TFEU. In theory at least, this would make it easier for the EC to identify and put a stop to illegal unfair conduct in the digital sphere.

It should be emphasised that the DMA is a proposal and is likely to be amended as it wends its way through the legislative process of scrutiny by the European Parliament and by the Member States in the Council of the EU. This process takes time, and so the proposed DMA is unlikely to be in force before 2023. In anticipation, dominant companies, particularly those in the high-tech field, would be well advised to take note of the provisions in the proposed DMA and to assess its likely future implications for their business practices.


C165/19 P, Slovak Telekom v European Commission; C152/19 P, Deutsche Telekom AG v European Commission,

25 March 2021

As the incumbent telecommunications operator in Slovakia, Slovak Telekom a.s. (ST) offers broadband services on its fixed copper and fibre optic networks. ST’s networks also include the local loop, i.e., the physical lines which connect the subscriber’s telephone termination point with the main distribution frame of the fixed telephone network. On 8 March 2005, ST was designated as an operator with significant market power on the wholesale market for unbundled access to the local loop by the Slovak national regulatory authority. As such, ST was obliged to grant alternative operators access to its local loop pursuant to Regulation (EC) No 2887/2000 of 18 December 2000 on unbundled access to the local loop and Directive 2002/21/EC of 7 March 2002 on a common regulatory framework for electronic communications networks and services.  

On 15 October 2014, the EC fined ST and its parent company, Deutsche Telekom AG (DT), for abuse of a dominant position on the Slovak market for broadband internet services. The EC found that ST and DT engaged in the following illegal practices:

  • They withheld from alternative operators network information necessary for the unbundling of local loops.

  • They reduced the scope of ST’s obligations regarding unbundled local loops. 

  • They set unfair terms and conditions in ST’s reference unbundling offer regarding collocation, qualification, forecasting, repairs and bank guarantees. 

  • They applied unfair tariffs which did not allow a competitor as efficient as ST relying on wholesale access to the unbundled local loops of that operator to replicate the retail broadband services offered by that operator without incurring a loss. 

As a consequence, the EC imposed a fine of just under EUR 39 million on ST and DT, jointly and severally, and a fine of just over EUR 31 million on DT. 

By judgments of 13 December 2018, the GCEU partially annulled the EC’s decision and set the fine for which ST and DT had been found jointly and severally liable at just over EUR 38 million, and the fine for which DT alone had been found liable at just over EUR 19 million. 

ST and DT appealed to the CJEU, requesting that the GCEU’s judgments be set aside in whole and that the fine imposed by the EC be cancelled or reduced. 

Both ST and DT pleaded that an infrastructure must be indispensable to competitors for a refusal to grant access to constitute a breach of Article 102 TFEU. On the issue of joint and several liability, DT also pleaded that it did not exercise a decisive influence over ST. 

The CJEU Judgment 

There Was No Need for the EC to Demonstrate that the Access to the Infrastructure Was Indispensable for the Competitors’ Activity 

Referring to the judgment of 26 November 1998, Bronner, ST argued that the GCEU erred in law when it considered that the EC was not required to prove that access to ST’s local loop was indispensable to alternative operators in order to qualify ST’s conduct as an abuse of dominance.

In the Bronner case law, the GCEU provided three conditions to determine if the refusal to grant access to infrastructure constitutes an abuse of a dominant position: 

  • The refusal to grant access must eliminate all competition from alternative operators requesting access.

  • That refusal cannot be objectively justified.

  • Such access must be indispensable for the activity of alternative operators, i.e., there is no other infrastructure or replacement for this infrastructure. 

The CJEU recalled that ST was obliged to grant alternative operators access to its local loop pursuant to a telecommunication regulatory obligation. In that context, the CJEU concluded that the practices at issue did not constitute a refusal of access to ST’s local loop but related to the conditions for such access. 

Indeed, ST did grant access to its infrastructure, according to the obligation set by the Slovak national regulatory authority in 2005. The practices did not constitute a refusal of access but an imposition of unfair conditions for such access.

Therefore, none of the conditions set out in the

Bronner case law applied in this context, and the CJEU confirmed the GCEU’s finding to the effect that the EC was not required to demonstrate that the access was indispensable to the activity of the competitors of ST.

The CJEU Rejected DT’s Argument Regarding Its Decisive Influence over ST 

As the parent company holding up to 51% of the capital of ST, DT argued that the GCEU erred in law when it considered that certain facts constituted indications that DT actually exercised decisive influence over ST, when these facts could only have demonstrated the possibility for the parent company to exercise a decisive influence. 

However, the CJEU rejected all DT’s arguments. In particular, the CJEU observed: 

. . . as regards ST’s reporting to the appellant, the [GCEU] did not err in law when it considered . . . that regular reporting, by a subsidiary to its parent company, of detailed information relating to its commercial policy was liable to establish awareness on the part of the parent company of its subsidiary’s conduct on the market and, consequently, to put the

parent company in a position to intervene in a more informed and therefore efficient way in the commercial policy of that subsidiary. Furthermore, while the fact that a subsidiary is required to send reports to its parent company concerning its commercial policy and financial results cannot in itself constitute an indication of an actual exercise of decisive influence by a parent company over its subsidiary, that fact can contribute to supporting such indications. Thus, the [GCEU] did not err in law by considering . . . that the regular reporting to [DT] of information concerning ST’s commercial policy was capable of contributing, along with other indicators, to establishing that those companies formed a single economic unit.”


In this case the CJEU confirmed the validity of the conditions provided in the Bronner case law in the event that a company in a dominant position refuses to grant competitors access to its infrastructure. However, it is interesting to note that these conditions do not apply where a company grants access but under unfair terms. Indeed, the CJEU specified that these practices did not equate to a refusal of access to the infrastructure but constituted a form of abuse in so far as they gave rise to anticompetitive effects. 

The ruling of the CJEU is also important as it considered that the EC was entitled to take into consideration, inter alia, the fact that a parent company could exercise decisive influence over its subsidiary as an indication of its actual exercise of decisive influence.



M.8181, Merck/Sigma-Aldrich (Fines), 3 May 2021

On 3 May 2021, the EC fined Sigma-Aldrich EUR 7.5 million for providing incorrect or misleading information during the EC’s review of Merck’s planned acquisition of the company.


On 21 April 2015, Merck notified the EC of its plan to acquire Sigma-Aldrich. On 15 June 2015, the EC approved the proposed acquisition subject to the divestiture of certain Sigma-Aldrich assets, which would address the competition concerns identified in markets for specific laboratory chemicals (Decision C

(2015) 4188 final of 15 June 2015 in Case M.7435, Merck/Sigma-Aldrich).

During the course of the divestment process, the EC became aware of a Sigma-Aldrich innovation project called iCap that had not previously been disclosed. The merging parties failed to mention it in remedy discussions. The parties also withheld information relating to this project in their responses to requests for information. Moreover, the EC inferred from internal documents that these omissions were intended to avoid the need to include the project as part of the business to be divested. 

The EC Decision

According to Article 14 (1) EUMR, the EC can impose fines of up to 1% of the aggregated turnover of companies that intentionally or at least negligently provide incorrect or misleading information. 

The EC found that Sigma-Aldrich committed three distinct infringements by providing, deliberately or at least negligently, incorrect or misleading information in the explanatory submission describing the remedy package and in the replies to two requests for information made pursuant to Article 11 (2) of the EUMR. In particular, the EC emphasised the importance of providing complete and accurate statements regarding R&D projects. R&D initiatives are normally not known publicly and therefore the EC must rely on the merging parties for information regarding such activities.

Although initially the EC began proceedings against Merck and Sigma-Aldrich, only Sigma-Aldrich was fined ultimately. Following hearings in 2020, the allegations against Merck were dropped. (As of this publication date, the official report of the EC’s fining decision had not been published.)


This is the third time since the entry into force of the EUMR that the EC has imposed fines against an undertaking for providing misleading information in the context of a merger control review: 

  • In 2017, the EC fined Facebook EUR 110 million for supplying incorrect and misleading information during the review of its acquisition of WhatsApp. In its merger notification, Facebook asserted that it would not be able to automatically link Facebook and WhatsApp user accounts. However, the EC discovered later that Facebook employees at the time were fully aware that the technical possibility of linking these user accounts existed. 

  • In 2019, the EC fined GE EUR 52 million for supplying incorrect and misleading information in the context of the review of its planned acquisition of LM Wind. GE had informed the EC that it did not have any offshore turbines in development beyond its existing six megawatt turbine, while the EC later learned from a third party that, in fact, GE was developing a turbine with a higher wattage and offering this turbine to customers.

These cases also reflect a broader global trend: competition authorities are increasingly sanctioning parties for procedural infringements of their merger rules—not only regarding incomplete or inaccurate submissions but also so-called “gun-jumping” (i.e., implementing a deal before required clearances) as illustrated by the Altice/PT Portugal case discussed on page 38. 

The EC relies heavily on the information submitted by merging parties in the Form CO to come to an informed decision when reviewing mergers, and to ensure the effectiveness of the merger control system. The three fining decisions adopted so far send a strong message that undertakings must disclose conscientiously all relevant information during merger control investigations. In addition, the EC has the power to revoke a decision approving a concentration if that decision was based on incorrect information for which one of the undertakings was responsible or where the decision was obtained by deceit. This possibility has not been used to date.  

Companies therefore must proceed with care in terms of the accuracy and completeness of the information they provide to the EC during a merger review, including in relation to any discussions about remedies.


M.9969, Veolia/Suez, 17 December 2020

On 17 December 2020, the EC rejected Suez’s gunjumping claim and provided further clarifications on the scope of the exemption to the standstill obligation in the case of two-step acquisitions encompassing a public bid. 


On 30 August 2020, the French company Veolia announced its intention to take over its main competitor in the French water supply market, Suez, with a view to creating “the great French world champion of the ecological transition.” 

Veolia structured its takeover project in two steps: first, a purchase of a minority stake of 29.9% of Suez’s shares from Engie, on 5 October 2020, followed by a takeover bid to acquire the remaining

70.1% of the share capital.  

On 16 October 2020, Suez wrote to the EC requesting it to find that, by acquiring the 29.9% holding in Suez on 5 October 2020, Veolia had infringed Article 7 (1) EUMR, which is the so-called “standstill” provision that prohibits implementation of a notifiable concentration until it has been cleared by the EC. The letter stated that it was a formal summons to act made pursuant to Article 265 TFEU. The EC did not agree that it was legally obliged to reply to Suez’s letter of 16 October 2020, but adopted a formal decision nevertheless, in the interests of clarity.


The EC Decision

First, Suez argued that the two steps (acquisition of Engie’s minority shareholding and the public bid to take control) should be viewed as a single concentration by which Veolia sought to acquire the sole control of Suez, so that Veolia should have obtained clearance from the EC before implementing the first step of the acquisition (i.e., the acquisition from Engie of 29.9% of Suez’s shares).

The EC agreed with Suez that the two steps of the acquisition pursued the same economic objective (the full acquisition of Suez) and were interdependent given that the first transaction would not have happened without the public bid. These two steps therefore constituted a “single concentration” within the meaning of Article 7 (1) EUMR. However, the EC went on to conclude that this single concentration was made up of “a series of transactions in securities, . . . by which control [was] acquired from various sellers” within the meaning of Article 7 (2) EUMR. Thus, the entirety of the concentration was exempted from the “standstill” requirement normally imposed by Article 7 (1) EUMR provided that the conditions laid down in Article 7 (2) were respected concerning notification to the EC without delay and non-exercise of voting rights. 

Second, Suez claimed that the first step of the transaction did not correspond to any of the situations where the derogation would apply: it was neither a “series of transactions in securities with various sellers” nor a public bid, which was supposed to be launched only after the EC’s clearance. Suez argued further that applying the public bid exemption would broaden the scope of that exemption, whereas derogations need to be interpreted strictly. 

The EC found no reason to exclude operations intended to acquire control from several vendors through both a (bilateral) transaction on securities and a public bid from the scope of application of Article 7 (2) EUMR, which provides a derogation from the standstill obligation in two situations: (1) a public bid, and (2) “ a series of transactions in securities, including those convertible into other securities admitted to trading on a market such as a stock exchange, by which control is acquired from various sellers.”

Rather, the EC considered the two steps as constituting “a series of transactions on securities” leading to the acquisition of control from several vendors. In fact, since the 29.9% block constituted a non-controlling minority interest within the meaning of competition law, suspending the first step of the transaction would have led the EC to suspend a transaction that did not result in a change of control, a result for which the EUMR does not provide.

Therefore the EC decided that it could not, on the one hand, consider that the two steps constituted one single concentration and then, on the other hand, consider that the two steps should be subject to two distinct regimes. A finding that there was a single concentration implied that the concentration needed to be appraised as a whole. The public bid exemption therefore had to be applied also to the first step of the concentration, even though this step was neither a public bid in itself nor a “series of transactions on securities”.

The EC observed that its reasoning was in line with the observations of the GCEU in Case T-704/14, Marine Harvest v Commission (EU:T:2017:753), in which the GCEU said that “it is possible that the acquisition of a minority stake which does not confer control of the target undertaking, followed by a public bid, may form part of a single concentration which falls within the scope of Article 7 (2)”. It should be noted, in passing, that the first step of the acquisition did not give Veolia the possibility of exercising decisive influence over Suez from a competition law point of view.

For these reasons the EC rejected Suez’s gunjumping complaint against Veolia.  


The EC’s decision is in line with the CJEU’s preliminary ruling in Case C-633/16, Ernst & Young P/S v Konkurrencerådet (EU:C:2018:371), in which the CJEU said that Article 7 (1) of the EUMR:

. . . must be interpreted as meaning that a concentration is implemented only by a transaction which, in whole or in part, in fact or in law, contributes to the change in control of the target undertaking.”

However, assessing when a change of control occurs can be difficult and complex in two-step acquisitions, notably in so-called “warehousing” schemes where a company is sold to a temporary buyer, usually a bank, on the basis of an agreement providing for the future resale of the business to a final buyer. For example, the EC imposed fines for infringement of the standstill obligation in Canon/Toshiba (Decision C (2019) 4559 final in Case M.8179, (Canon/Toshiba Medical Systems Corporation), but not in Odile

Jacob (Decision C (2003) 5277 final in Case No COMP/M.2978, Lagardère/Natexis/VUP). The difference between these two cases would appear to be as follows: in Odile Jacob, Lagardère did not acquire any right of veto or appointment, nor the possibility of benefiting from sensitive information, until after the second step in the acquisition, whereas in Canon/Toshiba, Canon obtained a right of veto as a result of the first step. 

In conclusion, when organising a two-step acquisition companies need to be very careful to avoid any change of control in the first step of the acquisition.

The other main practical impact of this decision is that the EC clarified that Article 7 (2) may be applied in a “hybrid” scenario such as the one at stake. There is indeed no reason for the EC to exclude from the scope of this provision a situation in which control is acquired from different sellers through the combination of a private transaction involving securities and a subsequent public offer for the outstanding shares.


M.10493, Illumina/GRAIL (Interim Measures), 29 October 2021

The EC’s review of Illumina’s acquisition of GRAIL raises two novel issues: the assertion of jurisdiction by the EC over transactions that do not meet any national or EU jurisdictional merger control tests, and the use of interim measures to stop parties closing a transaction where the EC has decided it wants to review the compatibility of the proposed concentration under the EU merger rules. 


On 26 March 2021, the EC published guidance that expands its jurisdiction to review transactions under the EUMR that do not meet any EU or national merger control thresholds via the rarely used referral mechanism enshrined in Article 22 EUMR (Guidance). The Guidance provides that the EC can review proposed concentrations upon referral from Member States even when they themselves do not have jurisdiction. 

The Guidance was introduced to capture transactions where the EC considered the competitive importance of an undertaking is not reflected in the turnover it generates, often called “nascent” competitors. Within a month of publication of the Guidance, the EC found the first transaction subject to its broadened jurisdiction: the US company Illumina’s acquisition of the US company GRAIL involving no EU turnover.

On 20 April 2021, the EC accepted France’s Article 22 EUMR referral request although neither the EU nor the French jurisdictional thresholds were met. Belgium, Greece, Iceland, the Netherlands and Norway joined France in the referral request, although their jurisdictional thresholds were also not met.

Rather, the EC’s press release noted that the planned acquisition “threatens to significantly affect competition within the territory of the Member States.” When detailing the potential effects of the acquisition, the press release did not further elaborate on how competition within the territories of the Member States would be affected, in particular in the absence of turnover in the EU. The press release merely stated that “the combined entity could restrict access to or increase prices of next generation sequencers and reagents to the detriment of GRAIL’s rivals active in genomic cancer tests following the transaction.”

The request by the EC that Illumina notify its planned acquisition to the EC meant that Illumina could not implement the transaction. Illumina submitted the notification on 16 June 2021.

Following an initial review, the EC, on 22 July 2021, opened a second phase investigation citing concerns regarding vertical input foreclosure. The EC raised concerns that Illumina’s leading position in next generation sequencing (NGS) for genetic and genomic analysis could mean that following the transaction, Illumina could foreclose competitors of GRAIL from using its NGS for the development of cancer detection tests. The EC also found that Illumina could have the economic incentive to foreclose such competitors. The foreclosure strategy could harm patients in Europe by hampering innovation and reducing choice for doctors, patients and health systems.

Following several suspensions of the review period, the EC should reach its final in February 2022. 

Merger Control Review Without Jurisdiction 

The EC has struggled to capture transactions concerning “nascent” competitors ever since Facebook acquired WhatsApp. However, the referral mechanisms in the EUMR have meant that NCAs, whose thresholds for jurisdiction are considerably lower, have been able to refer up to the EC transactions of European significance.

In the past, the NCAs only referred transactions up to the EC when they themselves could assert jurisdiction.

The new Guidance no longer requires the NCAs to have jurisdiction before referring a transaction. Illumina has challenged this before the European Courts. The GCEU is due to decide on the EC’s new interpretation of Article 22 EUMR under its Guidance early in 2022. Alongside Illumina, several NCAs have also pushed back on the EC’s new approach. Notably, Germany has publicly stated it will only refer transactions up to the EC when its own domestic merger control thresholds are met. 

Breach of the Standstill Obligation  

On 18 August 2021, Illumina publicly announced that it had completed its acquisition of GRAIL. Under the EUMR, companies must not implement concentrations that are subject to the EC’s review unless and until they have been cleared by the EC (standstill obligation). Separate from the second phase review on the merits, the EC issued a Statement of Objections on 20 September 2021 and found Illumina and GRAIL on a preliminary basis to have breached the standstill obligation. For the first time in the history of the EUMR, the EC therefore imposed interim measures to restore or maintain effective competition. 

On 29 October 2021, after having heard the parties, the EC adopted interim measures to prevent harm to competition following Illumina’s closing of the acquisition of GRAIL prior to obtaining a nonopposition decision from the EC. The interim measures provide that: 

  • GRAIL shall be kept separate from Illumina and be run by (an) independent Hold Separate Manager, exclusively in the interest of GRAIL (and not of Illumina).

  • Illumina and GRAIL are prohibited from sharing confidential business information, except where such disclosure is required to comply with the law or in line with the ordinary course of their supplier-customer relationship.

  • Illumina has the obligation to finance additional funds necessary for the operation and development of GRAIL.

  • The business interactions between the parties must be undertaken at arm’s length, in line with industry practice, hence without unduly favouring GRAIL to the detriment of its competitors.

  • GRAIL must actively work on alternative options to the transaction to prepare for the possible scenario whereby the deal would have to be undone should the EC declare the transaction incompatible with the internal market.


The Illumina/GRAIL case raises three unusual questions: the jurisdiction of the EC under its new interpretation of Article 22, the extraterritorial jurisdiction of the EC to review a merger between two US companies where the target does not have any turnover in the EU, and finally the imposition of a fine for blatantly closing a deal without clearance from the EC. The level of any ultimate fine will depend on how confident the EC is in asserting jurisdiction under its new Guidance, and is likely to be high. 

Generally, the EC has considered transactions that raise horizontal concerns to be more problematic. The issue in Illumina’s acquisition of GRAIL, however, revolves around vertical concerns. Going forward we may therefore very well see the EC focusing also on transactions that raise vertical concerns.


T-691/18, KPN v European Commission, 27 January 2021

On 27 January 2021, the GCEU confirmed the EC’s decision declaring the merger of the two main Dutch cable operators, Ziggo N.V. and Liberty Global Plc, compatible with the internal market, thus rejecting the action introduced by their competitor, KPN BV, a provider of retail television offers over a cable network.


By its decision of 10 October 2014, the EC authorised Liberty Global’s acquisition of control over its competitor Ziggo, subject to conditions (Decision C (2014) 7241 final in Case

COMP/M.7000, Liberty Global/Ziggo). Initially, the EC had concerns that the transaction would have negative effects on the market for the wholesale supply and acquisition of TV channels and subsegments of this market. 

The parties finalised the merger on 21 November 2014 after the divestment by Liberty Global, among other things, of Film1 channel to Sony, thus cancelling the horizontal overlap and reducing the possibility of adverse vertical effects. 

Challenged by KPN, the EC’s conditional clearance decision was annulled on 26 October 2017 by the GCEU in Case T-394/15, KPN v Commission (EU:T:2017:756). The GCEU considered that the EC had failed to comply with its obligation to state reasons under Article 296 of the TFEU. After annulment of its initial decision, and in order to comply with the GCEU’s judgment, the EC undertook a reassessment of the merger, as a result of which it adopted a new decision on 30 May 2018, which once again declared the merger compatible with the internal market, subject to conditions (Decision C (2018) 3569 in Case COMP/M.7000, Liberty Global/Ziggo).

The GCEU’s Judgment

KPN contested the EC’s new decision, filing a further challenge, which was ultimately dismissed by the

GCEU’s judgment of 27 January 2021 (Case T691/18, KPN v Commission, EU:T:2021:43).

In its first plea, KPN raised the EC’s alleged manifest error of assessment regarding the definition of the relevant market and criticised the lack of further segmentation, per channel concerned, of the market for the supply and wholesale acquisition of premium pay sports channels. This plea was rejected in its entirety since the GCEU considered that the relevant market should not be further segmented due to the substitutability of the channels, in particular Ziggo Sport Totaal and Fox Sports.

By its second plea, KPN alleged a manifest error of assessment by the EC when it found that the merger did not raise vertical competition concerns in the market for the supply and wholesale acquisition of premium pay-TV sports channels. The EC first recalled the three conditions necessary to demonstrate input foreclosure, namely, the merged entity’s ability to foreclose on inputs significantly, its incentive to do so, and the negative downstream impact of such a strategy. Next, the GCEU observed that since Fox Sports’s market share (70%-80%) was more than double that of the merged entity (20%-30%), it would not be able to foreclose significantly. Since the three conditions are cumulative, and the first one was not satisfied, the GCEU rejected KPN’s second plea. 

Moreover, KPN claimed a manifest error of assessment in relation to the vertical effects of the merger resulting from input foreclosure in the supply and wholesale acquisition of premium pay-TV movie channels, in particular in relation to HBO content.

The GCEU rejected this plea on the basis that the merged entity was not active on the market for the supply and wholesale acquisition of premium pay-TV sports channels since 2014. 

Finally, the GCEU considered that the EC had given sufficient reasons in its decision as regards the definition of the relevant market and the absence of vertical effects and therefore rejected KPN’s claim that the EC had not complied with its obligation to state reasons. 


This case illustrates the uncertainties and delays that can be introduced into a merger transaction if a competitor of the merging entities manages to challenge the merger successfully before the GCEU. Faced with annulment of its clearance decision, the EC is obliged by Article 266 TFEU to adopt a new decision, taking into account all the errors identified by the GCEU. In the end, the EC cleared the Ziggo/Liberty Global merger in a manner that was acceptable to the GCEU, but all this took six years and three months from the EC’s initial clearance decision.  

It is relatively rare for a third party to succeed in challenging an EC decision approving a merger or acquisition. KPN was successful in its challenge to the Ziggo/Liberty Global merger, but it was unsuccessful a few years earlier when, in Case T370/17, KPN v Commission (EU:T:2019:354), it challenged the EC’s decision of 3 August 2016 approving the Vodafone/Ziggo joint venture created by Vodafone and Liberty Global (Decision C (2016) 5165 in Case COMP/M.7978, Vodafone - Liberty Global - Dutch JV). The GCEU rejected all of KPN’s pleas in that challenge, both the pleas as to the substance and the plea alleging failure of the EC to state adequate reasons.


T-425/18, Altice Europe v European Commission, 22

September 2021

On 22 September 2021, the GCEU upheld a decision from the EC by which it fined telecommunications operator Altice for gun-jumping. In particular, the GCEU affirmed that the EC could impose two separate fines: a fine for implementing a concentration prior to its clearance, and a fine for implementing a concentration prior to its notification. In coming to those findings, the GCEU also clarified the appropriateness of certain pre-closing covenants and information exchanges.


In December 2014 Altice signed a share purchase agreement (SPA) with telecommunications operator

Oi to acquire PT Portugal. The deal was subject to EU merger control.

Prior to signing, Altice began communications with the EC to inform it of Altice’s intention to acquire PT Portugal. Shortly after signing the SPA, Altice sent a case-team allocation request to the EC and commenced pre-notification discussions. Altice formally notified the transaction in February 2015. In April 2015, the EC cleared the acquisition subject to commitments. A gun-jumping investigation arose following press reports of contacts between Altice and PT Portugal taking place before the adoption of the EC’s clearance decision in 2015.

Three years after clearing the acquisition, the EC concluded that Altice infringed both the notification obligation and the standstill obligation under the EUMR, and imposed two separate fines totalling EUR 124.5 million.

The EC found that Altice had the possibility of exercising decisive influence or had exercised decisive influence over PT Portugal before the adoption of the clearance decision and, in some instances, before notification. Certain pre-closing provisions included in the SPA gave Altice the right to veto decisions regarding PT Portugal’s commercial policy. Based on these provisions, Altice had been involved in the day-to-day running of PT Portugal in several instances.

The GCEU’s Judgment

Altice brought an action for annulment before the

GCEU, which was dismissed in part (T-425/18,

Altice Europe v Commission, EU:T:2021:607). The GCEU sided with the EC but reduced the fine relating to the infringement of the notification obligation by 10% (from EUR 62.25 million to EUR 56.025 million). The GCEU considered it appropriate to lower the fine because Altice had informed the EC of the concentration before the signing of the SPA, and it had sent a case-team allocation request to the EC shortly after signing.

The GCEU confirmed that breach of the notification obligation and breach of the standstill obligation can be subject to separate fines. The GCEU held that the notification obligation (obligation to act) and standstill obligation (obligation not to act) are separate obligations. Because each obligation was violated, the EC was entitled to impose two fines.

Furthermore, the GCEU held that pre-closing provisions included in a SPA must not afford a purchaser the possibility of exercising decisive influence over the target. EU merger rules do not preclude pre-closing provisions in a SPA aimed at protecting the value of the target between signing and closing. However, such provisions “can only be reasonably justified if they are strictly limited to what is necessary to ensure the maintenance of the target’s value and do not afford a purchaser the possibility to exercise decisive influence over the target”. In this matter, the GCEU determined that three types of preclosing provisions included in the SPA were problematic:

  • The appointment, dismissal or changes to the contracts of senior management: the GCEU stated that the possibility to co-determine the structure of the senior management usually confers the power to exercise decisive influence on the commercial policy of an undertaking.

  • Pricing policies and standard terms and conditions: the GCEU stated that the wording of the pre-closing provision was extremely broad, resulting in an obligation for PT Portugal to obtain Altice’s consent to any change in prices and giving Altice the possibility to object to any change in PT Portugal’s customer contracts.

  • Entering into, termination or modification of certain types of contracts: the GCEU stated that the limitations in these types of pre-closing covenants were numerous and broad, and the monetary thresholds were so low that they went beyond what was necessary to preserve the value of Altice’s investment.

The GCEU held that the powers afforded to Altice based on these pre-closing provisions constituted veto rights that went beyond what was necessary to preserve the target’s value until the closing of the transaction and gave Altice the ability to exercise decisive influence over the target as from signature of the SPA.

The GCEU also found that Altice had exercised decisive influence over PT Portugal before the adoption of the clearance decision and, in some instances, before notification.

Importantly, the GCEU determined that the existence of a mere possibility of exercising decisive influence is sufficient to constitute an infringement of the notification obligation and/or standstill obligation.

Moreover, information exchanges may contribute to the exercise of decisive influence. The GCEU confirmed that exchanges of business-related information between a potential acquirer and a vendor can be considered a normal part of the acquisition process if the nature and purpose of such exchanges directly relate to the potential acquirer’s need to assess the value of the business. In the present case, the information exchanges between the parties continued after the signing of the SPA and related to commercially and competitively sensitive information (note that the EC had also found that the information exchanges took place in the absence of any type of confidentiality or “clean team” arrangement, but the GCEU did not discuss this point). The GCEU held that such information exchanges were not justified by the aim to maintain the target’s value. As such, the EC could conclude that the information exchanges contributed to the exercise of decisive influence by Altice over certain aspects of PT Portugal’s business.


The GCEU’s judgment provides an important clarification on how pre-closing covenants can result in gun-jumping risks.

In light of the Altice judgment, parties to an M&A transaction should diligently draft pre-closing provisions and consult their antitrust counsel as to the permissibility of such provisions. Furthermore, information exchanges should be handled carefully and should be limited to the need to assess the value of the business. Preferably, information exchanges should take place with safeguards in place, such as “clean team” arrangements.



T-516/18, Grand Duchy of Luxembourg v European Commission; T-525/18, Engie Global LNG Holding Sàrl and Others v European Commission, 12 May 2021

On 12 May 2021 the GCEU confirmed the EC’s decision of 20 June 2018 that the tax rulings which allowed the Engie group to partially escape tax in Luxembourg constituted illegal State aid within the meaning of Article 107 (1) TFEU.  


As from September 2008, the Luxembourg tax authorities issued several tax rulings concerning the tax treatment of certain financial transactions between four Luxembourg companies of the Engie group (then GDF Suez). The EC was concerned that these tax rulings appeared to constitute State aid within the meaning of Article 107 TFEU. It therefore sent a request for information to the Grand Duchy of Luxembourg asking for all tax rulings in force in the previous 10 years which had been granted to Engie group companies between 2004 and 23 March 2015, the date of the EC’s request. 

Luxembourg replied on 23 June 2015, providing information about the tax rulings issued in favour of several companies of the Engie group (then GDF Suez) which were resident in Luxembourg. 

This information showed that, from 2008 to 2014, the Luxembourg tax authorities adopted two series of tax rulings concerning certain transactions between companies of the French group Engie, all resident in Luxembourg.  

Each transaction took place in three stages:

  • A holding company transferred assets to a subsidiary.
  • To finance the transfer of assets, the subsidiary took out an interest-free loan from an intermediary company, obligatorily convertible into shares, called ZORA. Under that agreement, the subsidiary that had taken out the ZORA had to repay the loan at the due date by issuing shares in an amount equivalent to the nominal amount of the loan, plus a premium representing all of the profits made by the

subsidiary during the term of the loan, namely the ZORA accretions, minus a limited margin agreed with the Luxembourg tax authorities.

  • The intermediary company financed the loan to the subsidiary by entering into a prepaid forward sale agreement with the holding company, at the end of which the holding company paid the subsidiary an amount equal to the nominal amount in exchange for acquiring the rights to the shares that the subsidiary would issue upon conversion of the ZORA.

The aim of the manoeuvre was for the holding company to end up owning the rights on the shares, which would include not only the nominal amount lent, but also any profits made by the subsidiary.

In its Decision (EU) 2019/421 of 20 June 2018, the

EC considered that the tax treatment allowed by the Luxembourg tax authorities constituted State aid. This decision also ordered Luxembourg to recover the illegal aid from the Engie group.

The EC Decision

In September 2016, the EC opened a formal in-depth investigation into the tax rulings granted to the Engie group. 

By its decision issued on 20 June 2018, the EC considered that the tax rulings satisfied all four conditions for the existence of State aid as laid down in Article 107 (1) TFEU: 

  • The tax rulings were the result of an intervention by the Grand Duchy of Luxembourg and relieved the Engie group companies from tax.

  • The tax rulings strengthened the competitive position of part of the Engie group, operating on various markets in several Member States, thus affecting intra-EU trade.

  • The tax rulings gave Engie more favourable treatment than under the standard Luxembourg tax rules that exempt from taxation income received by a shareholder from its subsidiary only if that income is taxed at the level of the subsidiary, which was not the case here. Since

the selective treatment resulted in a substantial lowering of Engie’s tax liability, Engie was granted a selective advantage. 

  • The tax rulings distorted competition on the internal market by freeing up financial resources for Engie which the latter could use to invest in its business operations or to undertake further investments.

The EC therefore concluded that the tax rulings constituted State aid incompatible with the internal market. Consequently, the EC ordered Luxembourg to recover EUR 120 million from the Engie group. 

Engie and Luxembourg each brought actions before the GCEU for annulment of the EC’s decision in its entirety. In the alternative, they sought annulment of the provision in the EC’s decision requiring recovery of the aid.   

The GCEU Judgment  

In their actions for annulment, Engie and

Luxembourg raised several grounds, all of which the

GCEU rejected as unfounded in Joined Cases T516/18, Grand Duchy of Luxembourg v Commission, and T-525/18, Engie Global LNG Holding Sàrl and Others v Commission, (T:2021:251).

The following analysis concentrates on certain grounds grouped by the GCEU under the heading

“errors of assessment and of law in the identification of a selective advantage.” In addition, this analysis examines the two grounds raised only by Engie, namely, that the tax rulings could not be imputed to the State of Luxembourg and that they were incorrectly classified as individual aid.

The Plea of Luxembourg and Engie that the EC Committed Errors of Assessment and of Law in the Identification of a Selective Advantage

The GCEU divided this plea into two parts. First, there was the allegation that the EC confused the conditions for finding an advantage and for finding that the tax rulings were selective by failing to conduct a clear assessment of those two conditions separately. Second, there were various arguments challenging the EC’s assessment that a selective advantage existed.

The Allegation that the EC Confused the Conditions for Finding an Advantage and for Finding that the Tax Rulings Were Selective

In considering this allegation, the GCEU began by recalling the relevant case law on advantage and selectivity. So far as advantage is concerned, the EC must show that the measure improves the financial position of the recipient (Case C-173/73, Italy v Commission, EU:C:1974:71, paragraph 15). So far as selectivity is concerned, the EC must show that the advantage is not enjoyed by other undertakings in a legal and factual situation comparable to that of the recipient in the light of the objective of the reference framework (Joined Cases C-78/08 to C-80/08, Paint Graphos and Others, EU:C:2011:550, paragraph 49).

The GCEU observed that in tax matters the examination of an advantage overlaps with the examination of selectivity. Moreover, it is apparent from the case law that these two criteria may be examined together (Case C-270/15 P, Belgium v Commission, EU:C:2016:489, paragraph 32).

The GCEU then examined the EC’s reasoning on the existence of both an advantage and selectivity. Three main points can be distilled from this reasoning.

First, the EC found that it was possible for the holding companies to be exempted from paying tax on participation income that should have been taxed under the Luxembourg corporate income tax system in the absence of the tax rulings. The EC concluded therefore that those tax rulings conferred an advantage and also derogated from the Luxembourg corporate income tax system.

Second, the EC reasoned that, in the absence of the tax rulings, the holding companies would not have been eligible for an exemption from tax on distributed income because this income had not been taxed at the level of the subsidiaries. The EC therefore made a finding of both an advantage as well as a derogation from the provisions on the participation exemption and the taxation of profit distributions.

Third, the finding of a derogation from the Luxembourg fiscal legislation on abuse of law simultaneously entailed the grant of an advantage. In the absence of the tax rulings—and assuming a correct application of the legislation on abuse of law—the EC argued that the holding companies would not have been eligible for an exemption from tax on the participation income.

In light of these considerations, the GCEU concluded that the EC did not confuse the conditions for finding an advantage and those for demonstrating the selectivity of the tax rulings. 

The Various Arguments Challenging the EC’s Assessment that a Selective Advantage Existed

Engie and Luxembourg raised many arguments challenging the EC’s assessment that a selective advantage existed that turned on the specific provisions of Luxembourg tax law and were very technical. The GCEU rejected all these arguments. The most interesting argument was the complaint by Engie and the Grand Duchy of Luxembourg that the

EC failed to take into account the practice of the Luxembourg tax authorities.  

On this argument, the GCEU found that the EC was right to refer to a circular of 1989 from the Luxembourg authorities and to judicial practice in Luxembourg, from which it identified the four criteria necessary for a finding of an abuse of law under Luxembourg law. These four conditions were the use of forms or institutions governed by private law, the reduction in the tax burden, the use of inappropriate legal means and the absence of valid non-tax-related reasons.

  • In the present case, it was not disputed that the first criterion, use of forms or institutions governed by private law, was satisfied.  

  • Concerning the second criterion, the Grand Duchy of Luxembourg and Engie argued that the tax rulings did not result in a reduction in the tax burden of the subsidiaries, the intermediary companies and the holding companies. The GCEU found that the EC did not err in finding that, as a matter of Luxembourg tax law, the tax rulings resulted in a reduction of the tax burden of these companies.  

  • As regards the third criterion, namely the use of inappropriate legal means, the Grand Duchy of Luxembourg and Engie argued that it was appropriate in the present case to have recourse to an indirect ZORA, namely one involving the intervention of an intermediary company, in order to finance the transfer of the business sectors concerned to the subsidiaries. The GCEU rejected this argument on the grounds that Engie’s chosen means of financing was in direct conflict with the intention of the Luxembourg legislature. That intention could not reasonably be the promotion of complex financial arrangements resulting, in real terms, in the double non-taxation of distributed income at the level of a subsidiary and of its parent company.

  • On the fourth criterion, the Grand Duchy of Luxembourg and Engie argued that the financing transaction did not pursue a purely tax-related aim and that it was motivated by valid economic reasons. The GCEU rejected this argument on the ground that “while a taxpayer cannot be criticised for choosing the least onerous legal means, that does not apply where other appropriate means are available but the legal means chosen have an exclusively taxrelated aim and actually result in no tax being levied.”

Engie’s Plea that the Tax Rulings Could Not Be Imputed to the State of Luxembourg

Engie argued that the tax rulings could not be regarded as involving an intervention by the Grand Duchy of Luxembourg. They were optional and strictly limited to drawing inferences from the application of Luxembourg tax law to a given situation.

The GCEU began by recalling the relevant case law. In order to assess whether a measure is imputable to the State, it is necessary to examine whether the public authorities were involved in the adoption of that measure (Case C-405/16 P, Germany v Commission, EU:C:2019:268, paragraph 49). The GCEU noted that, in the present case, the tax rulings were adopted by the Luxembourg tax authorities. The GCEU concluded therefore that it could not be reasonably disputed that the tax rulings were imputable to the Grand Duchy of Luxembourg.

Engie also argued that the tax rulings did not involve the commitment of State resources. 

The GCEU observed that measures which mitigate the charges that are normally included in the budget of an undertaking and which therefore, without being subsidies in the strict meaning of the word, are similar in character and have the same effect are considered to constitute aid (Joined Cases C-399/10 P, Bouygues and Bouygues Télécom v Commission and Others, and C-401/10 P, Commission v France and Others, EU:C:2013:175, paragraph 101).

On the facts, the GCEU observed that the tax rulings made it possible for the holding companies not to be taxed on some of their participation income. In other words, the tax rulings mitigated charges that are generally included in the budget of an undertaking. The GCEU concluded therefore that the condition relating to the use of State resources was also met.

Engie’s Plea that the Tax Rulings Were Incorrectly Classified as Individual Aid

Engie observed that, under identical tax rulings, other undertakings had benefited from the same financing structure. According to Engie the EC therefore had to identify an aid scheme, as it did in Commission Decision (EU) 2016/1699 of 11 January 2016 on the excess profit exemption State aid scheme SA.37667.

The GCEU observed that, according to the case law, the EC may consider a measure applying a general scheme as individual aid without first being required to demonstrate that the provisions on which that scheme is based constitute an aid scheme (Joined Cases C-71/09 P, C-73/09 P and C-76/09 P, Comitato

‘Venezia vuole vivere’ and Others v Commission, EU:C:2011:368, paragraph 63).

Moreover, the GCEU had already concluded that the EC had demonstrated to the requisite legal standard that the tax rulings granted a selective advantage to the holding companies. The GCEU therefore concluded that the EC did not err in law in considering the tax rulings as constituting individual aid.

The case is on appeal before the CJEU.


Executive Vice-President Margrethe Vestager said, in a press release:

The General Court has … confirmed that State aid enforcement can be a tool to tackle abusive tax planning structures that deviate from the objectives of the general tax system.”

The Engie judgment comes nearly a year after the Apple judgment in Case T-778/16, Ireland v

Commission, and Case T-892/16, Apple Sales

International and Apple Operations Europe v Commission (EU:T:2020:338), and a year and a half after the Starbucks judgment in Case T-760/15, Netherlands v Commission, and Case T-636/16,

Starbucks Corp and Starbucks Manufacturing EMEA BV (EU:T:2019:669). Both the Apple and Starbucks cases concerned transfer pricing arrangements which the EC found to constitute State aid. In both cases the EC decision was annulled for technical errors, and both cases are before the CJEU on appeal.  

After the GCEU’s judgment in the Apple case, Executive Vice-President Vestager emphasised that the EC would study the judgment and learn from its mistakes. In the light of the Engie judgment, it would seem that the EC has progressed but still has some way to go.


T-648/19, Nike European Operations Netherlands and Converse Netherlands v European Commission, 14 July 2021

On 14 July 2021, the GCEU dismissed the action brought by Nike and Converse against the EC decision to initiate a formal State aid investigation concerning certain advance tax rulings issued by the

Dutch tax authorities for the benefit of Nike and Converse.


In 2019, the EC decided to initiate a formal investigation procedure into advance tax rulings issued by the Netherlands tax administration to Nike European Operations Netherlands (Nike) in 2006,

2010 and 2015, and to Converse Netherlands (Converse) in 2010 and 2015. Nike and Converse are two Dutch subsidiaries of a Dutch holding company, Nike Europe Holding, owned by Nike Inc. 

These advance tax rulings validated a transfer pricing agreement concerning the level of royalties payable by Nike and Converse to other Nike group companies in return for the use of intellectual property rights. As such, these royalties were deductible from the taxable income of Nike and Converse in the Netherlands.

According to the EC’s provisional assessment, those advance tax rulings conferred a selective advantage in that the income tax for Nike and Converse in the Netherlands was calculated on the basis of an annual level of profit lower than it would have been if the royalties in question had been priced at arm’s length for tax purposes. In other words, the amount of these royalties did not correspond to the amount that would have been negotiated under market conditions for comparable transactions between independent companies.

It was against this background that the EC decided in 2019 to open a formal investigation procedure to determine whether there might be illegal State aid. Nike and Converse asked the GCEU to annul the EC’s decision to open the investigation. They put forward arguments alleging breach of the duty to state reasons, manifest errors of assessment and failure to respect procedural rights. 

GCEU Judgment

In its judgment, the GCEU did not accept any of the arguments put forward and dismissed the action in its entirety.

First, concerning the alleged breach of the obligation to state reasons, Nike and Converse argued that the contested EC decision was inadequately reasoned in that it failed to state the relevant issues of fact and law and the reasons for concluding that the measures at issue met the conditions laid down in Article 107 (1) TFEU.

The GCEU recalled that the requirements to be met by a statement of reasons depend on the circumstances of each case, in particular the content of the measure in question, the nature of the reasons given and the interest which the addressees of the measure, or other parties to whom it is of direct and individual concern, may have in obtaining explanations (Case T-308/00 RENV, Salzgitter v Commission, EU:T:2013:30, paragraph 112). 

The GCEU recalled the case law according to which it is not necessary for the statement of reasons to examine all the relevant facts and points of law, since the question whether the statement of reasons meets the requirements of the second paragraph of Article 296 TFEU must be assessed not only in the light of its wording, but also in the light of its context and of all the legal rules governing the matter in question (Case C-501/00, Spain v Commission,

EU:C:2004:438, paragraph 73; Case T-308/00 RENV, Salzgitter v Commission, EU:T:2013:30, paragraph 113).

The GCEU summarised the law in a nutshell by observing that:

In accordance with Article 6 (1) of Regulation 2015/1589, the contested decision must simply summarise the relevant issues of fact and law, include a preliminary assessment as to the aid character of the proposed measure and set out the doubts as to its compatibility with the internal market. The decision must give the Member State concerned and other interested parties the opportunity to participate effectively in the formal investigation procedure.”

Viewed against this test, the GCEU considered that the EC had given an adequate statement of reasons. Moreover, the GCEU noted that the EC’s assessment of the measures at issue was not definitive and could evolve during the formal procedure for obtaining additional information. Consequently, the GCEU concluded that Nike and Converse could not complain that the EC’s reasoning as regards the individual character of the measures at issue was incomplete. 

Second, the GCEU examined the alleged violation of procedural rights. Nike and Converse argued that the lack of clarity of the contested decision did not allow for effective intervention during the formal investigation procedure. The GCEU dismissed this plea expeditiously on the grounds that, as beneficiaries of the measures at stake and therefore as interested parties, Nike and Converse were in a position to submit their observations to the EC during the formal investigation procedure. Consequently, the contested decision did allow for effective intervention in the formal investigation procedure. Moreover, there was no basis to claim that the contested decision prevented exercise of this right.

Third, the GCEU examined the allegation of Nike and Converse that the EC made manifest errors of assessment, first, by failing to extend the preliminary examination to cover the existence of a possible aid scheme and, second, by mischaracterising the nature of an advance tax ruling in Netherlands law.

As regards the argument alleging the EC’s failure to identify an aid scheme, the GCEU considered that, in the context of an action brought against a decision to initiate a formal investigation, “review by the GCEU is limited to ascertaining whether or not the EC made a manifest error of assessment in forming the view that it was unable to resolve all the difficulties on that point during its initial examination of the measure concerned” (Joined Cases T-269/99, T-271/99 and T272/99, Diputación Foral de Guipúzcoa and Others v Commission, EU:T:2002:258, paragraph 49).

The GCEU observed that nowhere in the contested decision did the EC determine whether an aid scheme existed. Referring to Joined Cases C-71/09 P, C73/09 P and C-76/09 P, Comitato ‘Venezia vuole vivere’ and Others v Commission (EU:C:2011:368, at paragraph 63), the GCEU went on to observe that this fact could not give rise to a manifest error of assessment because:

The [EC] is entitled to treat a measure as being individual aid without being obliged to verify beforehand and as a matter of priority whether that measure may have been derived from such a scheme.” 

The GCEU also observed that this finding was not affected by the allegedly declaratory nature of an advance tax ruling in Netherlands law. Assuming that this declaratory nature was established, it could not preclude the EC from treating an advance tax ruling as being addressed to a taxpayer irrespective of whether the ruling was based on an aid scheme or not. 

As regards the argument alleging the EC’s mischaracterisation of the nature of advance tax rulings in Netherlands law, the GCEU repeated the observation made above that, even if an advance tax ruling were declaratory in nature, this would not preclude the EC from treating such a ruling as an individual measure. The GCEU also observed that the assessment of an advance tax ruling and its nature in Netherlands law raised “serious difficulties” that warranted a thoroughgoing examination by the EC. The GCEU examined the various elements that would have to be analysed and concluded that:

. . . having regard to the difficulties inherent in such an analysis, the initiation of the formal investigation procedure cannot reasonably be challenged.”

Fourth, the GCEU examined the allegation the EC had made an incorrect assessment of the selectivity of the measures at issue. According to Nike and Converse, the fact that an advance tax ruling was issued in favour of a single company did not enable the EC to conclude that it constituted individual aid. The GCEU concluded that the conditions for a provisional presumption of selectivity of the measures at issue were satisfied in the present case: 

  • The measures at stake were advance tax rulings concluded between the Netherlands tax authorities and Nike and Converse, and were intended to govern only the tax situation of these two companies. 
  • The EC had found provisionally that an economic advantage was conferred on Nike and Converse, that advantage resulting, in substance, from a reduction of the tax base of these two companies. Referring to Case T-314/15, Greece v Commission, (EU:T:2017:903, paragraph 79), the GCEU observed that, in the case of individual aid, as here:

. . . the identification of the economic advantage is, in principle, sufficient to support the presumption that it is selective. The presumption of selectivity operates independently of the question whether there are operators on the relevant market or markets which are in a comparable factual and legal situation.”  Fifth, the GCEU examined the allegation that the initiation of the formal investigation procedure had been premature. Nike and Converse criticised the alleged premature opening of the formal investigation procedure, at a stage when the investigation was not sufficiently established, and the difficulties encountered could be overcome on the basis of a thorough preliminary examination. The GCEU recalled that the EC can open a formal investigation procedure if it has serious difficulties after a preliminary assessment. Because the classification of a measure as State aid in a decision to open the formal investigation procedure is only provisional, the EC had fulfilled its obligation to initiate a formal investigation in the event of serious difficulties. 

Sixth and last, Nike and Converse claimed that the EC had violated the principles of good administration and equal treatment by arbitrarily choosing to examine whether the measures at issue were in compliance with State aid law, and by not extending its analysis to the general scheme on which the measures at issue were based and to its potential beneficiaries. The GCEU considered that the EC had not shown impartiality or lack of diligence by not extending its preliminary examination to the identification of a possible aid scheme. There is no rule in the TFEU nor in Regulation No 2015/1589 which requires the EC, in the case of an individual measure, to check first whether there is an aid scheme.


It is a general principle of EU law that an EC decision to open an investigation cannot be challenged before the courts because the investigation is a preliminary step that does not change individuals’ rights. Indeed, in Case C-60/81, IBM v Commission (EU:C:1981:264), the CJEU ruled, at paragraph 21, that “neither the initiation of a procedure nor a statement of objections may be considered, on the basis of their nature and the legal effects they produce, as being decisions within the meaning of Article 173 of the EEC Treaty which may be challenged in an action for a declaration that they are void.” Since the CJEU’s judgment in Case C-312/90, Spain v Commission -“Cenemesa” (EU:C:1992:132), it is well established that a decision to open an investigation pursuant to Article 108(2) TFEU does affect individuals’ rights and can be challenged before the courts. 

However, it will be difficult to succeed in obtaining annulment of an EC decision to open an investigation pursuant to Article 108 (2) if the EC is faced with serious difficulties as to whether the measure constitutes State aid within the meaning of Article 107 (1) TFEU. Initiating a formal in-depth investigation is a reasonable way to obtain the information necessary to resolve the matter.


C-337/19 P, European Commission v Belgium And Magnetrol

International, 16 September 2021

On 16 September 2021, the CJEU delivered its judgment on whether the Belgian system of excess profit rulings (EPRs) constituted a scheme or not. The CJEU set aside the GCEU’s judgment that annulled the EC’s decision qualifying the EPRs as an aid scheme. The CJEU referred the case back to the GCEU to rule on the remaining aspects of the case, principally, whether the EPRs constituted State aid. 


In 2016, the EC found that the Belgian system of EPRs constituted State aid in the form of an aid scheme, that was incompatible with the internal market and put into effect in breach of the standstill obligation. By means of an EPR, the profit recorded in Belgium in the Belgian financial reports of a multinational group that exceeded the arm’s length profit was not included in the taxable profit in Belgium.  

Belgium and one of the beneficiaries of an EPR, Magnetrol, brought an action for annulment before the GCEU. In 2019, the GCEU annulled the EC decision on the basis that the EC wrongly concluded that the EPRs constituted an aid scheme. Consequently, the EC brought an appeal before the CJEU.

The CJEU Judgment

The CJEU’s judgment concerned the issue of whether the EC could correctly classify the EPRs as a scheme. The CJEU did not rule on whether the EPRs constituted State aid.

First, the CJEU referred to the definition of an “aid scheme” as set out in Article 1 (d) of Regulation 2015/1589: “any act on the basis of which, without further implementing measures being required, individual aid awards may be made to undertakings defined within the act in a general and abstract manner.” On the basis of this definition, three cumulative conditions must be met to classify a State measure as an “aid scheme”: 

  • There is an “act” on the basis of which aid may be granted individually to undertakings.

  • No further implementing measures are required for that aid to be granted.

  • Beneficiaries of the aid scheme are defined “in a general and abstract manner.”

First Condition: There Is an “Act”

With regard to the first condition, the CJEU clarified that the concept of an “act” can also refer to a consistent administrative practice by the authorities of a Member State where that practice reveals a systematic approach.

In its judgment of 2019, the GCEU found that the legal basis for granting the EPRs resulted from a provision of the Belgian Income Tax Code and from the application of that provision by the Belgian tax authorities. However, according to the CJEU, the GCEU did not draw all the appropriate conclusions from that finding. The GCEU overlooked the fact that one of the essential characteristics of the scheme was a systematic approach applied by the Belgian tax authorities. The GCEU assumed that certain key elements of the scheme did not derive from the Income Tax Code as applied by the Belgian tax authorities, but from the rulings themselves. As a consequence, the GCEU found erroneously that further implementing measures were required and thereby misapplied the term “act”.

Second Condition: No Further Implementing Measures

The CJEU ruled that the second condition is linked to the first condition. It must be determined in light of the “act” whether the grant of individual aid requires further implementing measures. This second condition implies that the tax authorities have no discretion in adopting the measures, such as influence on the amount of the aid, its characteristics or the conditions under which the aid is granted. 

The CJEU ruled that the GCEU’s assessment of the second condition was incorrect. The GCEU overlooked the fact that one of the essential characteristics of the scheme was that the Belgian tax authorities had granted the EPRs systematically when certain conditions were satisfied. The CJEU held that this systematic approach was capable of constituting a relevant factor in order to show, where applicable, that the tax authorities did not in fact have any discretion in the application of the Income Tax Code.

Third Condition: Beneficiaries Defined “in a General and Abstract Manner”

The CJEU ruled that the third condition was linked to the two other conditions. Given the GCEU’s errors with respect to the two other conditions, the CJEU considered that the GCEU’s assessment of the third condition was also incorrect.

As a consequence, the CJEU concluded that the GCEU erred in its assessment of the three conditions that defined the concept of an “aid scheme.” The CJEU therefore set aside the GCEU’s judgment. Furthermore, the CJEU considered that the state of the proceedings did not permit final judgment to be given with regard to the question whether the EPRs actually constituted state aid, and the question whether the principles of legality and protection of legitimate expectations were infringed, in so far as the recovery of the alleged aid was incorrectly ordered, including from the groups to which the beneficiaries of that aid belong. These issues were therefore referred back to the GCEU.


The CJEU’s judgment provides important clarifications with regard to the concept of an “aid scheme.” The distinction between individual aid and an aid scheme is important because, in the case of an aid scheme, the EC can analyse the scheme’s characteristics in order to determine whether the scheme concerns State aid. As such, the EC does not need to analyse each individual grant made pursuant to the scheme in order to determine whether that grant constitutes aid. If the EC finds that a scheme constitutes State aid, all individual aid measures taken on the basis of that scheme will also constitute State aid. It is only at the stage of recovery of the aid that it is necessary to look at the individual situation of each beneficiary, and then only for the purposes of determining the amount of illegal aid granted.


C-362/19 P, European Commission v Fútbol Club Barcelona, 4 March 2021

On 4 March 2021, the CJEU rendered its judgment in Case C-362/19, European Commission v Fútbol Club Barcelona, concerning State aid arising out of the taxation of professional sports clubs in Spain. 

In July 2016, the EC found that Spain had unlawfully implemented aid in the form of a preferential corporate tax rate in favour of Fútbol Club Barcelona, Osasuna, Athletic Bilbao and Real Madrid. A Spanish law obliged all Spanish professional sports clubs to convert into public limited sports companies (SLCs), with the exception of professional sports clubs that had achieved a positive financial balance during the financial years preceding the adoption of the law. The only professional sports clubs that fell within the exception were the four football clubs mentioned above. The EC ordered the recovery of the difference between the corporate tax actually paid and the corporate tax the four football clubs would have been required to pay as SLCs.

Fútbol Club Barcelona applied to the GCEU for annulment of the EC’s decision.

The GCEU annulled the EC’s decision on the grounds that the EC did not prove the existence of an advantage to the requisite legal standard. In particular, the EC did not examine sufficiently whether the advantage resulting from the reduced tax rate was offset by a less favourable deduction rate for the reinvestment of extraordinary profits compared to the situation of SLCs. 

The EC appealed to the CJEU against GCEU’s judgment, raising a sole plea: infringement of Article 107 (1) TFEU by ruling that the EC should have examined whether the reduced tax rate was offset by the less favourable deduction for reinvestment of extraordinary profits.

CJEU Judgment

First, the CJEU recalled the case law according to which the EC must carry out a global assessment of the aid measure, according to the information available and developments foreseeable at the time when the decision to grant that aid was taken (Case

C-357/14 P, Electrabel and Dunamenti Erőmű v Commission, EU:C:2015:642, paragraph 104).

On the other hand, when the measure concerns an aid scheme, as in the case at hand, the EC may merely study the characteristics of the scheme and is not required to analyse the aid granted in individual cases. Contrary to the GCEU’s findings, the CJEU clarified that the EC only needs to demonstrate that the aid scheme is such as to favour beneficiaries, by being capable of resulting at the time of adoption in a lower tax liability than would have been the case if the general tax regime had been applied. It is only at recovery stage that the EC must determine whether that scheme has actually conferred an advantage.

The CJEU’s reasoning relied on the requirement for Member States to notify aid, which is essential to enable the EC to exercise its supervisory function. Member States breaching this obligation would be favoured over those which complied with the notification requirement if the EC were required to examine, on the basis of data collected after adoption of the scheme, whether the advantage actually materialised in all subsequent tax years or if it had been offset in certain tax years by the disadvantages recorded in other tax years. 

In the light of these considerations, the CJEU found that the EC was not required to examine the effect of the lower rate of deduction for reinvestment of extraordinary profits and whether it would level out the advantage from the reduced tax rate.

For these reasons the CJEU concluded that the GCEU had made an error of law by ruling that the EC should have examined whether the reduced tax rate was offset by the less favourable deduction of reinvestment of extraordinary profits. The CJEU therefore set aside the judgment of the GCEU and proceeded to give judgment itself on the application for annulment made by Fútbol Club Barcelona.

The CJEU considered that the application for annulment made by Fútbol Club Barcelona was unfounded, principally for the reasons explained above. As far as concerned the EC’s requirement that the Spanish government recover the aid, the CJEU observed that Fútbol Club Barcelona could not invoke a legitimate expectation that the aid was lawful. The aid had not been notified to the EC by the Government of Spain and consequently the EC had not granted approval for it. 


At a practical level, this case illustrates the dangers of receiving tax reductions that are granted only to a few companies and not to companies generally on objective criteria. Such tax reductions could very well constitute State aid within the meaning of Article 107 (1) TFEU. Such aid will be illegal if the Member

State concerned has not notified it in advance to the EC and obtained the latter’s prior approval. The EC can order the Member State to recover any illegally granted aid.

This case also illustrates how a Member State could try to confuse the issue by arguing, as in this case, that an unnotified “aid scheme” is really a collection of unnotified “individual aids.” If this interpretation were correct, the EC would have to open an investigation into each individual grant of aid as and when made, and demonstrate in each case that an advantage was conferred having regard for all the relevant circumstances of the individual beneficiary’s tax position. The CJEU’s judgment closes the way for such line of argument and spares the EC the time and effort that would be needed to examine each individual grant made pursuant to an unnotified aid scheme. The EC still must open an investigation and make a finding that a scheme is in fact an unnotified aid scheme. If this is the case, the aid scheme is illegal because it is unnotified and the EC can proceed to recovery of all illegal aid granted pursuant to that scheme.


C-562/19 P, European Commission v Republic of Poland; C596/19 P, European Commission v Hungary, 16 March 2021

On 16 March 2021, the CJEU delivered its judgments in Cases C-562/19 P, Commission v Republic of Poland (EU:C:2021:201) and C-596/19 P, Commission v Hungary (EU:C:2021:202), ruling that the Polish tax on the retail sector and the Hungarian advertisement tax do not breach EU State aid rules. Both these tax schemes provided for progressive rates by bands, based on the turnover derived from the taxable undertakings. 

In the EC’s view, these measures constituted illegal State aid because they discriminated between large undertakings (those with high turnover) and small undertakings (those with lower turnover), thereby conferring a selective advantage on the latter. The EC therefore adopted formal decisions to this effect. Poland and Hungary applied to the GCEU for annulment of the EC’s decisions. 

By judgment of 16 May 2019 in joined Cases T-

836/16 and T-624/17, Republic of Poland v

Commission (EU:T:2019:338), and judgment of 27

June 2019 in Case T-20/17, Hungary v Commission (EU:T:2019:448), the GCEU annulled the EC’s decisions. The EC subsequently appealed against the GCEU’s judgments before the CJEU, which sided with the opinion of Advocate General Juliane Kokott in the relevant cases, dismissed the EC’s appeals and upheld the judgments of the GCEU.

The CJEU Judgment

The CJEU began its analysis by referring to the threestep method of classifying a tax measure as selective, applied in the case law, namely the “derogation test”:

. . . the EC must begin by identifying the reference system, or ‘normal’ tax system applicable in the Member State concerned, and thereafter demonstrate that the tax measure at issue is a derogation from that reference system, in so far at it differentiates between operators who, in the light of the objective provided by that system, are in a comparable factual and legal situation.” 

Finally, if the Member State concerned proves that the tax differentiation is justified by the nature or legal structure of the tax system of which it forms part, that differentiation cannot constitute a selective advantage. 

The EC argued that the progressive tax rates based on the undertakings’ turnover rather than their profit must be excluded from the reference system, because they led to differences in the average tax rate and were intended to favour undertakings with low turnover. A normal turnover taxation reference system would only consist of a flat (proportional) rate for all undertakings.

However, the CJEU followed the rulings in Case C75/18, Vodafone Magyarország (EU:C:2020:139) and Case C-323/18, Tesco-Global Áruházak (EU:C:2020:140), and held that, without prejudice to the fields in which EU tax law has been harmonised, Member States have the discretion to establish the system of taxation that they deem most appropriate, under their fiscal autonomy, that discretion having, in any event, to be exercised in accordance with EU law. In particular, EU State aid law does not prohibit Member States from implementing progressive tax rates nor does it preclude progressive taxation from being based on turnover, a neutral criterion and indicator of the taxpayer’s ability to pay. It does not follow from any rule or principle of EU law that progressive tax rates can only be applied to profits.

Consequently, the CJEU upheld the GCEU’s view that the progressivity of the tax rates constituted an integral part of the reference system under which the existence of a selective advantage had to be analysed. The EC had not demonstrated that the characteristics of the tax measures at issue had been designed in a manifestly discriminatory manner with the aim of circumventing EU State aid law.

Practical Considerations

The CJEU’s judgments in these two cases should be reassuring for EU companies. They illustrate how the CJEU protects EU companies against illegal extensive interpretations of the TFEU by the EC.

Articles 107 and 108 TFEU are designed to prohibit

State aid that distorts the internal market by “favouring certain undertakings or the production of certain goods.” However, if this condition of selectivity were interpreted illegally, Articles 107 and 108 would become instruments for the EC to achieve all sorts of illegal ends, for example, the outlawing of any system of turnover taxation in which the tax rates were progressive.  



On 26 March 2021, the EC adopted its “Guidance on the application of the referral mechanism set out in Article 22 of the [EUMR] to certain categories of cases.” The adoption of this novel piece of guidance reflects recent concerns on the part of the EC that certain competitively significant transactions— particularly transactions involving nascent competitors in the digital and pharma sectors— unjustifiably escape EU merger control review. As a result, the EC now encourages Member States to refer certain transactions to it even if they do not meet national merger control thresholds. A prominent example of this new approach to the referral mechanism enshrined in Article 22 EUMR is the French Competition Authority’s request for, and the

EC’s acceptance of, a referral of the proposed USD 7.1 billion acquisition by Illumina (a US genomics company) of GRAIL (a US cancer test start-up), i.e., an acquisition between two non-EU companies discussed on page 35. 


At the time of the EUMR’s inception in 1989, the referral mechanism in Article 22 was known as the “Dutch clause” This is because the insertion of this provision into the EUMR came at the request of the Netherlands, which at the time did not have a merger control regime in place. Article 22 EUMR allows for one or more Member States to request the EC to examine any merger that does not have an EU dimension but meets the following cumulative conditions: it affects trade between Member States, and it threatens to significantly affect competition within the territory of the Member State or States making the request (Article 22 Conditions). Where the Article 22 Conditions are met, the EC may in its discretion decide whether to examine the transaction at hand.

Traditionally, the EC has discouraged the use of Article 22 EUMR in merger cases that were not notifiable under the laws of the referring Member State(s). This is principally because the EC considered such transactions unlikely to have a significant impact on the internal market. A concern of the EC over recent years (mainly since the acquisition of WhatsApp by Facebook for USD 19 billion), however, has been the number of mergers involving companies that play, or may develop into playing, a significant competitive role on the market despite having little or no turnover at the time of the merger. This development has been found to be particularly significant in the digital economy, where services regularly launch with the aim of developing a significant user base and/or commercially valuable data inventories, before the business is monetised, and in the pharma sector, where transactions have involved innovative companies conducting R&D with strong competitive potential, even if such companies have not yet finalised, let alone exploited commercially, the results of their R&D. Because of the absence of, or low, turnover of one of the parties to such transactions, such transactions have often escaped merger control review. 

With a view to capturing these prima facie problematic cases that technically escape review, the EC, with the promulgation of its new guidance, has therefore reversed its historical policy of discouraging the referral of cases under Article 22 when a Member State does not have jurisdiction, particularly, but not necessarily limited to, mergers in the healthcare and tech sectors. 

Candidate Cases for Referral 

The EC now encourages and is more amenable to accepting referrals in certain types of cases where the referring Member State does not have initial jurisdiction over the case, but where the Article 22 Conditions are met. Candidate transactions for referral are particularly (although not necessarily limited to) those where one party:

  • Is a start-up or recent entrant with significant competitive potential that has yet to develop or implement a business model generating significant revenues (or is still in the initial phase of implementing such business model)

  • Is an important innovator or is conducting potentially important research

  • Is an actual important competitive force

  • Has access to competitively significant assets (such as raw materials, infrastructure, data or intellectual property rights)

  • Provides products or services that are key inputs/components for other industries.

In exercising its discretion whether to encourage or accept a referral, the EC may also take into account whether the value of the transaction is particularly high compared to the current turnover of the target. This was the case in Illumina/GRAIL, where the EC considered that GRAIL’s competitive significance was not reflected in its turnover, especially in light of the USD 7.1 billion deal value. 


The EC will cooperate closely with the Member States to identify transactions that may constitute potential candidates for referral under Article 22. Merging parties may voluntarily approach the EC with a view to obtaining an “early indication” of whether a proposed transaction represents a candidate for referral. Conversely, where the EC becomes aware of a transaction considered to meet the criteria for a referral, it may inform the Member State(s) potentially concerned and “invite” them to make a referral request—a decision whether to refer lying within the discretion of such Member State(s). Further, third parties are encouraged to draw the EC’s and the Member States’ attention to potential candidate cases. 

If a referral request is under consideration, the EC will inform the merging parties thereof “as soon as possible”. While the merging parties are not required to delay closing upon receipt of such information, they may choose to do so until an actual decision has been rendered on whether a referral request will actually be made. Once the merging parties have been informed by the EC of a referral request having actually been made, they cannot close the transaction and must wait until EC clearance. 

A Member State that does not have jurisdiction to review the relevant merger must make a referral request within 15 working days of the merger being “made known” to it. Regarding the concept of “made known”, the EC’s guidance states that the Member State should possess sufficient information to make a preliminary assessment as to the existence of the criteria relevant for the assessment of the referral. Once a referral request has been made, the EC will “without delay” inform the other Member States and the merging parties thereof, and other Member States may then join the initial request within 15 working days of being informed by the EC of the initial request. At the latest 10 working days thereafter, the EC may decide to assert jurisdiction to examine the impact of the transaction within each of the Member States for which the referral request is accepted. If the EC does not take a decision within this period, it will be deemed to have adopted a decision to review the transaction in accordance with the request. 

While the referral is subject to the aforementioned deadlines, it bears noting that a transaction that has already been closed does not preclude a Member State from requesting a referral. The EC will generally not, however, consider a referral appropriate where more than six months after closing has passed. 


Transactions involving nascent competitors— especially in the digital economy and pharma sectors—are in the EC’s spotlight. As such, mergers that until recently were unlikely to face antitrust scrutiny because of the absence of, or low, turnover of the target now may be referred upwards to the EC for merger control review. Indeed, a potential eightweek delay may now have to be factored into any deal timeline in certain types of transaction. Further delays cannot be ruled out—when exactly is a merger “made known” to a Member State for the purposes of triggering the 15-working-day period to make a request for referral? Further, when the parties seek an early indication, there is no time limit for the EC to respond. Moreover, there is always the risk that the transaction in question will be referred to the EC post-closing. 

Given this uncertainty, merging parties may wish to approach the EC pre-emptively regarding their intended transaction with a view to soliciting an early indication from the EC. In certain cases, it would indeed be advisable to do so in view of deal timing implications, but also given the fact that an upward referral has in the past often required remedies as a condition for clearance. Companies that blindly close a transaction involving a nascent competitor, particularly in the digital and pharma sectors, in the hope that it will fly under the enforcement radar do so at their peril. Increased prudence is henceforth merited.


On 9 July 2021, following a thorough review process launched in October 2018, the EC published a draft revised Vertical Block Exemption Regulation (VBER) and draft revised accompanying Guidelines on vertical restraints (Vertical Guidelines). While the basic system underpinning the assessment of vertical agreements remains unchanged given its continued usefulness for businesses, the proposed revised rules seek to reflect new market realities characterised by the exponential growth of e-commerce and the use of online platforms. The draft revised rules are scheduled to enter into force on 1 June 2022, and the final VBER and Vertical Guidelines are unlikely to differ to any material extent from the drafts currently under consideration. 

Background: Application of the Current VBER and Vertical Guidelines

In principle, the EU competition rules prohibit anticompetitive vertical agreements. Certain vertical agreements are exempt from this prohibition, however. In particular, vertical agreements benefit from an exemption from the Article 101 (1) TFEU prohibition of anticompetitive agreements to the extent that they meet the conditions for exemption under Article 101 (3), i.e., they “contribute to improving the production or distribution of goods or to promoting technical or economic progress, while allowing consumers a fair share of the resulting benefit.” A vertical agreement is presumed to meet the conditions of Article 101 (3), and thus fall within the safe harbour of the VBER, where, inter alia

  • The market share of the parties to the agreement in question does not exceed 30%.

  • The agreement does not contain any so-called “hard-core restrictions”, e.g., resale price maintenance. The presence of a hard-core restriction removes the benefit of the exemption for the whole agreement.

  • The agreement does not contain “excluded restrictions”, e.g., non-compete obligations with a duration exceeding five years. Excluded restrictions remove the benefit of the exemption for that specific restriction although the remainder of the agreement continues to benefit therefrom.

That said, even if the conditions of the VBER are not satisfied, this does not necessarily mean that an agreement is anticompetitive and should be prohibited under Article 101 TFEU. Rather, the compatibility of a vertical agreement with Article 101 would then have to be assessed on a case-by-case basis in light of the teachings of the Vertical Guidelines. The Vertical Guidelines provide guidance to companies as to the compatibility of their vertical agreements with the EU competition rules where the agreement in question does not benefit from the safe harbour enshrined in the VBER.

Although much of the substance of the current VBER remains unchanged, the proposed revised rules essentially provide further clarity on certain types of vertical agreements given the perceived complexity of the current rules, and an update to the rules in light of the rise of e-commerce and the online platform economy since 2010, when the currently applicable VBER and Vertical Guidelines entered into force. Key changes under the proposed revised rules are as set out below.

Restrictions of Online Sales

Suppliers may, for various reasons, wish to impose online sales restrictions on distributors. However, the draft VBER does not apply to restrictions preventing buyers or their customers from selling their goods or services over the internet. The same holds true for total bans on the use of online advertising channels, such as price comparison tools or advertising on search engines. 

Online sales restrictions may be the result of direct obligations or indirect measures aimed at inducing the distributor not to sell online. The draft Vertical Guidelines provide, for example, that the following would not be eligible for exemption:

  • Geo-blocking measures, i.e., a requirement that the distributor prevent customers located in another territory from viewing its website through the automatic re-routing of customers to the manufacturer’s or other distributors' websites, or the termination of online transactions if consumers’ credit card data reveals they are located outside the distributor’s territory

  • Requirements that a distributor sell only from physical premises or in the physical presence of specialised personnel 

  • Requirements that a distributor seek the supplier’s prior authorisation to sell online.

On the other hand, the following actions by a supplier are permitted:

  • A supplier can set quality standards for selling online, e.g., standards relating to the “look and feel” of a website.

  • A supplier can restrict the use of a specific online sales channel, such as online marketplaces.

  • A supplier can demand that a distributor operate one or more brick-and-mortar shops or showrooms as a condition for becoming a member of the supplier’s distribution system.

  • A supplier can require that an absolute amount of products be sold offline.

  • A supplier can charge a higher price for products to be resold online than for products to be resold offline (dual pricing) as far as this incentivises or rewards appropriate levels of investments made online and offline respectively.

  • A supplier operating a selective distribution system may impose different criteria for online sales than for offline sales, as long as the criteria for online sales do not effectively prevent online sales. 

Dual Distribution

The draft VBER introduces stricter rules regarding dual distribution. According to the draft VBER, dual distribution agreements are exempt from the prohibition on anticompetitive agreements if the aggregate market share of the parties at the retail level does not exceed 10%. If the aggregate market share of the supplier and distributor exceeds 10%, but they individually do not have a market share exceeding 30%, the agreement would still be exempt except that any information exchanges between the parties would be subject to further analysis to see if they were compatible with the competition rules. 

There are two important exceptions to the safe harbour of the draft VBER: 

  • The safe harbour will not be available if the dual distribution agreement has as its object the restriction of competition.
  • An exception applies vis-à-vis providers of online intermediation services. If such providers sell goods or services in competition with companies to which they provide online intermediation services, the safe harbour will not apply.

Restrictions of Active Sales

As a general principle under the VBER, suppliers cannot restrict the territory or customer group to which a distributor can actively sell contract goods and services. However, the draft VBER provides for certain exceptions thereto, and clarifies when suppliers can impose active sales restrictions depending on the distribution system in place (exclusive, selective or neither of these). 

Generally, online sales are not considered as active sales, but rather as a form of passive sales (i.e., sales in response to unsolicited requests from individual customers). However, the draft VBER clarifies which means of online promotion can be considered as active selling:

  • Targeting customers by communication through online media, price comparison tools or advertising on search engines
  • Offering language options on a website different from the languages commonly used in the territory in which the distributor is established
  • Offering a website with a domain name corresponding to a territory other than the one in which the distributor is established. 

The difference between active and passive sales is important because suppliers can restrict active sales by distributors in territories or to customer groups exclusively allocated to another distributor. For example, a supplier can restrict a distributor from specifically targeting customers through online advertising within an exclusive territory or exclusive customer group allocated to another distributor (active sales). However, a supplier cannot restrict a distributor from general online advertising or promotion, intended to reach customers in the distributor’s territory or customer group, but which cannot be limited to that territory or customer group (passive sales).

Most Favoured Nation Clauses

Most favoured nation clauses (MFNs) require a contracting party not to offer more favourable sales conditions to competitors. The draft revised VBER does not always apply to MFNs in favour of online intermediation service providers, such as online marketplaces. Any obligation causing a buyer of online intermediation services not to offer, sell or resell goods or services to end users under more favourable conditions using competing online intermediation services will not benefit from the


Resale Price Maintenance 

Resale price maintenance, i.e., agreements or concerted practices having as their direct or indirect object the establishment of a fixed or minimum resale price level, remains a hardcore restriction. The imposition of a maximum retail price or the determination of a resale price recommendation does not constitute resale price maintenance, unless a maximum price or resale price recommendation is combined with incentives (e.g., reimbursement of promotional costs) to apply a certain price level, or disincentives to lower the sales price (e.g., a threat to cut further supplies). Furthermore, the fixing of the resale price in a vertical agreement between a

supplier and a buyer that executes a prior agreement between the supplier and a specific end-user, i.e., a “fulfilment contract”, does not constitute resale price maintenance where the end user has waived its right to choose the undertaking that should execute the supply agreement. 

Non-Compete Clauses 

In a similar fashion to the current rules, the draft VBER provides that non-compete clauses that are indefinite or longer than five years are an excluded restriction and therefore do not benefit from the VBER. However, the draft VBER provides that a non-compete clause that is tacitly renewable beyond a period of five years is covered provided that the buyer can effectively renegotiate or terminate the vertical agreement with a reasonable notice period and at a reasonable cost, thus allowing the buyer to effectively switch after the expiry of the five-year period. 


The current VBER and Vertical Guidelines have long been key pieces of guidance for companies to selfassess whether their vertical agreements comply with the EU competition rules. With the exponential rise of the digital economy, however, the current rules have in many respects become obsolete and therefore require an update. Moreover, up-to-date guidance is imperative for businesses in light of the EC’s ongoing enforcement activity in the digital arena, testimony of which is borne by the EC’s recent EUR 7.8 million fine on Valve and five publishers of PC video games for illegal geo-blocking practices. While the draft VBER sets forth a one-year transitional period until 31 May 2023 for vertical agreements which are already in force on 31 May 2022 and which meet the existing conditions for exemption, early assessment of the implications of the new rules and preparation therefor are already a must.


On 15 December 2020, the EC published a package of proposed measures on digital markets and digital services: 

  • A Digital Services Act (DSA) (a proposed Regulation on a Single Market for Digital

Services and amending Directive 2000/31/EC – COM (2020) 825 final)

  • A Digital Markets Act (DMA) (a proposed Regulation on Contestable and Fair Markets in the Digital Sector – COM (2020) 842 final).

According to the EC’s explanatory memorandum, the proposed DSA is a horizontal initiative focusing on issues such as liability of online intermediaries for third-party content, safety of users online and asymmetric due diligence obligations for different providers of information society services depending on the nature of the societal risks such services represent. In contrast, the proposed DMA is concerned with economic imbalances, unfair business practices by gatekeepers and their negative consequences, such as weakened contestability of platform markets. 

The proposed DMA would have a close relationship with competition law, notably the rules prohibiting abuse of a dominant position. The following discussion concentrates therefore on the provisions of the proposed DMA.

The Proposed Digital Markets Act

The proposed DMA would regulate certain providers of core platform services designated by the EC as “gatekeepers”. A provider of core platform services would be designated as a “gatekeeper” if it satisfied all three of the following qualitative criteria: 

  1. It has a significant impact on the internal market. 

  2. It operates a core platform service which serves as an important gateway for business users to reach end users.

  3. It enjoys an entrenched and durable position in its operations, or it is foreseeable that it will enjoy such a position in the near future.  

Each of these qualitative criteria would be deemed to be satisfied if certain quantitative thresholds were exceeded.

  • Qualitative criterion 1 would be presumed to be satisfied if the undertaking to which the operator belongs achieves an annual EEA turnover equal to or above EUR 6.5 billion in the last three financial years, or has an average market capitalisation or equivalent fair market value of at least EUR 65 billion in the last financial year, and it provides a core platform service in at least three Member States.

  • Qualitative criterion 2 would be presumed to be satisfied if the undertaking to which the operator belongs provides a core platform service that has more than 45 million monthly active end users established or located in the EU and more than 10,000 yearly active business users established in the EU in the last financial year. 

  • Qualitative criterion 3 would be presumed to be satisfied if the quantitative thresholds for qualitative criterion 2 were satisfied in each of the last three financial years.

An operator would be obliged to notify the EC if and when it exceeded all three of the relevant quantitative thresholds set out above, as a result of which the EC would designate that operator as a gatekeeper. The EC could also designate an operator as a gatekeeper of its own motion if, notwithstanding failure of the operator to notify the fact that it had exceeded all three of the quantitative thresholds, the EC had proof that all of these thresholds were in fact exceeded. Last of all (but more difficult), if the EC had proof that an operator satisfied all three of the qualitative criteria, it could designate that operator as a gatekeeper of its own motion. 

Conversely, an operator that exceeded all three of the relevant quantitative thresholds but did not satisfy all three of the qualitative criteria mentioned above could apply to the EC not to be designated as a gatekeeper.

When making a designation as a gatekeeper, the EC would list the relevant core platform services that were provided by that same undertaking and which individually served as an important gateway for business users to reach end users. Those services could be any of the following:

  • Online intermediation services

  • Online search engines 

  • Online social networking services 

  • Video-sharing platform services 

  • Number-independent interpersonal communication services 

  • Operating systems 

  • Cloud computing services 

  • Advertising services, including any advertising networks, advertising exchanges and any other advertising intermediation services provided by a provider of any of the core platform services listed above. 

The EC’s listing of the relevant core platform services provided by the undertaking would have important legal consequences: it would identify the core platform services in respect of which the designated gatekeeper would be subject to the prohibitions and obligations listed in Articles 5 and 6 of the proposed DMA. The designated gatekeeper would be obliged to adopt “effective” measures to achieve the objective of these prohibitions and obligations while, at the same time, complying with the General Data Protection Regulation (EU) 2016/679 and Directive 2002/58/EC on privacy and electronic communications, and with the legislation on cyber security, consumer protection and product safety.

The prohibitions and obligations listed in Articles 5 and 6 fall under the generic heading of Chapter III of the proposed DMA, “Practices of gatekeepers that limit contestability or are unfair.” The prohibitions and obligations in Article 5, entitled “Obligations for gatekeepers”, can be summarised as follows:

  • Prohibition against combining personal data sourced from core platform services with personal data from any other services offered by the same gatekeeper without the end user’s informed consent (and other similar prohibitions)

  • Obligation to allow business users to offer the same products or services to end users through third-party online intermediation services at prices or conditions that are different from those offered through the online intermediation services of the gatekeeper

  • Obligation to allow business users to promote offers to end users acquired via the core platform service, and to conclude contracts with these end users regardless of whether for that purpose they use the core platform services of the gatekeeper or not

(and other similar obligations)

  • Obligation to refrain from preventing or restricting business users from raising issues with any relevant public authority relating to any practice of gatekeepers

  • Prohibition against requiring business users to use, offer or interoperate with an identification service of the gatekeeper in the context of services offered by the business users using the core platform services of that gatekeeper

  • Prohibition against requiring business users or end users to subscribe to or register with other core platform services as a condition of access to, signing up or registration with any of the core platform services identified in the EC’s designation of the gatekeeper

  • Obligation to provide advertisers and publishers to which the gatekeeper supplies advertising services, upon their request, with information concerning the price paid by the advertiser and publisher, as well as the amount or remuneration paid to the publisher, for the publishing of a given advertisement and for each of the relevant advertising services provided by the gatekeeper.

The prohibitions and obligations listed in Article 6 would be subject to an additional provision that would not apply to those listed in Article 5. The EC would have the power to specify the measures that the gatekeeper concerned would be obliged to implement in order to ensure effective compliance with the prohibitions and obligations listed in Article 6. Before making such a specification, however, the EC would be obliged to initiate an investigation in which the gatekeeper concerned could exercise its rights of defence. The prohibitions and obligations listed in Article 6, entitled “Obligations for gatekeepers susceptible of being further specified”, can be summarised as follows:

  • Prohibition against using, in competition with business users, any data not publicly available that are generated by the activities of, or provided by, the business users of its

core platform services (or generated by the activities of or provided by the end user customers of these business users).

  • Obligation to allow end users to un-install any pre-installed software applications on its core platform service. The gatekeeper shall, however, be allowed to restrict uninstallation of software applications that are essential for the functioning of the operating system or of the device and which cannot technically be offered on a standalone basis by third parties. 

  • Obligation to allow the installation and effective use of third-party software applications or software application stores using, or interoperating with, operating systems of that gatekeeper, and to allow these software applications or software application stores to be accessed by means other than the gatekeeper’s core platform services. The gatekeeper shall not be prevented from taking proportionate measures to ensure that third-party software applications or software application stores do not endanger the integrity of the hardware or operating system provided by the gatekeeper.

  • Prohibition against ranking more favourably the services and products offered by the gatekeeper itself compared to similar services or products offered by a third party. There is a corresponding obligation to apply fair and non-discriminatory conditions to such ranking. 

  • Prohibition against technically restricting the ability of end users to switch between and subscribe to different software applications and services to be accessed using the operating system of the gatekeeper, including the choice of internet access provider. 

  • Obligation to allow business users and providers of ancillary services access to and interoperability with the same operating system, hardware or software features that are available or used in the provision by the gatekeeper of any ancillary services. 

  • Obligation to provide advertisers and publishers, upon their request and free of charge, with access to the performance measuring tools of the gatekeeper and the information necessary for advertisers and publishers to carry out their own independent verification of the advertisement inventory.

  • Obligation to provide effective portability of data generated through the activity of a business user or end user, and in particular to provide tools for end users to facilitate the exercise of data portability, in line with the General Data Protection Regulation (EU) 2016/679, including by the provision of continuous and real-time access. 

  • Obligation to provide free of charge to business users or to third parties authorised by a business user, effective, high-quality, continuous and real-time access and use of aggregated or non-aggregated data, that is provided for or generated in the context of the use of the relevant core platform services by those business users and the end users engaging with the products or services provided by those business users. As far as concerns personal data, there would be an obligation to provide access and use only where directly connected with the use effectuated by the end user in respect of the products or services offered by the relevant business user through the relevant core platform service, and when the end user opts in to such sharing with a consent in the sense of the General Data Protection Regulation (EU) 2016/679. 

  • Obligation to provide to any third-party providers of online search engines, upon their request, access on fair, reasonable and non-discriminatory terms to ranking, query, click and view data in relation to free and paid search generated by end users on online search engines of the gatekeeper, subject to anonymisation for the query, click and view data that constitutes personal data. 

  • Obligation to apply fair and nondiscriminatory general conditions of access for business users to its software application. 

The proposed DMA would give the EC power to suspend some or all of a gatekeeper’s obligations and prohibitions at the request of the latter. The EC would also have power to exempt a gatekeeper from certain obligations or prohibitions on the grounds of public morality, public health or public security. Last of all, the EC would have the power to update the DMA’s list of core platform services.  

The proposed DMA would also impose obligations and prohibitions of a general regulatory nature, that is to say, obligations and prohibitions that applied to all designated gatekeepers, whatever the core platform services provided. These obligations and prohibitions would be as follows:

  • A general prohibition against behaviour which undermines compliance with the prohibitions and obligations of Articles 5 and 6  

  • An obligation not to place business users at a disadvantage in their compliance with data protection rules  

  • An obligation not to make business users’ exercise of their rights and choices under Articles 5 and 6 more difficult, for example, by degrading the conditions and quality of services provided to them

  • An obligation to provide the EC with certain information about any concentration that the gatekeeper proposed to implement involving another provider of core platform services or of any other digital sector services,

whether or not the concentration was notifiable to the EC under the EUMR or to an NCA under the corresponding national legislation

  • An obligation to submit to the EC an independently audited description of any techniques for profiling of consumers that the gatekeeper applies to or across its core platform services, within six months of the operator’s designation as a gatekeeper, and updated at least annually.

The proposed DMA would confer authority on the EC to carry out market investigations for any of the following purposes:

  • To examine whether a provider of core platform services should be designated as a gatekeeper

  • To identify the core platform services of a gatekeeper 

  • To determine whether a gatekeeper has systematically infringed the obligations laid down in Articles 5 and 6 and has further strengthened or extended its gatekeeper position

  • To determine whether one or more services within the digital sector should be added to the list of core platform services or to detect types of practices that may limit the contestability of core platform services or may be unfair and which are not effectively addressed by the DMA Regulation.

The EC would be given investigative and enforcement tools similar to those under Regulation (EC) No 1/2003 and under the EUMR namely:

  • Power to request information

  • Power to conduct interviews and take statements

  • Power to conduct on-site inspections

  • Power to adopt interim measures

  • Power to accept commitments

  • Power to monitor compliance with obligations and prohibitions listed in Articles 5 and 6 as well as decisions ordering the adoption of measures to comply with the obligations and prohibitions listed in Article 6, decisions imposing structural or behavioural remedies, decisions adopting interim measures or decisions accepting commitments

  • Power to impose fines or periodic penalty payments.

Exercise of the EC’s power to adopt decisions imposing fines or periodic penalties would be subject to a limitation period of three years counted from the day on which the infringement was committed, or, in the case of a continuing or repeated infringement, the day on which the infringement ceased. The EC’s power to enforce such decisions would be subject to a limitation period of five years from the date of the decision, although this period would be suspended in certain cases, notably where time to pay was granted.

The CJEU/GCEU would be given jurisdiction to cancel, reduce or increase the amount of any fine or periodic penalty imposed by the EC (in the same way as under Regulation (EC) No 1/2003).

Last of all, the proposed DMA would contain the usual provisions on the right to be heard, access to the file and professional secrecy.


According to the EC’s explanatory memorandum, the proposed DMA would complement existing EU (and national) competition rules by addressing unfair practices by gatekeepers that either fall outside the existing EU competition rules or cannot be addressed effectively by these rules. Antitrust enforcement inevitably intervenes after the restrictive or abusive conduct has occurred and involves time-consuming investigative procedures to establish the existence of the infringement. The proposed DMA would minimise the detrimental structural effects of unfair practices ex ante, without limiting the ability of the EC and NCAs to intervene ex post. In practical terms, operators in the digital sector should understand that the proposed DMA would introduce an additional layer of regulation with which they must comply, while not forgetting the long-established rules that will continue to prohibit abuses of a dominant position by gatekeepers as, for example, in the case of Google Shopping (Case T-612/1, Google and

Alphabet v Commission (Google Shopping), EU:T:2021:763), discussed on page 26.

It should be emphasised that, at present, the DMA is only a legislative proposal made by the EC. It is currently passing through scrutiny and debate within the European Parliament, and parallel examination by the Council of the EU. Given the DMA’s innovative nature and far-reaching consequences for the digital sphere, it is highly likely that there will be substantial amendments as the text passes through the legislative process. Adoption and entry into force of a final legislative text is therefore unlikely in 2022.

Frederic Pradelles and Karolien Van Der Puttin also contributed to this update.

© 2022 McDermott Will & EmeryNational Law Review, Volume XII, Number 40

About this Author

Philip Bentley, QC, McDermott WIll Emery law Firm, Antitrust Attorney

Philip Bentley is a partner in the international law firm of McDermott Will & Emery/Stanbrook LLP based in its Brussels office.  He is a member of the Firm’s EU regulatory practice and European Competition and Trade Groups.  His practice focuses on EU anti-dumping, trade defense and customs, EU competition (including State aid and public procurement), EU regulatory matters, notably GMOs, and EU litigation.

32 2 282 35 27
Jacques Buhart, McDermott Will Emery Law Firm, Antitrust Attorney

Jacques Buhart is a partner in the law firm of McDermott Will & Emery and is based in the Firm’s Paris office.  He is head of the Paris office and focuses his practice on corporate mergers & acquisitions and EU/ French competition.

In his cross-border transactional practice, Jacques has represented, among others, Mitsubishi Heavy Industries in its attempt to acquire Alstom with Siemens, Canon in its acquisition of OCE, Total in its acquisition of Elf, Belgacom, Metsäliitto in various acquisitions and divestitures in Europe, Toyota Motor Corporation in setting up its plant in...

33 1 81 69 15 00
Hendrik Viaene McDermott Antitrust Attorney

Hendrik Viaene focuses his practice on competition law, regulated markets and regulatory law. He advises clients across a wide range of related issues including state aid, cartels, licensing agreements, merger filings, M&A negotiations, abuse of dominance cases, and distribution agreements.

Hendrik assists governments as well as privately and publicly owned companies. He has a wealth of experience in a range of sector such as energy, chemicals, paints, automotive, financial data, recycling and waste management, telecom, construction, renewables, media, private equity and the...

+32 2 282 35 93
David Henry, European Competition Attorney, McDermott Will Emery Law firm

David Henry is an associate in the international law firm of McDermott Will & Emery, based in its Brussels office.  His practice focuses on European competition law including merger control, cartels and abuse of dominance, and his clients include companies in the air transport, chemicals, electronics and semi-conductor products, food retailing and digital map industries.  He also advises clients in proceedings before the European courts. 

32 2 282 35 69
Mary Hecht Antitrust Law Attorney McDermott Law Firm Paris

Mary Hecht focuses her practice on EU competition and antitrust matters.

Before joining McDermott, Mary worked in a wealth management firm. She was an intern in several French law firms specialized in competition law and also interned at Samsung Electronics and at Euro Disney.