The rapid development of digital markets has been posing major challenges to competition authorities for some time. The ‘internet giants’ consisting of Google, Amazon, Facebook, Apple and Microsoft have as already known been able to gain and consolidate strong market positions faster than in analogue markets. The previous antitrust law was, according to the own diagnosis of the competition authorities and the German and European legislators, no longer sufficient to counter anticompetitive practices of the large digital platforms sufficiently quickly and efficiently.
Recently, on 24th of March 2022 shortly before midnight, the European Commission, the Council of Ministers and the European Parliament agreed on the eagerly awaited final version of the new Digital Markets Act in their Trilogue negotiations. The breakthrough has been achieved and much faster than originally expected. What had been said and written from all sides as the European Commission presented its draft on a DMA in December 2020! And when France announced its goal to adopt the DMA during the French EU Council Presidency in the first half of 2022, the surprise and the doubts were big. In January 2021, one year after the publication of the draft of a 10th Amendment to the German Competition Act (GWB), the German legislator had already come forward in a remarkable way with the new tool Section 19a GWB and took the lead in the fight against GAFAM (or better now MAAMA).
With the DMA, the European legislator’s answer to the recognised problems in dealing with the so-called ‘Gatekeepers’ has finally arrived. But not really. Surprisingly, the final text is still a long time coming. In the Brussels environment, however, two versions of the DMA are circulating (‘The leaked {almost final} DMA text’ from 8th of April 2022 and the European Council Version from the 11th of May 2022). The latter version is the basis for this article.
- The Challenges of Digitisation
Digital markets have according to empirical studies (such as the Furman Report, the ACCC Digital Platforms Report, the US House Report, the Competition policy for the digital era Report as well as the German Reports both on behalf of the German Ministry of Economy and Energy: The Market Power Study and the Commission Competition Law 4.0 Report) a tendency towards market concentration. Due to direct and indirect network effects, each of which is self-reinforcing in positive feedback loops, digital platforms can grow much faster than providers of analogue products and services. If users are also prevented from using several platforms at the same time (‘multi-homing’) and there are high barriers to switching due to lock-in effects, then the market can easily tip in favour of the biggest player.
Admittedly, such market dominance in the digital sector may simply be the result of great efficiency. After all, the users also have a great interest in network effects coming to full bloom (who would want to have a social network on which only one in ten friends is available?). However, the resulting reduction of competition is of course accompanied by negative consequences for consumers: in a competitive environment, consumers might have had to give up less privacy. In the absence of interoperability, competitive impulses can only come from regulation or disruptive innovation.
Digital platforms, unlike in the past (Yahoo search engine, Microsoft Internet Explorer), no longer give up these positions of power so easily. This is mainly due to the creation of so-called ‘Digital Ecosystems’. GAFAM & Co have used their power positions in platform markets or the buyout of emerging competitors (‘killer acquisitions’) to expand into neighbouring markets and thus better protect the overall system against individual failures. The greater the network effects and the fewer the opportunities to use other platforms at the same time (‘single-homing’), the greater are additionally to the above-mentioned tipping-effect the incentives for the intermediary to use its privileged access to consumers as a Gatekeeper with non-performance-related means to enter neighbouring markets. Competitors in neighbouring markets regularly lack the leverage to counter this behaviour.
Amazon, Apple, Facebook and Google, in particular, by virtue of their intermediary services, decide every day globally on the access of millions of commercial providers to billions of consumers each with transaction volumes in the trillions. In the process, the platforms themselves determine according to which rules and under which conditions the commercial users get access to the user base and whether and how these rules are enforced (‘platforms as rule-setters’), thereby controlling competition in these markets.
Through digital ecosystems, their special access to data and users as well as their great financial power (‘deep pockets’), the gatekeepers have secured their strong position in their central intermediary market even against downstream innovations in such a way that they currently do not appear economically vulnerable in these core markets in many countries.
For this purpose, the DMA aims to ensure ‘contestable and fair markets’. In the following, the author deliberately deals only with the essential provisions of the DMA and the questions most frequently discussed in connection with them, in order to provide a comprehensive overview of the new regulation.
- Competition Law or Regulation?
In addition to the DMA, national regulations already exist or are intended that also address large digital platforms. The most prominent example might be Section 19a of the German Act against Restrictions of Competition (GWB). A very intensively debated question is whether there is still room left for national rules after the DMA enters into force (which is most likely expected only after the summer break in 2022). The crucial question of whether there is primacy of application of the DMA as a European regulation over national regulations (for example Section 19a GWB) depends on the scope of the directly applicable DMA. In Art. 1 (5) and Art. 1 (6), the DMA itself specifies and limits its scope of application. This question is illustrated by the example of Section 19a GWB.
Art. 1 (5) DMA contains the so-called exclusion clause, according to which, in order to prevent many different laws applicable for the digital sector, there should in principle be no further national regulations on gatekeepers with the aim of ensuring contestable and fair markets. Outside the scope of the DMA (for non-gatekeepers), however, this should be possible.
The so-called opening clause according to Art. 1 (6) DMA, on the other hand, does not affect Art. 101 and 102 TFEU and their national counterparts. More importantly, however, other unilateral ‘national competition rules’ are applicable besides the DMA under certain conditions. More exactly, national competition rules on unilateral conduct apply to the extent that they oblige other undertakings than gatekeepers or provide for obligations going beyond the DMA. The DMA itself defines what is meant by competition law in recital 10: the core of competition law is ‘an individual assessment of market positions and behavior, including its actual or likely effects and the precise scope of the prohibited behavior’. Furthermore, competition law must provide for the possibility of efficiency or objective justification. By making a separate provision for competition law, the legislator has clarified that the rules to ensure contestable and fair markets according to Art. 1 (5) DMA have to be at least something different than competition law.
According to that, the decisive factor for the application of a national rule besides the DMA is, whether this rule (here: Section 19a of the GWB) can be designated as competition law (or alternatively as regulatory law). This is because the provisions of the DMA, which is not based on Art. 103 TFEU for Competition Law but on Art. 114 TFEU for Internal Market, appear not to be competition law and can rather be understood as regulatory law. In the event that the national rule aims at achieving objectives that run parallel to the DMA and has not the characteristics of competition law, it is subject to Art. 1 (5) DMA, which confirms the primacy of application of the DMA. If, on the other hand, the national regulation has a competition law character, the opening clause of Art. 1 (6) DMA is relevant.
The term ‘regulatory law’ is often used in the context of public economic law to describe the interrelated legal ‘regulation’ of economic sectors and network economies (e.g. telecommunications, railways or energy) that were often monopolized by the state. The goal of regulatory law is in the case of market failure or other structural functional weaknesses of the market, instead of the merely selective and reactive character of competition law, to create a competitive market and to secure it for the long term as well as to ensure an orientation towards the common good. (See Paper of the Research Service of the German Parliament, in German; See also here and here also in German).
Competition law also pursues the higher goal of limiting market power, but it is concerned with the protection of free competition against restrictions on a case-by-case basis. Regulatory law is thus to be understood more comprehensively than competition law: in addition to the prevention of negative effects (as in competition law), regulatory law rather also stands for a comprehensive proactive, positive design of market structure. Accordingly, the difference is usually also made in terms of an ex-post effect of antitrust law as opposed to an ex-ante effect of regulatory provisions. The DMA itself characterizes Art. 101 et seq. TFEU as classical antitrust rules in recital 5 as occurring ‘ex post’ and requiring an ‘extensive investigation […] on a case by case basis’, whereas the DMA shall reduce the negative structural effects of unfair practices in advance without limiting the possibility to act against incidents by means of European or national competition law. In this context, it should be kept in mind that Art. 101 and 102 TFEU also contain prohibitions that directly apply to the undertakings in the first place. Only the enforcement of the prohibitions takes place afterwards (ex-post).
The example of Section 19a GWB shows the problem of classification into regulatory and competition law. The rule combines instruments that can be assigned to both regulatory law as well as antitrust law. The possibility for the German Federal Cartel Office to issue a prohibition order even before the conduct has been carried out, speaks in favor of a regulation law provision, even if the provision requires an initial risk of infringement or a risk of repetition (which, however, can also be waived). On the other hand, the requirement of a case by case examination goes in the direction of classic antitrust law with the possibility of factual justification of the conduct in Sec. 19a (2) Sentence 2 GWB. Furthermore, there are similarities in content between the catalogue of norms of Section 19a (2) GWB and the catalogue of Section 19 (2) GWB, which is renown as classic antitrust law. There may also be overlaps between Section 19a and Sections 19, 20 GWB in certain cases, as already indicated in the legislative documents.
Insofar as Section 19a GWB is subject to the opening clause, an independent scope of application is likely to remain, since both the addressees and the individual offences can go beyond the provisions of the DMA in view of their width and flexibility. In the latter case, however, the question arises whether the opening clause covers completely different conducts only or captures moreover the situation that the national competition provision defines a qualitatively broader scope of application for ‘the same forbidden conduct’. For example, the DMA contains a prohibition of self-preference in Art. 6 (5) and a prohibition of tying in Art. 5 (9), both in specific cases. The German provisions in Section 19a (2) No. 1 GWB and Section 19a (2) No. 3 lit. a GWB, on the other hand, are much more general and thus go qualitatively beyond the DMA.
A final clarification of the parallel applicability of Section 19a GWB besides the DMA will probably only be provided by a legally binding decision of the Court of Justice, which can be forced by a preliminary ruling pursuant to Art. 267 TFEU.
- The new market dominance under the DMA: Gatekeeping
With the aim of meeting the particularities of the digital platform economy, a new and more suitable qualification of the norm addressees was looked for. Due to their position as intermediaries between providers and users, digital platforms often hold a sensitive position of power, which is accompanied by the possibility and also the incentive to use their market power for their own benefit without necessarily dominating a market. Although Art. 102 TFEU also applies to gatekeepers, it is limited to market dominance as particular form of market power, its enforcement is ex-post and it requires an ‘extensive investigation of often very complex facts on a case by case basis’ (Recital 6 DMA). The antitrust proceedings against digital gatekeepers pursued so far have taken too long, so that the market structure, which is particularly dynamic on digital markets, has regularly already been irrevocably shifted when a final decision is reached. The tendency of digital markets towards concentration plays into the hands of the unfairly profiting company, which thus becomes increasingly untouchable. Finally, Art. 102 TFEU does not cover intermediaries without a market dominant position, such as for instance duopolies or oligopolies with strong competition (e.g. in the areas of mobile phone operating systems and cloud computing services).
The European legislator has therefore introduced a new type of addressee in the DMA using the ‘gatekeeper’ model. Gatekeepers are characterized by a unique position in digital markets due to strong network effects, the setup of digital ecosystems, a special intermediary position which provides privileged access to data as well as users and suppliers, which results in a structurally low degree of contestability (Recitals 1 – 6 DMA).
According to Art. 2 (1) DMA, a gatekeeper is an undertaking that provides a core platform service and has been designated as such pursuant to Art. 3 DMA (the insertion of “undertaking” into the wording now makes clear the important fact that only undertakings are addressees of the DMA obligations). In view of its objectives to ensure contestable and fair markets, the DMA aims at widespread and commonly used digital services that mostly directly intermediate between business users and end users and where features like the specific characteristics of digital markets are the most prevalent (Recital 13 of the DMA). These are, according to Art. 2 (2) DMA, mostly online intermediation services, online search engines, online social networking systems, video-sharing platform services, number-independent interpersonal communications services, operating systems, cloud computing services, online advertising services and furthermore (added to the final text) virtual assistants as well as web browsers. The insertion of “connected TVs” as another planned amendment was, however, rejected (see the proposed amendment in the text of the European Parliament under Art. 2 [1] No. 2 lit. fc, here).
Similar to Section 19a GWB (paramount significance across markets), the DMA relies on qualitative criteria to determine who is a gatekeeper, Art. 3 (1) DMA. In addition, however, it also refers to quantitative criteria, the fulfilment of which leads to the presumption of the qualitative criteria, Art. 3 (2) DMA.
According to Art. 3 (1) DMA, a gatekeeper as such is required to satisfy three conditions, namely (a) having a significant impact on the internal market, (b) providing a core platform service which is an important gateway for business users to reach end users, and (c) enjoying, or foreseeably enjoying in the near future, an entrenched and durable position in its operations. With regard to the presumption of significant impact on the internal market, it was finally agreed in the Trilogue negotiations that this will be the case if the annual union-wide turnover exceeds EUR 7.5 billion in each of the last three financial years or if the market capitalization/market value in the last financial year exceeds EUR 75 billion, insofar as the same core platform service was provided in at least three member states, Art. 3 (2) lit. a. According to Art. 3 (2) lit. b, an important gateway for business users to end users exists if the core platform service (!) has had in the last financial year an average of at least 45 million monthly active end users across the Union and at least 10,000 yearly active business users established in the Union. Thirdly, the entrenched and durable position is presumed if the thresholds of (b) have been met in each of the previous three financial years. In this case, the Commission ‘designates’ the company as gatekeeper within 45 days pursuant to Art. 3 (4) DMA. At this point, the Council of Ministers has been successful in reducing the duration of the procedure from 60 days to the above-mentioned period.
However, under Article 3 (8) of the DMA, the Commission can also designate an undertaking as a gatekeeper in a separate procedure pursuant to Article 17 DMA, if the threshold values are not met. For this purpose, the DMA specifies criteria that should be considered in this process. These are, for example, the size of a core platform service provider, the number of business users dependent on the core platform service to reach end users, the number of end users, potential barriers to entry due to network effects and data-driven advantages, economies of scale and scope from which a core platform service provider benefits, the binding of business users or end users, conglomerate or vertical corporate structures as well as other structural market characteristics. In the opposite direction, a company that meets the quantitative thresholds of the presumption rule of Art. 3 (2) DMA can also plead that it is in fact no gatekeeper. However, it bears the burden of proof for the non-compliance with the qualitative criteria despite the fulfilment of the presumption thresholds, Art. 3 (5) DMA.
Overall, the designation of a gatekeeper on the basis of enumerated core platform services and quantitative thresholds is an approach that provides extra transparency for companies. In addition, this quantitative approach offers an easier information gathering, which causes less resources. Compared to Section 19a (1) GWB, however, a lack of flexibility stands out, as an extension of the list of core platform services in Art. 2 (2) DMA is only possible after a market investigation pursuant to Art. 19 DMA (duration: at least 12 months) and an elaborate legislative amendment. In addition, the scope of application of the DMA threatens to be overstretched due to the primary consideration of quantitative thresholds and the danger of over-enforcement arises. High user numbers do not necessarily indicate the existence of gatekeeper power, for example if users have the option of multi-homing.
- The heart of the DMA – Articles 5, 6 and 7 (See more general here)
Under Articles 5 and 6, the DMA regulates the obligations to be fulfilled by gatekeepers. In comparison to Section 19a GWB, the DMA does not only rely on prohibitions, but also on imperatives. Another special feature of the DMA is that these obligations are ‘self-executing’ and must therefore be followed without any further order, i.e. the European legislator has established a so-called ex-ante regulatory regime. Article 5 contains ‘obligations for gatekeepers’, while Article 6 sets out obligations of conduct that ‘may be further specified in Article 8′ by the solely competent Commission. Most of the newly introduced prohibitions can be linked to previous European antitrust proceedings. The most prominent examples are probably the cases of Google Shopping, Apple Pay, Google Android and Amazon Marketplace, to name only a few.
Art. 5 DMA provides (among others) a prohibition for gatekeepers to combine onsite data with offsite data (Art. 5 [2] lit. b, c and d), to use parity clauses (Art. 5 [3]), to oblige users to use the gatekeeper’s own payment service (etc.) (Art. 5 [7]), to require the subscription or registration for a core platform service in order to be able to use another platform service (Art. 5 [8]). Furthermore, it contains the obligation to enable business users to conclude contracts with end users also beyond the platform (Art. 5 [4]), to grant end users access to content, functions, etc. of business providers via the core platform services, even if the contract was not concluded via the platform service (Art. 5 [5]) as well as to provide information to advertisers and publishers regarding the placed online advertisement (Art. 5 [9 and 10]).
Art. 6 DMA contains (among others) a prohibition on gatekeepers to use non-public data from business users generated by the platform to compete with them on the platform (Art. 6 [2]), to give preference to their own services and products in ranking and related indexing and crawling (Art. 6 [5]) to restrict end users’ ability to switch (Art. 6 [6]), as well as the obligation to allow end users to uninstall pre-installed software and to freely choose default settings (Art. 6 [3]), to allow the installation of third party software (Art. 6 [4]), to ensure interoperability and access to hardware and software of the platform’s core service to enable the offering of ancillary services that interact with the operating system (Art. 6 [7]), to provide effective portability of data generated on the platform and continuous real-time access (by now only) for end-users (Art. 6 [9]) and to ensure effective, high quality, continuous and real-time access to this data in aggregated or non-aggregated form for business users (Art. 6 [10]), to provide commercial users with fair and non-discriminatory access to software application stores, online search engines and online social networking services (no longer just the App Store) (Art. 6 [12]), and to allow third party online search engine providers fair, reasonable and non-discriminatory access to the ranking, query, click and view data generated by a gatekeeper’s search engine (Art. 6[11]). Also, newly added to the DMA is the prohibition on gatekeepers from imposing disproportionate general conditions for the cancellation of a core platform service (Art. 6[13]).
An interesting aspect is the newly introduced Article 7 DMA, which is to apply specifically to number-independent interpersonal communications services and thus primarily targets social messenger services. The regulation obliges gatekeepers that offer number-independent interpersonal communications services to make at least the basic functions of their service interoperable to number-independent interpersonal communications services of another provider (Art. 7 [1] DMA). Following Art. 7 (2) DMA the sending of messages as well as images, voice messages, videos and other files between two individual users should be possible immediately (a). Within two years from the designation of the gatekeeper, the same functions shall be made accessible within and with the participation of groups of users (b). Within four years after designation date, this applies also to voice and video calls (c). This provision facilitates multi-homing for social messengers, as it is already available for email providers, and thus strengthens competition on the market for social messengers. As to what is meant by number-independent interpersonal communications services, reference can be made to Art. 2 No. 7 of Directive EU/2018/1972 on the European Electronic Communications Code: It is defined as an ‘interpersonal communications service which does not connect with publicly assigned numbering resources, namely, a number or numbers in national or international numbering plans, or which does not enable communication with a number or numbers in national or international numbering plans’. This is clearly the case for Facebook Messenger as well as presumably WhatsApp, which uses the phone number only for verification and identification.
The DMA follows the ‘one size fits all’ approach, which prohibits all above-mentioned practices by gatekeepers, irrespective of the individual business models, the various platform services with their individual characteristics as well as the particular conditions of competition. Obviously, no recourse was made to the original idea of introducing a blacklist, which covers certain types of conduct per se, and a greylist, which includes a justification exception. The prohibitions are worded precisely for the benefit of legal clarity and certainty, but as a result they also lack flexibility. In comparison, the provisions of Section 19a (2) of the GWB have a much wider scope and are more flexible in their application. For example, the list of obligations under the DMA can only be extended in accordance with Art. 12 (like the designation of a gatekeeper that does not fulfil the presumptions of Art. 3 [2], see above). Furthermore, the DMA does not cover a general objective justification apart from the exceptions for economic viability (Art. 9) or for reasons of public health and safety (Art. 10). On the one hand, this speaks in favor of the targeted acceleration and effectiveness of enforcement; on the other hand, however, the ‘one size fits all’ approach increases the danger of over-enforcement.
The ex-ante regulation is likely to have an accelerating effect and save resources, not only because of the elimination of any market analysis in order to define markets and examine market dominant positions, but also because of the elimination of the case by case examination of the conduct at issue and its effects on competition. However, it remains to be feared that comprehensive disputes between the parties will now focus on the classification as Gatekeeper and the necessary remedies, thus again contributing to a prolongation of the legal proceedings.
- Investigative Powers, Compulsory Measures, Enforcement and other matters
The Commission’s powers of investigation to prove an infringement of the DMA provisions (Articles 20 – 29 DMA) are similar to those of European antitrust law (Articles 17 – 22 of Regulation 1/2003). Now explicitly included is the power of the Commission to gain insight into the algorithms used by an undertaking (Art. 21 and 23 [2] lit. d). The latter is particularly interesting in order to review algorithm-based remedies if there is concern that the infringement has not been (sufficiently) stopped (see also the discussion in the Google Shopping case).
In the case of non-compliance, the Commission can impose a fine on the gatekeeper of up to 10% (in the case of less serious infringements such as information obligations: up to 1%) of its total worldwide turnover in the previous financial year in accordance with Art. 30 DMA. This substantially reflects Art. 23 of Regulation 1/2003. Newly included is the possibility to impose a fine of up to 20 % of the total worldwide turnover if there is a repeated violation in relation to the same core platform service which the gatekeeper has already been charged with in the eight preceding years, Art. 30 (2) DMA. In the event of systematic non-compliance with the gatekeeper obligations, the Commission has the right pursuant to Art. 18 DMA to impose not only behavioral but also structural remedies following a market investigation.
As mentioned above, the Commission is also provided with the tool of market investigation to designate an undertaking as gatekeeper (Art. 17 DMA; for example, in less clear-cut cases), to expand the list of core platform services (Art. 19 DMA) or the list of obligations of conduct (Artt. 12 and 19 DMA).
After the publication of the Commission’s draft of the DMA, national competition authorities, especially the German Federal Cartel Office which is equipped with Section 19a GWB, were concerned that they would no longer play or be able to play a role in proceedings against the digital giants. The urgent demand made by the national competition authorities to become more involved in the enforcement of the DMA seems to have been clearly rejected after the view of this version of the DMA. Even Articles 37 and 38 of the DMA on cooperation with the NCAs do not contain any concessions to the NCAs, but almost only further obligations for those. However, according to Article 41 DMA, a specified number of Member States are entitled to request the above-mentioned market investigations regarding systematic non-compliance with the gatekeeper obligations (at least one MS), the designation of gatekeepers, the extension of the lists of central platform services as well as the conduct obligations (at least three MS each) from the Commission, along with providing evidence. The Commission will then decide within four months as to whether it will comply with the request (Art. 41 [5]).
In addition, some national cartel authorities are involved by way of the European Competition Network, which according to Art. 40 DMA is part of a so-called High-Level Group chaired by the Commission. Its importance, however, is to be estimated at a lower level, since it has at most an advisory function by means of an annual report and the possibility to suggest amendments, deletions and additions of provisions. The so-called Digital Markets Act Advisory Committee (Art. 50 DMA) has a similar advisory function, in which all member states are represented, although they do not necessarily have to send a representative of the national competition authority. Finally, Art. 26 (2) of the DMA mentions the possibility for the Commission to mandate employees of national competition authorities to monitor compliance with the conduct obligations of the DMA.
Furthermore, the DMA only provides for public enforcement of the obligations set out in the regulation. Private enforcement, which has become increasingly important in recent years, remains probably possible. In its Q&A on the DMA, the Commission assumes the possibility of private enforcement and the DMA also mentions the cooperation relationship with national courts in Art. 39 (like Art. 15 of Regulation 1/2003). At least, private enforcement requires a national legal basis in each jurisdiction (in Germany this seems to be Sec. 823 [2] and Sec. 1004 of the German Civil Code).
- Conclusion
In conclusion, the DMA is a proper first step on the road to regulating digital gatekeepers. The goals of accelerating and making antitrust proceedings more effective, as well as ensuring fair and contestable markets and preventing a fragmentation of the legal system in the internal market, are valid and meet with a common understanding. However, the DMA as it stands now is not the ideal solution and has also some shortcomings. For example, due to the absence of the possibility of a factual justification, the provisions of the DMA give rise to the danger of over enforcement. Additionally, the lack of involvement of the national competition authorities regarding the enforcement of the DMA is at least to be questioned. Whether further-reaching national regulations remain applicable alongside the DMA can only be finally clarified by the Court of Justice.
Nevertheless, the DMA falls short of expectations, especially in view of its lack of flexibility to adapt to future relevant new core platform services and other anti-competitive practices, as well as its at least questionable impact on the acceleration of antitrust proceedings. When the DMA is come into force, due to the expected duration for designating gatekeepers we will probably see the first results at earliest in 2024.
***
The author is a Research Assistant and Doctoral Candidate at the Chair of Global Commercial Law, International Arbitration and Civil Law (Prof. Dr. Florian Bien) at the University of Wuerzburg, Germany.
The author gives special thanks for the valuable support to Mr. Moritz Schwemmer, Student Assistant with the mentioned Chair at the University of Wuerzburg, in preparing the manuscript.
________________________
To make sure you do not miss out on regular updates from the Kluwer Competition Law Blog, please subscribe here.
