By Kris Van Hove & Argyrios Papaefthymiou1
It is a truism to say that digitalisation has brought with it a revolution in economic life and consumer experience in the 21st century. Developments in digital technology unfold in rapid speed, markedly upgrading the quality of life, but also requiring constant review of legal and economic thinking to keep up with the pace of digital evolution.
Competition law has been at the epicentre of this re-evaluation process, as the field of law that moves most closely in tandem with economic thinking and business models. It is thus no surprise that there has been an onslaught of policy and legislative developments in recent years in the competition law space to tackle what is perceived by some as an “overexpansion” of the market power held by big tech companies (commonly identified as “GAFA” – Google, Amazon, Facebook, and Apple). In parallel with these developments in the policy and legislative space, competition authorities around the globe are increasingly stepping up their scrutiny of various practices implemented by digital companies.
The present article seeks to provide an overview of the key competition concerns identified in Europe in the digital sphere and the main approaches that are currently being discussed to tackle those concerns, with a focus on the EU. Although the digitalisation of the global economy presents specific challenges for competition law enforcement, which may require tweaking the existing regulatory framework to some extent, the article attempts to critically assess whether the reform proposals that have been put forward are necessary and proportionate to the perceived challenges. This assessment is made in light of the tools that are already available and that have been employed to address competition concerns in the digital sector by competition authorities in the EU. While the proposals that have been put forward centre on the rules on abuse of dominance and merger control, the analysis that follows is conducted through the prism of antitrust, with references to merger control as it relates to the interplay between ex ante enforcement (merger control) and ex post enforcement (rules on abuse of dominance).
A. Main Competition Concerns in Digital Markets: The View in Europe
Over the past couple of years, public authorities in Europe, both at EU and national level, have commissioned expert studies and reports on the application of competition law in the digital era. An indicative list includes: the European Commission’s report on “Competition Policy for the Digital Era,”2 the UK Furman Report on “Unlocking Digital Competition,”3 and the UK Competition and Markets Authority’s (the “CMA”) market study on online platforms and digital advertising,4 the German “Competition Law 4.0” report,5 the Italian6 and Franco-German7 reports on big data and competition law, the French study on competition and e-commerce8 and the French report on digital economy competition challenges,9 etc. From this body of expert studies and reports, it is possible to distill some key insights into the characteristics of digital markets and the perceived competition concerns they generate. These are addressed briefly below.
1. Characteristics of Digital Markets
Digital markets are often characterized by strong economies of scope, favoring the development of ecosystems and giving incumbents a strong competitive advantage. The economies of scope are considered to result from other features of digital markets, and in particular from:
- high returns to scale (meaning that the cost of producing a digital service is much less than proportional to the number of customers served);
- network externalities (meaning that the more users join a digital technology, e.g. a social media platform, the more convenient it becomes); and
- the importance of data as an input for many digital services (meaning that, where applicable, those with significant data collection and processing capabilities have a sizeable competitive advantage).
These features and the strong economies of scope they generate are increasingly considered in enforcement and policy circles as facilitating large incumbents’ market power to become ever more entrenched and as potentially allowing anti-competitive strategies to materialize unencumbered.
The European Commission’s Digital Experts’ Report defines “self-preferencing” as a practice engaged in by a platform operator that gives preferential treatment to the operator’s products and services when they compete with products and services of other entities using the platform. This practice has been at the forefront of many discussions on digital competition policy. Whether or not it is as such problematic from a competition perspective is the subject of ongoing debate. Yet, alongside data-related practices, it is the conduct most closely associated with competition concerns that big tech companies raise. Headline cases such as the EU Google Shopping case10 have featured this practice as the principal conduct under scrutiny.
3. Retail Most-Favored-Nation (“MFN”) Clauses
Retail Most-favored-nation clauses or MFNs11 are contractual obligations imposed by a buyer on a seller that require the seller to offer the buyer terms and conditions that are at least as favorable as the best terms and conditions offered by that seller to other buyers. The terms and conditions concerned may be either price-related or non-price-related. Such clauses are commonly divided into two categories:
- “narrow” Retail MFNs, which dictate that the seller may not offer better terms and conditions on its own website (but can do so on other platforms); and
- “wide” Retail MFNs, which dictate that the seller may not offer better terms and conditions on its own website or on any other platform.
Retail MFNs have been particularly pertinent in the digital context, with “platform MNFs” being scrutinized both at EU level (Amazon E-books MFNs case12) and at national level (antitrust investigations by the German, French, Italian and Swedish national competition authorities into MFNs in agreements between hotels and online travel agents (Hotel Reservation Service (HRS),13 Booking.com,14 and Expedia15). Generally speaking, “wide” Retail MFNs are considered more problematic from a competition law perspective than “narrow” Retail MFNs.
4. Switching Costs and Multihoming
Switching from a product or service offered by one firm to a functionally similar product or service offered by another firm may generate so-called “switching costs.” These are costs associated with certain investments made by the user in the ecosystem of the first firm (e.g. the time spent familiarizing with the particular features of that ecosystem) which have to be re-incurred when the user switches to the ecosystem of the second firm. Switching costs are also conceptually linked to “multihoming,” which is a consumer’s ability to use different competing online services simultaneously (e.g. platforms).
Depending on the circumstances, a dominant platform may try to retain its user base on an exclusive basis by making switching or multihoming more costly or more difficult otherwise. Such conduct could enable the platform to entrench its market position with potentially negative effects for competition.
5. Big Data and Data-Related Practices
“Big Data” is a dataset with three features which are commonly referred to as the “three V’s.” It consists of:
- large Volumes of data (that is, large amounts of data generated through online or offline transactions),
- data with significant Variety (it can be structured or unstructured data and human-generated or machine-generated data), and
- data generated with high Velocity (that is, incoming data flows are processed in real-time).
Big Data and its interplay with competition policy is a key focus in many competition authorities’ ongoing evaluation of the digital sector. Particular attention is paid to data leveraging strategies that, under given circumstances, may amount to anti-competitive unilateral conduct. For instance, a refusal to provide access to data that is unique and indispensable for a competitor to provide services (so, data that constitutes an “essential facility”) may be considered to amount to an exclusionary abuse. Another example of potentially abusive conduct is discrimination by a vertically integrated platform with downstream activities in the provision of access by retailers to competitively strategic data. Furthermore, competition authorities are increasingly sensitive to issues of “data portability” and “interoperability.” “Data portability” (giving end-users the freedom to transfer their data to alternative digital service providers) as well as “interoperability” (ensuring compatibility between software of different ecosystems) are considered valuable means to keep switching and multihoming viable for end-users. But these are not the only issues competition authorities are looking into. For instance, at EU level, the European Commission has launched a formal investigation into Amazon’s use of sensitive data obtained from independent retailers selling on Amazon’s marketplace and competing with Amazon’s retail business (July 2019).16 The Commission is also conducting antitrust enquiries into Facebook’s and Google’s data-gathering and monetization practices (since 2019).
An additional concern that recently came to the fore in relation to data-related practices centers on the interaction between privacy protection (legislated at EU level through the General Data Protection Regulation – GDPR17) and competition law. This interaction is highlighted in the German Facebook case. In that case, the German competition authority found that Facebook had abused its dominant position in the market for social networks by making access to its social network conditional on the collection of user data from multiple sources, including third-party websites. The German authority thus ordered Facebook to cease applying the abusive clauses.18 The authority’s decision was appealed to the Higher Regional Court of Düsseldorf, which issued a suspension order against the decision, noting serious doubts about its legality.19 The decision was, however, reinstated by Germany’s Supreme Court20 pending a final decision on the merits of the case from the Higher Regional Court. The German Facebook case has sparked an intense debate on the theory of harm put forward by the German competition authority, with critics finding its focus on consumer protection problematic for the purpose of identifying the competition concerns. It is expected that the final outcome of this case will have significant implications for the use of data-related theories of harm in competition law enforcement.
6. Killer Acquisitions
A strongly contested practice that some consider is taking place in the tech sector in recent years is the acquisition by established players of start-ups that have the potential of strongly competing with the incumbents, with a view to “killing off” future competition from those start-ups. Such acquisitions are commonly referred to as “killer acquisitions.” The concept originated in the pharma sector, where – according to some – the practice of buying up start-ups, only to subsequently eliminate pipeline projects that might have increased competition for the incumbent’s offerings, would have been empirically observed.21 In the context of the digital sphere, however, this concept appears to mean something different, as the acquisitions considered by some to be controversial (e.g. Facebook/WhatsApp,22 Facebook/Instagram) have not resulted in the “killing off” of the acquired firm’s offerings. Rather, what is meant by “strategic acquisitions” in digital markets relates to a counterfactual where the acquired firm, in light of its subsequent evolution into a strong market player, might have exerted intense competitive forces on the acquiring incumbent, with positive results for competition and consumers, had it not been acquired. The existence of harm to competition arising from these acquisitions is not beyond dispute. On the one hand, arguably, the acquisition of innovative start-ups can in some cases reduce future competitive constraints on the incumbent. On the other hand, for many start-ups, being acquired by an established market player at a lucrative price point can serve as a key incentive for innovative research and development.
In any event, in various European jurisdictions there is a strong current towards reforming the rules on merger control to address what is perceived to be an excessively lax approach to controlling acquisitions in the tech sphere. Reform proposals have taken various forms. One commonly pronounced suggestion is to lower the jurisdictional turnover thresholds to account for the fact that, at the time of acquisition, the turnover of many innovative start-ups is too low to trigger a notification. Another proposal is meant to address the fact that turnover may not be the most appropriate metric to assess the need to exercise merger control over a tech acquisition. It consists in adding an alternative threshold based on the value of the transaction, where a high purchase price (despite low turnover) triggers mandatory notification due to its indications of innovative potential. This proposal has in fact already been adopted in certain jurisdictions and notably Austria and Germany. Finally, at the more “radical” end of the reform spectrum, some jurisdictions (such as France) have been considering to implement an ex post merger control regime, whereby certain transactions (particularly in the digital sector) may be re-evaluated even years after they have been consummated, should market developments in the transaction’s aftermath indicate that the transaction might have generated adverse effects on competition.
The discussion presented above has direct implications for the debate on digital-specific reforms of the competition rules: if the concerns delineated have merit, then more ex post scrutiny may be required. For example, a history of relatively lax merger control could be viewed as a factor that has contributed to the proliferation of self-preferencing cases which focus on perceived anti-competitive differential treatment by vertically integrated businesses. The intensifying review of data-related abusive practices could perhaps also be partially attributed to a perceived overexpansion of big tech firms.
B. Reform Directions: A Critical Overview
Competition authorities in Europe have been engaged in intensive public discussions on how to approach the perceived problems identified above for competition in the digital sector. From the flurry of policy activity and legislative initiatives on this subject, it is possible to extract the key proposals presently on the table, and to critically assess their necessity and proportionality. Each of these proposals is addressed below.
1. Mandating Pro-Competitive Conduct and Prohibiting Anti-Competitive Conduct under an Ex Ante Regulatory Regime
In June 2020, the European Commission launched a public consultation to gather views on two legislative proposals. A first proposal relates to a “Digital Services Act.” This proposal would introduce, inter alia, a targeted, sector-specific ex ante framework regulating the conduct of large online platforms that benefit from significant network effects and that act as “gatekeepers” of their respective platforms. A second proposal relates to a “New Competition Tool.” This proposal would give the European Commission new powers to investigate entire sectors of the economy and impose remedies where it finds dominance-related or market structure-related competition problems (on this second proposal, see section 2 below).
The Digital Services Act package considered by the Commission in its Inception Impact Assessment23 puts forth different policy options for enhancing the “fairness and contestability” of markets where large online platforms operate. While two of these policy options focus on enhancing transparency obligations for platforms in the context of the Platform-to-Business Regulation24, the third option – which has received the most public attention – proposes the adoption of a new ex ante regulatory framework that governs specific trading practices of large online platforms benefiting from significant network effects and acting as “gatekeepers.”25 The platforms concerned would be identified on the basis of predetermined criteria such as significant network effects, the size of their user base and/or their capacity to leverage data across sectors.
Two separate components are considered for the ex ante framework. A first component would lay down predetermined prohibited trading practices (“blacklisted practices”) and predetermined obligations. An example of a blacklisted practice subject to a horizontal prohibition would be intra-platform self-preferencing. A second component would provide for the possibility for the regulator to impose tailor-made remedies. Examples of such remedies would be the imposition of platform-specific non-personal data access obligations and the imposition of specific interoperability and data portability requirements. The two components are not mutually exclusive and could be introduced as complements. The idea of ex ante prohibitions of certain platform conduct (e.g. self-preferencing) is also endorsed in the proposed 10th amendment of the German Act Against Restraints on Competition,26 as well as in the French competition authority’s study on competition and e-commerce and the French report on digital economy competition challenges.
Similar to the notion of ex ante regulation, the UK Furman Report puts forth the idea of drawing up binding codes of conduct for certain platforms with “strategic market status” (broadly defined as a position of enduring market power over a strategic gateway market, giving the platform a powerful negotiating position and making other undertakings dependent on it). Adherence to the codes of conduct would be monitored and enforced by a dedicated Digital Markets Unit within the competition authority (such a unit has already been established within the French competition authority, with a mandate to assist in cases with a significant digital dimension27). As opposed to the focus of the ex ante regulatory approach on the prohibition of specific types of conduct in addition to required compliance with specific obligations, the focus of the codes of conduct approach would be only on mandating pro-competitive conduct. This more flexible form of regulation could be seen as an incentive for large platforms to work in partnership and share their input in drawing up codes and as a means perhaps of avoiding exogenously imposed obligations and prohibitions that may go “over the top” in their response to the perceived problems they seek to address. The idea of codes of conduct specifically for large online platforms has also been supported in the German Competition Law 4.0 Report.
It further bears observing that, for a practice to qualify as an ex ante prohibition, it is rather logical to assume that such practice’s harm to competition must be significant, acknowledged to a degree of (near-)consensus, and context-independent – almost, in a sense, like a “by object” restriction under EU competition law. However, certain of the practices identified in policy proposals, such as the Digital Services Act and the relevant German and French studies, are still the subject of much debate as regards their anti-competitive nature. The most indicative example in this respect is self-preferencing, which some commentators view as a natural consequence of vertical integration, and its horizontal condemnation as near-equivalent to competition law engaging in the imposition of specific business models on companies.28 Furthermore, depending on the context in which it is observed, self-preferencing could be viewed through the lens of classic theories of harm that have already been successfully applied by EU courts to deal with instances of integrated firms favoring their own affiliates, such as a refusal to supply an indispensable input (Bronner, IMS Health, Slovak Telekom) or tying (Microsoft I).29 In any event, the uncertainty that surrounds self-preferencing as a self-standing theory of harm might call for a more nuanced approach than outright banning practices that can be labelled as self-preferencing.
Of course, self-preferencing is not the only conduct that is targeted by the ex ante regulation approach and, for instance, proposals mandating interoperability and data portability in specific circumstances may be less controversial. However, these practices could just as easily be imposed through codes of conduct. This approach would arguably be preferable, as it may provide a step – rather than a leap – in the exploration of ways to address competition concerns relating to the behavior of large incumbents in digital markets. Codes of conduct are likely more easily adjustable to account for developing market dynamics and insights drawn from their enforcement practice than formal legislation (especially at the EU level where the legislative process includes up to three different institutions, i.e. the European Commission, the Council of the EU, and the European Parliament).
2. Leveraging Market Investigation Tools
In parallel with the Digital Services Act addressed above, the European Commission has been gathering views on a proposed “New Competition Tool” (or “NCT” for short). This tool would allow the Commission to investigate entire sectors of the economy and impose behavioral or structural remedies to address what it perceives to be dominance-related or market-structure-related risks to competition. The NCT could be used pre-emptively, that is, before actual risks to competition materialize. Its scope of application would be limited to markets within the digital sphere, or be open-ended.30 If the tool would be open-ended and targeting market structure risks, it would resemble a classic market investigation mechanism, such as the one available in the UK as well as in certain other jurisdictions worldwide (e.g. Greece, Mexico, Israel). Although the European Commission already has a sector-wide probe tool at its disposal, i.e. the sector inquiry, that tool can result only in findings but not in remedies, which may be imposed only in subsequent individual infringement proceedings.31 In contrast, an NCT investigation could result in the immediate imposition of market-wide remedies.
A market investigation mechanism would be a very powerful tool in the hands of the European Commission, with significant implications for competition enforcement in the EU at large. Cases that have centered on challenging theories of harm could be dealt with through an NCT-like process in an easier manner, dispensing with the hurdle of proving to the requisite legal standard the elements of a “traditional” antitrust infringement (e.g. specific abusive conduct). The Commission’s antitrust investigations in landmark digital cases have often spanned multiple years (the Google Android32 case took over three years and the Google Shopping case took seven years for the Commission to complete). Critics have argued that the duration of these investigations, especially in rapid-changing markets, has made it impossible to effectively reverse the damage that the conduct at issue has allegedly caused to competition. By the time a decision is issued, the undertaking concerned would have derived most, if not all, potential competitive advantages from the infringing behavior, and would have further entrenched its dominant market position. The duration factor may have played a significant role in motivating the Commission to seek market investigation powers.
The NCT proposal has, however, not been met with unequivocal approval. Stakeholders have expressed concerns that the tool is too far-reaching, disproportionate to the issues it seeks to address, and legally questionable (its legal basis is at least partially the antitrust provisions of the Treaty on the Functioning of the European Union, despite the fact that the NCT’s application – as envisaged – cannot result in a finding of infringement). Also, the design of the NCT itself is raising questions: the profound impact the tool may have on markets and market players, especially in its broadest possible form, requires meticulous design of the different parameters for its use, namely: when use of the tool would be appropriate, who would be empowered to request the launch of an NCT investigation, what the legal thresholds and substantive standards of assessment would be, and what precise form the rights of defense would take. Yet, most notably, in the 40+-page questionnaire of the NCT public consultation, only one question addresses rights of defense, which is arguably the most burning concern for stakeholders. Since the NCT is evidently intended to facilitate swifter intervention, there is a pressing need to ensure that this speedy and more flexible approach does not discount on any of the rights of defense that are available to companies in regular antitrust proceedings as provided for in Regulation 1/2003. 33 In fact, beyond the uncertainty as to what form the rights of defense would take, there may be a broader systemic issue in play here: due to the relatively loose concepts involved compared to those in regular antitrust enforcement (e.g. “structural competition concerns” vs “abuse of a dominant position”), market investigation instruments have the potential to go much further with respect to the conducts targeted and the remedies that can be imposed. This far-reaching nature justifies strong checks and balances and recourse to an efficient and effective judicial oversight system.
A final observation with regard to the NCT relates to its scope of application: while two of the four policy options put on the table by the Commission are limited to certain digital or digitally enabled markets, the other two are horizontal. As the rationale presented by the Commission for the need of an NCT is heavily focused on the reform of competition enforcement in the digital sector, the proposal of a horizontal tool is somewhat puzzling. Yet, the horizontal option may be sensible in a “future-proofing” strategy: as the EU legislative procedure is rather cumbersome, and expanding the tool’s scope in the future, should the need arise, would be time- and resource-consuming (as well as uncertain), choosing the broadest possible option from the outset could make practical sense. However, the fact remains that, based on its reasoning to date, it is questionable whether the Commission has made a convincing case for a horizontally applicable market investigation tool.
3. Facilitating the Burden of Proof
The European Commission’s Digital Experts’ Report observes that highly concentrated markets with strong network effects and high barriers to entry may require the imposition on the incumbent of proving the pro-competitiveness of its conduct. In terms of specific conduct to which this burden of proof shift would apply, the Report suggests that, where a dominant platform tries to expand into neighboring markets, a presumption should exist in favor of a duty to ensure interoperability and the burden to reverse that presumption should be on the platform. The Report further notes that a presumption of interoperability could also be justified where a dominant platform controls specific competitively relevant sets of user or aggregated data that competitors cannot reproduce but need in order to compete and again the burden to reverse that presumption should be on the platform. With regard to self-preferencing, the Report makes the proposition that, if a platform serves as an important intermediary with a “regulatory” function on its platform, then that platform should bear the burden of proving that a given self-preferencing practice has no long-run exclusionary effects on product markets.
The proposals for shifting the burden of proof from the investigating authority to the investigated undertaking constitute a significant departure from established norms in EU competition law. As provided for in Regulation 1/2003, the burden of proof in antitrust proceedings falls on the party making the claim at issue. It is therefore unclear how these proposals could come to pass without revision of this Regulation. It is little surprise that proposals on altering the burden of proof allocation have not been widely endorsed in other policy reports or legislative proposals in this area. Furthermore, the need for such a significant change to a fundamental procedural rule in a selective manner for a specific group of companies is not apparent in practice: the European Commission as well as national competition authorities in the EU have managed to issue decisions with findings of infringement by big tech companies in the past without having presumptions of illegality at their disposal (e.g. the EU Google Shopping case).
4. Adapting Existing Antitrust Tools to the Digital Era
Certain EU jurisdictions have recently been examining amendments to their competition laws aimed at facilitating a more effective enforcement in the digital context. The most notable and coherently laid out proposal for reform to date in the EU has come from Germany, which in late 2019 introduced for deliberation a 10th amendment to its Act Against Restraints on Competition, which is suitably dubbed “Digitalisation Act.” This draft legislation introduces modifications to German competition law focusing on its update for the digital era. It encapsulates many of the reform directions that have been put forth in the policy discourse on the subject.
The German draft amendment proposes to supplement the criteria for assessing dominance with two new criteria, namely access to data relevant for competition, and “intermediary power.” The first criterion in essence expands the “essential facilities” doctrine to expressly cover data that is necessary as a resource for effective competition. The second criterion attempts to capture the importance of intermediation services in multi-sided markets for access to purchase and supply markets. This modification targets specifically the bottleneck power of digital platforms that act as “gatekeepers.” Their market power is considered particularly enhanced due to their ability to control market access for suppliers that use the intermediary services through features such as search result rankings.
In addition to these adjustments of the dominance criteria, the 10th amendment proposes to expand the German competition authority’s powers so that it will be able to investigate in a time-efficient manner undertakings with “paramount significance for competition across markets” and the business practices they engage in. To assess whether an undertaking holds such “paramount significance,” the authority will in particular take account of:
- the undertaking’s dominant position in one or more markets;
- its financial strength or access to other resources;
- its vertical integration and activity in interconnected markets in ways other than through vertical integration;
- its access to data relevant for competition; and
- the importance of the undertaking’s activities for access to purchase and supply markets by third parties, as well as the influence of those activities on business activity.
What is particularly notable about these expanded powers is that, to sanction certain practices that are considered anti-competitive, the competition authority will no longer have to prove an undertaking’s dominant position in a particular market. Rather, it will be sufficient for the authority to look at the undertaking’s overall position across multiple markets, and if it meets the paramount significance threshold, the authority will be able to block practices considered anti-competitive at an early stage, even before the undertaking entrenches a dominant position in the market at issue. The provision is considered to be targeting specifically the big tech companies which, due to their particular features such as network effects and data advantages, may outpace the competition authority when leveraging their strong market positions in certain markets to rapidly build up such positions in neighboring markets so that regulatory intervention is no longer effective. By dispensing with the arduous process of establishing an undertaking’s dominance in a specific market, this amendment’s apparent aim is to give the German competition authority powers to curb leveraging practices at a much faster pace.
Once the authority has issued a decision that finds a particular undertaking to hold “paramount significance for competition across markets,” it will be able to order that undertaking to refrain from certain practices, namely:
- engaging in “self-preferencing,” that is, in its capacity as provider of access to purchase and sales markets (i.e. in its capacity as a platform), providing preferential treatment to its own products or services to the disadvantage of competing products and services;
- directly or indirectly hindering its competitors in markets where the undertaking in question can rapidly expand its position even without being dominant, if the hindrance is likely to significantly impair the competitive process;
- using competitively relevant data collected in a market where the undertaking in question is dominant to establish or enhance barriers to entry in markets where the undertaking is not dominant;
- making the interoperability of products or services or the portability of data more difficult, thereby hindering competition; and
- inadequately informing commercial users about the scope, quality or success of the service provided or commissioned, thereby making it difficult for them to assess the value of this service.
By expressly listing these practices in the legislation as potentially the subject of a prohibition, the competition authority will be able to dispense of developing detailed theories of harm, thereby facilitating faster intervention.
The 10th amendment also proposes to broaden the German competition law concept of “relative market power.” This concept is the German law manifestation of the “economic dependence” doctrine, which – irrespective of market dominance – seeks to prevent abusive practices by undertakings on which other undertakings are business-dependent. While the application of the current version of the provision is triggered only when abusive practices take place vis-à-vis economically dependent small-and-medium size enterprises (“SMEs”), the proposal expands the doctrine to business-dependent undertakings of all sizes.
Finally, the draft amendment proposes to facilitate the ability for the competition authority to impose interim measures (a sort of injunctive action). It does so by changing the threshold for such measures from the risk of serious and irreparable damage to competition to the predominant probability of an infringement finding coupled with the necessity to prevent harm to competition or to avoid an imminent and serious threat to another undertaking. This change in effect amounts to a lowering of the threshold, as the constitutive elements of the new test are likely more easily met than the stringent requirement of proving serious and irreparable damage to competition. In this regard, it is interesting to note that the draft amendment was proposed around the time the European Commission decided to impose interim measures for the first time in 18 years (this was in the Broadcom case).34 This “resurrection” of interim measures at EU level in a sector closely associated with the digital sector (chipsets for TV set-top boxes and modems) is no accident: in the context of the broader debate on the efficacy and speed of antitrust enforcement, the digital sector (equally characterized by rapid evolution of market dynamics) has been earmarked as the prime candidate for increased imposition of interim measures. As noted by Competition Commissioner Vestager after the Broadcom interim measures decision: “[I]nterim measures are one way to tackle the challenge of enforcing our competition rules in a fast and effective manner. And this is why they are so important. Especially in fast-moving markets. Whenever necessary, I am therefore committed to making the best possible use of this important tool.”35
5. Introducing Ex Post Merger Control
The idea of introducing an ex post merger control regime, which has recently been circulating among policy circles in the EU, is arguably closely related to the perceived inadequacies of the abuse of dominance rules. These inadequacies, it is claimed, would have contributed to the creation of “mega-players” in the digital sector.
In 2018, the French competition authority launched a public consultation on a proposal to introduce a new ex post review mechanism into the French merger control regime. Under this system, unreported past mergers that did not meet EU or national filing thresholds would be open to investigation by the French competition authority if the latter would determine after implementation that the transaction is raising substantial competition concerns in France.36 The French authority originally proposed a post-transaction deadline of between six months and two years, after which ex post intervention would no longer be possible. However, more recently Isabelle de Silva, head of the French competition authority, has expressed support for the possibility of reviewing “killer acquisitions” in certain sectors, most notably the digital sector, years after the conclusion of the transaction, invoking as inspiration the US merger control system which allows for a review of mergers years after their completion.37
Although the view that the entrenchment of certain firms’ market positions is a fait accompli that can perhaps only be undone by essentially rewinding the clock on those firms’ years-long expansion may be understandable in some cases, the introduction of an ex post review would constitute a significant shift for merger control standards in the EU. Merger control has traditionally operated within the EU in an ex ante fashion, and for good reason: the legal uncertainty that an ex post regime entails could be too significant and therefore have an inhibiting effect, especially if the deadline for intervention post-transaction is very long. On the other hand, to effectively take account of the real effects of a merger on competition, a longer time horizon may be necessary.
Striking a balance between these conflicting considerations is certainly not easy. However, other proposals that have been voiced on the subject – although not perfect themselves – could be more prudent. For example, a wider introduction of transaction-value-based thresholds would be a more modest way of capturing acquisitions of start-ups by big tech firms that fall under the radar in a pure turnover-based threshold system. Another way of tackling the specific concern of “killer” acquisitions, although indirect, is offered in the UK Furman Report: for companies with the “strategic market status” designation, all acquisitions could be made subject to mandatory notification. A more modest approach to address “killer” acquisitions has very recently been advanced by Competition Commissioner Vestager suggesting that the European Commission could be empowered to review transactions referred to the Commission by national competition authorities even if those authorities themselves do not have the power to review these transactions because they fall below the national turnover thresholds.38 Finally, a more vigorous application of interim measures in the context of abuse of dominance investigations could contribute to a more speedy and efficient antitrust enforcement, thereby preventing the risk of systematic consolidation and expansion of such firms’ market power.
The ongoing vigorous policy debate on reforming the application of antitrust law in the digital sector is not without reason. The digitalization of the economy has brought with it notable challenges for maintaining competitive conditions in digital and digitally enabled markets. However, certain challenges in the digital sphere are not novel and have already been dealt with by competition authorities and courts in both digital and non-digital markets using the existing antitrust toolset with success. While the digital economy’s particularities may require some tweaks to the enforcement mechanism, it is important to assess each proposal against robust necessity and proportionality standards, while placing the need to guard procedural defense rights – and, in a sense, the rule of law – at a prominent place in that reform process.
Some proposals and general policy directions for reform are uncontroversial, such as enhancing and facilitating the application of interim measures. However, before implementing concrete reforms, it is equally important to carefully study the implications of each initiative that might bring about deeper and lasting changes to the foundations of competition law enforcement in the EU. Ex ante prohibitions, the establishment of expansive market investigation powers and ex post merger control are just some of the proposals that require intense scrutiny to avoid generating chilling effects: first in their rationale, and second, if considered desirable, also in their design. It remains to be seen how this delicate balancing act will play out in practice.
Click here for a PDF version of the article
1 Kris Van Hove is a competition partner at the law firm Van Bael & Bellis, Brussels. Argyrios Papaefthymiou is an associate in the competition practice of the law firm Zepos & Yannopoulos, Athens. The views expressed in this article are those of its authors and should not be attributed to their respective law firms or clients.
2 Available at https://ec.europa.eu/competition/publications/reports/kd0419345enn.pdf.
3 Available at https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/785547/unlocking_digital_competition_furman_review_web.pdf.
4 Available at https://assets.publishing.service.gov.uk/media/5dfa0580ed915d0933009761/Interim_report.pdf.
5 Available at https://www.bmwi.de/Redaktion/DE/Publikationen/Wirtschaft/bericht-der-kommission-wettbewerbsrecht-4-0.pdf?__blob=publicationFile&v=12.
6 Available at https://www.agcm.it/dotcmsdoc/allegati-news/Big_Data_Lineeguida_Raccomandazioni_di_policy.pdf.
7 Available at https://www.bundeskartellamt.de/SharedDocs/Publikation/DE/Berichte/Big%20Data%20Papier.pdf?__blob=publicationFile&v=2.
8 Available at https://www.autoritedelaconcurrence.fr/sites/default/files/concurrence-commerce-en-ligne-en.pdf.
9 Available at https://www.autoritedelaconcurrence.fr/sites/default/files/2020-02/2020.02.28_contribution_adlc_enjeux_num.pdf.
10 Google Search (Shopping), Case AT.39740, available at https://ec.europa.eu/competition/antitrust/cases/dec_docs/39740/39740_14996_3.pdf. As most other Commission decisions finding an abuse of dominance, the Commission’s decision in Google Shopping is the subject of an appeal before the EU Courts, see Case T-612/17, Google and Alphabet v. Commission, case not yet decided. The investigations the European Commission has recently opened into Apple’s App Store would reportedly also involve a form of conduct that would favour Apple’s own services and disadvantage rivals: Apple – App Store Practices (music streaming), Case AT.40437, Apple – App Store Practices, Case AT.40716, and Apple – App Store Practices (e-books/audiobooks), Case AT.40652, press release, June 16, 2020, IP/20/1073, available at https://ec.europa.eu/commission/presscorner/detail/en/ip_20_1073.
11 Such clauses are also referred to as “most-favoured-customer” or “retail parity” clauses.
12 E-book MFNs and related matters (Amazon), Case AT.40153, available at https://ec.europa.eu/competition/antitrust/cases/dec_docs/40153/40153_4392_3.pdf.
13 Bundeskartellamt Decision of December 20, 2013, B 9-66/10, available at http://www.bundeskartellamt.de/SharedDocs/Entscheidung/EN/Entscheidungen/Kartellverbot/B9-66-10.pdf?__blob=publicationFile&v=3; upheld by Higher Regional Court of Düsseldorf, available here https://www.olg-duesseldorf.nrw.de/behoerde/presse/archiv/Pressemitteilungen_aus_2015/20150109_PM_HRS_Entscheidung/index.php (press release).
14 Décision n.º 15-D-06 du 21 avril 2015 sur les pratiques mises en œuvre par les sociétés Booking.com B.V., Booking.com France SAS et Booking.com Customer Service France SAS dans le secteur de la réservation hôtelière en ligne, available at https://www.autoritedelaconcurrence.fr/sites/default/files/commitments//15d06.pdf; Decision of the Swedish Competition Authority, no. 596/2013, April 15, 2015, available at http://www.konkurrensverket.se/globalassets/konkurrens/beslut/13-0595e.pdf; Decisione dell’Autorità Garante della Concorrenza e del Mercato, April 21, 2015, available at https://www.agcm.it/dotcmsDOC/allegati-news/I779_chiusura.pdf; Decision of the German competition authority, December 22, 2015, available at https://www.mlex.com/Attachments/2020-08-27_6FK3QP0LX76X258S/BGH%20Facebook%20kvr__69-19.pdf (This decision was reversed on appeal by the judgment of the Düsseldorf Higher Regional Court, June 4, 2019, available at https://www.justiz.nrw.de/nrwe/olgs/duesseldorf/j2019/Kart_2_16_V_Beschluss_20190604.html (a further appeal is pending before the German Supreme Court). In August 2020, the German competition authority issued a critical report on the effects of narrow MFNs on online sales. This report is available at https://www.bundeskartellamt.de/SharedDocs/Publikation/EN/Schriftenreihe_Digitales_VII.pdf?__blob=publicationFile&v=3. The report is heavily criticised by Booking.com: see “Booking.com spars with German antitrust regulator over 2015 ban of ‘narrow parity clauses,’” MLex (August 28, 2020), available at https://www.mlex.com/GlobalAntitrust/DetailView.aspx?cid=1219000&siteid=190&rdir=1.
15 After Booking.com undertook commitments vis-à-vis the French, Italian and Swedish competition authorities, Expedia also removed wide MFN clauses from its contracts, leading to the authorities closing their respective investigations into Expedia’s MFN practices.
16 Amazon Marketplace, Case AT.40462, press release 1 IP/19/429, 17 July 2019, available at https://ec.europa.eu/commission/presscorner/detail/en/IP_19_4291.
17 Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation), available at https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32016R0679&from=EN.
18 German competition authority case summary available here https://www.bundeskartellamt.de/SharedDocs/Entscheidung/EN/Fallberichte/Missbrauchsaufsicht/2019/B6-22-16.pdf?__blob=publicationFile&v=4.
19 Facebook v. Bundeskartellamt, VI-Kart 1/19 (V), available at https://www.olg-duesseldorf.nrw.de/behoerde/presse/archiv/Pressemitteilungen_aus_2019/20190826_PM_Facebook/20190826-Beschluss-VI-Kart-1-19-_V_.pdf.
20 German Supreme Court press release available at https://www.bundesgerichtshof.de/SharedDocs/Pressemitteilungen/DE/2020/2020080.html; the ruling is available at https://www.mlex.com/Attachments/2020-08-27_6FK3QP0LX76X258S/BGH%20Facebook%20kvr__69-19.pdf .
21 Cunningham, Colleen, F. Ederer & S. Ma, “Killer Acquisitions” (April 19, 2020), available at https://poseidon01.ssrn.com/delivery.php?ID=295096073027113074076094112099117107055064063038093061070015090027073100081000116023041124051002057017044101124073080026096029017010070089022094092093126028087072035047004004099026015001093098031030020084025102096115105031001105107007077024067064096&EXT=pdf.
22 Facebook/WhatsApp, Case M.7217, available at https://ec.europa.eu/competition/mergers/cases/decisions/m7217_20141003_20310_3962132_EN.pdf.
23 Available at https://ec.europa.eu/info/law/better-regulation/have-your-say/initiatives/12417-Digital-Services-Act-deepening-the-Internal-Market-and-clarifying-responsibilities-for-digital-services.
24 Regulation (EU) 2019/1150 of the European Parliament and of the Council of 20 June 2019 on promoting fairness and transparency for business users of online intermediation services, available at https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32019R1150&from=EN.
25 The concept of “gatekeeper” originates in the work of social psychologist Kurt Lewin and has been used in different contexts in various fields (information science, communications, management). Within the broader regulatory context, Emily Laidlaw identifies “gatekeepers” as “non-state actors with the capacity to alter the behaviour of others in circumstances where the state has limited capacity to do the same.” See Emily Laidlaw, “A framework for identifying Internet information gatekeepers” (2010) 24(3) International Review of Law, Computers & Technology 263, 265. The term is, however, not defined in the Inception Impact Assessment or in any of the other policy documents mentioned above that use the term.
26 Available at https://www.bmwi.de/Redaktion/DE/Downloads/G/gwb-digitalisierungsgesetz-referentenentwurf.pdf?__blob=publicationFile&v=10.
27 See the French authority’s relevant press release here: https://www.autoritedelaconcurrence.fr/en/press-release/autorite-creates-digital-economy-unit.
28 Pablo Ibáñez Colomo, “Self-Preferencing: Yet Another Epithet in Need of Limiting Principles,” available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3654083.
30 See the Commission’s Inception Impact Assessment, available at https://ec.europa.eu/info/law/better-regulation/have-your-say/initiatives/12416-New-competition-tool.
31 A recent example of the use of this tool is the opening by the European Commission of an inquiry into the sector of consumer Internet of Things, press release IP/20/1326, July 16, 2020, available at https://ec.europa.eu/commission/presscorner/detail/en/ip_20_1326.
32 Google Android, Case AT.40099, available at https://ec.europa.eu/competition/antitrust/cases/dec_docs/40099/40099_9993_3.pdf. On appeal: Case T-604/18, Google and Alphabet v. Commission, case not yet decided.
33 Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, available at https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32003R0001&from=EN.
34 “Antitrust: Commission imposes interim measures on Broadcom in TV and modem chipset markets,” October 16, 2019, press release available at https://ec.europa.eu/commission/presscorner/detail/en/IP_19_6109.
35 “Statement by Commissioner Vestager on Commission decision to impose interim measures on Broadcom in TV and modem chipset markets,” available at https://ec.europa.eu/commission/presscorner/detail/en/STATEMENT_19_6115.
36 “Réforme du droit des concentrations et contrôle ex post,” available at https://www.autoritedelaconcurrence.fr/sites/default/files/note_controle_expost.pdf.
37 “Online platforms’ mergers should face review even if years have passed, De Silva says,” MLex (July 14, 2020), available at https://www.mlex.com/GlobalAntitrust/DetailView.aspx?cid=1207836&siteid=190&rdir=1.
39 Commissioner Vestager Speech at the International Bar Association 24th Annual Competition Conference, September 11, 2020, “The future of EU merger control,” available at https://ec.europa.eu/commission/commissioners/2019-2024/vestager/announcements/future-eu-merger-control_en.