digitisation is making it harder to keep competition working the way that it should …. There can be a constant risk that by abusing their power, big companies will push a market to the tipping point, where competition is gone forever. That the spring will be stretched so far that it no longer snaps back – that just ordering companies to stop their harmful behaviour is no longer enough, to bring competition back to these markets.Footnote 1
In her mission letter from December 2019, European Commission President von der Leyen gave Executive Vice-President Vestager the task to make sure that “competition policy and rules are fit for the modern economy” as well as to “strengthening competition enforcement in all sectors.”Footnote 2 The European Commission then identified situations where the existing legal framework would not be sufficient to deal with this task:
The enforcement experience of the Commission and national competition authorities, and the reflection process on the fitness of the existing competition rules have helped identify certain structural competition problems that these rules cannot tackle (e.g. monopolisation strategies by non-dominant companies with market power) or cannot address in the most effective manner (e.g. parallel leveraging strategies by dominant companies into multiple adjacent markets).Footnote 3
As a result, Vice-President Vestager proposed to introduce market investigations as a “New Competition Tool” (NCT) aimed at complementing existing legal instrumentsFootnote 4 and whose goal and scope was summarized thus:
This new tool would let us deal with structural problems with competition, when we can’t do that with the powers that we already have – or when those existing powers would be much less effective. It would let us investigate markets, in the same rigorous way that we already look into individual cases …. And if we did find problems, we could intervene. Not by fining companies, or finding them guilty of breaking the rules – but by imposing obligations that would protect competition. That could involve duties to behave in a certain way …. As a last resort, it could even mean breaking up companies, to protect competition.Footnote 5
What are the competition problems that existing competition tools fail to catch, which lead to harm to consumers and society? Among others, it is about firm behavior in a gray zone of competition law; that is, firm behavior that harms consumers but is not explicitly prohibited under competition law. One such example is tacit collusion. Again, quoting from the Vestager speech,
it would mean we could deal with concentrated markets where companies keep their prices in line, not by agreeing to form a cartel, but just by taking care not to undercut each other. That tacit collusion can be especially easy in a transparent digital world like ours, where companies can constantly watch each other’s prices, and change their own very quickly in response. When there’s no agreement, that’s not illegal under the rules as they are now – even though it means that consumers end up paying more than they should. But with this new tool, we could tackle this problem.Footnote 6
There are markets with lack of competition because of institutional features or industry-wide business practices in which interventions may lead to more competitive outcomes to the benefit of consumers and society. While interventions to rule out certain opaque business practices may be considered to be the realm of consumer protection policy, some of these practices harm consumers because of the associated market power they give to firms. If none of these firms is deemed to have significant market power, existing competition law will find it difficult to deal with their practices. What is more, some institutional features are largely outside the control of individual firms; interventions that remove obstacles to competition could then be part of the remedies of a market investigation.A more recent concern is the winner-takes-all feature of some markets and the tendency of tipping with one firm obtaining an entrenched position – that is, that it becomes difficult to be replaced. This mostly applies to “digital” markets when economies of scale and scope or network effects are pronounced. As Vestager mentions in her speech:
We could also deal better with markets that are prone to tip. We could move a lot faster to prevent this from happening, by stepping in as soon as a company starts to misuse its power to drive out competition – even if it isn’t yet big enough to have broken the competition rules as they stand. And with just one investigation, we could deal with a situation where a gatekeeper platform is using its power to drive out competition in a whole series of markets.Footnote 7
In light of the challenges in some markets called digital, one possible path was to restrict market investigations to such markets. In her speech, Vestager discussed this possibility:
Many of the biggest issues that this tool could help us resolve are linked to digital markets. But I doubt that it would make sense to apply it only to these markets – instead of covering the whole economy, as our existing competition powers do.Footnote 8
Vestager gives two reasons not to follow such a narrow approach. The first is conceptual: “The sort of issues I’ve discussed come up in many other markets as well. In fact, the Greek, Icelandic and British competition authorities have so far only used this type of power in markets that aren’t digital.” The other reason is that it is difficult to draw a meaningful dividing line between “digital” and “non-digital.” As Vestager pointed out, “the digital transition is affecting pretty much every industry there is. So it’s hard to draw the line between what’s digital and what isn’t – especially when you consider that the rules we come up with now should be ready for the future, when that line may get even more blurred.”Footnote 9
As we put this book together, looking across the Channel, a recent market investigation in the UK is about funeral services.Footnote 10 Do they belong to the “digital” market in times of the COVID-19 pandemic?
Thus, the stage had been well prepared. A new competition tool for Europe: market investigations that allow the European Commission to address structural competition problems in an effective and efficient way. Why do we need a book on this new competition tool even before the tool has been implemented? Well, one way of reading the course of events is that legislative efforts have come to a halt. Another reading is that they have been redirected in a very specific way. Either way, market investigations as a horizontal competition tool appear not to be politically viable at the moment,Footnote 11 with recent legislative proposals aiming at limiting the use of such a tool exclusively to the digital industry (however defined).
However, we do believe that market investigations merit a full academic discussion as a new horizontal competition tool. Our aim is to revive this discussion.Footnote 12
A core reason for the failure to introduce a general market investigation regime at the European level has apparently been the difficulty to find a workable legal basis for its introduction – a question directly linked to the relation between the NCT and EU competition law. Article 103 TFEU empowers the Council to pass appropriate regulations or directives giving effect to the principles set out in Articles 101 and 102 TFEU, on a proposal from the Commission and after consultations with the European Parliament. A European market investigation regime would, however, reach beyond Articles 101 and 102 TFEU. The fact that market investigations are not broadly used in the Member States so far stands in the way of using Article 114 TFEU as a legal basis for a European market investigation regime. The recent debates and studies on the NCT would appear to back up an argument that a European market investigation can be considered a necessary instrument to effectively ensure undistorted competition in the internal market, such that Article 352 TFEU would provide a viable legal basis. But Article 352 TFEU would require the Council to act unanimously.
Strong arguments are required, therefore, to complement the existing competition rules by a novel market investigation regime. The economic considerations are summarized below and further explored in Chapter 2 of this volume. On the legal side, the interaction between the long-established legal framework and a novel empowerment of the Commission to intervene ex ante and on a significantly broader basis must be explored. So far, Articles 101 and 102 TFEU have not only served as a basis for the Commission, national competition authorities, and courts to intervene to protect competition, but they have likewise defined the limits of intervention. A European market investigation instrument would provide the Commission with a much wider discretion to meddle in the functioning of markets and to engage in pro-active market design. Ongoing debates about an increased politization of competition policy – be it with a view to industrial policy or to sustainability goals – illustrate the risks associated with such an empowerment.
Furthermore, the interaction between a European market investigation regime and Articles 101 and 102 TFEU needs further exploration. This is so, in particular, because market investigations might have some overlaps with the existing competition rules, rather than purely complementing them. Apart from addressing certain gaps of competition law, this new tool should enable the Commission to engage in a more holistic analysis of possible market failures and to more effectively address certain market features that tend to produce anti-competitive outcomes in the longer run but will typically go beyond the reach of remedial action in infringement proceedings: Article 7 Regulation 1/2003 remedies are focused on putting an end to an infringement, not on addressing the features of the market that hamper competition. A future market investigation regime, on the other hand, would be justified by its ambition to preserve or restore the possibility for competition in reaction to the special features of the market identified that may hamper competition. Given the vaster remedial powers of the Commission in the context of a market investigation regime, concerns could arise that it would tend to displace traditional competition law enforcement and induce the Commission to engage in potentially far-reaching “market engineering,” bypassing some of the procedural and judicial checks that come with Articles 101 and 102 TFEU.
Where a new market investigation instrument would come with pros and cons from a legal and institutional perspective, much turns on an economic analysis: Which forms of malfunctioning of competition are currently not, or not effectively, addressed by competition law rules, and how could market investigations contribute to closing these gaps?
In some sectors, market mechanisms do not function as well as they should, for at least two reasons. First, they may be characterized by market features that are not necessarily caused by the firms’ behavior (although they may be reinforced by it) – such as scale or scope economies, (direct or indirect) network effects, switching costs and lock-in effects, asymmetric information, and behavioral biases by consumers. Second, they may be adversely affected by the conduct of the firms themselves – such as (tacit or explicit) collusion, other (horizontal or vertical) agreements, contractual clauses imposed on consumers, and business practices that may be deemed abusive. Note also that these two reasons may coexist within the same sector.
“Traditional” competition law tools in the EU would not allow to restore effective competition in markets that do not function properly due to the market features mentioned above, because such features are not, or not entirely, the product of firms’ actions. Further, competition law provisions might not allow to take care of firms’ conduct that may nonetheless have anti-competitive effects. For instance, competition law does not preclude firms from tacitly colluding (and for good reasons); and as things stand, there are few obstacles for rivals to have minority shareholding in each other, or to have common owners, although this may dampen market competition and/or possibly promote collusive outcomes. Arguably, there may also exist business practices that may be anti-competitive but for which the intervention threshold is for various reasons very high or the timing of intervention very long, thereby making it difficult to use competition law in a timely and effective way. This implies that only a subset of the potentially anti-competitive conduct of firms may be covered by the competition law provisions contained in Articles 101 and 102 TFEU and the Merger Regulation.
For these reasons, an economic argument can be made that it would be desirable to integrate the existing competition tools with a new instrument consisting of market investigations, which may help promote effective competition in situations where markets do not work properly.
Furthermore, there may exist markets that are not currently experiencing problems but that for different reasons (whether conduct by the incumbent or other market features, or both) may be at risk in the near future. In such cases, it is conceivable that market investigations might provide a preventive tool of intervention that is currently not available under EU competition law.
A market investigation should identify the mechanisms that lock competition in the market and hence the interventions that should possibly neutralize those mechanisms and unlock competition. Whatever the theory of harm that may justify such an investigation, in order to address consumer harm in a meaningful way, the European Commission must have the power to implement suitable remedies, including structural ones.
This does not mean, however, that market investigations are the panacea of markets or that they should replace the central role of Articles 101 and 102. Markets are complex, and different market features and firms’ conduct interact to determine market outcomes in ways that are not always easy to foresee. In other words, considerable uncertainty may exist about the impact of a concrete intervention. In particular, while some remedies have virtually no likely “side-effects” and hence could be imposed relatively safely, others may have adverse consequences, not only on the firms at issue but also on consumers – and hence trade-offs should be carefully considered before remedies are imposed.
More generally, a market investigation will employ the scarce resources of an antitrust authority and may have significant costs (for instance, in terms of uncertainty) for the firms operating in the sector. A necessary condition for opening a market investigation should therefore be that the potential consumer harm is sufficiently large (due to long-run effects or to immediate serious harm), and one could foresee that there may be feasible and appropriate remedies. Furthermore, in each specific case, the European Commission will also have to consider whether a market investigation is the preferred instrument to solve the problem and is superior to traditional competition law tools such as Articles 101 or 102 or even sectoral regulation (to the extent that they are applicable).
The Commission’s consultation on the NCT revealed some support in favor of the introduction of a general, cross-sectoral market investigation regime “UK style.” However, with its proposed legislative package of December 15, 2020, the Commission has opted for a narrow scope of the market investigation. The proposed Digital Markets Act (DMA), which has a quasi-regulatory approach and comprises a list of “dos” and “don’ts” for the digital platforms with “gatekeeping power,” foresees the use of a market investigation tool in various respects. But its scope is tied to the scope of the DMA itself: market investigations shall be used as an instrument to make the DMA future proof, namely, to expand the list of “core platform services” as set out in Article 2(2) of the draft DMA; to designate gatekeepers that meet the requirements of Article 3(1) of the draft DMA, but do not satisfy the quantitative thresholds of Article 3(2) of the draft DMA; or to update the list of obligations for gatekeepers as laid down in Articles 5 and 6 of the draft DMA (see Article 10 DMA).
The draft DMA is conceived as an ex ante regulation of the provision of core platform services offered by entities that qualify as “gatekeepers” within the meaning of Article 3 of the draft DMA. “Core platform services,” as defined in Article 2(2) of the draft DMA, shall encompass online intermediation services, online search engines, online social networking services, video-sharing platform services, messenger services, operating systems, cloud computing services, and advertising services provided by a provider of any of the prementioned core platform services. A provider of such services shall be designated as a “gatekeeper” if it has a significant impact on the internal market, serves as an important gateway for business users to reach end users, and enjoys an entrenched and durable position in its operations or will foreseeably enjoy such a position in the near future (Article 3(1) of the draft DMA).
These preconditions shall be presumed to be satisfied if the core platform services are provided in at least three Member States and certain quantitative thresholds are met, namely, if the undertaking to which the provider of core platform services belongs achieves an annual EEA turnover of no less than 6.5 billion Euros or where the average market capitalization or fair market value amounted to at least 65 billion in the last financial year, if the core platform service has more than 45 million monthly active end users established or is located in the Union and has more than 10,000 yearly active business users established in the Union in the last financial year, and if these thresholds were met in each of the last three financial years (Article 3(2) of the draft DMA). According to Commissioner Thierry Breton, an estimated 10–15 platforms may meet these criteria.
But a provider of core platform services may qualify as a “gatekeeper” even if it fails these thresholds, provided that it meets the criteria set out in Article 3(1) of the draft DMA (see Article 3(6) of the draft DMA). In order to designate a gatekeeper in such cases, the Commission must conduct a “market investigation.” Relevant criteria for the gatekeeper qualification include the size of the platform services provider; the number of business users depending on the core platform service to reach end users and the number of end users; the entry barriers deriving from network effects and data driven advantages, economies of scale and scope, including, with regard to data, a business user or end user lock-in; and other structural market characteristics.
In the provision of core platform services that serve as an important gateway to business users to reach end users, all gatekeepers will then be required to comply with the rules of conduct listed in Articles 5 and 6, where Article 5 contains those obligations that the Commission considers to be self-explanatory, and Article 6 lists those obligations that may need further specification. In essence, the obligations set out in Articles 5 and 6 of the draft DMA are a collection of the remedies that have been imposed upon dominant digital platforms in various competition law proceedings at the EU or at the Member States level over the last couple of years or are currently being considered for imposition. Following a “one size fits all” approach, this code of conduct shall apply to the provision of all core platform services that serve as an important gateway for business users to reach end users and are provided by designated gatekeepers, and both Article 5 and Article 6 are designed as per se rules: with one exception – namely, designated gatekeepers that do not yet enjoy an entrenched and durable position but will foreseeably do so in the near future (see Article 15(4)) – the draft DMA does not foresee a tailoring of the rules of conduct to the individual gatekeepers, the possibility of an objective justification, or an efficiency defense. According to Article 10 of the draft DMA, the Commission shall be empowered to adopt delegated acts to update the lists of obligations where, based on a market investigation, it has identified a need for doing so to address practices that limit the contestability of core platform services or that are “unfair” in the same way as the practices currently addressed in Article 5 and 6.
The envisioned market investigation procedure is further specified in Chapter IV of the draft DMA. Any market investigation shall start with an opening decision that, inter alia, specifies the purpose of the investigation. The draft DMA distinguishes between a market investigation for designating gatekeepers (Article 15), market investigations into new services and new practices (Article 17), and market investigations into systematic non-compliance (Article 16). Different time lines shall apply in the different settings – for example, a market investigation for designating gatekeepers that do not meet the thresholds set out in Article 3(2) shall be concluded within twelve months from the opening decision, with preliminary findings being communicated within six months, whereas a market investigation following an alleged gatekeeper’s claim that it does not fulfill the requirements of Article 3(1) despite meeting the Article 3(2) thresholds shall be concluded within five months from the opening decision, with preliminary findings being communicated within three months (see Article 15). A market investigation into new services potentially to 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 are not effectively addressed by the DMA shall result into the issuing of a public report within 24 months from the opening decision. Where appropriate, the report shall be accompanied by a proposal to amend the DMA to include additional core platform services or by a delegated act amending the rules of conduct in Articles 5 and 6 DMA (see Article 16). A market investigation into a gatekeeper’s systematic non-compliance with the obligations set out in Articles 5 and 6 shall be concluded within twelve months from the opening decision, with a statement of objections being communicated to the gatekeeper concerned within six months, but with a possibility for extension of no more than six months. The non-compliance market investigation shall establish whether a designated gatekeeper has systematically infringed the rules laid down in Articles 5 and 6 and has further strengthened or extended its gatekeeper position thereby. If this is so, the Commission can impose behavioral or structural remedies that are proportionate to the infringements and necessary to ensure compliance with the DMA. A systematic infringement is presumed where at least three non-compliance or fining decisions have been issued within the last five years (see Article 16).
As this brief overview shows, market investigations of some sort are an important element of the proposed DMA: it partially adopts the core idea of the market investigation instrument to identify structural features of the market that create particular risks to competition and to react to such risks by imposing rules of conduct, regardless of an initial finding of a position of dominance and an infringement of Article 102 TFEU. But the DMA market investigations are very much part of the regulatory regime that the DMA proposes to establish in the first place, rather than a flexible instrument to identify and remedy competition problems that are currently not effectively addressed by EU competition law. And they are of course limited to the digital sector only.
Given these limitations of the market investigation mechanism in the Digital Markets Act, the debate whether such an instrument will be needed to complement EU competition law in the future remains open. The following chapters provide context and shed light on important issues in this debate.
Chapter 2, by Massimo Motta and Martin Peitz, assesses the possible role of a market investigation tool endowed with broad remedies, when a market suffers from competition problems and infringement cases under 101 TFEU and 102 TFEU would be infeasible or ineffective. The chapter lays out a number of theories of harm, that is, reasons why certain market features or behavior by market participants may lead to consumer harm compared to a relevant counterfactual. The report identifies theories of harm (i) in markets in which none of the firms is dominant and (ii) in markets with a dominant firm but Article 102 TFEU is not effective or applicable or there may be a dominant firm in the future. It also argues that the European Commission should look for simple measures as “intervention triggers” for a market investigation and identifies some possible triggers. While some of the identified harms are more likely or more pronounced in digital markets, the authors argue that a presumption that narrowing the tool to primarily address competition problems in digital markets would be misguided. Finally, when sector regulation is, in principle, applicable, the authors see the market investigation tool as filling a gap between standard competition tools and sector regulation.
Chapter 3, by Heike Schweitzer, discusses the institutional set-up and procedural design of a possible future European market investigation regime. The institutional framework and procedural rules must be tailored to promote the core goals of the new instrument, namely, to address competition problems that do not primarily follow from conduct but from “features of the market,” such that a European market investigation – contrary to traditional infringement proceedings – will not be of a quasi-criminal nature but a purely administrative proceedings and will allow for a particular timely intervention. Against this background, the following questions are raised: How does the market investigation regime interact with Article 101 and 102 TFEU enforcement and sector inquiries at the EU level and at the national level? How can the procedure be structured such as to allow for a timely and effective intervention? What can be done to allow for a less adversarial and more participative interaction between the Commission and market actors? Furthermore, the remedial regime, voluntary commitments, the possibility for interim measures, and judicial review are discussed.
Chapter 4, by Pierre Larouche and Alexandre de Streel, reviews the relationship among competition law and the various sectoral regulatory regimes making up EU economic law and, on that basis, make recommendations regarding how a wide and a narrow version of a market investigation could be integrated in EU economic law. The authors consider the integration of the broad market investigation to be easy. At the systemic level, this broad version could usefully close regulatory gaps. At the substantive level, it should rest on economic analysis, yet without being straitjacketed within specific competition law analysis. At the institutional level, a close transversal cooperation between the authority in charge of the market investigation and the relevant NRA(s) is deemed necessary. However, with the European Commission’s draft DMA, a narrow version of the market investigation is on the table that could also easily be integrated in EU economic law. At the systemic level, the DMA complements competition law where practice has shown that competition law is ineffective in solving competitive problems. At the substantive level, the standards, the criteria, and the indicators used to implement the three types of DMA market investigations are not straightjacketed within competition methodologies but should be applied and interpreted with sound economic analysis. At the institutional level, the Commission would acquire concurrent regulatory and competition powers and it should explain the criteria it will use to choose between those different powers when addressing the conducts of the digital gatekeepers. Moreover, given the possible parallel application of the DMA by the Commission and national competition law by the NCAs, Larouche and de Streel consider it key that the cooperation between the Commission and the NCAs is ensured.
Chapter 5, by Richard Whish, provides an account of the market investigation provisions on the UK Enterprise Act of 2002, which enables the Competition and Markets Authority (CMA) to investigate markets and to determine whether any “features” of a market prevent, restrict, or distort competition. If the CMA discovers “adverse effects on competition,” powers are available to achieve as “comprehensive a solution as is reasonable and practicable” through the imposition of remedies, up to and including mandatory divestiture. This chapter describes these powers, explains the institutional regime within which decisions are made, and the procedure that the CMA follows in market investigation cases. Some other jurisdictions possess similar powers to those contained in the Enterprise Act. In particular, the chapter describes the powers available to the competition authorities in Greece, Iceland, Mexico, and South Africa.
Chapter 6, by Tembinkosi Bonakele, Reena das Nair, and Simon Roberts, assesses the South African record with market inquiries, which are the equivalent of market investigations in other jurisdictions. The authors find that market inquiries have been an important tool for the Competition Commission of South Africa to address competition problems reflecting entrenched positions of firms with market power in key sectors. Inquiries have been completed into banking, private healthcare, liquefied petroleum gas, grocery retail, and data services, with inquires in passenger transport and online intermediation platforms underway. In a comparative review of the inquiries, the authors find that the inquisitorial process has been very valuable in identifying and remedying competition issues more expeditiously than the adversarial process in enforcement investigations. This has been the case where independent panels were used to conduct the inquiries as well as where the inquiries were undertaken by the Commission’s own team. Issues relating to competition policy questions such as barriers to entry and the ability of smaller firms to compete have also been canvassed in inquiries. Recommendations on these “competition-plus” issues have tended to require improved regulation and/or related policy measures, for which inquiries have played an important agenda-setting role. Inquiries have also been used to address wider questions of public policy, often at the request of government, and here, while important data and analysis have been brought into the public domain, the impacts are less clear.
Chapter 7, by Gregory Crawford, Patrick Rey, and Monika Schnitzer, provides an assessment of the economic merits of the NCT, as it was considered in Summer 2020 by the European Commission as a tool to address structural competition problems in a timely and effective manner. The NCT has strong analogies to the UK’s “markets regime,” which empowers the UK competition regulator, the CMA, to initiate market studies and investigations. The authors review the UK’s markets regime and survey some of the competition concerns the regime is intended to address. This includes a selective review of UK market studies and investigations to illustrate some of the ways these concerns have been explored and a description of the remedies imposed or proposed (in the case of market studies or ongoing investigations). The authors conclude with a critical evaluation of the functioning of the UK’s markets regime in light of this evidence and offer seven recommendations regarding the merits and design of market investigations as an NCT for the EU. In sum, they see a strong case for the introduction of an NCT to address factors that prevent effective competition in markets, and they see no benefit to limiting it to specific sectors and to dominant firms.
Chapter 8, by Amelia Fletcher, considers the pros and cons of market investigations in the context of the UK regime. The author sees them as a valuable addition to the standard competition law toolkit and concludes that this would likely be true also at EU level, both when applied to digital platforms and more widely. Since the tool is potentially so powerful and flexible, the author argues in favor of strong procedural checks and balances. These should guard against confirmation bias and politicization. While powerful and flexible, the tool also has important limitations and thus, as the author argues, should not be viewed as a full solution to the issues raised by digital platforms but rather as a valuable complementary tool alongside new ex ante regulation. Interoperability is discussed as one example where the tools could valuably be used alongside each other.