Gatekeeper Regulation in the Digital Economy – The Pitfalls (and Opportunities) Ahead?

Intense regulatory scrutiny on the digital economy and “Big Tech” companies has led to a raft of expert reports and regulatory proposals across the world relating to the digital economy. Regulators have increasingly focused on identifying digital businesses who play a “gatekeeper” role as the basis of regulatory intervention. However, the exact delineations of how such a “gatekeeper” designation should be determined remain amorphous. There are complex questions where further debate and consideration should continue, in particular: how should a “gatekeeper” designation be formulated, and is a “blacklist” or principles-based approach to restricted conduct more suitable in the dynamic digital economy? Beyond the form such regulation takes, what types of “gatekeeper” conduct should be regulated and how will this impact longer-term competition and innovation incentives?

By Rod Carlton, Rikki Haria & Caroline Chew¹

I. THE (RE-)FORMULATION OF THE CONCEPT OF “GATEKEEPERS”

With the digital economy and “Big Tech” companies under intense regulatory scrutiny across the globe, regulators have increasingly coalesced around identifying digital businesses who play a “gatekeeper” role as the basis of regulatory intervention. Many regulators have focused on identifying as “gatekeepers” those digital platforms who play a “dual” role in the market, in that they provide a crucial upstream service while at the same time competing with other third parties in a downstream market (or otherwise have reached a certain size that third parties are allegedly ‘reliant’ on them to access customers in other markets). For example, the European Commission (“EC”) has imposed over €8 billion in fines on Google in relation to its Shopping, Android and AdSense cases² and has opened investigations into Apple³ and Amazon.⁴

While stakeholders are in broad agreement that digital innovation has provided significant benefits to consumers and businesses,⁵ vociferous public and political pressures have resulted in a particular urgency on the regulators’ part to test and determine the remit of the gatekeeper concept, as well as any potential avenues for heightened regulation and enforcement. Against this backdrop, a raft of reports have been published which deal with the question of how best to regulate the digital economy. Prominent examples include: the UK Competition and Markets Authority’s (“CMA’s”) Final Report in its market study on Online Platforms and Digital Advertising,⁶ the U.S. House Judiciary Subcommittee’s report on its Investigation of Competition in Digital Markets,⁷ the report commissioned by the EC on “Competition Policy for the Digital Era,”⁸ and the Australian Competition & Consumer Commission’s Final Report in its Digital Platforms Inquiry.⁹

From a practical perspective, a number of significant proposals are being rapidly developed in the near-term, including:¹⁰

• The United Kingdom’s proposed Digital Markets Unit (“DMU”): the UK set up a Digital Markets Taskforce (“DMT”), which published its advice to the UK Government in December 2020 on digital regulation (the “DMU Proposal”). A key part of the DMT’s recommendations revolves around the designation of firms with “Strategic Market Status” (“SMS”), defined as firms with “substantial, entrenched market power in a particular digital activity” which provides the firm with “a strategic position,” meaning the “effects of its market power are likely to be particularly widespread or significant.”¹¹ As well as proposing a separate merger control regime for SMS firms, firms’ activities which are deemed to have SMS status will be subject to an enforceable, principles-based code of conduct,¹² as well as possible “pro-competitive interventions” (such as data mobility and interoperability).¹³
• The EC’s proposed Digital Markets Act (“DMA”): in December 2020, the EC published its legislative proposals imposing ex ante regulation to regulate “gatekeeper” platforms in order to promote the contestability of digital markets (the “DMA Proposal”).¹⁴ The DMA Proposal applies to platforms which, on a cumulative basis, provide “core platform services,” have “a significant impact on the internal market,” “serve an important gateway for business users to reach their customers” and “which enjoy, or will foreseeably enjoy, an entrenched and durable position.”¹⁵ There are quantitative thresholds for each of these criteria; if met, there is a rebuttable presumption that the company is a “gatekeeper.” In addition, the EC is able to conduct a market investigation to determine, on a qualitative case-by-case assessment, whether companies should be deemed to be gatekeepers or otherwise may constitute “emerging gatekeepers.”¹⁶ In contrast to the DMU Proposal, which suggests keeping the burden of proving SMS status on the DMU, the DMA Proposal goes further in that it reverses the burden of proof to the benefit of the EC. There is an obligation on companies themselves proactively to self-assess whether they meet the quantitative thresholds and provide such information to the EC. It sets out prohibited practices to tackle unfair practices by gatekeeper platforms, with a list of prohibited conduct (i.e. a “blacklist”).¹⁷ Platforms will be subject to sanctions in the event of non-compliance.

However, on either approach, the exact delineations of how such a designation will be determined remain amorphous. Rather than providing a view on the hotly debated topic of whether ex ante regulation is actually needed or desirable in relation to large digital businesses, this article seeks to explore three issues that go to the fundamentals underpinning any potential regulation of selected digital businesses, namely:

• how a “gatekeeper” designation should be formulated, particularly in light of the wide-ranging ramifications of being subject to ex ante regulation;
• whether the approach of having a “blacklist” of restricted conduct, or conversely a principles-based approach, for “gatekeepers” (however defined) is suitable in the digital economy; and
• what types of “gatekeeper” conduct should be regulated.

 

II. HOW TO IDENTIFY AND DESIGNATE “GATEKEEPERS”

The well-established EU regulatory framework for telecommunications is a case study in how to designate a “gatekeeper.” Under this framework, national regulators are tasked with identifying the markets which contain characteristics justifying the imposition of regulation, identifying whether there are any undertakings that should be subject to such regulation (which is based on whether any such undertaking has “significant market power” or “SMP”), and then determining what regulatory measures are appropriate and proportionate on a case-by-case basis. National regulators may consider that the imposition of regulatory remedies on a relevant national market is defensible if (upon investigation):

• there are high and non-transitory structural, legal or regulatory barriers to entry;
• there is a market structure which does not tend towards effective competition within the relevant time horizon (keeping in mind the state of infrastructure-based competition and other sources of competition behind the barriers to entry); and
• competition law alone is insufficient to adequately address the identified market failure.

The telecommunications sector has a number of features in common with the digital economy, such as network effects and economies of scale, which suggest that certain lessons can be drawn from the existing regulatory framework for telecommunications, particularly in terms of process and procedure (as noted below).

However, key differentiating features remain that make the use of regulation more challenging in relation to large digital businesses. For example, large telecommunications players in the EU typically operate critical (and non-replicable) infrastructure, are vertically integrated and hold substantial and often enduring market power as a result of their former position as State monopolies. These factors underpin the case for a telecommunications regulatory framework, but they do not necessarily read across to the digital economy. In addition, the EU telecommunications sector remains characterized by national markets and predominantly national players.

In contrast, large digital businesses are typically active across the EU (and globally), with the markets they operate in typically being at least EU-wide in scope. This means that there is a greater risk, as compared to the telecommunications sector, that differing approaches at a national level (e.g. between EU member states) will hinder the ability of digital products to function seamlessly and consistently across national borders and, in practice, require digital businesses to abide with the most stringent national regulation with respect to their activities across the EU (and possibly even globally). There is therefore a greater need for transnational coherence and cooperation to enable effective regulation, to avoid digital businesses facing materially different regulatory obligations in different countries. If that situation were to arise, any local consumer benefits specific to each national regime risk being outweighed by EU (or global) consumer detriment, potentially hindering innovation on a global scale and resulting in adverse outcomes for consumers in certain countries (e.g. the offer of more limited digital services or not rolling out new features in certain countries). In addition, the digital sector is fast-paced and dynamic, making regulation inherently more difficult to implement and “future-proof” in an effective way.

A. The need for a robust analysis of digital markets and services to devise clear, objective criteria for gatekeepers

Given this context, it is crucial that regulators carry out rigorous analysis as to what types of digital markets or services should be the points of regulatory focus. In particular:

• When considering which digital markets and services should be captured by a gatekeeper regulation, regulators should go further than identifying certain market factors which might conceptually lead to gatekeeper-type situations, such as the presence of network effects, economics of scale and scope or the ability to engage in self-preferencing behavior. It is crucial for regulators to analyze which markets or services are in practice producing adverse outcomes for customers. In light of the extensive benefits that digital services have provided for consumers, businesses and wider society (highlighted most recently in the ongoing COVID-19 pandemic), any assessment must balance robustly evidenced (rather than speculative) consumer harm against such benefits.
• Moreover, given any gatekeeper regulation would cover a variety of different markets (many of which are interconnected), it is important that the designation process does not occur in a way that provides certain businesses (or business models) with an advantage over others. This would distort competition. By way of example, the UK’s recent DMT advice suggests that the DMU should initially prioritize firms active in “online marketplaces, app stores, social networks, web browsers, online search engines, operating systems and cloud computing services.” However, little explanation is provided around: (i) the reasons why these sectors should be prioritized over other digital markets; and (ii) whether the designation process would be undertaken in parallel across interconnected digital markets to avoid certain businesses being placed at a regulatory disadvantage due to an earlier designation.

In the interests of legal certainty and fairness, clearly articulated objective criteria are needed to determine which firms should be subject to regulation. The current proposals differ in this respect:

• For example, the EC’s DMA Proposal includes quantitative criteria in relation to designating gatekeepers. While this seeks to draw a bright line between companies that fall within the scope of the DMA Proposal and those that do not, this largely binary approach risks inadvertently distorting competition among competitors who hold a fairly similar market position.
• The DMU Proposal takes a different approach and provides much greater discretion to the DMU. When explaining the SMS concept, the UK’s DMT notes that a firm’s market power must furnish the firm “with a strategic position,” in which its effects are “likely to be particularly widespread and / or significant.” However, the relevant factors – for example, a “very high proportion of the population,” the facilitation of “large” values of transactions and/or the ability to “extend market power from one activity into a range of other activities” – are not particularly clear and make it difficult to identify which digital markets and firms should be considered to have SMS.

Given that gatekeeper designation-based regulatory interventions being proposed globally are likely to have far-ranging implications for a significant number of digital players, it remains important for regulators to formulate clear, objective criteria which go beyond simply identifying that businesses are large or that many customers have chosen to use a particular business. These criteria should be determined as a result of careful, evidence-based analysis, avoiding the trap of being crafted to target certain business models. Failing to produce such objective criteria will not only result in widespread legal uncertainty but will risk undermining innovation incentives and competition in the long-term.

B. Procedural fairness is key – defensible, logical processes with appropriate rights of appeal

In addition to setting out clear, objective legal tests for the designation of gatekeeper status, it is also crucial that any designation process runs on a statutory timeline, building in opportunities for a firm to respond to the objections of any regulator to its practices. For example, under the EU telecommunications regulatory framework, national regulators are only able to impose an ex ante designation of a firm as having SMP following an investigation into each relevant market. Similar mechanisms should function in a gatekeeper designation process, with the burden of proof lying with the regulator. As part of this process, firms should be allowed to propose remedies and solutions to any regulatory concerns raised in a collaborative manner. From this perspective, it is encouraging that both the EC’s DMA Proposal¹⁸ and UK’s DMU Proposal¹⁹ recognize the need for the timely review of the designation of gatekeepers, with the opportunity for companies to be heard as part of the process. This would assist in assessing whether the gatekeeper designation and associated activities continue to be aligned with rapid developments characteristic of digital markets.

Following a regulator’s decision, firms should have recourse to robust appeal procedures against a designation as a gatekeeper and/or the additional regulatory burdens that flow from such designation. This is in line with both the EU telecommunications regulatory framework²⁰ and the CMA’s powers to make a market investigation reference,²¹ which both contain a judicial review mechanism.

Beyond the designation process itself, regulators should also commit to a process of reviewing their decisions and assessments on a periodic basis. Market reviews in the telecommunications space are usually taken periodically every five years, with only the undertakings which continue to have SMP being subject to further regulatory interventions, and only to the extent necessary to address such SMP. Bearing in mind the fast-paced, dynamic nature of digital markets, a shorter review period may be desirable in relation to any gatekeeper regulation.

 

III. A “BLACKLIST” APPROACH vs. A PRINCIPLES-BASED APPROACH

One of the differences that has arisen amongst regulators in how to formulate any ex ante regulation lies in whether to take a rules-based or principles-based approach. For example, the EC’s proposed DMA has set out a list of clear “blacklisted” practices that are forbidden if they are put in place by gatekeeper firms. These obligations are split into two categories: (i) those which apply outright and where the Commission would ensure compliance via sanctions; and (ii) those which are “susceptible of being further specified” and where – in addition to sanctions – the Commission would be able to ensure compliance by specifying specific implementation measures. Some of the more controversial obligations fall within this second category of obligations, including those relating to data access, interoperability, data portability, and the use of data from business users in activities that compete with those business users. While there is currently little detail around what further specification is envisaged in relation to the second category of obligations, this approach indicates that considerable further work will be needed after any adoption of the DMA Proposal to work out what these obligations mean in practice and how different gatekeepers would comply with them.

In contrast, the UK’s DMU aims to put into place a principles-based, enforceable code of conduct, which is meant to “provide flexibility to address a wide range of practices…[which] could be difficult to capture within a narrow ‘blacklist’ of restrictions.”²²

Any suitable approach to gatekeeper regulation must ensure flexibility, particularly given the dynamic, fast-changing nature of the markets in question and the different business models through which tech firms operate. In this respect, advice from the UK’s DMT recognizes that a rules-based (or “blacklist”) approach may provide greater clarity to firms on compliance but notes certain downsides, including that “they tend towards a one-size-fits-all approach which is not tailored to where there is evidence of harm” and “given the pace with which firms’ conduct evolves in digital markets, [rules] are likely to need frequent review and updating to remain effective.”²³ It seems that a “blacklist” approach may be too blunt an instrument for fast-moving markets; history has shown that such an approach to regulation can often lead to unintended consequences.

Moreover, many of the practices that regulators are attempting to restrict – such as “self-preferencing” and how firms use data across their activities – encompass a wide range of possible conduct. It is widely recognized by regulators, practitioners and academics that these types of practices can be pro-competitive and generate benefits for consumers.²⁴ Accordingly, blanket prohibitions on these types of practices would be untargeted, disproportionate and carry a substantial risk of distorting competition and hindering innovation. The better view might therefore be to avoid or at least restrict the specification of black-listed practices.

While it may be ostensibly tempting to impose blanket bans on certain types of conduct for administrative ease, it would be preferable to apply remedies on a case-by-case basis, taking into account the overriding need for such remedies to be proportionate (in line with well-established EU principles). This targeted approach may result in more efficient outcomes as remedies will be tailored the specific market failures as they arise in relation to a certain gatekeeper. Drawing on the EU telecommunications regulatory framework, this approach of basing regulation on an objective assessment of markets, with due consideration to the proportionality of remedies proposed, provides a more stable and predictable regulatory environment. It would further be appropriate to provide for a periodic review of these remedies, with the designated regulator reviewing the effectiveness of such remedies to ensure that they are appropriate and keep pace with any market changes. This approach would minimize the loss of innovation, avoid the imposition of unnecessary regulatory burdens and limit the potential for other unintended consequences.

 

IV. APPROPRIATE TYPES OF CONDUCT TO RESTRICT BY WAY OF REGULATION

Differences of view remain as to what types of conduct should be captured by ex ante regulation that applies to “gatekeepers.” For example:

• Where the conduct overlaps with, for example, Article 102 of the Treaty of the Functioning of the EU (“TFEU”), this has the practical effect of replacing an in-depth, case-by-case ex post assessment of the impact of such conduct with an outright ex ante ban (or otherwise a reversed burden of proof). This leaves regulators vulnerable to the charge that they are simply seeking to side-step the need to demonstrate that the conduct in question results in anti-competitive effects, such that their approach is disproportionate.
• Certain concerns that have been raised by regulators and stakeholders, such as a lack of transparency and consumer “choice,” appear to be industry-wide issues that are not limited to businesses of a particular size. Regulation aimed at tackling these concerns should be applied more broadly to be effective. Imposing regulatory obligations only on firms that have been designated as “gatekeepers” runs the risk of missing many harms, without necessarily alleviating the underlying concerns justifying the use of regulation.

There is a strong case that conduct identified under any gatekeeper regulation should be limited to practices where the harmful nature of the conduct is easily identified, and the conduct is not capable of being assessed as pro-competitive. Unlike the Article 102 TFEU mechanism, where the categories of abuse are not “closed” and thereby provide flexibility to regulators, categories of conduct captured by ex ante regulation should be more specifically and clearly demarcated in order to guarantee legal certainty. This approach avoids the risks of restricting competition on the merits, which would adversely affect the incentives for market players to deliver valuable innovations to consumers.

In this respect, questions arise over whether “leveraging” practices, in which a company uses its market power in a market where it is dominant to boost its position in a vertically-related or adjacent market where it may not have market power, should be included within the remit of any gatekeeper regulation. It is commonly acknowledged that new entry into markets is generally pro-competitive and the conditions that lead to consumer harm in these scenarios tend to be highly case-specific (and linked to the concept of “tipping” markets). However, the line between legitimate regulatory intervention and regulatory overreach that hinders pro-competitive market entry is perilously thin. And it is notoriously difficult to identify and predict markets that are likely to “tip”; indeed, this predictive exercise is largely unproven in the digital sector. Over-regulating in this area therefore risks re-instating “efficiency offence”-type theories of harm that have been widely rejected in the merger context.

If regulators consider it necessary to enforce against leveraging practices prior to any competitive damage being done, it could be that some limited reversal of the burden of proof may be appropriate. This would require gatekeeper firms to prove that rivals will still be able to compete in the vertically-related or adjacent market. However, if such a potentially significant measure were to be introduced as part of any gatekeeper regulation, this would need to be accompanied by: (i) more rigorous evidence gathering to ensure regulatory remedies are justified and proportionate in the particular circumstances, and not outweighed by any harm to competition and innovation; and (ii) greater procedural safeguards, such as a more stringent approach to the designation process and the ability to appeal the imposition of any regulatory remedies to an independent body or court for a “merits review” (i.e. not limited to judicial review grounds).

 

V. CONCLUSION

The issues that lie at the heart of designing effective, proportionate and pro-competitive regulation are manifold and complex, particularly in light of the need to balance competing interests in fast-changing digital markets and to avoid dis-incentivizing the innovation that has enabled the substantial benefits delivered by the digital economy. Given this backdrop, there can be no pretense of easy or clear-cut answers; any such pretense carries significant risks and costs for consumers. However, given the imminent introduction of regulation on multiple fronts, it is key that all stakeholders engage in further debate and consideration to ensure that the most constructive regulatory approaches are adopted.

---

1 Rod Carlton (partner), Rikki Haria (senior associate) and Caroline Chew (associate) are members of Freshfields Bruckhaus Deringer’s Antitrust Competition and Trade Group. The views expressed in this article are the authors’ own.

2 See https://ec.europa.eu/competition/antitrust/cases/dec_docs/39740/39740_14996_3.pdf, https://ec.europa.eu/competition/antitrust/cases/dec_docs/40099/40099_9993_3.pdf and https://ec.europa.eu/commission/presscorner/detail/en/IP_19_1770.

3 See https://ec.europa.eu/commission/presscorner/detail/en/ip_20_1073.

4 See https://ec.europa.eu/commission/presscorner/detail/en/ip_20_2077.

5 For example, the CMA’s Final Report in its market study on Online Platforms and Digital Advertising acknowledges that platform services funded by digital advertising bring “substantial benefits to consumers,” with “consumers plac[ing] great financial value on a range of online services”; and that the “current COVID-19 pandemic has emphasised the critical importance of digital services for consumers’ well-being and prosperity” (see Box 2.1, available at: https://assets.publishing.service.gov.uk/media/5efc57ed3a6f4023d242ed56/Final_report_1_July_2020_.pdf); see also the Advice of the Digital Markets Taskforce, para 2.2, available at: https://assets.publishing.service.gov.uk/media/5fce7567e90e07562f98286c/Digital_Taskforce_-_Advice_–.pdf).

6 See https://assets.publishing.service.gov.uk/media/5efc57ed3a6f4023d242ed56/Final_report_1_July_2020_.pdf.

7 See https://judiciary.house.gov/uploadedfiles/competition_in_digital_markets.pdf.

8 See https://ec.europa.eu/competition/publications/reports/kd0419345enn.pdf.

9 See https://www.accc.gov.au/system/files/Digital%20platforms%20inquiry%20-%20final%20report.pdf.

10 In addition to those listed below, there are active legislative proposals worldwide, including the establishment of a special unit within the Australian Competition and Consumer Commission to proactively enforce, monitor and investigate competition and consumer protection in digital markets (see https://treasury.gov.au/sites/default/files/2019-12/Government-Response-p2019-41708.pdf); the endorsement by the German Federal Government of draft legislation introducing the concept of “undertakings with paramount significance for competition across markets” (see https://www.bundeskartellamt.de/SharedDocs/Meldung/EN/Pressemitteilungen/2020/25_02_2020_Stellungnahme_10_GWB_Novelle.html); and the establishment of a Japanese Headquarters for Digital Market Competition, which released an interim report proposing ex ante regulation for digital platforms (see https://www.kantei.go.jp/jp/singi/digitalmarket/index_e.html).

11 See paras 12 and 4.4 of the Advice of the Digital Markets Taskforce.

12 This code of conduct is based on the principles of Fair Trading, Open Choices and Trust and Transparency (see paras 4.37-4.41 of the Advice of the Digital Markets Taskforce).

13 For example, focusing on transparency, consumer controls, interoperability and data access and separation.

14 See https://ec.europa.eu/info/sites/info/files/proposal-regulation-single-market-digital-services-digital-services-act_en.pdf.

15 See Articles 3(1) and 3(2) of the DMA Proposal, which also set out the cumulative quantitative thresholds which create a rebuttable presumption of gatekeeper status as follows:

(i) EEA turnover of at least €6.5bn in the last three financial years or market capitalisation or equivalent fair market value of at least €65bn in the last financial year;

(ii) more than (a) 45 million monthly active end users established / located in the EU and 10,000 yearly active business users established in the EU in the last financial year; and

(iii) providing a core platform service in at least three EU Member States.

16 Emerging gatekeepers are companies which will foreseeably have an entrenched and durable position in the future. Emerging gatekeepers would only need to comply at most with certain obligations (i.e. those in Article 5(b) and Article 6(1)(e), (f), (h) and (i) of the DMA Proposal) and only those which are “necessary and appropriate to ensure that the company does not achieve by unfair means such entrenched and durable position.”

17 See Article 5 and 6 of the DMA Proposal.

18 Based on the current DMA Proposal, the EC will review, at least once every 2 years, whether designated gatekeepers continue to satisfy the threshold requirements or whether new providers of CPS do so; as well as whether a gatekeeper’s list of affected CPSs requires adjustment (see Article 4(2) of the DMA proposal).

19 The DMU’s SMS designation process is proposed to be open, transparent and consultative. Once a decision is made, the SMS designation would persist for a fixed time period (e.g. 5 years); however, where there has been a material change in circumstances, the DMU could receive applications from companies to remove the designation (subject to the DMU’s own review).

20 For example, see para 76 of Directive (EU) 2018/1972, which states that “[a]ny party subject to a decision of a competent authority should have the right to appeal to a body that is independent…Member States should grant effective judicial review against such decisions” (available at: https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32018L1972&from=EN).

21 See para 87, Guidelines for market investigations (CC3 (Revised)), available at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/284390/cc3_revised.pdf).

22 See para 4.36 of the Advice of the Digital Markets Taskforce.

23 See para 20 of Appendix C of the Advice of the Digital Markets Taskforce.

24 For example: (i) the European Commission’s guidelines on the assessment of non-horizontal mergers state that “the integration of complementary activities or products within a single firm may produce significant efficiencies and be procompetitive”; and (ii) the OECD has noted that “self-preferencing is in many ways simply a softer variant of in-sourcing, and it is well-established that in-sourcing, even by government monopolists, can often be more efficient than out-sourcing when there are transaction costs” (OECD, Lines of Business Restrictions – Background Note by the Secretariat, (8 June 2020), available at: https://one.oecd.org/document/DAF/COMP/WP2(2020)1/en/pdf, paragraph 7).