The allegedly entrenched positions of certain dominant companies in the digital economy and the alleged inability of antitrust enforcement to prevent potential anticompetitive conduct by these firms have led several European jurisdictions to consider an overhaul of their competition policy approach. A common theme is giving competition authorities or newly created digital agencies the power to regulate firms with gatekeeper status. This article discusses the criteria that have been proposed in the EU, UK, and Germany for identifying the firms, or gatekeepers, that are to be subject to such regulation. It also highlights advantages and disadvantages of different approaches as well as challenges for practitioners in their implementation, and discusses some potentially far-reaching consequences due to inconsistencies in the proposed criteria across jurisdictions.
By Can Çeliktemur, Arnd H. Klein, Vivek Mani & Marc Rysman1
There has been extensive commentary on the allegedly entrenched positions of GAFA – Google, Amazon, Facebook, and Apple – and the inability of ex post enforcement of competition law to prevent potential anticompetitive conduct by these firms.2 Several European jurisdictions are currently considering an overhaul of their competition policy approach towards dominant companies in the digital economy. A common theme is that competition authorities or newly created digital markets agencies could be given the power to regulate dominant companies with a gatekeeper status. As an important first step in this new regulatory approach, policymakers in various jurisdictions have specified the criteria for identifying the firms, or gatekeepers, that are to be subject to regulation.
This article discusses the criteria that have been proposed in the EU, UK, and Germany for assigning a special gatekeeper status to digital platforms. We highlight advantages and disadvantages of different approaches as well as challenges for practitioners in their implementation. We conclude by discussing some potentially far-reaching consequences due to inconsistencies in the proposed criteria across jurisdictions.
II. THE EUROPEAN APPROACH TO GATEKEEPERS
A. The EU’s Gatekeeper Status
The EU recently published a proposed regulation of tech firms, the Digital Markets Act.3 A key component is the designation of gatekeeper status to tech firms that operate at least one so-called core platform service and have a lasting, large user base in multiple countries in the EU. This designation triggers behavioral regulation.
A wide range of economic activities within the digital sphere could potentially be under scrutiny: The core platform services targeted by the regulation include “(i) online intermediation services (incl. for example marketplaces, app stores and online intermediation services in other sectors like mobility, transport or energy)[,] (ii) online search engines, (iii) social networking[,] (iv) video sharing platform services, (v) number-independent interpersonal electronic communication services,4 (vi) operating systems, (vii) cloud services and (viii) advertising services …”5
The proposed necessary conditions for designating the gatekeeper status to a firm include the following three cumulative criteria:6
- The firm “has a significant impact on the internal market,” with a presumption of significance if, among other criteria,
- it “achieves an annual [European Economic Area] turnover equal to or above EUR 6.5 billion in the last three financial years,” or
- “where the average market capitalisation or the equivalent fair market value of the undertaking to which it belongs amounted to at least EUR 65 billion in the last financial year.”
- The firm “operates a core platform service which serves as an important gateway for business users to reach end users,” which is presumed if “it provides a core platform service that has more than 45 million monthly active end users established or located in the Union and more than 10 000 yearly active business users established in the Union in the last financial year,” and
- The firm “enjoys an entrenched and durable position in its operations or it is foreseeable that it will enjoy such a position in the near future,” which is likely to be the case where the above user threshold criteria are met “in each of the last three financial years.”
- The firm “has a significant impact on the internal market,” with a presumption of significance if, among other criteria,
If the above criteria hold, the firm is required to notify the Commission of the fact, with an opportunity to rebut the presumption of strategic market status. In addition, the Commission would be allowed to designate gatekeeper status even when some of these thresholds are not satisfied, for example, based on an assessment of:7
- “the size, including turnover and market capitalisation, operations and position of the provider of core platform services;
- the number of business users depending on the core platform service to reach end users and the number of end users;
- entry barriers derived from network effects and data driven advantages, in particular in relation to the provider’s access to and collection of personal and non-personal data or analytics capabilities;
- scale and scope effects the provider benefits from, including with regard to data;
- business user or end user lock-in; [and]
- other structural market characteristics.”
The Digital Markets Act specifies a comprehensive catalogue of behavioral remedies for designated gatekeepers. While many of the remedies are very specific in that they apply only to certain types of gatekeepers, others are more general. Examples of general requirements include the prohibition of the use of proprietary data generated by the gatekeeper’s business users in the gatekeeper’s activities that compete with the business users; the prohibition of self-favoring in rankings when products are sold by the gatekeeper; the obligation to provide data portability; the prohibition to combine personal data across services; prohibitions of price parity clauses; the obligation to allow business and end users that matched through the platform to conclude contracts outside the platform; and the obligation to notify the Commission of any acquisition of a provider of digital services irrespective of the regular notification requirements.8
The requirements that are more specific to the type of platform services offered by the gatekeeper include, among others, requirements for operation system providers to allow the de-installation of pre-installed software, and the facilitation of the installation and use of third-party apps and app stores; requirements for advertising platforms to provide advertisers and publishers with price information; and requirements for search engines to provide certain data to competing search engines.9
A. Alternative Proposals From Other Jurisdictions
- UK: Strategic Market Status
In the UK, the concept of strategic market status (“SMS”) was introduced by the Furman Report. It proposes to designate SMS to a company that has enduring market power over a strategic gateway or bottleneck in a digital market, where it controls market access by others.10 The Furman Report further proposes that companies with SMS would be subject to certain forms of ex ante regulation by a newly established Digital Markets Unit (“DMU”).
The UK Competition and Markets Authority (“CMA”) recently published a market study11 in which the UK government accepted the strategic recommendations of the Furman Report, in particular the concept of SMS and a DMU as the enforcer of an ex ante regulatory regime.12 The CMA interprets SMS as a “position of enduring market power or control over a strategic gateway market with the consequence that the platform enjoys a powerful negotiating position resulting in a position of business dependency,”13 concluding that SMS would “include firms that have obtained gatekeeper positions and have enduring market power over the users of their products”14 and would be based on evidence such as:15
- “measures of shares of supply in the consumer-facing market;”
- “the extent of reach across consumers;”
- “[the] share of digital advertising revenues;”
- “control over the rules or standards which apply in the market;” and
- “the ability to obtain and control unique data that is applicable outside the market.”
- According to the CMA, the SMS “should apply to the corporate group as a whole (i.e., including all businesses with the same ultimate owner) … to … ensure that the DMU would have the ability to address concerns in markets that are adjacent to those where the firm has market power.…”16
The regulatory authority of the DMU would include measures with varying degrees of invasiveness. As the least invasive measure, the DMU could develop a code of conduct for platforms with SMS that would be aimed at safeguarding fair access to the platform and limiting the creation of barriers to switching to alternative platforms. This code would be developed in collaboration with market participants and other stakeholders.17 As a more drastic intervention, the DMU would have the authority to mandate mobility of and access to data.18 Importantly, companies with SMS would be required to make the CMA aware of all intended acquisitions.19
- Germany: Paramount Importance for Competition Across Markets
In January 2021, Germany revised its Act Against Restraints of Competition, the federal antitrust law.20 Most significantly, the law now includes a new statute that allows the Federal Cartel Office (Bundeskartellamt) to impose behavioral rules on platforms that it considers to have paramount importance for competition across markets (“PIFCAM”).21 The German PIFCAM concept is similar to the UK’s SMS, but emphasizes the cross-market implications of gatekeepers.
With the revision, the German federal antitrust law defines a set of factors that the Federal Cartel Office is supposed to take into account in determining whether a platform has PIFCAM status. These include:22
- dominance in one or more markets,
- its financial or other resources,
- vertical integration or activities in otherwise linked markets,
- access to data relevant for competition, and
- the importance of its business activities for access to input and output markets by third parties, and the resulting influence on business activities of others.
Per the revision, the Bundeskartellamt can impose a set of prohibitions on companies with PIFCAM. The conduct that can be prohibited includes, among others, self-preferencing of its own offerings in intermediating access to input and output markets, the use of data collected as the gatekeeper to raise entry barriers or hinder competitors in other markets, requiring users to agree to the consolidation of data across different services provided by the company, tying of separate services, limiting interoperability and data portability, and limiting information on quality or success of its services. Firms have the ability to object to these prohibitions by providing an (efficiency) justification.23
A. EU’s Threshold-based Approach
The criteria proposed by the EU on the one hand, and the UK and Germany on the other hand to identify gatekeepers differ substantially. Under the EU criteria, gatekeeper status may be triggered simply by exceeding thresholds for market size, and the numbers of users and business participants. The criteria of Germany (“dominance in one or more markets … and the importance of its business activities for access to input and output markets by third parties”) and the UK (“market power or control over a strategic gateway market”) depend on an economic analysis of an individual firm’s market position. There are advantages and disadvantages to these approaches.
One of the advantages of the EU’s proposal is that it is objective, that is, it does not leave too much room for interpretation and discretion of the regulatory body. While the gatekeeper status can be rebutted once triggered, firms can expect that they will, with a certain likelihood, be subject to regulatory scrutiny when exceeding these thresholds.24
The EU’s threshold-based approach also has certain disadvantages. The existence of a fixed threshold may create perverse incentives for platforms to “stay under the radar.” While this may not be an issue for regulating currently incumbent gatekeepers, it may be relevant in the future when those gatekeepers whose growth the new regulatory approach intends to foster will have reached a point when they will have gotten close to the threshold. This may lead platforms to actively limit the number of users (notably limiting the network effects that make these platforms so beneficial) to avoid regulatory scrutiny, for example, by limiting innovation or marketing activities. It could also lead to incentives to restructure. For example, a formerly pan-European gatekeeper could split into platforms at the member-state level and evade the EU’s thresholds, while not improving the state of the competition in each member state.
Using thresholds for presumptions also runs the risk of designating firms as gatekeepers when the firms do not necessarily possess sufficient market power, potentially resulting in false positives – the risk that non-problematic behavior is falsely prohibited. For example, an important aspect of platform competition that may significantly limit platforms’ market power is multi-homing. Consider a situation in which buyers offer products and sellers make purchases on two competing online e-commerce platforms.25 In this situation, one of the platforms may very well exceed the proposed user thresholds to be designated gatekeeper status. However, given that both business and end users are already present on a competing platform, the designated gatekeeper may risk losing both sides to its competitor if it worsens the terms (price or quality of service) to its customers. Moreover, it may be the case that the platform with a larger user base offers the more efficient, more consumer-friendly core platform service compared to the competing platforms, and designating it a gatekeeper status may put constraints on the platform’s ability to offer the better services to a wider user base. While such factors do not appear to be explicitly accounted for in the EU’s threshold-based approach, they will presumably be accounted for in the alleged gatekeeper’s attempt to rebut the gatekeeper designation.
The application of thresholds may also result in several firms in the same industry triggering the gatekeeper status, independent of the degree of competition between them. For example, consider a situation in which there are two large firms offering smartphone operating systems. These two companies would likely exceed the EU’s thresholds in terms of market size and users, that is, a sufficiently large number of active business users (app developers) and active end users (phone owners). Under the EU’s threshold-based approach, the gatekeeper presumption may be triggered for both firms, independent of the degree of competition between them. A blanket approach to apply thresholds independent of structure or conduct may also lead to false positives in cases where firms closely compete over prices or quality of services and act as strong competitive constraints on each other. The firms would have to engage in a rebuttal to assert that they are in fact competing.
B. UK’s and Germany’s Analysis-based Approaches
The UK’s and Germany’s analysis-based approaches act as an additional filter against false positives. While the proponents of a regulatory approach may disregard the risk of false positives in favor of avoiding false negatives (i.e. permitting problematic behavior),26 the introduction of at least some analysis in the determination may shift the balance more towards what is economically reasonable. The more “uncertain” approach in the UK and Germany may also be less prone to potential perverse incentives arising with the EU’s threshold-based approach.
However, the UK’s and Germany’s proposals are much more subject to the difficulties that arise in ex post enforcement of competition law in digital markets. In particular, in the UK and Germany, firms might have to form a view on the analysis of the relevant market and market power by the regulatory body. This is an inherently complex exercise in two-sided markets due to the challenges associated with market definition and the analysis of market power in two-sided markets. For example, such an exercise will need to contend with much debated issues relating to market definition for a two-sided platform – should one define a relevant market for each side separately or include both sides in a single relevant market?27 Furthermore, while there is some practical guidance on how to implement a SSNIP test in a two-sided market (be it separate for each side or a joint one), these tests can be difficult to implement due to their data requirements or infeasible when prices to one side are zero.28
Finally, the role of indirect network effects in the analysis of market power is ambiguous – for example, in the case of positive indirect network effects, on the one hand, they may serve as a competitive constraint to pricing to one side; on the other hand, they may strengthen the market power of incumbent platforms.29 As such, there may be differing views on the proper analysis of the relevant market and market power in the designation of the gatekeeper status when this designation is based on an economic analysis of the firm’s economic position. Notably, the difficulty of this analysis has been cited as a motivation for using a regulatory approach in the first place.30
The above proposals vary not only in the specific criteria for special status designation, but also in their reliance on economic analysis. While all of these approaches may capture currently incumbent platforms (i.e. GAFA), the differences between jurisdictions may create substantial burdens when today’s entrants will have grown to a size that makes them subject to regulation in some – but not all – jurisdictions. This is significant because platforms often employ one platform design across all jurisdictions they operate in.31 Consequently, the strictest legislation that triggers a firm’s gatekeeper status the earliest may have an impact on the conduct of the gatekeeper across jurisdictions, or result in the gatekeeper ceasing operations in a particular jurisdiction to avoid a costly adaptation of its platform to the regulatory requirements in that jurisdiction.32,33
1 Can Çeliktemur is a Principal at Cornerstone Research, London. Arnd H. Klein is a Manager at Cornerstone Research, San Francisco. Vivek Mani is a Principal at Cornerstone Research, London. Marc Rysman is a Professor of Economics at Boston University. The views expressed in this article are solely those of the authors, who are responsible for the content, and do not necessarily represent the views of Cornerstone Research.
2 See e.g. Jacques Crémer et al., “Competition Policy for the Digital Era,” Report for the European Commission, 2019 (“Crémer Report”), pp. 12–13, https://ec.europa.eu/competition/publications/reports/kd0419345enn.pdf; Jason Furman et al., “Unlocking Digital Competition,” Report of the Digital Competition Expert Panel, 2019 (“Furman Report”), p. 22, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/785547/unlocking_digital_competition_furman_review_web.pdf.
3 European Commission, “Proposal for a Regulation of the European Parliament and of the Council on Contestable and Fair Markets in the Digital Sector (Digital Markets Act),” 2020 (“Digital Markets Act”), https://ec.europa.eu/info/sites/info/files/proposal-regulation-single-market-digital-services-digital-services-act_en.pdf.
4 “‘[N]umber-independent interpersonal communications service’ means an interpersonal communications service which does not connect with publicly assigned numbering resources, namely, a number or numbers in national or international numbering plans, or which does not enable communication with a number or numbers in national or international numbering plans.” See Directive (EU) 2018/1972, Art 2.7. This appears to cover instant messengers.
5 Digital Markets Act, pp. 2, 34–35 (Art. 2.2).
6 Digital Markets Act, pp. 36–37 (Art. 3.1, 3.2).
7 Digital Markets Act, pp. 37–38 (Art. 3.6).
8 Digital Markets Act, pp. 39–40, 44 (Art. 5, 6, 12).
9 Digital Markets Act, pp. 39–41 (Art. 5, 6).
10 Furman Report, p. 55.
11 Competition and Markets Authority, “Online Platforms and Digital Advertising,” 2020 (“CMA Market Study”), https://assets.publishing.service.gov.uk/media/5fa557668fa8f5788db46efc/Final_report_Digital_ALT_TEXT.pdf.
12 CMA Market Study, ¶ 7.2.
13 CMA Market Study, ¶ 7.55.
14 CMA Market Study, ¶ 7.56.
15 CMA Market Study, ¶ 7.57.
16 CMA Market Study, ¶ 7.65.
17 Furman Report, pp. 57, 61.
18 Furman Report, pp. 57, 65.
19 Furman Report, p. 12.
20 Philipp Bongartz, “Happy New GWB!,” blog post, January 14, 2021, https://www.d-kart.de/en/blog/2021/01/14/happy-new-gwb/. The adopted revision is available in German as Deutscher Bundestag, Drucksache 19/25868, November 13, 2020, (“German Revision”), https://dip21.bundestag.de/dip21/btd/19/258/1925868.pdf. See “ Wettbewerbsrecht novelliert und Kinderkrankengeld-Regelung ausgedehnt,” press release of the German Bundestag, January 14, 2021.
21 German Revision, pp. 19-22 (§ 19a).
22 German Revision, p. 19 (§ 19a (1)).
23 German Revision, pp. 20–22 (§ 19a (2)).
24 The burden of proof for the rebuttal is on the firms. See Digital Markets Act, p. 37 (Art. 3.4).
25 The ability of sellers and buyers to multi-home is facilitated through tools that allow sellers to post their products on multiple platforms (multiple listing services) and price comparison websites.
26 Crémer Report, p. 51.
27 In the United States, the Supreme Court has clarified that two-sided transaction markets be appropriately captured in a single relevant market including both sides. See Ohio v. American Express Co., 585 U.S. ___ (2018). However, there is also substantial critique to this approach. See e.g. Michael Katz, “Ohio v. American Express: Assessing the Threat to Antitrust Enforcement,” CPI Antitrust Chronicle, June 25, 2019. In Europe, case law has been inconsistent. See Lapo Filistrucchi et al., “Market Definition in Two-Sided Markets: Theory and Practice,” Journal of Competition Law and Economics 10, no. 2 (2014): 293–339 (“Filistrucchi (2014)”). Notably, in Germany, the Bundeskartellamt has provided substantial guidance on how it will define relevant markets in two-sided markets and noted that, ultimately, “the question of whether one should define one or two markets therefore needs to be decided on a case-by-case basis.” See Bundeskartellamt, “The Market Power of Platforms and Networks, Executive Summary,” working paper, June 2016, p. 6, https://www.bundeskartellamt.de/SharedDocs/Publikation/EN/Berichte/Think-Tank-Bericht-Zusammenfassung.pdf?__blob=publicationFile&v=4.
28 Filistrucchi (2014); Crémer Report, pp. 42–45; Furman Report, p. 89; Marc Rysman, “The Empirics of Antitrust in Two-Sided Markets,” Competition Policy International 3, no. 1 (2007), pp. 197–209 at 204.
29 A platform experiences positive indirect network effects when users on one side benefit from having a greater number of users on the other side. The presence of these positive indirect network effects implies that a price increase that leads to a reduction in the number of users on one side also causes users on the other side to leave; on the other hand, positive indirect network effects may make it harder to compete against incumbent platforms with large user bases and may contribute to concentration.
30 Digital Markets Act, p. 8 (“However, the Commission considered that Article 102 TFEU is not sufficient to deal with all the problems associated with gatekeepers, given that a gatekeeper … may not be captured by Article 102 TFEU if there is no demonstrable effect on competition within clearly defined relevant markets.”).
31 For example, following an investigation by Germany’s Bundeskartellamt, Amazon changed the contract terms for its marketplace worldwide. See Natasha Lomas, “Amazon Amends Seller Terms Worldwide after German Antitrust Action,” TechCrunch, July 17, 2019, https://techcrunch.com/2019/07/17/amazon-amends-seller-terms-worldwide-after-german-antitrust-action/?renderMode=ie11. As another example, in reaction to the introduction of the GDPR in Europe, Microsoft changed its contract terms on a worldwide basis. See “ Microsoft’s Commitment to GDPR, Privacy and Putting Customers in Control of Their Own Data,” blog post, Microsoft, May 21, 2018, https://blogs.microsoft.com/on-the-issues/2018/05/21/microsofts-commitment-to-gdpr-privacy-and-putting-customers-in-control-of-their-own-data; Natasha Lomas, “Microsoft Announces Changes to Cloud Contract Terms Following EU Privacy Probe,” TechCrunch, November 18, 2019, https://techcrunch.com/2019/11/18/microsoft-announces-changes-to-cloud-contract-terms-following-eu-privacy-probe. More recently, Microsoft announced the implementation of certain policies for the Microsoft Store (its app store for Windows) that do not appear to be country-specific. See Rima Alaily, “10 App Store Principles to Promote Choice, Fairness and Innovation,” blog post, Microsoft, October 8, 2020, https://blogs.microsoft.com/on-the-issues/2020/10/08/app-store-fairness-caf-interoperability-principles/.
32 Google recently announced it could exit the Australian online search market in reaction to the newly proposed Australian regulation regarding royalty payments to media companies for content shared on Google’s platform. Damien Cave, “An Australia With No Google? The Bitter Fight Behind a Drastic Threat,” The New York Times, January 22, 2021.
33 This abstracts from the problem that regulations themselves may be inconsistent and – if a firm is designated gatekeeper in multiple jurisdictions – may impose contradicting requirements on the gatekeeper.