The merits of ex ante regulation have sparked intense debate on both sides of the Atlantic. But their potential challenge to consistent Transatlantic competition policy and market regulation requires further consideration. The potential introduction of myriad new regulation and enforcement regimes across Europe, Germany, the UK, the U.S., and other jurisdictions is likely to put significant strain on an area of competition policy that has otherwise seen signs of increasing convergence. We explore here both the risk regulatory divergence on digital markets and the degree to which such divergence is likely to result in regulatory conflict or impose practical costs on market participants.
By Will Leslie & John Eichlin[1]
Large digital platforms have captured the zeitgeist of current competition policy. The introduction of ex ante regulation for the sector represents one of the most significant changes in competition policy in decades. This proposed regulation is far from consistent, however, covering a slew of enacted and proposed legislation with a variety of mechanisms for achieving a myriad of policy goals. Its introduction is also expected to trigger a wave of litigation as market participants adjust to the new regulatory environment.[2]
Of the many challenges, the risk of regulatory divergence has so far received little scrutiny. It carries, however, real practical significance. Competition regulators have over the last three decades sought to strengthen cooperation and, where possible, reach consensus on competition policy.[3] EU and U.S. competition authorities have recently affirmed the need for cooperation in the tech sector as part of the wider EU – U.S. technology council aimed at preserving open technology markets.[4] The prospect of regulatory divergence as some jurisdictions press ahead ex ante regulation and others don’t, and differing conceptions of what such rules should do, risks undermining the Transatlantic regulatory convergence that has taken place to date.
These risks are all the more pertinent for digital services which are frequently global in nature. These economic linkages introduce the practical risk of regulatory friction: ranging from the costs of adapting services to comply with different legal rules through to the risk of regulatory conflict where firms are subject to incompatible regulatory obligations.[5]
We explore the prospects and practical risks of ex ante regulation causing regulatory divergence. That ex ante regulation will result in divergence is straightforward: the introduction of the EU’s Digital Markets Act and the absence of similar legislation in the U.S. in the near term means there will inevitably be divergence. However, there are a number of salient questions to understand the implications. What, if anything, will drive potential divergent outcomes rather than simply divergence in regulatory instruments? How great is such divergence likely to be? And what practical impact will it have? Are imminent rules in Europe likely to ultimately have a “Brussels Effect” in setting de facto global regulatory standards, notwithstanding that most of the companies are headquartered in the U.S.?[6] Or will companies adopt a patchwork of business models adapted to each relevant jurisdiction or set of jurisdictions?
I. EX ANTE REGULATION OF LARGE DIGITAL PLATFORMS: COMPARING (REGULATORY) APPLES, PEARS, AND ORANGES
The ex ante label is applied to the EU’s digital markets act (hereafter the “DMA”), S.19A of Germany’s competition act (hereafter “S.19A GWB”) and the UK’s legislative proposals for empowering its digital markets unit (hereafter the “DMU”).[7] In the U.S., lawmakers in both the Senate and the House have introduced legislative proposals aimed at similar policy goals, but the prospects for this legislation are far less certain.[8] Other regimes sit somewhere across this spectrum. For example, China has pursued an evolving approach to its own unique landscape of digital platforms, most recently shifting its focus to a prevention regime aimed at developing bespoke solutions tailored to particular platforms.[9]
These initiatives address similar perceived deficiencies in competition law’s ability to regulate digital markets effectively: namely insufficiently fast intervention, weak deterrent effect, and a perception that traditional enforcement either does not permit intervention to address all of the substantive issues occurring in digital markets or that the thresholds for intervention are too burdensome. Many of these rules seek to address the special market features and common practices that proponents identify as raising barriers to effective competition, including network effects that feed into economies of scale and scope. They also rely on a policy recalibration that the risk of harm to competition outweighs the potential chilling effect on innovation or introducing security risks.
These legislative instruments are, however, far from a uniform model of market regulation. Their commonality has two main dimensions.
- First, they use a model whereby a regulator must designate firms and their digital services as “gatekeepers,” “platforms of paramount importance” or having “strategic market status” and, in turn, are subject to additional rules. The model is similar, albeit not identical to sectoral regulation such as telecoms and public utilities.[10]
- Second, they employ effectively per se rules or presumptions of competitive harm to specific conduct by those designated firms, mechanisms previously reserved for conduct that always or almost always achieved anticompetitive outcomes.[11]
Their commonality, however, largely ends there.
The regulations pursue, in the first instance, various differing underlying policy objectives. The DMA focuses on goals of “contestability” and “fairness” as explicitly distinct from the objective of undistorted competition underpinning EU competition law. S.19A GWB forms part of German competition law. The DMU proposal seeks to protect consumers. U.S. legislation is primarily rooted in traditional competition enforcement aimed at protecting consumers, but includes an undercurrent of fairness considerations that pervades current enforcement rhetoric.[12]
Their form equally differs. For example, the DMA applies an ostensible one-size-fits-all model, with the same rules across all designated gatekeepers even where the logic of particular rules for some gatekeepers may not be clearly established. Conversely, on the basis of the legislative proposal, the DMU would tailor codes of conduct to firms designated as having strategic market status. And, unlike the ex ante regulatory approach in other jurisdictions, the U.S. legislative proposals tend to be framed more as law enforcement measures that shift traditional enforcement burdens.
Finally, the various regulations envisage different obligations and remedies. Most fundamentally, the DMA and S.19A GWB are primarily intended to regulate designated firms’ conduct rather than effect structural change. The DMU and some of the U.S. legislation,[13] on the other hand, would permit structural remedies such as the break-up of different business lines within a firm if necessary to achieve the desired objectives.
II. IS THE INTRODUCTION OF EX ANTE REGULATION LIKELY TO RESULT IN REGULATORY DIVERGENCE
Regulatory divergence occurs where firms in the same economic position are treated differently by rules serving the same underlying purpose or firms in different economic position are treated the same. If ex ante platform regulation is viewed as regulating economic power, the introduction of ex ante platform regulation will inevitably result in some degree of regulatory divergence from competition law and between different ex ante regulations. The salient question is rather the degree of regulatory divergence.
Ex ante regulation is inherently set to diverge from competition law at home and abroad. The DMA provides, for example, for the designation of incipient “gatekeepers” (i.e. firms operating platforms that are likely to be become gatekeepers): thereby encompassing non-dominant platforms in contrast to competition law. All of the proposed regulations also envisage more intrusive obligations on platforms. The DMA, S.19A, the DMU, and to a lesser extent the U.S. legislation all contemplate, for example, requiring digital platforms to interoperate with their rivals, provide access to data which they collect and provide FRAND access to services.[14] These are novel obligations which competition law in Europe and the U.S. has so far either never mandated or required only in limited circumstances.
The prospect of regulatory divergence is, moreover, not a bug but a function of legislative intent for at least some jurisdictions. By pursuing different objectives, the DMA is intended to have different regulatory outcomes to competition law. This does not hold to the same degree for S.19A GWB and the DMU since the two regulations are, in contrast to the DMA, intended to be competition law. The different underlying objectives also mean that the DMA serves ostensibly distinct purposes and hence is simply different to competition law. The challenge is that it equally seeks to regulate economic power and, more practically, its rules are derived from competition law. S.19A GWB and the DMU in contrast are intended as competition law but will nevertheless impose obligations that go beyond those typically imposed. The practical effect is however the same: platforms will be subject to new rules focused on protecting effective market functioning that previously did not exist.
There is also a question of whether and to what degree the experience of ex ante regulation in one jurisdiction will influence future enforcement or regulatory outcomes in another. The potential divergence in regulatory rules will create an unprecedented natural experiment. The lessons of this experiment may in practice drive further convergence or divergence. As new ex ante regimes come into force, there is a prospect that the answers to some of the “what-ifs” that have cautioned prudence in many jurisdiction rules enforcement rules will be answered. For example, new access rules are intended to open up previously closed systems or require access to particular technologies. The results could either confirming the potential risks of intervention or undermine arguments on the counterfactuals. Those lessons will undoubtedly influence the confidence of courts and lawmakers in the risks of predicting outcomes of imposing enforcement remedies.
III. WHAT ARE THE IMPLICATIONS FOR REGULATORY DIVERGENCE?
The practical implications of different regulatory approaches across jurisdictions depend on the nature and degree of potential conflict. At its most substantial, divergence can arise from direct conflict between irreconcilable obligations across different regimes. While this is perhaps unlikely to occur strictly within competition policy, an example would be where companies are faced with access rules in one jurisdiction that conflict with data privacy rules in another. Other areas of divergence occur across a spectrum. In some instances, rules in one jurisdiction may have de facto extra-territorial effect where the proverbial lowest common denominator applies to all jurisdictions. In others, firms may be able to reasonably comply with different standards in different jurisdictions at varying difficulties and costs of compliance.
Despite some global convergence in competition law, regulatory divergence is already a feature of transatlantic rules for competition enforcement involving unilateral conduct. Article 102 TFEU encompasses exploitative abuses as well as exclusionary abuses, unlike Section 2 of the Sherman Act. Furthermore, differences in case law (as well as competition authorities’ selection of cases) mean that there are already differences in competition law standards across jurisdictions. The EU’s Google Shopping and Google Android cases are not, for example, mirrored by similar enforcement in the United States, which has not clearly adopted similar standards for self-preferencing as a theory of harm. As a result, Google has chosen to implement the EU’s decisions in the EU alone.[15] In short, some regulatory divergence is manageable, albeit potentially imposes additional costs for business.
The risk of extra-territorial effect where one jurisdiction’s rules de facto impose an outcome for other jurisdictions is higher where regulation addresses market structure rather than conduct. For example, merger control which regulates the nature of the firms permitted to participate allows the most stringent regulator to impose the most restrictive outcome on all jurisdictions. This has in the past raised political hackles. The UK Competition & Markets Authority’s extraterritorial merger enforcement initiatives in the digital sector was presaged by earlier examples such as European Commission’s prohibition of General Electric’s proposed acquisition of Honeywell, which elicited significant criticism in the U.S. where both parties were headquartered.[16] Since the DMA and S.19A GWB regulate conduct, the risk of true conflict with other jurisdictions is unlikely to arise. Albeit the political challenge of regulating firms predominantly operating from other jurisdictions remains. The DMU on the other hand is currently expected to allow for structural interventions, which carry a greater risk of direct regulatory conflict.
Provided that structural remedies do not feature significantly in the future, the impact of divergence in digital platform regulation may largely drive increased regulatory cost for businesses. On the one hand, businesses can maintain a single business model. In this case the lowest common denominator applies: firms will need to adapt the business model to the jurisdiction with the most onerous rules – the so-called Brussels effect. Alternatively, firms can potentially offer different business models in different jurisdictions to comply with the relevant rules. In which case, the effect is the emergence of different regulatory ecosystems. The outcome is not, however, binary. Firms can choose different models for different products. There is also already evidence of such divergence. Google has, for example, publicly announced that it will adopt a different app store model for the EEA versus that used for consumers elsewhere.
IV. CONCLUSIONS
Ex ante regulation is only just starting to move from drawing board to practical reality: whether we will see regulatory consensus or frictional divergence will play out in the coming months and years. Its significance is also likely to turn on the extent to which jurisdictions converge in relation to the substance of ex ante regulation, the approach of regulators to their new powers, and the degree to which companies seek to preserve a single operating business model, or mix and match between their different operating regions.
The practical application of the rules is, moreover, likely to be dynamic rather than static: regulators will not take decisions in a vacuum and will learn lessons from each other as the relationship between ex ante regulation and competition law as well as with ex ante regulation in other jurisdictions works itself out. So while the outcome is uncertain, the prospect of more regulatory cooperation aimed at easing the hard edges of regulation seems somewhat inevitable.
[1] Counsel, Linklaters, Brussels and New York, respectively.
[2] Foo Yun Chee, Top EU judge expects a wave of litigation from tech giants against new tech law, March 17, 2023, Reuters, https://www.reuters.com/technology/top-eu-judge-expects-wave-litigation-tech-giants-against-new-tech-law-2023-03-17/.
[3] See in particular, the Agreement between the Government of the United States of America and the Commission of the European Communities regarding the application of their competition laws OJ 1995 L 95/47, https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:21995A0427(01)&from=EN. Agreement between the European Communities and the Government of the United States of America on the application of positive comity principles in the enforcement of their competition laws OJ 1998 L 173/28
[4] EU-U.S. Joint Technology Competition Policy Dialogue Inaugural Joint Statement between the European Commission, the United States Department of Justice Antitrust Division, and the United States Federal Trade Commission
[5] See e.g. Alden F. Abbott, ‘Competition Policy and Its Convergence as Key Drivers of Economic Development, https://www.ftc.gov/system/files/attachments/key-speeches-presentations/2009unctapaper.pdf.
[6] Anu Bradford, The Brussels Effect: How the European Union Rules the World (OUP 2020).
[7] Regulation (EU) 2022/1925 of the European Parliament and of the Council on contestable and fair markets in the digital sector and amending Directives (EU) 2019/1937 and (EU) 2020/1828; S.19A (Abusive Conduct of Undertakings of Paramount Significance for Competition Across Markets) of Germany’s Competition Act in the version published on June 26, 2013 (Bundesgesetzblatt (Federal Law Gazette) I, 2013, p. 1750, 3245 (as amended); BEIS, ‘A new pro-competition regime for digital markets – government response to consultation’ Command Paper: CP 657
[8] See e.g. American Innovation and Choice Online Act (AICOA), S. 2992, 117th Cong. (2021) (creating a cause of action for the federal agencies prohibiting certain forms of discriminatory conduct, subject to certain affirmative defenses). See generally Heidi M. Silton, Craig S. Davis & Halli Spraggins, Congressional Antitrust Bills Seek to Regulate a New Internet Era, Antitrust (Spring 2022) (providing an overview of recent legislation).
[9] State Administration for Market Regulation (“SAMR”), Press Release, SAMR to Further Specify the Detailed Anti-Monopoly Rules for Internet Platform Rules and Enhance Normalised Supervision Abilities, April 13, 2023, https://finance.sina.com.cn/jjxw/2023-04-13/doc-imyqfnhp3263465.shtml.
[10] Directive (EU) 2018/1972 of the European Parliament and of the Council of 11 December 2018 establishing the European Electronic Communications Code (Recast)
[11] Article 102 TFEU previously included some categories of abuse that were per se. See, for example, T-286/09 – Intel v. Commission [2014] ECLI:EU:T:2014:547, paras 72 – 94 in relation to exclusivity rebates. The Court of Justice subsequently overturned the judgment.
[12] Federal Trade Commission, Prepared Statement of the Hearing on Fiscal Year 2024 Budget, Before the Committee on Energy and Commerce Subcommittee on Innovation, Data, & Commerce, U.S. House of Representatives (April 18, 2023), https://www.ftc.gov/system/files/ftc_gov/pdf/p210100testimonyfy2024budget.pdf.
[13] See e.g. Ending Platform Monopolies Act (“EPMA”), H.R. 3825, 117th Cong. (2021) (prohibiting covered platforms from operating multiple lines of business that generate certain conflicts of interests).
[14] DMA, Articles 6(10), 6(12) and 7; Augmenting Compatibility and Competition by Enabling Service Switching Act of 2022 (“ACCESS Act”), S. 4309, 117th Cong. (2022).
[15] See e.g. Hiroshi Lockheimer, Google in Europe: Complying with the EC’s Android Decision, The Keyword Blog, Oct 16, 2018, https://www.blog.google/around-the-globe/google-europe/complying-ecs-android-decision/.
[16] Tobias Buck, Court upholds EU ban on GE-Honeywell merger, Financial Times, Dec. 14, 2005, https://www.ft.com/content/420694e0-6c8b-11da-90c2-0000779e2340.
