Competition agencies should build on the existing and successful case-by-case cooperation framework used in assessing mergers that are notified in multiple jurisdictions and develop similar working practices when regulating competition in digital markets. In merger control we see an example of cooperation, coordination, and convergence that serves to streamline individual enforcement efforts and the mutual trust that is thereby generated can spill-over into discussions about the shape of competition law enforcement. This system also serves as a platform for learning about shared competition concerns and innovative methods for assessing competition concerns. Ex post review of regulatory solutions can serve to identify superior practices which each agency can then adopt.
By Giorgio Monti & Jasper van den Boom[1]
I. INTRODUCTION
There is an increasing interest in regulating competition in digital markets, with amendments to some antitrust laws and new regulatory frameworks emerging in a number of jurisdictions. Some firms will face a kaleidoscope of rules and regulators controlling their conduct as agencies try and make markets work to achieve more competition and improve consumer welfare. It has been recognized by the G7 that some form of cooperation can serve to assist firms and regulators alike as they apply the new rules, in particular as the enforcers in these countries will be in the front line. What sort of cooperation framework might be adopted?
In the field of merger control, we see a successful global system by which transactions are notified in multiple jurisdictions, subjected to synchronized and broadly convergent regulatory scrutiny, and frequently addressed with multiple agencies agreeing on solutions and announcing these on the same day. What is it that has made cooperation in the field of mergers a success? And what can we learn from it when we move to facilitate cooperation in digital market regulation?
In this paper we claim that the voluntary multilateral analysis of discrete cases by multiple agencies is the most immediate and useful step to facilitate global coordination. The story of cooperation in merger control reveal agencies are able to coordinate both on how to address competition concerns and also learn from each other in shaping merger control in light of new developments. The mutual trust that is generated by working on single cases, we suggest, creates an environment that allows for wider strategic reflections as well as deeper collaboration than is possible under other institutional settings.
II. COOPERATION IN MERGER CONTROL – THE INGREDIENTS FOR SUCCESS
If we trace the history of international antitrust relations, the first significant moment of formal cooperation started with a bilateral agreement between the EU and the U.S., which was motivated principally by the arrival of the EU Merger Regulation.[2] The 1991 EU-U.S. agreement regarding the application of their competition laws provides a process by which coordination in enforcement is facilitated by ensuring that each party notifies the other well in advance “to enable the other Party’s views to be taken into account.”[3] The agreement recognizes that enforcement coordination is particularly useful when it can allow for a more efficient use of resources, improve the information available to reach a decision and thereby make law enforcement more effective. It is also recognized that cooperation can reduce costs incurred by persons subject to the enforcement activities of multiple agencies.[4] A significant number of mergers have been reviewed by the two agencies jointly. Even relatively early on during the dot com bubble for example, the two collaborated in exploring competition risks in emerging digital markets.[5] Frequently, agencies report that these interactions are “a model of international cooperation between the United States and the European Commission.”[6]
Cooperation was stepped up in 2002 with a document setting out Best Practices on Cooperation in Merger Investigations.[7] This is much more salient than the much-discussed transatlantic spats in Boeing/MDD and GE/Honeywell, which remain isolated cases in a set of merger decisions where the agencies coordinate and agree on how to handle mergers. The Best Practices are a reaction to the divergences identified at the time and indicate a willingness to coordinate timetables, to extend coordination with other agencies that may be reviewing the same merger, to organize joint calls with parties, and to engage in joint discussions about market delineation, theories of harm and remedies.
The document reveals the mutual trust the two agencies place on each other’s capacities and their commitment to reach convergent decisions whenever possible. This is evidenced by findings that contacts between the two occur virtually on a daily basis.[8] There is no commitment to align in every instance but even tricky cases raising new issues like Bayer/Monsanto which was reviewed by fifteen different agencies, allowed for discussions on relevant markets and theories of harm. Cooperation also extends to supervising remedies, so in ThermoFisher/Life Technologies the FTC and EU also cooperated in approving GE Healthcare as the divestiture buyer on the same day.
A further feature of merger control that helps oil cooperation is that parties often waive confidentiality so as to allow in-depth inter-agency communication, facilitating all elements of analysis, from market definition to merger remedies. Parties also often anticipate competition concerns by offering divestitures early and contact the competition authorities in advance of notification. This reveals that a further ingredient to successful transnational cooperation is the involvement of the parties during the review process. A good example of this is the InBev/SAB Miller merger, which was notified in a large number of jurisdictions, the parties ready with proposed divestitures in those markets which raised competition concerns.
Cooperation on discrete cases is also supported by the Mergers Working Group instituted in 1999. Dialogue with the U.S. assisted the EU in drafting its remedies notice and the new test for the merger regulation in 2004, while in the recent revision of the U.S. Guidelines on Vertical Mergers one detects that lessons from the EU were drawn.
Moreover, global merger regulation is not frustrated by certain jurisdictions applying somewhat different standards of merger assessment. For example, in InBev/SAB Miller the remedies package in South Africa was more extensive because of the wider range of public interest grounds considered when clearing a merger, in particular concerns about the economic participation of historically disadvantaged persons. Insofar as these differences are known and applied in a consistent and transparent manner, competition agencies will know the limits of cooperation and work within these.
In addition, when divergences are identified, like in the assessment of mergers in the pharmaceutical sector, a new multilateral working group for this sector was formed in 2021 which includes the European Commission, the Canadian Competition Bureau, the Federal Trade Commission, and the UK Competition and Markets Authority. Its ambition is to build a new approach to pharmaceutical mergers.[9] This quick setup of a working group was only possible given the longstanding interactions among agencies across a wide range of merger decisions.
III. LESSONS FROM COOPERATION IN MERGER CONTROL
What we see in merger control then are repeated instances of collaboration in discrete cases, the incremental development of good practices to deepen collaboration, an atmosphere of mutual trust among agencies, a collaborative attitude from firms during the regulatory process, and a mechanism that tolerates some regulatory divergence. Can one reproduce the cooperation we see in merger control to antitrust more generally, and to the regulation of digital markets more specifically? At first sight, some barriers seem to exist.
The first is that mergers are compulsorily notified ex ante while anticompetitive conduct is identified ex post and prosecution is discretionary. This makes coordination less feasible because one cannot generate the same kind of frequent interaction. However, this might become less of an issue as more jurisdictions move to ex ante regulation, and the same firms may be designated for ex ante regulation in multiple jurisdictions. Even in the context of ex post enforcement, the first agency is expected to notify the others of its competition concerns, which may motivate similar enforcement efforts or start a discussion among agencies about the relevant markets being examined and the competition concerns. Moreover, merger control will continue to apply to transactions in digital markets, which can serve as a platform to identify competition concerns for the purposes of ex post enforcement as well. A working group to discuss mergers of firms whose business model is data driven can easily be created in light of the differences seen in the Google/Fitbit merger.[10] By widening cooperation in the analysis of mergers in digital markets, agencies will become aware of issues that are relevant also in antitrust and regulation: notions of market definition and market power as well as competition concerns identified in ex ante merger analysis are just as relevant when examining unilateral conduct.
Second, there is a punitive element in antitrust which will make parties less willing to waive confidentiality and collaborate with agencies. However, if large platforms are expected to provide a remedy to a competition or regulatory concern that is shared among a number of jurisdictions, then there can be incentives on the part of the firm to waive confidentiality at the stage of remedy implementation to secure a global package of remedies. Firms might prefer settlement on terms that are less disruptive to their business model than a long running battle that leads to divestitures. Agencies might likewise prefer a less than perfect remedy today than an optimal, court-sanctioned remedy decades later when market conditions have changed beyond recognition.
Another fundamental issue is that there is likely to be more division about how to regulate unilateral conduct than there is in merger control. However, at the time of writing there is some convergence – the many expert reports that have been produced in many jurisdictions exhibit a shared understanding of how digital markets work and the market failures that need to be addressed. Agencies may differ when it comes to remedying market failures but there is more consensus on the competition concerns than in the past.
In sum, while the regulation of competition in digital markets does not have all the attributes which made cooperation in merger control the success story that it is today, there are signs that some G7 jurisdictions will prioritize antitrust and regulatory efforts in certain digital markets and if so, agreements to keep each other informed can stimulate coordination in investigations (think of coordinated dawn raids in cartel investigations), collaboration in identifying market failures and further cooperation in imposing remedies. Firms might prefer to engage in regulatory dialogue with multiple regulators as a means of avoiding lengthy legal processes if they challenge each agency in turn. Moreover, as we discuss below, cooperation in handling specific cases can yield a number of learning opportunities. However, before moving to consider this, we reflect on the possible role existing transnational institutions might play and why we favor the pragmatic collaboration just described.
IV. THE LIMITS OF EXISTING INSTITUTIONS
Transnational bodies like the International Competition Network (“ICN”) or the Organization for Economic Cooperation and Development (“OECD”) might be seen as natural orchestrators of global digital market regulation, but both have limits.
The ICN has some 140 members globally. This virtual network relies on Working Groups consisting of its members that operate produce recommendations on best practices. While these recommendations provide guidance to ICN members, they are free to implement or disregard recommendations as they see fit. The creation of a new Working Group – or subgroups – on digital markets may aid the deepening of cooperation. However, it is questionable whether this forum can achieve the depth of desired cooperation within this large group of countries. It is also doubtful whether all ICN members have digital regulation as a priority; conversely the ICN might need to extend its remit to include other regulatory bodies from all these jurisdictions to be fully effective.
The depth of cooperation through the ICN may be hindered by the nature of collaborative efforts and the products delivered by the ICN. As ICN members are free to implement or disregard any outcomes, it is questionable whether reliance on ICN recommendations leads to a higher level of convergence between competition policy and regulatory frameworks in digital markets without collaboration on the ground in discrete cases. There is a risk that any agreed substantive standard may be pitched at a high level of generality. Procedurally, membership of the ICN does not necessarily add much to facilitate case-by-case collaboration. Moreover, it has been argued that legal and practical barriers for cooperation are difficult to resolve through cooperation within the ICN itself.[11]
The deepening of cooperation between G7 members specifically is further limited by the global nature of the ICN. Firstly, Working Groups should be open to participation by all ICN members.[12] It is undesirable that G7 members form closed subgroups or working groups within the ICN which may lead to the exclusion of other global competition authorities. Alternatively, having G7 members set an agenda on digital markets within the ICN may lead to frictions with other ICN members which maintain different competition norms, have different interests or different access to resources as G7 members.[13] It may also lead to criticisms that it is again a small group of ICN members that dictates the discourse on how to proceed with novel and complex issues.[14]
In sum, while the ICN is a successful and dynamic transnational network, its membership is very extensive, and it is not clear that many competition authorities will prioritize enforcement in digital markets or introduce fresh regulations. Accordingly, while the ICN facilitates conversations about convergence, informed divergence, and cooperation in identifying best practices, it might be too large a forum to facilitate this type of discussion.
Similar observations can be made regarding collaboration through the OECD. While the OECD will continue to facilitate conversations, this forum is complementary to the daily work of enforcers. The OECD does provide the in-depth research and extensive dissemination of information that will aid competition authorities in creating convergence in their Guidelines to competition assessments for digital markets. This may in turn minimize legal and practical barriers for cooperation. In 2022, the OECD has released a background paper on forms and methods for international cooperation in competition policy and beyond. This background paper also suggests the introduction of new multilateral (binding) agreements between parties to facilitate cooperation.[15] However, it recognizes that political will is a prerequisite for a binding agreement that could facilitate deeper levels of cooperation among members.
Moreover, any further-reaching form of cooperation requires that the differences in legal interpretations and legal constraints on information sharing are minimized. Multilateral agreements between the parties could allow for highly intense cooperation including the exchange of sensitive information and evidence (similar to the waiver of confidentiality in EU-U.S. cooperation on mergers) or even mutual recognition of (aspects of) competition law decisions. The European Competition Network (“ECN”) is an example of far-reaching cooperation between the European Commission and the EU’s National Competition Authorities.
The underlying ECN+ Directive confers powers on these competition authorities to facilitate cross-border cooperation and convergence on several issues related to competition law enforcement.[16] A multilateral agreement between G7 members could establish similar rules on collaboration and coherent enforcement. However, the collaboration achieved by the ECN required the development of a shared competition culture through many years of gradual converge, the alignment of substantive laws some fifteen years before the ECN+ Directive was proposed and a broadly comparable legal and institutional framework among the Member States. Can we really expect to see the depth of collaboration among EU agencies to emerge quickly at global level even if there is political will?
V. LEARNING BY DOING
Returning to our favored model of collaboration via regular bilateral and multilateral contacts based on live cases, this holds an additional advantage: agencies can benefit from the experience of others. There are three types of learning opportunities that our framework facilitates.
The first is that agencies can stimulate each other to consider similar types of conduct found problematic in one jurisdiction. For example, the Commission’s three Google decisions have led U.S. Federal and State authorities to investigate similar conduct.[17] In their complaint vis-à-vis general search services, the DOJ notes similar concerns as expressed by the Commission in their Google Shopping and Google Android cases.[18] The case by the DOJ is building on findings of fact and evidence accrued by the Commission in a decade long investigation, potentially speeding up antitrust proceedings.
While the complaint by the DOJ follows the Commission’s decisions relatively quickly, an example of belated repeated investigations can be found in the complaint by the state of California against Amazon. This complaint relates to Amazon’s use of parity clauses to protect as part of a monopolization strategy. The European Commission has investigated Amazon’s use of parity clauses in their role as a bookseller as early as 2017.[19] Enhanced cooperation between different G7 members may help the detection of anti-competitive conduct across jurisdictions. When competition law policy across jurisdictions becomes increasingly converged, this could lead to faster enforcement actions as it becomes more likely that behavior which is considered anti-competitive in one jurisdiction is also likely to be found anti-competitive in another. This is of course without prejudice to the ability of the competition authority to assess the facts in light of national circumstances.
Second, where competition policy between jurisdictions is divergent, there is room to learn from one another in how investigations are conducted. The recent complaints against Facebook by the Bundeskartellamt (“BKA”) and the FTC for example reveal different ways of assessing market power. The FTC complaint introduces novel considerations on how social media markets should be defined. The FTC primarily relies on Facebook’s connection to a “social graph” to distinguish it from other forms of online networks and identify market power.[20] The BKA relied on alternative measures, considering the number of users, time spent on the platform and other indicia such as the strength of network effects. This competition authority was also more ambiguous towards whether other products such as YouTube, Snapchat or Twitter should be included in the relevant market.[21] The substance of the issue in the German case against Facebook also differed strongly. Where the FTC focused on monopolization in the market for social media, the German case intervened against excessive collection and combining of data.
Exchanging information and experiences on these different approaches to investigating competition law and harms may help to develop new insights and best practices. This could result in convergence over time. The question remains however whether each competition authority has the interest in pursuing certain cases or intervening against certain types of behavior. This depends on the objectives pursued by competition policy and the framing of the legal text.
Third, and more ambitiously, authorities can learn from the effects of regulation that each of them achieves with their intervention. Rather than counting judicial victories or measuring the degree of convergence, a helpful exercise would be for each authority to identify the impact that intervention has had on the market. If, as is foreseeable, there will remain some degree of divergence in how agencies select cases or apply remedies, this creates a setting where superior practices can emerge. This is something which is signaled as a task for the working group on pharmaceutical mergers where one of the points of cooperation is to explore what divestitures work and what makes for a successful divestiture buyer. This harnesses divergences that result from the application of different standards as a mechanism for identifying, ex post, what has worked best and encourages the refinement of national approaches in the light of lessons learned from ex post review.
VI. CONCLUSION
Too often talk of cooperation leads to lofty statements of ambition and vague discussions. The key point we make here is that nothing facilitates cooperation like agency officials working together and delving into the details of how markets are to be regulated by considering live cases. Coordination with other agencies in this process can enrich the regulatory process and dialogue with the regulated entities can reduce information asymmetries and lead to better informed intervention that balances the policy of opening markets with the economic freedom of the firms being regulated.
The way forward we suggest is to use existing communication channels that competition agencies have developed, including sector regulators when these enforce specific rules to regulate competition, and coordinate on cases. When there are overlapping investigations, agencies will have the incentive to collaborate closely to ensure a coherent response. As the set of comparable cases grows, this can be used to reflect on how to achieve greater convergence in analysis as well as learning among agencies: learning about standards for assessment and about what interventions produce superior market outcomes.
Thus, rather than creating a novel legal institution (either in the form of a multilateral treaty or another transnational organ) or extending the ICN to include digital regulators, it is argued that much progress can be achieved via a voluntary, multilateral engagement among agencies that prioritize digital market regulation. The successful modalities for cooperation, coordination and convergence that have been discovered in merger practice during the past thirty years can be replicated when multiple agencies confront the challenges of regulating competition in digital markets.
[1] Professor of Competition Law, Tilburg University, Tilburg Law and Economics Center.; and PhD candidate, Tilburg University, respectively.
[2] R. Brandenburger “Transatlantic Antitrust: Past and Present” (2011) 2(1) Journal of European Competition Law and Practice 78.
[3] Agreement between the Government of the United States of America and the Commission of the European Communities regarding the application of their competition laws [1995] OJ L95/47, Article II(3).
[4] Ibid. Art IV.
[5] Report from the Commission to the Council and the European Parliament on the application of the agreements between the European Communities and the Government of the United States of America and the Government of Canada regarding the application of their competition laws 1 January 2000 to 31 December 2000 (2002) 45 final.
[6] U.S. Dept of Justice, “Justice Department Will Not Challenge Cisco’s Acquisition of Tandberg” (29 March 2010).
[7] This was updated in 2011 and is available here: https://competition-policy.ec.europa.eu/system/files/2021-06/US-EU_cooperation-in-mergers_best_practices_2011_en.pdf.
[8] Report on the application of the agreements between the European Communities and the Government of the United States of America and the Government of Canada regarding the application of their competition laws 1 January 2001 to 31 December 2001, COM (2002) 505 final p.3.
[9] Federal Trade Commission, FTC Announces Multilateral Working Group to Build a New Approach to Pharmaceutical Mergers (Press Release, 16 March 2021).
[10] The merger was cleared by the European Commission, but it is subject to analysis as a completed merger in Australia.
[11] H. Abu Karky, “The Impact of the International Competition Network on Competition Advocacy and Global Competition Collaboration” (2019) 40 ECLR 10.
[12] M. Coppola, E. Kraus, C. Lagdameo, P. O’Brien & R. Tritell, “(Nearly) A Century with the ICN,” Competition Policy International (2020).
[13] H.M. Hollman & W.E. Kovacic, “The International Competition Network: Its Past, Current and Future Role,” (2011) 20 Minnesota Journal of International Law 274.
[14] Abu Karky (above n 11) states that some agencies he sampled expressed the view that the ICN needs “fresh air.” In some respects the Network appears to be directed by the same group of people and certain agencies and people with critical opinions on matters pertaining to its mandate and operation are not heard. It should be noted that the group of respondents to this research was relatively small (eighteen members), and that these respondents were often smaller countries with younger competition authorities. The results however remain important in showing the limits of global organizations coordinating many jurisdictions.
[15] OECD, Thinking out of the Competition Box: Enforcement Co-operation in Other Policy Areas, OECD Competition Policy Roundtable Background Note (2022).
[16] Directive 2019/1 of the European Parliament and of the Council of 11 December 2018 to empower the competition authorities of the Member States to be more effective enforcers and to ensure the proper functioning of the internal market (ECN+ Directive).
[17] For a preliminary comparative assessment see Monti “Taming Digital Monopolies: A Comparative Account of the Evolution of Antitrust and Regulation in the European Union and the United States” (2022) 67(1) Antitrust Bulletin 40 and Monti and Ruiz Feases, “The case against Google: Has the U.S. department of justice become European?” (2021) Antitrust 26.
[18] For instance, the positioning and representation of Google Shopping results, the use of anti-fragmentation and Mobile Application Distribution Agreements, as well as revenue sharing agreements to ensure de facto exclusivity of Google Search as the access point for internet for consumers. conduct; United States v. Google LLC, Case 1:20-cv-03010, filed 20 October 2020.
[19] CASE AT.40153, E-book MFNs and related matters (Amazon), Commission Decision of 4 May 2017, C(2017) 2876 final.
[20] FTC, v. Facebook Inc., Case No.: 1:20-cv-03590-JEB (doc 75-1), filed 19 August 2021.
[21] B6-22/16, Facebook, Exploitative business terms pursuant to Section 19(1) GWB for inadequate data processing, Decision by the Bundeskartellamt of 6 February 2019.
