By Dr. Thomas Weck (German Monopolies Commission)1
[Abstract:] Digital platform markets pose novel issues when it comes to enforcing the competition rules. A draft bill to amend the German Competition Act addresses two major issues – tipping and ecosystems. A third issue is being discussed at the EU level: information advantages that large platform operators may enjoy vis-à-vis other market participants and vis-à-vis enforcers. The German Monopolies Commission made a proposal to deal with this latter issue.
Both in the EU and in Germany, there is a growing consensus that the rules to counter the abusive conduct of large digital platforms need to be revised. In Germany, the Federal Ministry for Economic Affairs and Energy recently proposed a bill for a 10th amendment to the Competition Act to that end.2 Taking into account the discussion at EU level, the question arises: what’s in this amendment for the EU? Well, the short answer is that it could be more!
In order to understand which legal reforms are promising, one has to understand why digital platforms raise novel questions regarding competition enforcement (Section II). On this basis, one has to consider what can be done with the existing abuse prohibition (Article 102 TFEU) to address the questions of substance, and where there may be a need to supplement the existing rules (Section III). The same exercise is necessary regarding any procedural obstacles to enforcement (Section IV). To conclude, the German draft identifies the right issues, but the proposed solutions fall short of what’s actually needed (Section V).
II. What are the Issues?
Digital platforms pursue a business model that can be ambivalent from a competition policy perspective. Platform markets are quite often highly concentrated. These markets may, on the one hand, be highly efficient. On the other hand, however, competition problems may arise, as large platforms may enjoy structural advantages that their competitors are not able to match, and that also give the platforms a head start towards the authorities in competition proceedings.
Digital platforms target several groups of different users that interact with each other via the platform. Some platforms, for example social networks, mediate between users of the same type (“one-sided”). They can also bring together different user groups who either offer online content or search for it, as is the case with search or trading platforms (“multi-sided”). Digital platforms with a data-driven business model derive their market power not just from (direct and indirect) network effects, but also from the competitive relevance of the data aggregated on the platform. This is at least true where the platform uses the data exclusively.3
The larger a platform is, the more difficult it is for others to compete with it. The size of a platform therefore has both positive effects for its users, but also negative effects due to reduced competition. Platform markets are not exclusively found in digital services. However, competition issues can arise here due to the often hardly geographically limited markets and rapid and sustainable growth of platforms due to the exploitation of network effects.
The competition issues provoked by large platforms can be of a structural nature and be twofold:
- One potential issue is that the market may permanently “tip” in favor of a platform, meaning that the market position of the platform is not contestable anymore.
- Another potential issue is that digital platforms leverage their market power into other markets to create more-or-less unassailable “ecosystems.”
To be sure, ecosystems may be welcome in many cases. Take the ecosystem created by Google around Android, for example. There, consumers and mobile device manufacturers benefit from a large number of apps compatible with the operating system and, conversely, app developers benefit from high consumer demand. A critical view of ecosystems is justified, however, where such an ecosystem squashes competition from outside the ecosystem. This concerns both (i) the ecosystem itself, i.e. competitors find it harder to establish an ecosystem once consumers are locked in to the ecosystem of a dominant provider, and (ii) markets that are part of the ecosystem, as a dominant provider can foreclose competitors by leveraging its market power to these markets.4
A potential third issue is related to the exclusive access to data that large platforms with a data-driven business model possess. In practice, this is less an issue of abusive behavior (in substance), but more an issue of finding out about any abusive behavior (in competition proceedings). Especially dominant platforms following a data-based business model may have information advantages, both vis-à-vis its competitors and its users, and vis-à-vis the authorities. The competition authorities certainly have comprehensive powers to gather information in administrative proceedings. However, it may still be difficult for them, as outsiders, to understand for which purposes a dominant platform uses the data available to it. In addition, market conditions in the digital sector are changing rapidly, and market players are constantly adapting their behavior to the changing circumstances. This increases the difficulties for the authorities to keep pace with the behavior of the platform companies and other market participants when investigating the facts of the case (i.e. it is a moving target).
III. Tipping and Ecosystems – The German Ministry’s Answer
In its existing form, the German legislative project focuses on the substantive issues flagged above. The draft bill of the Ministry takes up scientific impulses and findings from both EU and German case practice. On this basis, the Ministry proposes two new rules to address the issues of “tipping” and of “ecosystems.”5
Regarding the “tipping” issue, the Ministry proposes a new Section 20(3a) that will come into play where competitors of companies with superior market power are prevented from achieving economies of scale themselves. The new provision will prohibit creating any “unfair impediment” for competitors. Section 20(3a) is meant to apply before dominance within the meaning of Article 102 TFEU has developed.6 It is intended to keep markets open and prevent dominant positions.7 The new provision corresponds in principle to a recommendation of the expert opinion on “modernizing the law on abuse of market power.”8
An open question, however, is what the authorities can do if they miss the right moment to intervene and the market permanently tips in favor of the relevant platform. Imposing a fine may not be a convincing remedy in that case. Unfortunately, the draft bill does not address this issue.
Regarding digital “ecosystems,” with a new Section 19a, the draft bill aims to prevent platforms from using their market position and the economic power in certain markets strategically to restrict competition in other markets.9 The proposed rule is meant to address problems that may arise when the behavior of platform operators fosters the establishment of anti-competitive structures, for example in new markets, without the platform operators necessarily being already dominant in all these markets.10
The proposed Section 19a combines many individual recommendations of expert reports and the outcomes of various (traditional) abuses of platform companies investigated by the competition authorities, but in some cases also goes beyond this.11 In order to establish a possible infringement under Section 19a, the Federal Cartel Office (“FCO”) will have to conduct a two-stage procedure. It will first have to determine the “overriding importance of the relevant company for the overall market,” and then any abuse that may be prohibited. The concept of “paramount cross-market significance” (para. 1) is so far unknown to German abuse law.12 The prerequisites for such a cross-market significance are, however, only outlined in Section 19a(1) with reference to criteria that are also relevant for establishing a dominant position. In addition, it is left to the FCO to develop criteria for a differentiation from traditional market power. The abuse groups listed in Section 19a(2) are broadly defined and address very different types of behavior.
The most notable addition to existing abuse law is certainly Section 19a(2) No. 1, which prohibits self-preferencing behavior without requiring a finding of exclusionary effects. This prohibition goes beyond the Google Shopping case after which it has been modeled.13 Indeed, in a market that is permanently tipped in favor of one dominant platform operator, it may be impossible to reverse the foreclosure effects operating to the detriment of other companies. Thus, it can be presumed that self-preferencing in such a market can bar competition on the merits.14 The dominant platform operators are here in a similarly strong position as a network operator in the position of a natural monopolist. Many of the other abuse groups (Section 19a(2) Nos 2-4) overlap more or less with the existing abuse prohibition in Article 102 TFEU. The only exception is the last prohibition (Section 19a(2) No. 5), which relates to information asymmetries and, like Section 19(1) No. 1, does not require any showing of an impact on competition whatsoever.15 In this case, however, it remains obscure why the provision is included in a statute protecting competition.
The large variety of abuse groups means that it will largely be left to the FCO to determine what the common characteristics of abuses under Section 19a(2) are. It must be expected that, in the absence of a clear and uniform approach and due to the resulting legal uncertainty, the new provision will keep the courts busy for years until clarification may be eventually reached.
With regard to the legal consequences where an abuse has been established under the proposed Section 19a, the draft bill explicitly refers to the already existing enforcement powers of the FCO. However, it is unclear whether these powers are sufficient to be able to take effective countermeasures where abusive behavior by platform companies has had a negative impact on the market structure. It is also conceivable that it might be necessary to foresee ongoing monitoring and regulation of the companies’ behavior with instruments that go beyond the existing competition act rules.
After all, while the proposed Section 20(3a) is rather straightforward, the proposed Section 19a raises a number of fundamental questions. It is doubtful whether the provision can be operationalized at all in a way that will contribute to effective abuse control.
IV. Platform Information Advantages – The Monopolies Commission’s Answer
An expert report prepared for Commissioner Vestager notes that platforms, due to their mediating function, can act as rule-setters for the platform users. If users interact with platforms in a dominant position, it is difficult for them to switch to alternative offers. Dominant platforms can therefore exploit their rule-setting function to the detriment of their users.16 At the same time, transparency can likewise pose a competition policy issue.17 As highlighted before, dominant platforms with a data-based business model may have information advantages, both vis-à-vis their users and vis-à-vis the authorities.
In the view of German Monopolies Commission, an expert body advising the German government and legislature, it could be justified to extend the obligations of the relevant companies to provide information in the case of requests for information by the competition authorities. The Monopolies Commission made a proposal according to which the relevant companies not only would have to provide the data explicitly requested by the authorities, but they would also be obliged to a reasonable extent to provide information from their sphere of influence which the competition authority cannot determine on its own initiative. If these obligations were not fulfilled, the law could empower the competition authority to draw conclusions from the free assessment of evidence and to decide on the merits of the case without taking into account the information not provided.18
To conclude, the draft bill of the German Federal Ministry for Economic Affairs and Energy is a notable contribution to the discussion, but it does not address all issues that could have been tackled concerning large digital platforms. The proposed provision of Section 20(3a) is welcome as it may help to prevent “tipping,” and to keep markets contestable. Open questions at this stage include how a similar provision could be implemented at EU level, and what the right remedies would be if competition authorities miss the right moment to intervene. The new Section 19a, in contrast, represents a crude mixture of concepts relating to the development of platform “ecosystems.” If anything, then, the proposed prohibition of self-preferencing appears to be the most interesting addition to existing abuse law. Regarding the information advantages of platforms with a data-based business model, it may be justified to limit the investigative obligations of the authorities by shifting the burden of providing information. The German Monopolies Commission has made a proposal to that end, which could serve as a model also for the EU.
1 Dr. Thomas Weck, LL.M., is Lead Analyst at the German Monopolies Commission in Bonn. This contribution is based on the German Monopolies Commission’s Policy Brief No 4, “10th Amendment to the Competition Act–Meeting Challenges in Digital and Regional Markets!”, January 2020. Nevertheless, the contribution reflects exclusively the personal views of the author.
2 German Federal Ministry for Economic Affairs and Energy, Referentenentwurf eines Zehnten Gesetzes zur Änderung des Gesetzes gegen Wettbewerbsbeschränkungen für ein fokussiertes, proaktives und digitales Wettbewerbsrecht 4.0 (GWB-Digitalisierungsgesetz), date of last edit: July 10, 2019, 18:14 (“Ministerial Draft Bill”).
3 German Federal Court of Justice (Bundesgerichtshof), Judgment of October 8. 2019, KZR 73/17 – Werbeblocker III, headnote b; see also paras. 24 ff.
4 See ECJ, Judgment of February 21, 1973, Europemballage Corporation and Continental Can Company v. Commission (6/72, ECR 1973 p. 215), ECLI:EU:C:1973:22, paras. 26, 29; see also Judgment of October 3, 1985, CBEM / CLT and IPB (311/84, ECR 1985 p. 3261), ECLI:EU:C:1985:394, para. 26 (at the end).
5 Schweitzer, Haucap, Kerber & Welker, Modernising the Law on Abuse of Market Power, Final Report of August 29, 2018; Commission “Competition Law 4.0,“ A New Competition Framework for the Digital Economy, Report of September 2019.
6 Regarding the requirements of a dominant position, see ECJ, Judgment of February 14, 1978, United Brands v. Commission (27/76, ECR 1978 p. 207), ECLI:EU:C:1978:22, para. 65.
7 Ministerial Draft Bill, supra, pp. 85-86.
8 Schweitzer, Haucap, Kerber & Welker, supra note 4, pp. 62-64.
9 As regards “economic power,” cf. Art. 2(1) lit. a and b of the EU Merger Regulation (Reg. No 139/2004) for the area of merger control.
10 Ministerial Draft Bill, supra, p. 73.
11 See Ministerial Draft Bill, supra, p. 75 ff., expressly referring to recommendations of the Commission “Competition Law 4.0” and EU Commission, Case No. 39.740 – Google Shopping; FCO, Case No. B6-22/16 – Facebook.
12 See, however: UK Government, Unlocking Digital Competition, Report of the Digital Competition Expert Panel, March 2019, paras no 2.10, 2.17, 2.24 ff.: “establish a digital platform code of conduct, based on a set of core principles. The code would apply to conduct by digital platforms that have been designated as having a strategic market status”; further, see Ministerial Draft Bill, supra, p. 57; Steinberg in: Wirtz & Steinberg, WuW 2019, 606, 611.
13 Ministerial draft bill, supra, p. 78.
14 Commission “Competition Law 4.0,” supra, p. 51.
15 See also UK Government, supra, p. 61 and p. 64, regarding the principle of “rankings and reviews […] on a fair, consistent, and transparent basis.” The Ministerial draft does not explicitly refer to in the UK Report in this context.
16 See Crémer, de Montjoye & Schweitzer, Competition Policy for the Digital Era, Final Report, 2019, pp. 60-62.
17 Id., p. 64.
18 German Monopolies Commission, Policy Brief No 4, supra, p. 5.