Targeted at Big Tech, Section 19a of the Competition Act, Germany’s new antitrust tool for dealing with large digital platforms, rebalances power in favor of the German competition authority. Under the new tool, the authority may declare that a firm is of “paramount significance for competition across markets” and prohibit it from certain specified practices presumed to be unlawful. The addressed firms carry the burden of proving the practice’s countervailing procompetitive or efficiency-enhancing effects. Decisions by the German competition authority under Section 19a can only be challenged at the German Federal Court of Justice as the first and only avenue of appeal. We identify the advances and shortcomings of this new tool, as well as the opportunities and risks, when it comes to employing it in the field.
By Jens-Uwe Franck & Martin Peitz1
Around the globe, legislatures have become active in providing competition authorities with new tools and resources to curb the market power of Big Tech. Germany is one of the countries at the forefront of these developments. On January 18, 2021, the German legislature finally adopted the Tenth Amendment to the German Competition Act2 (“Gesetz gegen Wettbewerbsbeschränkungen”), which includes a number of legal changes aimed at protecting competition in times of digitalization. Its major innovation is the competition instrument enshrined in section 19a, which will give new powers to the Bundeskartellamt, the German competition watchdog, when dealing with large digital platforms. The new tool deviates in substance and procedure from traditional competition law and approaches the role of a regulatory instrument targeting the digital platform industry. In essence, section 19a of the Competition Act is the functional equivalent of the Digital Markets Act (“DMA”)3 proposed by the European Commission.
I. THE MECHANICS OF THE NEW “19a TOOL”
As a first step, the Bundeskartellamt needs to decide whether a firm is of “paramount significance for competition across markets.”4 Such a decision will be effective for five years.5 Once it has done so, it can then prohibit the firm from engaging in certain types of conduct perceived to be anticompetitive.6 The declaratory decision and the prohibition decision can be combined. The new law contains an exhaustive list of seven types of practices the German competition authority may prohibit:
(1) Self-preferencing by vertically integrated firms;
(2) Hindering supply or sales activities of other firms (even if they are not competitors);
(3) Hindering competitors in markets where the 19a firm is not dominant but where it can rapidly expand its position (“envelopment”);
(4) Using collected data to raise market entry barriers or requiring users’ permission for such use;
(5) Hindering competition by impeding interoperability or by making data less portable;
(6) Withholding information on the 19a firm’s performance — this is particularly relevant for intermediation services, concerning information on consumers’ click behavior or parameters determining how the firm ranks goods and services;
(7) Exploiting business customers.
Except for numbers (5) and (6), the descriptions of these types of prohibitable conduct are each complemented by two illustrative examples in the law. For example, tying and bundling may be prohibited to prevent envelopment strategies by a 19a firm. However, the firm can try to demonstrate that behavior that comes under this list is “objectively justified” — it carries the burden of proving this.
The Bundeskartellamt’s decisions under section 19a can be challenged before the Bundesgerichtshof (“BGH”), the German Federal Court of Justice. In these cases, the court will decide as the first and only avenue of appeal.7
A 19a firm that infringes a prohibition decision commits an administrative offense8 and may be fined by the Bundeskartellamt. Other market players may bring actions for injunctions9 or damages.10
II. GETTING TO THE HEART OF THE MATTER: CONCEPTUAL DEVIATIONS FROM TRADITIONAL COMPETITION LAW
If we compare the new competition tool enshrined in section 19a of the Competition Act with traditional instruments of competition law, four distinct features stand out in particular:
(1) Traditionally, competition law has addressed market power. Article 101 TFEU and section 1 of the German Competition Act and other equivalent provisions under Member States’ laws prevent the establishment and use of market power through agreements and other forms of coordination between firms that are used to restrict competition. Article 102 TFEU and section 19 of the German Competition Act, as well as other equivalent domestic positions, target the unilateral conduct of firms that dominate a defined market.11 In contrast, section 19a of the Competition Act addresses unilateral practices by digital ecosystems or platforms because of their specific position as intermediaries and gatekeepers, as firms that control an interface between markets, regardless of whether they actually dominate one defined market.
(2) Pursuant to the general rules for competition enforcement under German law, as a matter of principle,12 in abuse cases the competition authority is entrusted with the task of determining, measuring, and balancing pro- and anticompetitive effects and/or the efficiency losses and gains of a scrutinized practice. On a case-by-case basis, firms may be obliged to provide information under procedural obligations to cooperate and, if they fail to do so, must accept that the possible procompetitive effects or efficiency gains are not (fully) appreciated by the authority. In contrast, section 19a of the Competition Act provides for an explicit shift of the burden of proof applicable to an extensive list of broadly formulated practices that are presumed to be abusive.
(3) Conventional competition law provides for standards and rules, established by the legislature and refined through authorities’ and courts’ practice, which are directly binding on the market players that are addressed and which are directly applicable and, therefore, can also be directly invoked and enforced by private parties before the courts. This is different in section 19a of the Competition Act, where the prohibition of the listed practices in an actual case depends on an intervention by the Bundeskartellamt, which enjoys discretion in this regard. The role of the competition authority thus comes close to that of a regulatory authority.
(4) While decisions by the Bundeskartellamt are usually subject to a two-level system of judicial review, the authority’s 19a decisions can only be reviewed by the Federal Court of Justice acting as the court of first and last instance.
III. A TOOL ONLY MEANT FOR LARGE DIGITAL PLATFORMS?
Section 19a allows firms that are not (yet) dominant in any market to be targeted. In the legislative memorandum13 accompanying the finalized version of the Tenth Amendment to the Competition Act, the potential addressees of the 19a tool are described as firms that, “for example due to their financial, technical or data-related resources or as cross-market digital ecosystems or platforms, are particularly capable of extending their position of power across market boundaries or securing their unassailable position.” Here and elsewhere, it becomes clear that the legislature introduced the new tool to address risks to competition in digital markets. While some of the aspects listed under section 19(1) of the Competition Act are not specifically linked to the digital economy (dominance on one or several markets; financial strength and access to other resources; vertical integration and activities on otherwise related markets), such a connection is apparent in relation to the “access to data relevant for competition” and “intermediation power” criteria.
More to the point, the reference to section 18(3a) of the Competition Act clarifies that only firms that are “active to a significant extent” as two-sided platforms or networks may be targeted by the 19a tool. As explained in the memorandum accompanying the original draft, the “criterion of significance” is supposed to ensure that only firms “with a focus on digital business models are subject to the rule.”14 What is more, the legislative memoranda repeatedly emphasize that the provision only targets a “small group of firms” or “digital ecosystems.”15 Ultimately, therefore, one can assume that the legislature indeed had only Big Tech in mind when drafting the 19a tool. Yet, since the criteria are formulated with a certain flexibility, it will be interesting to watch which digital firms will actually find themselves in the crosshairs of the German competition authority over the next few years.
IV. EXTENSIVE LIST OF CONDUCTS PRESUMED TO BE ABUSIVE
The types of behavior that the Bundeskartellamt may prohibit 19a firms from engaging in are deliberately drafted very broadly. They also include scenarios in which these practices may be procompetitive and consumer welfare-enhancing. For instance, final consumers may benefit from vertically integrated offers that are shown prominently and provide a minimum quality of service such as quick delivery, adequate packaging, or authentic products. Thus, self-preferencing may to a certain extent be in the interest of final consumers.
Another example is the use of bundling as a strategy to enter a market. In the presence of network effects, the best hope for limiting dominance by a firm in one market might be that (another) 19a firm challenges it. More specifically, if Microsoft falls under section 19a, its hands may be tied with respect to its search engine, Bing, which in any case has a hard time challenging Google Search in many countries.
Yet another example concerns allegedly exploitative behavior. A search engine that prioritizes news outlets that provide unrestricted access or allow snippets to be placed in conjunction with the search results may be disliked by some media outlets but has clear consumer benefits. It is quite remarkable that in the legislative memoranda these and other potentially procompetitive or pro-consumer scenarios are hardly mentioned or inadequately treated. For instance, with regard to self-preferencing it is acknowledged that exclusive integration of a firm’s own offers may be justified if necessary “for the utility of core functions of the hardware,” such as the “telephone, camera, message feature or file management of a mobile phone.” The example of the mobile phone reveals that consumers’ expectations of what constitute a product’s “core functions” evolve dynamically and are in fact the result of competition between more- or less-integrated product designs. Therefore, the concept of “core functions” merely begs the question of where to draw the line between consumer welfare-enhancing and procompetitive integration of a firm’s own offers and abusive self-preferencing.
Whether the 19a tool will lead to socially desirable outcomes will very much depend on whether the Bundeskartellamt strikes the right balance between the pro- and anticompetitive effects of these behaviors. Ideally, as a result of the authority’s practice and the case law of the Federal Court of Justice a subset of rules will develop over the years that specifies which conduct is prohibited under section 19a(2) of the Competition Act, taking into account in particular the considerable differences in the monetization models pursued by the digital platforms addressed.16 However, we see the risk of a rather mechanical assessment that certain practices fall under the law and will therefore be prohibited based on a quasi–per se rule that does not sufficiently take into account the circumstances of the individual case. The hope is that this will not happen and that the reversal of the burden of proof both opens the door to anticompetitive effects receiving the consideration they deserve and addresses the asymmetric information problem the Bundeskartellamt has faced elsewhere. The law, however, is rather unclear about the basis on which to do the balancing.
In particular, considering the case of two-sided platforms that cater to business users and final consumers — for example, on Amazon Marketplace the business users are the sellers — the question will often be whether or when there should be a netting of countervailing welfare effects for these two groups. The reform is a missed opportunity to address this issue and to provide guidance. The legislature essentially left it at the somewhat vague comment that in the context of “objective justification” a balancing of interests will be required, considering, on the one hand, the goal of protecting free competition and, on the other hand, the legitimate freedom of business and possible procompetitive elements of the conduct in question.
The memo that accompanied the original draft of the Tenth Amendment had stated more in detail that particular weight should be given to the long-term objectives associated with section 19a — limiting economic power, keeping markets open, and protecting the competitive process — as opposed to, in particular, short-term efficiencies for the benefit of businesses and consumers.17 Interestingly, this passage was not taken up by the parliamentary committee in its final memorandum on the bill. Nevertheless, there is a widespread perception that 19a firms that put forward procompetitive effects and/or efficiency gains should find it difficult to rebut the presumption of abusiveness. We reiterate that it would be a serious flaw if the practices included in section 19a were to evolve into quasi–per se prohibitions.
V. PROMPT AND EFFECTIVE INTERVENTION: ABRIDGED JUDICIAL REVIEW
In principle, decisions of the Bundeskartellamt can first be challenged before the Düsseldorf Higher Regional Court (Oberlandesgericht), where six specialized divisions have been set up to handle these cases. The Higher Regional Court’s decisions may then in turn be appealed against before the BGH.
It is most remarkable that this well-established legal protection mechanism will not apply to decisions under section 19a of the Competition Act. Bypassing the Düsseldorf Higher Regional Court, the Federal Court of Justice will decide as a court of appeal in the first and last instance. The legislature has gone to great lengths to justify this abridged judicial review.18 It is argued that the particularities of digital markets require particularly swift intervention to be effective. It is assumed that not even the Bundeskartellamt’s option to order immediate enforceability of its decisions suffices to do justice to this necessity. The legislature refers at this point to the current Facebook litigation: the firm succeeded with its application for an order of suspensive effect before the Higher Regional Court,19 which, however, was overturned by the Federal Court of Justice.20 Facebook then initiated a second summary proceeding, whereupon the Higher Regional Court again ordered suspensive effect by an interim decision.21 After all of that, Facebook has now withdrawn its application, so the Bundeskartellamt’s order22 is temporarily enforceable, albeit after almost two years of interim legal proceedings. In light of this, the legislature saw the risk that, given the options for judicial review currently available, the enforceability of the Bundeskartellamt’s future 19a decisions could remain in dispute for a number of years.23
The single-tier judicial review of decisions by public authorities is nothing unheard of under German law and appears not to conflict with constitutional law.24 Yet, what is lost by reducing judicial review to one appeal instance must not be overlooked or underestimated. The Federal Court of Justice will have to rule on decisions of the Bundeskartellamt, which will significantly interfere with the entrepreneurial freedom of the platform operators concerned and which will heavily influence the development of digital markets in Germany and (possibly) beyond. As the legislature itself emphasized, the application of the new competition tool will typically involve complex questions of fact and unresolved legal questions. What is more, the number of cases is expected to remain small.25 Therefore, for a prudent development of the law under section 19a of the Competition Act, we submit that it would have been fruitful for the Düsseldorf Higher Regional Court, which has been familiar with competition proceedings for many years, to deal with these cases first. In any case, it would not be a good reason to bypass the court just because it has occasionally been quite critical of the competition authority’s views, especially in the course of the aforementioned Facebook proceedings.
VI. TOWARD EVER-INCREASING REGULATORY FRAGMENTATION IN THE EU INTERNAL MARKET?
By moving ahead and enacting section 19a of the Competition Act, the German legislature has deliberately put pressure on the EU to more effectively address the perceived competition problems caused by the big digital platforms. The recently proposed EU Digital Markets Act (“DMA”) is intended to be the functional equivalent of Germany’s new competition tool. To be sure, it is by no means settled in which form and when or whether at all the DMA will enter into force. Yet, in the EU’s legislative process, the Commission’s new powers to regulate Big Tech under the DMA will be benchmarked against section 19a of the Competition Act. Suffice it to mention here two differences that seem to be apparent: first, the circle of potential addressees of the DMA appears to be wider, in particular in view of the qualitative criteria, which trigger a (rebuttable) presumption that the firm may be designated as a gatekeeper.26 Second, the regulatory design of the DMA is much closer to conventional ex ante regulation, as a firm that is designated as a gatekeeper is automatically obliged to comply with the obligations laid down in the DMA,27 while the gatekeeper may apply for a suspension where compliance “would endanger, due to exceptional circumstances beyond the control of the gatekeeper, the economic viability of the operation of the gatekeeper in the Union”28 or request an exemption for overriding reasons of public interest.29
What is more, with the 19a tool, the German lawmakers have opened up a testing ground for digital platform regulation. Ideally, other rule-makers, particularly in the EU, will watch and learn from the wanted and unwanted effects of the Bundeskartellamt’s future 19a interventions.
It is foreseeable, however, that national competition enforcement and ex ante regulation will lead to increasing differences in digital platform regulation across EU Member States. While the EU could react to such fragmentation by striving for a full harmonization of domestic laws including competition law, this is not to be expected. The EU’s DMA proposal aims at harmonizing Member States’ gatekeeper regulation, which is adopted “for the purpose of ensuring contestable and fair markets.”30 Pursuant to the proposal, Member States will remain free, however, to adopt stricter gatekeeper regulation to pursue “other legitimate interests.”31 Further, national competition law will also remain unaffected by the DMA.32 For the foreseeable future, digital platforms and their users will therefore have to bear the costs of fragmented rule-making in the EU’s internal market. When deciding in a particular case whether or not to impose a certain rule on a digital platform, national competition authorities cannot be expected to consider these costs in any meaningful way. However, we foresee that, in the medium term, some “soft” EU-wide convergence will result from exercise, exchange, and experience.
VII. CONCLUDING REMARKS
The 19a tool is a revolution in German competition law. It sits between traditional competition law and sector regulation. Targeted at Big Tech, it aims to rebalance the power between the Bundeskartellamt and powerful firms. In particular, the competition authority benefits from a reversal of the burden of proof. We see this a positive innovation in light of the information asymmetry between the watchdog and Big Tech and the resources available to the latter.
However, bypassing the specialized competition law court and making the German Federal Court of Justice the first and only appeal instance does not appear to us to be the last word in wisdom. What is more, drafting the practices covered in wide terms that may be interpreted as per se prohibitions might lead to outcomes that particularly hurt final consumers because special interest groups obtain undue considerations. Whether the tool is employed for good use will mostly depend on the Bundeskartellamt. The authority already took the opportunity to put its new instrument to the test and initiated its first 19a proceedings against Facebook, investigating the linking of Oculus, a subsidiary of Facebook that produces virtual reality devices, with Facebook’s social network.33 Only time will tell whether the 19a tool has sharp teeth and to what effect it will be used.
Notably absent from Germany’s reform is an update of merger control rules to better meet the challenges posed by digital platforms. The new section 19a aims at prohibiting certain behaviors, but the amended Competition Act does not provide sharpened tools to go against acquisitions by Big Tech firms. While the Commission seems to be equally reluctant to engage in a substantial reform of the EU Merger Regulation, it is remarkable that, under the proposed DMA, designed digital gatekeepers will have an obligation to notify any intended acquisition regardless of whether it is notifiable under the EU Merger Regulation or national merger rules.34 This fits in with the announcement by Commissioner Vestager that from mid-2021 the Commission will accept referrals by Member States even if the mergers in question would not be notifiable under national law.35 In the light of this new policy as regards Article 22 of the EU Merger Regulation, it would have made sense to introduce a corresponding information obligation under German competition law. However, one may assume that the Commission will find a way to signal to a Member State when it considers a referral request to be appropriate.
What are the repercussions for platform users in Germany and beyond? On the one hand, Germany as a mid-sized country may well be important enough for the platforms to comply with its legal peculiarities and to adapt their business models instead of simply withdrawing from the market. On the other hand, stricter (and conceivably too far-reaching) regulation in Germany is unlikely to significantly influence platforms’ global investments in new services — unless the 19a tool triggers a domino effect in other countries.
1 Jens-Uwe Franck is Professor of Private Law, Commercial Law and Competition Law at the University of Mannheim and Senior Member of the Mannheim Center of Competition and Innovation—MaCCI. Martin Peitz is Professor of Economics at the University of Mannheim and a director of the Mannheim Centre for Competition and Innovation—MaCCI. He is a research fellow of CERRE, CEPR, and CESifo and a research associate of ZEW. A shorter version of this article has been published under the title “Taming Big Tech: What Can We Expect from Germany’s New Antitrust Tool?” on promarket.org, the competition blog of the Stigler Center. The authors acknowledge financial support by the Deutsche Forschungsgemeinschaft through CRC TR 224 (project B05).
2 Bundesgesetzblatt (Federal Law Gazette), January 18, 2021, Part I No. 1 pp. 2 et seq. Available at https://www.bgbl.de/xaver/bgbl/start.xav?start=//*%5B@attr_id=%27%27%5D#__bgbl__%2F%2F*%5B%40attr_id%3D%27bgbl121001.pdf%27%5D__1611317574622.
3 Proposal for a Regulation of the European Parliament and of the Council on contestable and fair markets in the digital sector (Digital Markets Act). COM/2020/842 final.
4 Section 19a(1) 1st and 2nd sentences of the Competition Act.
5 Section 19a(1) 3rd sentence of the Competition Act.
6 Section 19a(2) of the Competition Act.
7 Section 73(5) of the Competition Act.
8 Section 81(2) No 2a) of the Competition Act.
9 Section 33 of the Competition Act.
10 Section 33a of the Competition Act.
11 Note that section 20(1) of the Competition Act extends the applicability of the prohibition of exclusionary practices (as embodied in section 19(1) and (2) no. 1 of the Competition Act) to firms with mere relative market power. Remarkably, due to the recent reform, the provision now applies more broadly against abusive practices by (large) digital practices because: first, the provision can be invoked now by any undertaking and not only by small and medium-sized firms and, second, the concept of relative market power has now been extended to include also the notion of “intermediation power.”
12 A reversal of the burden of proof with regard to a possible objective justification also applies in the case of discriminatory practices covered by section 19 of the Competition Act.
13 Deutscher Bundestag, Drucksache 19/25868, 13.11.2021, Beschlussempfehlung und Bericht des Ausschusses für Wirtschaft und Energie (9. Ausschuss), p. 113. See https://dip21.bundestag.de/dip21/btd/19/258/1925868.pdf.
14 Referentenentwurf des Bundesministeriums für Wirtschaft und Energie, Entwurf eines Zehnten Gesetzes zur Änderung des Gesetzes gegen Wettbewerbsbeschränkungen für ein fokussiertes, proaktives und digitales Wettbewerbsrecht 4.0 (GWB-Digitalisierungsgesetz) of 24.1.2020, pp. 76–77. See https://www.bmwi.de/Redaktion/DE/Downloads/G/gwb-digitalisierungsgesetz-referentenentwurf.pdf?_blob=publicationFile&v=10.
15 See Beschlussempfehlung (n 13), p. 114 and p. 120 (section 19a applies only to a “very small circle of potential addressees”). See also Referentenentwurf (n 14), p. 76.
16 Economic theory has shown that platform incentives with respect to certain practices may well depend on the platform’s monetization model (see, e.g. Teh, 2019, “Platform governance,” available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3521026). This is reflected by a recent opinion piece by Caffarra & Scott Morton (“The European Commission Digital Markets Act: A translation,” voxeu.org, Feb. 5, 2021) in the context of the proposed DMA and also applies to section 19a of the Competition Act.
17 See Referentenentwurf (n 14), p. 80.
18 See Beschlussempfehlung (n 13), pp. 120–122.
19 OLG Düsseldorf August 26, 2019, VI-Kart 1/19(V) – Facebook I, Juris.
20 BGH June 23, 2020, KVR 69/19 – Facebook, Juris.
21 OLG Düsseldorf November 30, 2020, Kart 13/20 (V), Juris.
22 Bundeskartellamt February 6, 2019 B6-22/16, available at https://www.bundeskartellamt.de/SharedDocs/Entscheidung/EN/Entscheidungen/Missbrauchsaufsicht/2019/B6-22-16.pdf?__blob=publicationFile&v=5.
23 See Beschlussempfehlung (n 13), p. 121.
24 The German Constitutional Court (Bundesverfassungsgericht–BVerfG) held that the Constitution does not guarantee a multitiered judicial review. See BVerfG April 30, 2003, 1 PBvU 1/02, BVerfGE 107, 395, 402, Juris, para. 18.
26 Article 3(2) of the DMA proposal (n 3).
27 Articles 5 and 6 of the DMA proposal (n 3).
28 Article 8 of the DMA proposal (n 3).
29 Article 9 of the DMA proposal (n 3).
30 Article 1(5) 1st sentence of the DMA proposal (n 3).
31 Article 1(5) 2nd sentence of the DMA proposal (n 3).
32 See Recital 9 of the DMA proposal (n 3). In particular, it is specified that the proposed “Regulation is without prejudice … to other national competition rules regarding unilateral behavior that are based on an individualised assessment of market positions and behavior, including its likely effects and the precise scope of the prohibited behavior, and which provide for the possibility of undertakings to make efficiency and objective justification arguments for the behavior in question.” Section 19a of the Competition Act appears to meet these criteria and would therefore be exempt from harmonization through the DMA.
33 See the Bundeskartellamt’s press release of January 28, 2021, available at https://www.bundeskartellamt.de/SharedDocs/Publikation/EN/Pressemitteilungen/2021/28_01_2021_Facebook_Oculus.pdf?__blob=publicationFile&v=2.
34 Article 12 of the DMA proposal (n 3).
35 Commissioner Vestager, “The future of EU Merger Control,” International Bar Association 24th Annual Competition Conference, September 11, 2020, available at https://ec.europa.eu/commission/commissioners/2019-2024/vestager/announcements/future-eu-merger-control_en.