By Rupprecht Podszun & Fabian Brauckmann (Heinrich Heine University)1
Germany’s Ministry of Economics has published its draft for the 10th amendment of the German competition act, the Gesetz gegen Wettbewerbsbeschränkungen (“GWB”). While the starting point for the amendment is the implementation of the ECN+ Directive public attention focuses on the rules for digital players. In this contribution, we give an overview regarding the most important changes to be expected. The draft bill, that has been published in the blog D’Kart,2 is supposed to enter into the legislative process soon. People responsible in the Ministry of Economics and Energy expect the bill to be passed in the first half of 2020.
While former amendments of the German competition act were largely of a rather technical nature and did not arouse public interest so much, this is different with the 10th amendment, proposed by the German minister of Economics, Peter Altmaier, a close ally of chancellor Angela Merkel. He and his competition law crew are determined to enact a modern, new competition law that sends out signals to the public that this government takes up the fight for a competitive framework for digital markets. It is also the German ambition to set a precedent for reforms on the European stage.
A History, of Sorts
With the 9th amendment Germany had already introduced rules for platforms, e.g. mentioning the special characteristics that give market power to platforms, such as access to data. Germany had also introduced a transaction value based merger threshold (400 m EUR) so as to catch the next Facebook/WhatsApp merger. The EU does not yet have such a merger threshold and had been advised by the Special Advisers not to introduce one.3
The new reform comes at very interesting times: While the 9th amendment was ahead of time, somewhat, this one now stands in the shadows of the big cases: The three Google cases of the European Commission, the much-discussed Facebook case of the Bundeskartellamt and the Amazon investigations of both institutions. How does a legislator react to what we have witnessed so far?
To complicate matters, Germany was thrown into trouble by a widespread feeling that the German companies, once the pride of the nation, are falling behind in the data economy: SAP was the last big IT champion from Germany, and that company has been around in pre-Internet times already. Car manufacturers seem to have lost touch with latest developments, instead fighting in the self-inflicted Diesel saga. Merger cases of companies like Siemens/Alstom or Tata/Thyssen where the European Commission found that competition from China was not strong enough to allow the mergers to proceed met with criticism in Germany. This is the scene set for the reform of competition laws.
Implementing ECN+
The reform was triggered by the ECN+ Directive 2019/1 of the EU which required changes in procedural law and institutional matters for the antitrust enforcement framework in all Member States. The largest part of the draft is dedicated to implementing these rules. Changes in practice may not be particularly significant. One point where practitioners are troubled is that the Bundeskartellamt, the national competition agency, shall have powers of interrogation during dawn raids in a way that – in their reading – exceeds the requirements of Article 6(1) e) of the Directive (cf. section 59b (3) of the draft). There may be a conflict looming with the right against self-incrimination (nemo tenetur). It has to be noted in this context that in Germany individuals are liable to fines as well as companies. Yet, while the right to remain silent may help the individuals as they may be granted a guarantee by the Bundeskartellamt not to be prosecuted in person, then they may be required to incriminate their company.
The main goal of ECN+, namely to make competition authorities more independent, has already been realised for the Bundeskartellamt. It already works very independently (sometimes to the dismay of politicians), yet some amendments in this regard were still necessary, e.g. a strengthening of the position of the Bundeskartellamt in court proceedings when fines are dealt with. The rules on leniency and fines are now incorporated in detail in the law. The current version of the draft does not solve the conflict of the Düsseldorf Higher Regional Court and the Bundeskartellamt in how to calculate the fine. The Düsseldorf court is often even stricter than the agency since it does not start from the turnover relevant for the offence, as does the agency in its guidelines (that are not binding for the courts).
New Thresholds for Merger Control
There is also considerable news in merger control including amended thresholds. In particular, the high number of non-problematic merger notifications (around 1300 per year) is to be reduced:
“The provisions on formal merger control are being revised in order to make merger control more effective and to enable the Bundeskartellamt to focus on the most relevant mergers. (…) The central objective here is to further relieve small and medium-sized enterprises of their notification obligations.” (p. 59) With the proposed changes, the Ministry expects 20 percent less notifications of mergers.
To this end, the second domestic turnover threshold is raised from 5 million EUR to 10 million EUR. An extra-rule currently contained in the law for smaller independent undertakings that are sold becomes superfluous with this amendment.
The rule in merger control that there is no prohibition of a merger if it only relates to minor markets is extended. Minor markets (or Bagatellmarkt in German) are now markets that have less than 20 million EUR market volume (as opposed to 15 before). Yet, markets with free goods or services are not included in the definition.
Merger control for press publications is eased by raising the turnover thresholds. So far, the turnover of publishers had been multiplied by 8 for the calculation of the turnover thresholds. Now, this is reduced to a multiplication by 4. The multiplication as such serves the aim of having a tighter control of media concentrations. For TV and radio, it remains with factor 8.
Plans that had been around to have some more freedom in picking merger cases with a view to smaller mergers have not made it into the law, and there has been no provision proposed so far on “killer acquisitions.” It may be up to legislators in Parliament to step forward with a proposal.
The time frame for a phase II investigation in Germany is prolonged from four months to five. At the same time, extensions of deadlines in merger proceedings may not exceed one month in total.
The ministerial authorisation procedure that had become an easy exit for undertakings that had lost their merger case with the Bundeskartellamt will have a new requirement. Now, you can only turn to the Minister for his approval of a banned merger if you sought help from the Düsseldorf court before – at least in interim proceedings.
The Proposed Digital Regulatory Framework
According to the Ministry of Economics, “the present draft contributes to the creation of a digital regulatory framework.” For a start, three minor things:
- In section 19 (equivalent to Article 102 TFEU), the new law cuts the causal link of dominance and abuse in some way. At least the new wording (only visible in the German version) may be interpreted in this sense.
- In section 18, dealing with market power, access to data is acknowledged as a factor for that, not just for platforms, but for all sorts of undertakings.
- There is also a special mention for intermediary services in section 18 now (as had been suggested by a preceding study commissioned by the Ministry).4
Regulating GAFA
The most remarkable change is a new rule in section 19a of the act that implements a competition-oriented regulation for “undertakings with paramount significance for competition across markets” (“UPSCAM”). The rule has been dubbed as “revolutionary” by observers, and it seems to be inspired by the notion of undertakings with strategic market status, as identified in the UK Furman Report.5
Paragraph 1 of section 19a goes into more detail who shall be targeted by the rule. The requirement to be fulfilled is “paramount significance for competition across markets” The draft says in its explanatory part: “It is to be expected, that the determination of a paramount significance for competition across markets can only be made for a few companies and the rule will therefore have a narrowly limited circle of addressees.” (p. 73) It instantly becomes clear who is meant by this rule – the GAFA companies, Google, Apple, Facebook, Amazon. The key component, “across markets,” remains rather vague, there is no further explanation of that. But there are indications. The first requirement is that the undertakings are active to a significant extent on multi-sided markets or with networks. So, this is primarily really a rule for digital players (although other companies can be active on multi-sided markets as well). Indicators for paramount significance across markets are dominance, financial strength, access to resources, vertical integration, activities on other related markets, access to data and “the importance of its activities for third parties’ access to supply and sales markets and its related influence on third parties’ business activities.” This can be interpreted as meaning that the undertaking in question must have a gatekeeping position. Companies operating “digital ecosystems,” locking customers and suppliers into their business model, are the ones targeted here.
The Bundeskartellamt has to establish that an undertaking is in such a position, which brings it close to a traditional provider of infrastructure, a utility. It is not entirely clear from the wording how that procedure actually takes place and what the value of that is (e.g. for private enforcement cases), but the competition agency has to decide this by order and this order may be challenged in courts.
The explanatory memo states: “The background to this is that markets in the digital economy can exhibit strong and rapid concentration tendencies, particularly due to network effects, data advantages and the associated self-strengthening effects, which require early intervention in the event of undesirable developments.” (pp. 72-73)
If an undertaking is identified as an UPSCAM there is a distinct set of powers that the Bundeskartellamt holds to apply (paragraph 2 of section 19a): The UPSCAM may not:
- discriminate in intermediary services;
- impede competition in markets where it is not yet dominant;
- use data for making market entry more difficult;
- demand terms and conditions that allow the use of data of others;
- make portability of data more difficult;
- withhold information from other companies about their success in markets.
In accordance with the objective of the provision, the abuses referred to in paragraph 2 are aimed at those types of conduct which have an increased potential for competitive harm, particularly if they are used by undertakings with paramount significance for competition across markets. In particular, such undertakings have the possibility to use their power positions and resources from other markets to restrict competition on additional markets, thereby also promoting their market position there and ultimately further consolidating their paramount significance for competition across markets. (p. 73)
Undertakings may prove that the behaviour in question is justified, yet they carry the burden of proof for this. The agency may not just work with plain prohibitions but also with interim measures and remedies for section 19a.
Paragraph 3 clarifies that Sections 19 and 20 GWB remain unaffected in their entirety. Conversely, any conclusion to the effect that conduct not prohibited under paragraph 2 would thus also be permitted under sections 19 and 20 of the act is also excluded.
There is some concern among practitioners, that the order that establishes a company as an addressee of section 19a and the prohibition order regarding specific abuses may coincide. The company would only learn of its paramount significance for competition across markets once there is already a prohibition order hitting.
The rule is on the defining line of sectoral regulatory rules on the one hand and standard competition law on the other. It certainly stimulates the imagination of people that favour a sturdier approach in enforcement, yet the success will largely depend on the judiciary’s willingness to make the rule effective.
Access to Data
Access to data has been a major topic of discussions in Germany for some time. This is owed to a mix of motivations. Companies need access to data for innovation, particularly in the field of Internet of Things or smart applications. German companies seem to be particularly data-hungry in this respect – they have all it takes from a hardware perspective, but lack the necessary data. At the same time, the Facebook ruling and the whole controversy surrounding this pointed at the lack of consumer sovereignty in this field and the privacy issues associated with personal data (and regulated by the GDPR). Former head of the Social Democrats, Andrea Nahles, had proposed a Data for all-law including the suggestion to hand out non-personal data as a public good, oblige “data monopolies” to share, and incentivise data sharing models.
It seems that the industrial players were most influential with their plea to the legislator. In section 19 (2) No. 4, the equivalent to Article 102 TFEU, the draft amendment aims at easing the essential-facility-rule. The definition of “essential facility” is now explicitly extended to include data, or networks.
The explanatory memorandum explains the shift from physical infrastructure to a broader notion, including intangible assets: “The previous wording of number 4 was characterised by a narrow understanding of an abusive refusal of access, particularly related to physical infrastructure. (…) The new version, with its more open wording, is intended to make it clear that a refusal of access to platforms or interfaces can also be abusive, as can a refusal to license intellectual property rights.” (pp. 71-72)
This is a significant amendment even though follow-on problems (such as on remuneration and updates) will keep on troubling courts and parties.
Even More Help with Data
Germany has tougher rules, than European competition law, on abusive practices in section 20. In particular, for abusive practices it largely suffices to have superior bargaining power or relative market power. This provision is strengthened, and again it is particularly tailored for access to data.
Firstly, companies may rely on section 20 not just if they are small or medium-sized enterprises. So far, bigger companies were not able to rely on section 20(1). This shall now be changed, so that, just as an example, Allianz, the insurer, or BASF, the chemical company, could use the provision to tackle Google or Amazon (there still needs to be a power asymmetry). The explanatory memo justifies this change with a move from traditional German protection of the Mittelstand to the protection of competition as a process:
Section 20(1) is not primarily intended to protect small and medium-sized businesses, but to protect competition as a process and an institution. (…) The extension of the scope of protection applies to all sectors of the economy, not only to the digital economy, but a particularly great benefit of this change for competition is to be expected there. (…) Section 20 (1) may thus be of particular importance, for example, in the case of obstruction strategies for digital platforms with a gatekeeper position, because large enterprises may also be dependent on such platforms. (pp. 78-79)
Accordingly, it is further clarified that intermediary companies are bound not to impede other companies when they act as gatekeepers for supply and sales markets (new section 20(1) sentence 2).
Thirdly, dependency may also be there if a company depends on access to data for its activities. The draft states that this is even possible if access to such data had not yet been commercialised, i.e. commerce had not yet been opened up at all. This will be a feast for all those who wish to go into fields where they have everything but the data, and where these data are not available on the market.
A Special Provision on Tipping
The answer to many digital cases may be the new section 20(3a), that aims to reduce competition problems through tipping of markets:
“It shall also be an unfair impediment (…) if an undertaking with superior market power on a market in the sense of section 18(3a) impedes the independent attainment of distinctive positive network effects by competitors and thereby creates a serious risk that competition on the merits is restricted to a not inconsiderable extent.”
It aims at digital companies again (section 18(3a) is a rule for multi-sided markets and networks), and it seems to entail a very broad notion of impediment.
On Interim Measures, Cooperation Agreements, Damages, and Consumer Protection
The Bundeskartellamt – just like the European Commission up to recently – has not made use of interim measures. The government sees the cause in a lack of practicability of section 32a in its current version. Accordingly, “[t]he amendment contains a moderate lowering of the conditions for the use of interim measures in order to enable the cartel authorities to intervene more quickly (in particular in the digital economy)” (p. 84).
Whereas up to now section 32a required a prima facie infringement (as expressly required by the parallel European provision in Article 8 of Regulation 1/2003), now a “predominant probability” suffices. This allows the Bundeskartellamt to base its assessment on whether, at the time when the interim measure is adopted, it appears more likely than not, that an infringement will also be established in the main proceedings.
Section 32c reintroduces the comfort letter. Up to 2004, as some readers may remember, the European Commission’s DG COMP championed in issuing comfort letters. The notification system for restrictive business practices had led to the request by parties to rubber-stamp their contracts. The Commission decided that this was to no avail since usually cartels were not notified, but everything else was and the Commission simply was no longer able to carry the burden. This meant that companies lost a lot of legal certainty since there were no more decisions on the day-to-day business of cooperation agreements. Guidelines and case law did not always help.
Today, this has grown into an even bigger problem since some forms of cooperation are new to the undertakings, their lawyers and the agencies. The standard example is data pooling or data sharing. Are you allowed to cooperate in order to make a smart app run – or does that amount to an illicit information exchange? This is often hard to tell. So companies can now turn to the Bundeskartellamt and ask for a decision – to be given within six months! – that the competition authority does not see grounds to take action.
The legislator also used the chance of the amendment to clarify minor aspects regarding damage claims, thereby “correcting” findings of courts after the implementation of the Damages Directive.
Other than expected, the proposal has no new powers for the Bundeskartellamt to deal with consumer protection issues (e.g. under the Unfair Commercial Practices Directive 2005/29).
Concluding Remarks
In summing up, the changes in merger control, procedure and other parts already add up to a significant change. Yet, at the heart of this amendment is a far-reaching, partly even “revolutionary” change of abuse provisions. Germany intends to set a European standard for a digital framework, by on the one hand “taming the tech titans” (as the Economist once put it), and on the other hand helping European companies with their access to the data economy. While the rules may trouble more conservative laissez-faire antitrust scholars, they are still far from being radical. There is no talk of prohibiting killer acquisitions, no breaking up of the GAFAs, no consumer-related enforcement powers. It is largely up to practice by the agency and the courts how far the changes will reach in practice. And before this implementation, there is still the whole legislative procedure coming up – we may well expect to see changes in the draft. This makes it clear that it is very timely now to join the debate!
Click here for a PDF version of the article
1 Rupprecht Podszun is a professor for civil law and competition law at Heinrich Heine University Düsseldorf and affiliated research fellow with the Munich Max Planck Institute for Innovation and Competition. Fabian Brauckmann is a researcher at Heinrich Heine University Düsseldorf.
2 Including an English translation of key provisions, see https://www.d-kart.de/en/blog/2019/10/14/der-referentenentwurf-zum-gwb/.
3 Cf. Crémer/de Montjoye/Schweitzer, Competition policy for the digital era, 2019, pp. 113 ff.
4 Schweitzer/Haucap/Kerber/Welker, Modernisierung der Missbrauchsaufsicht für marktmächtige Unternehmen, 2018.
5 Digital Competition Expert Panel, Unlocking digital competition, 2018 (“Furman Report”).
