The German Federal Supreme Court has ruled against Facebook and in favour of the Bundeskartellamt, the German competition authority, in preliminary proceedings on a leading competition/data case. This goes into the heart of Facebook’s business model. For the Bundeskartellamt, it is the most important case in the era of the data economy. The decision is, at the very least, a surprise. Rupprecht Podszun has first thoughts on it.
You may read this post in German, too, by switching the language button.
Before today’s decision by the Bundesgerichtshof, the German Federal Supreme Court (BGH) in the Facebook case, we in my team had calculated the chances for the Bundeskartellamt, the national competition agency. Let’s put it this way: I would not have bet anything like a good original ORDO volume that the Bundeskartellamt would win. Reason is: The Düsseldorf Higher Regional Court with its 1st Senate under Jürgen Kühnen had set the bar damn high for the BGH. We had calculated that it would be very difficult to overrule the OLG for two reasons.
But first a reminder for the not so regular readers of this blog: in February 2019, after an investigation lasting about three years, the Bundeskartellamt issued a decision against Facebook that was over 300 pages long. We had reported about it in this blog here and here and here and here. In August 2019, the Cartel Office’s decision was overturned by the Düsseldorf Higher Regional Court in summary proceedings involving the order of suspensive effect – with pithy words for which the 1st Cartel Senate is well known. We reported on this here. This decision was now subject to review by the BGH, still in interim proceedings.
Against all odds
There were two reasons why it was not very likely that the BGH cartel senate would decide differently from the lower court:
Firstly, the Düsseldorf Senate had not really been happy with the decision of the Sixth Decision Division of the Bundeskartellamt and had ordered the suspensive effect for a number of reasons (lack of causality, no damage to competition, wrong remedies). After reading the straight-forward attack, even well-meaning observers of the Bundeskartellamt – such as myself – had doubts as to whether everything in the authority’s decision was the last word of wisdom.
Secondly, the standard of review in interim proceedings before the Federal Court of Justice is extremely limited: The BGH examines whether the serious doubts of the lower court are justifiable. This is a mere plausibility check provided for such preliminary proceedings – see for example BGH 2007 in Lotto im Internet. There it says (at para 17):
„According to Section 65 (3) no. 2 GWB [the Germa competition act], the standard for reviewing the legality to be carried out by the Court is whether there are „serious doubts about the legality of the contested decision“. The court only examines the result found by the lower court for legal plausibility. The appeal decision is therefore valid in the appeal proceedings to the extent that it proves to be justifiable.”
Well, it seemed to me that some of the objections of the OLG Düsseldorf were justifiable. (But maybe my standards have also slipped: As a university teacher I am constantly correcting exams, and you wouldn’t believe what I find „just about justifiable“… but – oh, let’s drop that). I am quite curious to see how the BGH will define its own examination mandate in the decision.
So far, I have only a press release from the BGH, so that all statements made here are based on a thin foundation. However, what can be deduced from this press release is, at least, „spectacular“.
The result alone is impressive. After the defeat that the office had suffered with the Düsseldorf Court, I had the impression that the fizz was gone. The Facebook case suddenly looked like a nice idea which unfortunately did not work. Now, okay, it served as some sort of justification that the tools of the Bundeskartellamt do not suffice and the planned 10th amendment to the GWB, the German act, is necessary.
So, today’s decision is a reversal of thrust, also in emotional terms. The antitrust law-loyal Twitter community and the data protection foxes had been eagerly awaiting the decision, and every tweet from Politico reporter Simon van Dorpe, the only one to tweet faithfully from the „media workspace“ in Karlsruhe, was gratefully absorbed. A decision was then announced for 4 p.m., after apparently three hours of intensive negotiations starting at 9.30 a.m. This decision is a small, subtle vulgarity for the data protection foxes – more on this later. For the antitrust lawyers, however, it is an important impulse. In substance, it means that the suspensive effect of seeking a legal decision which the OLG had ordered is no longer given.
Of course, the pulse of those who sat in the court room in Karlsruhe was racing probably much more than that of Twitteritos. Let me take some of the work off the shoulders of the esteemed reporters from JUVE, the glossy German legal magazine, and report here on the team line-up: the Bundeskartellamt’s tried and tested head of the litigation department, Jörg Nothdurft, had taken the stand. At his side: Irene Sewczyk and Sandro Gleave as well as Silke Hossenfelder, head of the policy department. Three law firms were in for Facebook: Latham Watkins had already dealt with the antitrust aspects and Wilmer Hale with the data protection issues. In addition, Gleiss Lutz, another well-proven warhorse of antitrust law, joined as second adviser. (It is reported that they sat in corona-like seating arrangements, i.e. at a distance. This is perhaps not a bad idea for rival yet cooperating lawyers, even without corona.) The Cartel Senate was chaired by Peter Meier-Beck, framed by his deputy Wolfgang Kirchhoff, Patricia Rombach, Jan Tolkmitt and Birgit Linder. It is said that the importance of the case was in the air when the judgement was handed down. However, it is not every day that 30 million euros in dispute and headlines in the New York Times are at stake.
The next steps
There was no audible cheering after judgement was given, but the Bundeskartellamt, which had held back before the OLG decision, will now, it can be assumed, begin to implement its decision. To do so, Facebook must first present a plan, and then start implementing it. However, it cannot be ruled out that before the reform of the Facebook data policy, a decision will be made in the main proceedings in the meantime. The ball is now once again in the field of the Düsseldorf Higher Regional Court. The readers of this blog are well aware that the bench there does not always follow the course of the BGH, so that a decision in favour of Facebook cannot be ruled out even after this day. (I repeat this once again for our foreign readers: The lower court will continue the main proceedings, and it is conceivable that this court, which has so far supported Facebook, will continue to do so). Then the BGH will be allowed to go for it again.
There is another unknown variable: In the main proceedings (which Facebook is planning to go through according to an initial announcement), the European Court of Justice could also be called upon. A word from Luxembourg could possibly be a game changer for the senate in Düsseldorf or the one in Karlsruhe, which have both taken a pointed stand on this matter. During the hearing the Bundeskartellamt apparently suggested such a submission to Luxembourg. But how the proceedings will play out is ultimately speculation. If, at some point, a final decision in favour of the Bundeskartellamt is reached, then the creation of user “super-profiles” with data from all sorts of sources will come to an end for the time being. That would, in my perception, be a very painful blow to Facebook.
A spectacular reasoning
The Bundeskartellamt’s decision had a serious catch. It was not quite clear where the competitive damage would occur, for which Facebook would be held liable. The Düsseldorf Higher Regional Court (OLG) had also perceived it that way. „Where is the competitive harm in that case?“ became the battle cry of those who found this case esoteric from the beginning.
The ratio of the Office in its (abbreviated) form seemed very much like a simple equation:
Violation of the General Data Protection Regulation = violation of the prohibition of abuse, if market dominance
The Bundeskartellamt had used the infringement of non-competitive legal norms – the GDPR – as a yardstick for an exploitative abuse and justified this with decisions of the Federal Court of Justice (BGH) such as VLB Gegenwert and Pechstein. However, these two decisions are not in themselves the most easily digestible dietary dishes on the menu of competition law.
The data protection standard posed several problems:
- Firstly, it remained unclear what the case had to do with the protection of competition and where the boundary should be in the future: Would any breach of a rule by a dominant company constitute an abuse of market power?
- Secondly, the infringement of the GDPR remained disputed. There are serious privacy experts who believe that Facebook has not committed such violations. The OLG had not looked into this question.
- Thirdly, it was repeatedly objected that the Bundeskartellamt was not competent at all because the GDPR provides for an exclusive system of intervention.
But the Federal Court of Justice has now elegantly circumnavigated all these adversities with its ruling, because the Senate is not even getting involved in the entire GDPR problem. It constructs the abuse of market power differently, namely by restricting the freedom of choice of Facebook users. Sorry, data guys: This leading case of the data economy is none of your business. Please go to your playground, thank you!
Peter Meier-Beck explained this at the oral pronouncement of judgement:
This choice would have to be between a model with very intensive personalization of the user experience and unrestricted data access on the one hand, and less personalization and correspondingly limited access to user data on the other. This would be the solution, in Meier-Beck’s words, that would emerge if there was competition.
The consumer’s freedom of choice and the user’s decision-making sovereignty are becoming a benchmark for abuse. The Senate draws a clear line from autonomy to economy, as Meier-Beck also expressly stated in his oral statement.
Sovereignty in competition law
So is antitrust law perhaps about freedom of choice after all? Of course it is. For the European Court of Justice, the principle of independence in decision-making is a central component of free competition. This basic principle of the protection of competition, which has been emphasized again and again, states: Market participants must be able to make their own decisions independently and freely; only then do markets function. Up to now, however, the independence formula had mainly been evoked in cartel cases with undertakings. The BGH now asserts the sovereignty of decision in an abuse case and demands such authority for consumers, who are of course also market participants. Where their freedom of choice is suppressed by companies, the competitive process is damaged. (I am delighted with this argument, as I defended this autonomy-based interpretation of antitrust law for Art. 102 TFEU in a paper last year, which you can read at SSRN (end of self-promotion)). The concept makes sense to me: The demand side decides on success in competition, so this referee must also be able to make free and independent decisions. It is a piece of digital autonomy that is also a prerequisite for functioning markets.
The return to the consumer’s decision-making sovereignty is not a revolution, but it is remarkable. In the past decades it had become out of fashion to name concepts such as consumer choice, autonomy, protection of the competitive process as objectives of competition law. Post-Chicago economists‘ models do not reflect these fundamentals of markets very well. However, they are of course in line with the tradition of ordoliberalism or, more generally speaking, of an antitrust law that has not completely ceded its territory to number-crunching.
The rest is then almost the small basics of the abuse argument: the special responsibility of the dominant company for competition, the balancing of interests, the effects also on the (in this case rather hypothetically existing) competitors. Having said that – I think I have heard another exciting point in the reasons for the judgement.
What are the affected markets? In the press release the BGH writes unequivocally:
The focus here is therefore on the German market for social networks. But the Senate also seems to have considered the multi-sided nature of Facebook’s activities. At one point in the giving reasons for the ruling, Meier-Beck speaks of „two theoretically distinguishable market sides“, and it can be heard that he does not exactly attach immense practical importance to this „theoretical“ distinction. Rather, according to the Senate, two markets are very closely intertwined: the market for social networks and the market for online advertising. This entanglement is created by the personalisation of advertising thanks to data power. From the press release:
„Facebook’s access to a considerably larger database further reinforces the already pronounced „lock-in“ effects. In addition, this larger database improves the possibilities of financing the social network with the proceeds from advertising contracts, which also depend on the scope and quality of the available data. Finally, due to the negative effects on competition for advertising contracts, an impairment of the market for online advertising cannot be ruled out.”
If the BGH deepens this idea in the decision, the bridge from exploitative abuse to a hindering abuse will probably be built; and also the role of data for the business model will be painted in very vivid colours. The fact that the financing aspect also plays a role here, and that the „deep pockets“ of the Silicon Valley companies are alluded to, is a further nice give-away of this decision – who does not think of killer acquisitions here!?
Year 4 (and counting)
What is known so far seems to me – on first reflection a few hours after the announcement of the decision – quite plausible. (I reserve the right to change my opinion at any time, since this is a major qualification for being a good academic.) It remains to be seen whether the decision of the BGH is within the scope of the standard of examination and whether the aspects indicated here will be taken into account in the written opinion. In any case, the decision is a liberating blow in the direction of the GDPR friends. At the same time it does not make things too easy for the Bundeskartellamt: the exact yardstick by which the restriction of choice leads to an exploitation is likely to be further specified.
Nevertheless, it remains somehow unsatisfactory: the Bonn-based competition watchdog started its investigations in 2016, we are in our fourth year with this case and a final decision is still a long way off. Meanwhile, businesses on the Internet are growing at lightning speed. The market power of the GAFA companies is strong and growing. Once again, this case shows that competition law is a great tool for analysing economic phenomena. All parties involved make their contribution to this, and the judges in Düsseldorf and Karlsruhe have fired up the competition for the right legal assessment with their trenchant reasoning. One could almost say: Here the procedure works as it should! A strong authority with lawyers and economists, a first-class team of lawyers, clever judges in all instances and a wild academic community on the sidelines. But it takes too long. And so the process is perhaps above all a source of inspiration for legislation, whether it be the 10th amendment to the German competition bill or the European initiative for a „new competition tool“.
Rupprecht Podszun, the author of this post, is a law professor and a Director of the Institute for Competition Law at Heinrich Heine University Düsseldorf. The Institute runs this blog.