Relaxing whilst doing Competition Law is not an Oxymoron

Archive for October 6th, 2020

To Comment or not to Comment on the Ex Ante Rules for Gatekeepers (+ 9 Other Questions on the Draft Proposals)

with one comment

Until today, I had avoided commenting publicly on the ex ante regulatory instrument that the Commission is considering for “platforms acting as gatekeepers”. There were three reasons for this. First, unlike the New Competition Tool, I did not see this initiative as a threat to competition law as we know it. Second, I do understand that there is a certain public anxiety about digital markets (to some degree justified, to some degree exaggerated by interested stakeholders), and a margin for regulation to legitimately address that anxiety and improve things. Third, my opinions could be legitimately criticized as biased because an important part of my work is to advise and represent companies targeted by this initiative. My thinking was that there were already enough people with professional interests making noise in all directions for me to contribute to the cacophony.

Alas, my reasoning on those three fronts has changed after reading the bold leaked drafts that circulate widely since last week. First, I now see a risk that the ex ante rules might have a much greater impact on competition law than I had anticipated. Second, it seems that the current plans might not always be addressed at the issues causing public anxiety, but to a large extent target issues at the core of pending competition cases. Third, I will not falsely pretend my opinions are neutral, but I hope they might add some value to the debate. Experts working against my clients have also  authored some of the influential reports that the leaked documents cite as evidence supporting the need for intervention and, to be sure, I don’t think their views should be disqualified. In addition, given that the rules appear to be crafted to affect only a remarkably limited number of services, it is likely that a majority of stakeholders will feel relieved and/or might not have incentives to voice out concerns, so you might not be exposed to many contrarian views. And even if you are, a lot of the commentary out there seems somewhat radical.

As in other matters, there is excessive polarization here, and even a tendency to look at things through myopic and binary (progressive vs conservative) lenses. Some partisans of regulation invoke “progressive” attitudes as a reason to favor these initiatives. Others claim that they oppose them in line with “conservative” principles. In reality, though, this debate has very little to do with politics. There are media companies who support Trump, Brexit and deny climate change that also support these initiatives (provided they only target their rivals, of course). There are also conservatives who want more regulation and antitrust intervention (even if arguably not always for the right reasons; e.g. at a recent antitrust hearing a Republican Congressman claim Google should be regulated because it favours WHO health advice over Trump’s…). And then there is a vast majority of other perfectly reasonable companies or stakeholders who have legitimate commercial/professional interests in these initiatives passing, or not passing, regardless of politics.

Instead of simply shouting “this is great” or “this is rubbish”, I’ll try to reason my concerns through. Confronting (hopefully) reasonable views should (hopefully) contribute to progress. The public debate should not belong only to those shouting the most (and you should see my Twitter feed…). I said before that we can do better, and I will try to play my part by being assertive, but as constructive as possible. It is not easy to decide where to start, but here are some questions:

1-.Why The Focus on 4 Key Services?  The current drafts indicate that the broader options have already been discarded, and that the plan is to focus only on 4 companies “key services”, namely “online intermediation services (market places, app stores and social networks), (ii) online search engines, (iii) operating systems and (iv) cloud services”. Queries: What were the criteria to concentrate on these? Can other “information society services” be excluded from the potential scope of application of the rules, even in the presence of “gatekeeping” features?  Are these the services of the economy where consumers are getting the worst deal in terms of price/innovation/quality/R&D investment? Are these the services with the greatest risk of consumer lock-in or with the highest switching costs? Are these services that have escaped competition law scrutiny? Or are these the intermediary services that have succeeded in creating greater opportunities for third party business users? 

2-. Unavoidable = Indispensable = Market Power? Perhaps the questions above are irrelevant, because the drafts show that the concern is that some services may be “de facto unavoidable for business users”. That does not necessarily mean the initiative is less legitimate, but it raises questions worth asking. Queries:

·   The assessment of market power depends on the competitive constraints faced by the company under examination. Under this alternative standard, however, “gatekeepers” subject to these rules would be identified judging solely from the perspective of whether (all? some? how many?) business partners have (other? equally convenient?) alternatives, regardless of the possible exercise of market power. Should that be the right approach? Should we adapt our thinking about identifying market power, or should we only make a carve-out for these 4 services? And if the concern is not market power but the “economic dependency” of certain users, can the proposed remedies go beyond that and constrain, for instance, product design decisions?

·   Does “de facto unavoidable” mean the same as indispensable within the meaning of EU Law, also beyond competition? Consider, for example, paragraph 55 of the recent CJEU Airbnb ruling. Would accomodation intermediaries, for example, be covered by these ex ante rules in spite of the Court’s observations? 

.   Rather than simply identifying “key services” and designating them as “gatekeepers” (which could arguably risk looking like reverse-engineering), would it make sense to craft some sort of normative or objective criteria to decide whether a given service might have “gatekeeper status” (e.g. on the basis of market shares, shares of traffic or other objective parameters)? What would be the process to identify the services with gatekeeper status? Would companies have opportunities to make their views known? Should there be consultations like in telecom?  How often would this status be reviewed?

3-.A Gap in Competition Law or An Overlap With Competition Law? I had the impression that the ex ante rules would cover matters falling outside the scope of competition law, and that this is what could lend legitimacy to the initiative (competition law is not everything after all). I was wrong. The list of “black” and “grey” practices in the leaked documents cover three categories: (i) data practices; (ii) self-preferencing; and (iii) tying and bundling. The practices in the list pretty much coincide with the same very practices that the Commission and NCAs have recently investigated, or are currently investigating, in cases concerning mainly Amazon, Apple, Booking, Facebook, Google and Microsoft. 

This means that (unless the EU Courts rule otherwise), this is all conduct that currently falls within the scope of the competition rules. Some of the practices in the draft lists in fact contain the actual wording of some of the examples included in Art. 102; others reproduce the wording and terminology used in recent decisions. Queries: To the extent that the rules would apply to practices falling within the scope of the competition rules, and only to companies that the Commission believes are dominant, then would this initiative fill a gap? Or would it rather replace case-by-case evidence-based assessments by outright bans when it comes to a handful of companies? Given the clear overlaps, wouldn’t it make sense for the enforcement of these rules to sit with DG Comp in order to ensure consistency with existing enforcement tools and with a possible NCT?

4-. Why Now? The Commission has spent years building cases, presumably challenging what it considers to be the most egregious instances of “black-listed” conduct. It is perfectly reasonable for the Commission to build its legislative proposal on “evidence derived from competition enforcement practice” as the drafts seek to do. But shouldn’t we then wait until those cases are over? The Commission has often said that what lends legitimacy to the fact-finding and enforcement process is judicial review. And EU Courts have now been called to render Judgments assessing the effects of some of these practices on the basis of actual evidence. Queries: Would upcoming Judgments concerning practices in the black list be rendered irrelevant before they are decided? Would that be that an unintended consequence? Would that be good thing? Would we accept this process in all areas of the law, or should we have an exception for matters that affect only a few companies? Would regulation be better or worse if it were also informed by the findings of the Courts in relation to the cases brought by the Commission? 

5-. Harmonization as the Goal? According to the leaked documents, “the objective of the intervention is to harmonise rules in Member States relating to unfair behaviour in gatekeeper platforms”. Query: To the extent that competition laws, unfair competition laws, the P2B Regulation, and other initiatives at the national or EU level would remain in place, would the ex ante rules harmonize national legislation or would they rather be adding one more tool to the toolkit?

6-.What About the NCT? If these rules were to enter into force, an NCT would arguably not add much value with respect to these 4 “key services”. Most people thought, however, that these were precisely the services that an NCT would target. Query: Does this mean that the NCT will be deployed mostly in relation to other services?

7- The “black list”. The list is “based on CNET/GROW” evidence gathering. The legitimacy of the initiative would arguably be enhanced if that evidence were identified and made public. As explained above, it would appear that most of the practices in the black list target conduct at the core of recent or pending cases. It is commonly accepted that many of those practices (e.g. MFNs) are competitively ambiguous; they might have pro- or anti-competitive implications depending on the circumstances. Not even complainants in “self-preferencing” cases argued that self-preferencing should be prohibited in all circumstances. The drafts recognize that this practice is widespread and states that “in offline situations, such behaviour is not generally considered anticompetitive”. Queries: Should practices like self-preferencing be treated differently in offline vs online situations? Are practices included in the black list because (a) there is evidence that procompetitive aspects disappear, or are always outweighed, (only) in relation to these 4 key services/companies?; or (b) we can’t be bothered to assess this question in relation to the 4 key services/ companies? 

8-. The added value of the “grey list”? The grey list contains a list of practices “where intervention by the competent regulator is required”. Query: Would intervention in these cases take place through standard procedures (e.g. a competition investigation?) subject to established standards? The merit of a grey list may arguably streamline and accelerate the process and could provide some sort of ex ante guidance, but then again, which would be the competent regulator? Given the clear overlaps with possible competition cases (and the need for coordination with national authorities), wouldn’t it be preferable for DG Comp to be in charge of enforcing the rules? That would also enable the Commission to coordinate with national authorities through tested channels like the ECN.

9-. Are you trying to say that there is nothing to do and that we should never regulate? Not at all. There is probably merit in the idea of having some sort of ex ante rules. The digital economy, like any sector of the economy, needs rules. Like in any sector of the economy, rules should be objective, not subjective, and a result of careful evidence-based reflection. Rules do not need to be only (or mostly) about efficiency, there are certainly other higher values. Rules can and should change, they can be creative; they can go against precedents and even be harsh when needed. But they should ideally be sensible, they should benefit the most (not the few), and they should be crafted to pursue their goals while minimizing the risk of undesirable outcomes. Leaving room for nuance and flexibility is often wise, particularly in the case of forward-looking rules. It would be a mistake to craft these rule to target only specific practices seen so far, because these will also be the applicable rules for future platforms. As Pablo explained in a previous post, building on the experience acquired in telecoms, a business-neutral, principles-based regime and the possibility of case-by-case assessments would seem much more flexible, adaptable and sensible than the alternative “black list” approach.

Written by Alfonso Lamadrid

6 October 2020 at 5:28 pm

Posted in Uncategorized