News, views and reports

Almost one year to the day after taking office, the European Commission delivered on one of President Von der Leyen’s central promises, and one of the world’s most widely anticipated digital policy initiatives: the proposal of a Digital Services Act (DSA), supposed to update the EU’s ground rules for the digital economy and to govern what Commissioner Thierry Breton calls “the information space”.

Compared with the initial plan laid out in Von der Leyen’s “Political Guidelines”, the Commission even over-delivered, adding to it a proposed Digital Markets Act specifically aimed at regulating the practices of large online platforms acting as “gatekeepers”.

By EU standards, the speed at which the proposals have been published is a feat. Yes, discussions on the need to update the 2000 E-Commerce Directive had started years ago and had largely been prepared by the Juncker Commission; and yes, the need for specific ex ante regulation of large platforms had become a central feature of the discussion on how to adapt competition rules to the digital era, being addressed in a multitude of expert reports and even in some legislative initiatives such as Germany’s competition law reform. Yet the Commission didn’t explicitly endorse the notion of “gatekeeper” until its overarching digital policy Communication of February 2020. And drafting two fully fledged draft legislations within a few months on issues of such importance and sensitivity, while running a public consultation that received over 3,000 responses, is certainly unusually fast.

This leads to a paradox: while the review of the E-Commerce Directive was certainly the more mature project in terms of the legal preparatory work that preceded it, the DMA is undoubtedly the more ambitious and consequential reform. For all the discussions on reinforcing online platforms’ responsibility for the content they host, the DSA proposal leaves the central principle of limited intermediary liability largely untouched, focusing instead on introducing flanking requirements e.g. in terms of transparency, due diligence, and harmonised notice and take-down procedures.

By contrast, the DMA introduces an entirely new regulatory regime for the designation and regulation of “gatekeeper platforms”. Procedures and obligations are clearly inspired by antitrust enforcement but with the major difference that they are not based on the demonstration of market power and abusive conduct, but on sheer size only (specifically: significant impact in the single market, gatekeeper position and durably entrenched position).

Many competition lawyers and even some DG COMP officials had expressed concerns with the implications of such an approach, which by definition departs from the basic principles of ex post antitrust enforcement but also implies an admission of failure of the EU Big Tech antitrust investigations conducted over the past decade. But let’s bet that both sides are eventually going to be comforted by the significant increase in Commission powers and the whole new branch of legal work this proposal entails.

The paradox, however, is only apparent, and the outcome is in fact quite logical from a political perspective. Despite proposals to revise the limited intermediary liability principle, the Commission never hid its desire to avoid a repeat of the controversies around “upload filters” and the “end of the free internet” that surrounded the 2019 Copyright Directive’s adoption. And with the DSA conceived from the start as a horizontal instrument covering all actors of the internet, it was always going to be a challenge to obtain specific measures for specific types of content.

Conversely, regulating “gatekeepers” might be legally more challenging (e.g. in terms of the appropriate legal basis), but it is certainly politically more acceptable, especially when this is first and foremost about Big Tech. One of the most important developments these past years was the shift operated by “liberal” countries such as the UK and, crucially, The Netherlands, towards targeted gatekeeper regulation. Alongside France’s and Germany’s push for a new industrial policy and technological sovereignty agenda, The Hague’s support for ex ante regulation played an important role in convincing the Commission (whose internal market and digital Commissioner Thierry Breton has long been an advocate of strict stance on Big Tech) to move forward with the plan, presumably while maintaining a largely untouched framework in the DSA.

So what does the Commission seek to achieve with this approach? As Margrethe Vestager’s and especially Thierry Breton’s frequent references to the banking sector show, at heart lies the (certainly correct) notion that some platforms have de facto become fundamental infrastructure for the entire economy, which justifies a special societal responsibility (going beyond the “special responsibility” associated with market power in competition law) and the need for targeted, asymmetric regulation.

This focus on the special responsibility of systemically relevant platforms is reflected in the DSA, with very large platforms facing additional obligations e.g. in terms of enhanced transparency and addressing systemic risks.

And it is, of course, the very essence of the DMA. It is reflected in some of the obligations that gatekeeper platforms will face, such as bans on “dual platforms’” utilization of their business users’ data (art. 6.1.a) and on “self-preferencing” their own additional services (art. 6.1.d), echoing recent competition investigations into Amazon and Google that some experts have dubbed “common carrier antitrust”. Overall, and in line with the Commission’s broader strategy, the DMA seems to pursue much more of an industrial agenda than a consumer one, with a focus on tackling big platforms’ “unfair practices” vis-a-vis business users and on controlling (including through the notification of all acquisitions) their expansion into an ever-growing number of markets and sectors.

As the Commission tries to get the ball over the line in the upcoming legislative procedure, the main challenge is unlikely to come from opposition to the general approach to ex ante regulation (which enjoys broad support in the European Parliament and from the most powerful member states) but from the devil in the details. To name just three crucial issues:

• How do you define who is big enough to be regulated ex ante? According to many observers, the potential scope of the proposed regulation, both in terms of the list of 8 “core platform services” it captures and of the quantitative turnover and user thresholds it sets, could be significantly broader than initially thought. While this could be useful to fend off the accusation of anti-US, anti-Big Tech bias, it is also likely to generate backlash from (European) companies and their member states that will argue that they should certainly not be put in the same basket as the GAFAs.
• What kinds of specific obligations and prohibitions do you set? On this point, some of the most ardent supporters of ex ante regulation, especially in the European Parliament, might feel that the DMA does not go as far as they would have liked. For example, the data-related obligations listed in articles 5 and 6 seem to be largely focused on the possibility for business users to access and use the data generated through their own use of the gatekeeper platform, without including broader data sharing obligations in cases where data is considered an “essential facility” to compete with the gatekeeper platform (except, it seems, in the specific area of search engines). Similarly, the possibility to “break up” Big Tech companies, long called for by the European Parliament and much touted by Thierry Breton in the run up to the DMA’s publication, remains a remote, last-resort possibility that does not seem to add much to what is theoretically already feasible under competition law (see in particular the limitations in terms of procedure and proportionality enshrined in art. 16 of the proposed DMA).
• Finally, member states will pay particular attention to the articulation of the DSA / DMA with their own national initiatives, be it in the area of content liability (France) or of competition and ex ante regulation. The latter point will be particularly relevant for Germany, which is busy passing its own ex ante regulation of systemic platforms, and where the statement that “Member States shall not impose on gatekeepers further obligations by way of laws, regulations or administrative action for the purpose of ensuring contestable and fair markets” might explain why Economy Minister Altmaier felt the need to reaffirm this reform in his reaction to the DMA publication.

These are likely to be some of the questions at the heart of the legislative procedure that will begin in a few weeks. M. Vestager said during her press conference that she expected the procedure to take between 18 and 24 months. If the GDPR and Copyright are any indication of what’s to come, the Commission should be happy to see these proposals adopted before the end of its mandate.