EU Regulatory Initiatives and Competition Policy in the Digital Era: What has been achieved and what to expect in the next mandate [i]
The tech revolution rages through the world, providing new opportunities and challenges in in a globalised world. In Europe, the European Commission (the Commission) has claimed that achieving a connected Digital Single Market (DSM) can contribute EUR 415 billion per year into the economy. Shaping EU communications and media regulations, in parallel with EU competition policies, in this new digitised era has undoubtedly been on the top of the agenda of EU institutions. In light of the recent elections of the European Parliament (the Parliament), it is appropriate to take stock of the development of the EU's DSM and competition policy.
In this article, we will discuss:
- The state of play of key regulatory initiatives, in particular:
- The EU's accomplishments in the DSM (Section 1.1)
- The outlook for the next Commission and Parliament (Section 1.2)
- Shaping EU competition policy in the digital era, where we will look successively at certain developments in merger control (Section 2.2), anticompetitive practices (Section 2.3), abuse of dominance (Section 2.4) and State aid (Section 2.5).
1. State of Play of Key Regulatory Initiatives
The EU's accomplishments in the DSM
Launched in May 2015, the Digital Single Market strategy has been one of the principal political priorities of the Juncker Commission. It is designed to create an area where businesses and consumers have unrestricted access to digital goods and services across EU Member States, underpinned by the free flow of data across borders, with the goal of creating growth in the digital economy. The DSM strategy comprises 16 key initiatives with some 30 separate legislative proposals. To date, 28 of these legislative proposals have been agreed by the co-legislature.
The European Electronic Communications Code
As one of the centrepieces of the DSM strategy, the Code focuses on the development of very high capacity networks, consisting of full-fibre infrastructures or other infrastructures with equivalent performances. In addition, the Code:
- extends regulation to over-the-top (OTT) services such as Skype and WhatsApp;
- extends the scope of regulatory obligations that could be imposed on providers who do not have significant market power (SMP);
- encourages co investment by stating that an SMP provider which enters into agreements with its competitors may be subject to lighter touch regulation, or even be exempted from regulation;
- provides for wholesale only operators and related light touch regulatory regime; and
- sets out rules for use of harmonised radio spectrum, including coordinated assignment of 5G spectrum.
While the Commission aimed to achieve maximum harmonisation, the Code does leave significant room for Member States to take specific national circumstances into account when implementing the new rules. Furthermore, the final impact of the Code can only be assessed fully when the Commission adopts a series of delegated acts and implementing acts between December 2019 and 2022, clarifying issues such as a single maximum Union wide mobile voice termination rate and a single maximum Union wide fixed voice termination rate, and specifications for small area wireless access points. Remarkably, a retail cap for intra-EU calls has also been mandated.
The P2B Regulation can be regarded as one of the first concrete signs of a more interventionist EU approach to regulating online platforms. The new rules are aimed at creating a fair, transparent and more predictable environment for businesses and traders, when using online platforms. This legislative initiative comes after a complaint to the Commission by traders selling online via marketplaces, hotels using booking platforms and app developers, particularly SMEs, regarding what they saw as the unfair practices of the online platforms they use to reach consumers. The final text of the P2B Regulation, which will be directly applicable at the national level, was adopted at first reading by the Parliament on 17 April 2019. The Council's final endorsement is expected around mid-June [ii]. The new P2B Regulation will enter into force 12 months after publication in the EU Official Journal, and is already being hailed by the Commission as a first of its kind anywhere in the world.
The Regulation will apply to online intermediation services and online search engines which provide their services to business users as well as corporate websites established in the EU and which offer goods or services to consumers located in the EU. This is likely to encompass e-commerce market places and app stores, social media for business, price comparison tools and general online search engines.
While the Commission considered its original approach to be "light touch", the final agreement does bring with it some notable new obligations for online platforms including: a ban of certain practices deemed to be unfair; greater transparency; and a dispute resolution mechanism.
DSM Copyright Directive
The passage of the P2B Regulation has been relatively smooth in comparison to the controversial DSM Copyright Directive. Following its adoption by the Parliament in March, the proposal has been formally adopted by the Council in April and was published in the EU Official Journal on 17 May 2019. Member States now have 24 months in which to implement the Directive into national law. The newly adopted Directive will tip the overall balance of the copyright framework in favour of certain rightholders and creators (although not all), while attributing more responsibility to online platforms and service providers.
Among the most lobbied provisions in the Directive are Articles 15 and 17. The first provision aims to provide the legal recognition of press publishers as rightholders. Specifically, press publishers are granted a two year term of protection in copyrighted works, for its reproduction and for making it available to the public, to offer protection against online uses by internet service providers and aggregators. The use of individual words or "very short extracts" is not covered by this provision, although we expect further litigation regarding the exact meaning of "very short extracts". Article 15 does not apply to private or non-commercial uses of press publications by individual users, or to hyperlinking.
The source of the greatest dissent, Article 17, addresses what the Commission has characterised as the legal uncertainty regarding whether online content sharing services that provide access to a large amount of content uploaded by their users (e.g. YouTube) engage in copyright relevant acts, and would therefore need to obtain authorisations from rightholders. Article 17 stipulates that online content sharing service providers perform an act of communication to the public, or an act of making available to the public, when they give the public access to copyright protected works uploaded by its users. OCCSPs (like YouTube) should therefore obtain authorisations from right-holders before in order to lawfully perform this communication. If no authorisation is granted, OCCSPs can be held liable, unless they demonstrate that they have: (i) made "best efforts" to obtain authorisation; (ii) made "best efforts" to ensure the unavailability of specific works highlighted by rightholders; and (iii) acted "expeditiously" to remove access to notified works preventing their future upload. Therefore, under the framework of this new provision, they will not be able to invoke anymore the limitation of liability enshrined in the e-Commerce Directive.
The forthcoming e-Privacy Regulation, which was intended to complement the General Data Protection Regulation (GDPR), has also proven to be difficult to get over the finish line. The proposed Regulation focuses on the confidentiality of communications and rules regarding tracking and monitoring. Discussions have focused on issues such as the confidentiality of machine-to-machine communication in the Internet of Things as well as the confidentiality of individuals' communication on publicly accessible networks such as public Wi-Fi.
Directive on Online Transmissions
Other initiatives, like the Directive on Online Transmissions[iii], had to be greatly reduced in scope in order to reach political consensus. The Directive, which aims to enhance cross-border access to TV and radio programmes by simplifying copyright clearance for online services (such as simulcasting and catch-up services), is now confined to news and current affairs, or programmes fully financed and controlled by the broadcaster, excluding any sports events. The Directive was also initially conceived as a Regulation but had to be redrafted as a Directive due to opposition of the Member States, who were reluctant to alter national legislation significantly. The reduced proposal was endorsed by the Council on 15 April 2019 and published in the EU Official Journal on 17 May 2019.
1.2 Outlook for the next Commission and Parliament
At this late stage in its mandate, the Juncker Commission appears to be grappling with the issue of how to apply "European values" to the digital economy. Although it will not be stated so overtly in Commission position papers, the current DSM Strategy appears to have been motivated by concern that the EU is falling behind in innovative technologies and infrastructure, in comparison to the U.S. and China.
In its search to find some answers to the issues raised by the new digitised economy, the Commission has created a number of expert groups charged with gathering information that will help to shape the work programme for the next mandate. The Commission last year appointed:
- an expert group for a new Observatory on the Online Platform Economy;
- a High-Level Expert Group on Artificial Intelligence (AI), which issued "Ethics Guidelines for trustworthy AI" on 8th April;
- a High-Level Expert Group on the Impact of the Digital Transformation on EU Labour Markets;
- an Expert Group on Business-to-Government Data Sharing (B2G); and
- three Special Advisors on the future challenges of digitisation for competition policy (see below).
The EU institutions have been circling the issue of online platform regulation, not only via the P2B Regulation, but also with certain provisions in the DSM Copyright Directive, as well as the proposal for a Regulation on "preventing the dissemination of terrorist content online", which is still under discussion. It is likely that the next European Parliament and College of Commissioners will attempt to go further with platform regulation and in this context a potential review of the e-Commerce Directive could prove to be a real flash-point.
The next mandate of the Commission and Parliament may also be expected to push forward with plans to ensure maximum connectivity, so businesses and individuals can participate fully in the digital economy. Whether we are looking towards the future of connected cars, infotainment, smart cities and the Internet of Things, all of these sectors of the innovative economy rely on 5G technologies. However, there are already marked differences in approach between Member States when it comes to the allocation of suitable spectrum for 5G. According to the Code, this allocation has to take place by the end of 2020 at the latest, but the Commission has already acknowledged it is expecting delays in certain Member States. In addition, existing 5G auction processes are already hotly contested.
To sum up, some DSM initiatives to date have resulted in questions rather than providing any clarity, while others risk adding a layer of regulation to the existing legal framework that could arguably further complicate contract negotiations in the digital environment. It could also be said that some DSM proposals contradict key elements of existing legislation, such as the uneasy relationship between the liability provisions in the DSM Copyright Directive and the e-Commerce Directive.
Finally, the DSM strategy has been criticised by third countries behind closed doors for being protectionist in nature, or for attempting to compensate for lack of risk-taking among European entrepreneurs. In short, the DSM strategy, with its strengths and obvious weaknesses, cannot compensate for any shortcomings in the entrepreneurial culture in Europe.
2. Shaping EU competition policy in the digital era
2.1. Another seminal year for competition law and policy
The regulatory developments in the DSM are supplemented by a debate on how competition policy can contribute to an innovative and fair European digital market. The debate is ongoing, and we anticipate that the coming year will be another crucial year for the application of EU competition law focusing on topical issues, such as whether the Commission should allow the creation of "European champions", whether the recent multi-billion euro fines imposed on tech companies for alleged exclusionary practices (also, for example, involving the use of algorithms) are a prelude for increased enforcement, or whether the availability of hundreds of millions of euro in State funding for broadband development will have implications to State aid enforcement
In addition, a fundamental debate has arisen in the competition law community about the role and effectiveness of competition law in today's society. Some argue that big tech giants are the Standard Oil of our day, calling for them to be broken up in line with what happened to the big U.S. trusts more than a century ago. However, others argue radically against that notion citing the increased benefits for consumers which the tech wave has brought about.
Against the background of that policy debate, Commissioner Vestager last year appointed a panel of three external special advisors, namely Professors Heike Schweitzer, Jacques Crémer and Assistant Professor Yves-Alexandre de Montjoye. The special advisors published a report on 4 April 2019 regarding the future challenges of digitisation for competition policy. For a more detailed discussion of the expert report, we direct the reader to our article on the topic.
In the below sections, we will look back on the last Commission's policy for each of the major pillars of competition law.
2.2. Merger control
In the Commission mandate which is now coming to an end, several merger control decisions have been adopted relating to the digital sector. An analysis of these decisions reveals certain trends.
First, there is the ongoing debate as to whether the current merger control rules are able to capture all important transactions in the market. While there is consensus that high-value transactions involving low-turnover companies are currently not sufficiently covered by the merger rules, the debate continues on how the rules could be change. Germany and Austria have already introduced new value-of-transaction thresholds, albeit with limited success, and Hungary has introduced lower, “soft thresholds” focusing on the expected competitive effects of the transaction. The Commission's experts however advised against amending the rules of the EU Merger Regulation. In addition, critics are concerned that increased scrutiny of transactions could negatively affect development in the digital sector as more regulatory intervention may lead to fewer transactions, which in turn could result in a loss of financing for start-ups.
Second, recent decisional practice shows that the Commission's understanding on how to apply the merger rules to online markets is maturing. Previous decisions such as Facebook/WhatsApp were heavily criticised, including from within the Commission's own ranks, as naive. The recent Apple/Shazam transaction can however be seen as somewhat of a shift with regard to the Commission's approach to data aspects in merger control. In this deal, Apple acquired the popular music identification app, Shazam. While the transaction was cleared unconditionally, the Commission decision could serve as a blueprint for its approach towards data-heavy mergers. In its decision, the Commission mainly focused on two elements: (i) whether Shazam would be an important additional tool for Apple to acquire customers for the latter's music streaming business; and (ii) whether Shazam's data on users' preferences could be considered unique. Neither point seemed problematic in Apple / Shazam, but the decision nevertheless reveals the Commission's focus: ability to attract customers and uniqueness of data.
Third, in the telecommunications sector, mobile-to-mobile mergers remain a hot topic. After two blocked/abandoned transactions in the UK and Denmark, the Commission only cleared an Italian transaction after receiving commitments which enabled the entry of a fourth player. This led the competition law community to posit whether the Commission had a "magic number" in the sense that four-to-three mobile mergers generally would not be cleared without remedies enabling a new player to enter the market. The Commission has always denied this, and now points to last year's Tele2 NL/T-Mobile Netherlands case as proof that the "magic number" theory does not hold water. The transaction, which was cleared unconditionally, is however somewhat particular: the Dutch market was dominated by two large, vertically integrated multi-play players (KPN and VodafoneZiggo) and the market shares of the second and third mobile-only players (T-Mobile Netherlands and Tele2 NL) were much smaller. The joining of the latter two companies effectively allowed for the creation of a viable third party as a competitor to KPN and VodafoneZiggo. It remains to be seen whether the Commission would also clear a more traditional four-to-three mobile transaction without market entry commitments.
Finally, the Commission's recent telecommunications mergers show that it has not stood in the way of consolidation in the sector. At the same time, decisions in other sectors show that the Commission is not moved by political arguments related to the creation of "European champions". In the recently prohibited Siemens / Alstom transaction, heavy-weight countries Germany and France had urged the Commission to approve the deal, which would see the two companies' railway business combined. The Member States argued that a prohibition would leave the EU industry unable to compete with Chinese competition. Commissioner Vestager did not budge, stating that it is not her business to create "European champions" to counter a threat that may never materialise. In recent speeches, she pointed at other areas of EU law which are, in her view, more appropriate to develop an EU industrial policy, stating that "the EU must make full use of the trade defence instruments at its disposal".
2.3. Anticompetitive practices – Article 101 TFEU
There have been relatively few decisions or judgments on anticompetitive practices in the digitised world. In Eturas, the Court of Justice held that imposition of maximum rebates via an online platform for travel agents constituted an anticompetitive practice. The Court held that while the travel agents would technically have been able to override the limitation, it would have involved several additional steps on the platform and, in practice, none of them appeared to have done so. In addition, some of the travel agents claimed not to have been aware of the limitation. The Court held that platform users can be presumed to have been aware of (and to have consented to) such a limitation, unless they can prove that they actively distanced themselves from the practice. The judgment mainly relates to a classic hub-and-spoke situation and the conditions under which a party can distance itself from anticompetitive behaviour, but the only "digital element" in that case was that the anticompetitive practices occurred via an online platform.
Platforms have also been an important topic for Article 101 cases in the context of e commerce. In Coty, the Court held that, while it is not possible completely to restrict the sale of goods via online platforms, a supplier may limit its distributors' online sales to the sales made through their own e-shop and to those made through a third-party, insofar as the involvement of that third party is not discernible to the public. The Commission was also actively pursuing manufacturers who restricted online sales: it fined consumer electronics manufacturers for imposing fixed or minimum resale prices on their online retailers and sports merchandise producers for restricting cross-country merchandise sales within the EEA.
The cases show that the Commission's decisional practice and the Court's case law on anticompetitive agreements in the digital sector remains fairly limited and we also expect that it will be some time before there will be a consistent stream of digital cases.
2.4. Abuse of dominance – Article 102 TFEU
Platforms, algorithms and apps
The Commission's practice on abuse of dominance in the digital sector is, in contrast, more extensive. It has even received widespread attention in mainstream media. The three cases involving Google are of particular interest.
In Google Shopping, the Commission found that Google had abused its dominant position by placing its own price-comparison results in a more favourable position than those of other price comparison websites when users searched for products on Google's search engine. The Commission required Google to amend the algorithm which it used to organise the results. In Google Android, the Commission found that Google had also abused its dominant position by requiring or incentivising manufacturers of Android phones to pre-install the Google Search and Chrome apps, and by preventing manufacturers wishing to pre-install Google apps from selling even a single smart mobile device running on alternative Android forks which were not approved by Google. In Google AdSense the Commission fined Google for imposing certain restrictive clauses in Google’s agreements with third party websites which allegedly prevented Google's rivals from placing their search adverts onto these websites.
Rather than imposing a fixed solution, the Commission allowed Google to comply with the decisions as the latter saw fit, subject to scrutiny by the Commission. For Google Android, phone producers would be free to use Google's version of Android as well as competing versions. In addition, Google no longer requires manufacturers using its version of Android to pre-install Google Search and Chrome as a condition for access to Google's app store. Instead, they can pay a fee to license Google's app store. Google does however still offer incentives for the manufacturers to install Google Search and Chrome. As a remedy in the Google Shopping case, Google changed a search box at the top of its site that appears when a consumer searches for a product on Google's general search engine. It also introduced an 'auction system' which allowed other price comparison websites to bid for the chance to appear in the ads banner at the top of Google Search. Google Shopping also has to participate in these bids. In Google AdSense, the company had already terminated the relevant practices, so the Commission just prescribed that it should refrain from similar practices in the future.
The remedies in Google Android and Google Search have received considerable criticism, and questions are raised as to their effectiveness. In Google Android, critics say that it is too little, too late: consumers are accustomed to the presence of Google's apps on Android phones, and Android device producers will continue to have to include Google's applications in order to compete. The solution in Google Shopping has also attracted strong opposition as Google Shopping's bids are, according to critics, meaningless internal accounting measures, paid from one Google entity (Google Shopping) to the other (Google Search). In addition, critics claim that as long as the slots are populated by auction rather than by relevance, participants are required to bid away the majority of their anticipated profit due to the overpopulation of the auctions. In March 2019, Google introduced a new set of links in the result boxes at the top of certain searches improving their competitors' visibility. Top Commission officials, including Nicholas Banasevic, an EU official leading the Google probe, Johannes Laitenberger and the Commissioner herself have said that Google's changes to its search site are increasing traffic to online shopping services and that the "remedy is working".
Burden of proof and the presumption of innocence
In Intel, the Commission imposed a EUR 1.06 billion fine on Intel for abusing its dominant position. The Commission found that Intel had granted loyalty-enhancing rebates to its customers (such as Hewlett-Packard, Lenovo or Dell) in order to exclude its competitor AMD from the market. However, Intel brought an action for annulment against the decision before the General Court, which then rejected the action and upheld the Commission decision. Intel then lodged an appeal before the Court of Justice, which quashed the General Court’s judgment on procedural grounds. The case has now been referred back to the General Court, which must rule again on the merits of the case.
This case is particularly interesting in the debate on whether or not to shift the burden of proof in digital cases or whether to apply per se prohibitions. That point was reflected in the speech of Mr Tirole at the European Commission's conference on shaping competition policy in the digital world (17 January 2019), in which he claimed that due to an absence of information, it is hard to prove suppression of competition in digital cases. He argued that the burden of proof should therefore in some instances be shifted from the regulator to the undertakings concerned. This is also one of the recommendations featured in the special advisors' recent report.
Mr Tirole's position has been heavily criticised, especially when it comes to anticompetitive agreements and abuse of dominance cases (in contrast to merger control), as courts have consistently held that such cases are quasi-criminal in nature. Accordingly, shifting the burden of proof to the defendant in such cases is likely to run counter to the presumption of innocence and the rule of law. It will be interesting to see how these two divergent schools of thought will further develop.
2.5. State aid
In 2013, the Commission estimated that EUR 250 billion would be required to achieve its 2020 broadband targets, and in addition EUR 500 billion to achieve the 2025 broadband targets. It comes as no surprise that State aid policy is therefore an important factor in relation to these targets and the digital economy in general. The Commission has approved several aid measures in recent years.
In the Lithuanian RAIN 3 case, the Commission approved a EUR 50 million extension of the Lithuanian governments RAIN funding, for the development of broadband networks. The approval was subject to the condition of third party access to the network on equal and non-discriminatory terms. In addition, the infrastructure was limited to remote rural areas in Lithuania where no equivalent structures are currently in place or planned by private investors. In the Bavarian Gigabit project, the Commission approved under EU State aid rules a national project to deploy very high capacity networks in six municipalities. The aid aims at bringing very fast broadband to customers in areas where the market does not provide such access, in line with the EU broadband connectivity goals. This decision is relevant because for the first time the Commission has looked at a support measure in the context of the objectives of the Gigabit Communication and, in particular, is the first support measure involving a “step change”. In the Greek Superfast Broadband (SFBB) Project the Commission approved, for the first time, a national demand aid scheme providing vouchers to support increased take-up by covering part of the costs for the set-up as well as the monthly fee for a maximum of 24 months.
With respect to the so-called digital switchover, which is the migration to digital broadcasting, the Commission in principle supports the idea to grant State aid to achieve a quick switchover to free up spectrum for alternative uses. However, aid must be necessary and appropriate and should not impact platform competition between terrestrial, cable and satellite TV providers.
That was not the case in Comunidad Autónoma de Galicia, Redes de Telecomunicación Galegas Retegal SA (Retegal) v Commission. In this case, the Commission considered that the aid to switch from analogue to digital terrestrial TV in Spain was not compatible in part because the "vast majority" of tenders were not technologically neutral. This was confirmed by the Luxembourg Court: in the absence of robust studies justifying the choice for a particular technology, tenders that are not technologically neutral cannot be allowed.
Despite a number of positive Commission decisions, broadband goals are far from being achieved and it is likely that more aid will be needed to fulfil the targets. In that regard, the European Court of Auditors has stated in a report that the Commission should clarify the way in which it interprets the State aid rules in the broadband sector, as some Member States seem to adopt a more restrictive approach, resulting in less aid being granted. At the same time, Johannes Laitenberger stated that "of 38 billion euros' worth of projects cleared under State aid since 2009, only 30% resulted in spending and the rollout of broadband infrastructure in Member States". The question is therefore also whether more aid will necessarily be enough, or whether better implementation plans are needed.
Critics claim that the competition rules are not adapted to deal with the challenges of the digital economy. However, in our view, competition law is an open-ended and flexible tool which can adapt to the changing circumstances in the digital world.
In addition, it should be kept front of mind, as Commissioner Vestager warned that "competition rules are not a panacea to address all challenges raised by the digitisation of the economy." She posited that we need rules that go further than competition and also indicated that trade defence instruments, rather than competition law, should be the first option to safeguard European industry from unfair practices by third countries.
Of course, while certain challenges exist, tweaks to the competition law framework may be necessary. Traditional approaches, such as leniency applications, may for example become less effective, or evidence of infringement may be harder to find (e.g. because algorithms do not keep a record of pricing decisions). This may require authorities to adapt their enforcement toolbox and adopt novel methods. Finally, in the area of merger control, the rules may have to be amended to capture important data-driven transactions and to avoid killer acquisitions. Importantly, the interplay between regulation and competition policy remains as topical as ever. The choice between regulation of the digital sector via regulatory measures or via competition law remains heavily debated.
These are interesting times and the challenges with which the current Commission is confronted will remain a point of contention for the new mandate.
[i] This article is based on the Report on the Status of the Regulation of the Digital Economy in the EU by the Communications and Digital Working Economy working group of the International Bar Association's Communication Law Committee. That report was prepared under the leadership of Hein Hobbelen, Francine Cunningham and Baptist Vleeshouwers and is available here.
[ii] The vote in the Council can be regarded at this stage as a mere formal step, as the text had already been informally approved by the two European co-legislators (Parliament and Council) before its official adoption.
[iii] This Regulation-turned-Directive is also colloquially known as 'SatCab' Directive