Key Takeaways

Contents

Key Takeaways 1

Introduction. 2

The Digital Markets Act Increases Regulatory Fragmentation. 3

The DMA as a Pan-European Regulation. 3

The Risk of Decentralized Enforcement 4

The Digital Markets Act Ignores Legitimate Justifications 6

Blanket Prohibitions as Economically Harmful 8

Blanket Prohibitions as Fundamentally Unethical 9

The DMA’s Impossible Implementation. 12

Conclusion. 15

Endnotes 16

Introduction

The EU is about to adopt the DMA, which the European Commission proposed in December 2020.[1] The DMA is now expected to be applicable in 2023.[2] This legislation intends to regulate competition in the digital sector by imposing many obligations and prohibitions on a limited number of designated “gatekeepers.”[3]

The DMA falls into the broader DSM strategy outlined in 2015, which has led to multiple regulatory proposals for the digital sector.[4] The DMA’s rules will allegedly foster fair competition and contestability in digital markets.[5] However, they will also generate considerable unintended consequences at the expense of both innovation and consumers not only in Europe but globally.[6] Indeed, the DMA imposes ex ante regulatory obligations that depart from traditional ex post antitrust liability.[7] This departure will inevitably lead to many false positives: The DMA will prohibit pro-innovative practices only because they are carried out by designated “gatekeepers.”[8]

The DMA’s inherent bias toward the status quo together with its shift toward ex ante regulatory rules suggest that the DMA embeds a precautionary approach against innovative companies.[9] It favors static competition over dynamic competition and fairness over disruption; in other words, precaution over innovation.[10] The DMA has three major flaws.

First, the DMA does not help achieve the “DSM,” the goal of which is not regulatory harmonization—that is the means. Rather, the goal is increased digital innovation, efficiency, and productivity. As such, the DMA will work counter to that goal by imposing restrictive regulatory obligations that will harm innovation. Further and ironically, the design of the DMA provides nation states with enforcement powers, which will increase regulatory fragmentation.

Second, the DMA contains per se prohibitions that are not subject to pro-efficiency and pro-innovation justifications. This contradicts the EU’s principle of proportionality and is likely to increase compliance costs and reduce innovation.

Third, the DMA lays down numerous obligations and prohibitions that may contradict other EU legislation. This will inevitably lead to litigation and uncertainty—potentially deterring innovative companies.

The Digital Markets Act Increases Regulatory Fragmentation

Regulatory fragmentation is costly: The EU’s fragmented digital markets make it hard for firms to scale. Indeed, existing regulatory barriers impede the creation of an integrated DSM, hindering companies’ efforts to innovate and create greater efficiencies. The European Commission considers that “Europe has the capabilities to lead in the global digital economy, but [it is] currently not making the most of them. Fragmentation and barriers that do not exist in the physical Single Market are holding the EU back.”[11]

The European Commission assessed some years ago that the completion of the DSM with the removal of regulatory barriers would generate, over a few years, €415 billion.[12] Although quantifying such counterfactuals may prove to be a herculean task, it is reasonable to expect that the cost of an uncompleted DSM amounts to hundreds of billions of euros per year.

Indeed, scalability is crucial in general and, in particular, in the digital sector.^[13] The European Commission noted that “in many cases, without these economies of scale, an online business may not be viable, either because there is not enough demand in a single Member State or because lower production increases prices to unaffordable levels.”[14] Scale economies also allow for cross-subsidization, a key characteristic of platforms’ two-sidedness: Absent sufficient size and product diversification, platforms can hardly engage in cross-subsidization by offering consumers certain products for free in exchange for leveraging capabilities in adjacent markets.[15] The “extreme returns to scale” lead to larger platforms providing cheaper and better services than smaller platforms can.[16]

Against this background, the DMA is intended to limit regulatory fragmentation by producing an EU-wide regulation that would preclude national regulatory initiatives on digital competition. The Commission has already noted national regulations on digital competition taking place in Belgium, Bulgaria, France, Germany, Hungary, and Italy.[17] It stated that Member States “have begun to take regulatory initiatives to address these effects, potentially fragmenting the Internal Market.”[18]

Consequently, the Commission justified a DMA because it believes that “in the absence of action at EU level, action may be taken in different ways by Member States, risking the fragmentation of the single market and increased costs and frictions for platforms business users seeking to do business cross-border.”[19]

The DMA as a Pan-European Regulation

Although the DMA intends, according to its Article 1(1), to “contribute to the proper functioning of the internal market by laying down harmonised rules,” it does not achieve this.[20] While this would and should have been the only tangible and positive contribution of an EU-wide regulation concerning digital markets, the DMA in fact permits additional national regulatory initiatives, does not remove national regulatory barriers on how companies can operate online, and increases the risk of decentralized and disparate enforcement.

For example, Article 1(6) details that the Member States can set new rules and obligations on gatekeepers as long as they pertain to “national competition rules.”[21] Since the DMA lays down new competition rules and the Member States can lay down further competition rules on gatekeepers, the overlap will likely increase regulatory fragmentation. Germany’s passage of the “Digitalization Act,” which entered into force on January 19, 2021, amending the “Act Against Restraints of Competition (GWB),” illustrates this trend. While German law essentially adopts similar obligations and prohibitions as those contained in the DMA, it creates the risk of diverging interpretations. Given Germany’s leading role in the EU, it is inevitable that the other Member States will be inspired to follow suit and adopt their own “national DMAs.”

The potential for “overlap and inconsistency” between the DMA and DMA-like national rules is widely shared among commentators.[22] Hence, the estimated €92 billion in economic gains the European Regulatory Scrutiny Board anticipates from a decrease in internal market fragmentation will not occur because the “EU Member States will not need to introduce national legislations.”[23]

The Risk of Decentralized Enforcement

The DMA’s decentralized enforcement also threatens the DSM. During negotiations, The European Parliament added a provision to Article 31a that creates a “European High-Level Group of Digital Regulators” composed of a representative of the Commission, a representative of relevant Union bodies, representation of national competition authorities and representation of other national competent authorities.[24]

This group will advise the Commission on how to involve national competition authorities (NCAs) in decentralized enforcement of the DMA. Article 31c, which the European Parliament added, further confirms this, describing “the role of national competition authorities and other competent authorities,” this article states that NCAs “shall support the Commission in monitoring compliance with and enforcement of the obligations laid down in this Regulation.”[25]

Consequently, the DMA will be in the hands of the NCAs, thereby opening the doors to decentralized enforcement, which runs counter to the goal of minimizing regulatory fragmentation.

We already see signs of this decentralized approach. The “Friends of an Effective Digital Markets Act” coalition, composed of Germany, France, and the Netherlands, has issued a position paper titled “Strengthening the Digital Markets Act and Its Enforcement.”[26] Despite the name of this self-proclaimed coalition, its suggested ideas would undermine the very effectiveness of the DMA by encouraging divergent enforcement at the national level.

Among other suggestions, the coalition has proposed that member states be able to set and enforce national rules around competition law because the “importance of the digital markets for our economies is too high to rely on one single pillar of enforcement only.” They also state that “a larger role should be played by national authorities in supporting the Commission [in enforcing the DMA].”[27]

The coalition also advocates for “private” enforcement of the DMA—namely, firms should be able to sue gatekeepers to enforce obligations.[28] This reflects a misguided view that such private enforcement will increase the effectiveness of the DMA rather than open the door to efforts by powerful rivals to stifle competition and innovation.[29]

The pressures exercised by the powerful states represented in the Friends of an Effective Digital Markets Act coalition have paid off. Executive Vice President of the European Commission Margrethe Vestager now promotes national authorities’ role in the DMA enforcement.[30] Germany is already asserting its power and influence in determining how to enforce the DMA.[31] Additionally, the European Competition Network of NCAs issued in June 2021 a joint paper on “how national competition agencies can strengthen the DMA.”[32]

On September 8, 2021, the Friends of an Effective Digital Markets Act intensified its pressure during the negotiation process of the DMA with the publication of a second joint position paper, in which they suggested introducing tailor-made remedies.[33] Still, and most importantly, they have advocated for a more decisive role for the Member States and especially NCAs in the enforcement of the DMA. While acknowledging that centralizing regulatory powers at the EU level “improves effectiveness and prevents fragmentation,” the Friends nevertheless suggest that NCAs “may complement” the Commission’s enforcement actions regarding the DMA.[34]

National pressure to decentralize enforcement of the DMA with a more significant role for the NCAs has made considerable inroads in the latest version of the DMA. Indeed, the political agreement reached between the EU institutions has entirely reformed Article 31d(1) of the DMA. As drafted by the European Parliament, this article imposed limits and requirements on the role of Member States in enforcing the DMA. The new version of the article has now removed any restrictions, simply stating that “the Commission consult national authorities where appropriate, on any matter relating to the application of the Regulation.”[35] This absence of conditions provides considerable freedoms for the Member States to press for more frequent and regular consultations. Furthermore, the new article includes cooperation not only on the enforcement of the DMA but also of “available legal instruments”—meaning Member States can press for weaponizing other national laws in order to achieve a wide range of objectives.[36]

The new Article 32(a)(1) of the DMA clearly mentions that NCAs can “launch an investigation on gatekeepers based on national laws” and, in such case, the NCAs only have to inform the European Commission of such action.[37] Any NCA can adopt interim measures against designated gatekeepers according to Article 32a(3); the relevant NCA has the sole duty to inform the Commission before or after adopting such a measure.[38] Clearly, these mere reporting requirements will not deter NCAs from enforcing the DMA through the weaponization of the other “national laws” of competition, since the alleged anticompetitive practice can be presumed. Indeed, a practice prohibited by the DMA will de facto be anticompetitive under national competition laws. Correspondingly, the NCAs can, through Article 32a(1), enforce at a decentralized level the DMA, with the Commission left with little information and coordination capacities, since NCAs only have to report after their decisions and actions are made.

Article 32a(6) of the latest version of the DMA explicitly provides for the direct enforcement of the DMA by NCAs without recourse to the disguised use of national competition rules. Indeed, this Article states, “Where it has the competence and investigative powers to do so under national law, a competent authority of the Member States enforcing the rules referred to in Article 1(6) may on its own initiative investigate a case of possible non-compliance with Article 5, 6, and 6a of this Regulation on its territory.”[39]

Such unilateral and local enforcement of the DMA’s obligations and prohibitions is only subject to the duty of the relevant NCA to inform the Commission of such investigation. In other words, the latest version of the DMA imposes virtually no limit on decentralized enforcement of the DMA, in line with the recommendations made by the Friends of an Effective Digital Markets Act. However, given the lack of coordination, such decentralized enforcement of the DMA at the EU level will be less effective and enforcement at the national levels more prone to judicial actions—let alone be a convincing legal basis for NCAs to enforce a regulation initially designed to remove regulatory fragmentation.[40]

The role of Member States in enforcing the DMA has increased considerably in the latest version of the DMA from the Commission’s original proposal. First, the Commission and NCAs will coordinate their respective enforcement actions under the DMA and competition rules regarding designated gatekeepers. Second, NCAs will be able to initiate their investigations on gatekeepers suspected of violating obligations listed in Article 5 and Article 6 of the DMA. Third, NCAs will be active in the High-Level Group instituted by the DMA. Fourth, Member States will be involved in the Digital Markets Advisory Committee, also created by the DMA.

Finally, three or more Member States will be able to request the Commission to open investigations against a potential gatekeeper or add new obligations to the DMA. Still, only one Member State will be authorized to ask the Commission to open an investigation against a gatekeeper for alleged systematic non-compliance with the DMA. The prospect of regulatory fragmentation will prevent companies and consumers from reaping the benefits of the lack of fragmentation in the DSM—the very basis for the adoption of the DMA in the first place.

The Digital Markets Act Ignores Legitimate Justifications

Commentators have criticized the DMA for arbitrarily imposing numerous obligations and prohibitions on certain companies while carving out exemptions for their rivals to be exempt from the DMA’s ambit.[41]

The arbitrary line-drawing rules of the DMA regarding size threshold and qualitative criteria used to designate the so-called “gatekeepers” makes little sense economically.[42] It permits treating firms with similar market positions differently based merely on whether they fall within the DMA’s ambit. In addition, the DMA does not define markets and disregards fundamental competition law notions such as “market dominance.”[43] Thus, nondominant companies will be subject to these new competition rules, while some dominant companies will be exempted from these very rules. This is the DMA’s paradox of potentially treating market challengers more stringently than it does incumbents.[44] Recital 5 of the DMA explicitly states that designated gatekeepers “are not necessarily dominant in competition-law terms.”[45]

Various examples demonstrate how the DMA will prevent nondominant companies from competing—turning principles of competition law on their head. For instance, within the music streaming industry, the DMA will regulate Apple Music (which has 15 percent market share) while exempting Spotify (31 percent market share).[46] In the travel industry, the DMA will impose restrictions on Google Flights while exempting such incumbents as Expedia. Within the social media industry, the DMA will regulate Facebook while exempting Twitter.

Beyond the highly questionable size thresholds encapsulated in the DMA, the regulation is detrimental, as it lays down ex ante rules, which are per se rules of prohibitions. Per se rules allegedly are fast and easy to enforce. However, rules of reason (i.e., where legitimate justifications can be balanced against regulatory obligations) best prevent false negatives and better identify pro-competitive practices.[47] Thus, as a “per se rule regime,” the DMA detrimentally disallows defendants’ justifications at the expense of consumers benefitting from pro-competitive and pro-innovative practices.[48]

The DMA’s blanket prohibitions imposed on designated gatekeepers cannot be rebutted through gatekeepers’ assertions of economic justifications. The only exceptions the DMA will accept are justifications based on “public morality, public health or public security.”[49] Consequently, the prohibited practices are presumed to harm competition irrespective of efficiency considerations raised by the defendant, such as improving consumer welfare or product improvement by technological innovation.[50]

This shift from ex post antitrust enforcement to ex ante regulatory rules is tantamount to introducing the precautionary principle in antitrust matters: Early regulatory interventions intend to prevent disruptive behaviors and favor the status quo.

Regulators no longer fear the Type I error costs (i.e., the cost of false positives due to over-enforcement). Instead, they fear Type II error costs (i.e., the cost of false negatives due to under-enforcement). With the reversed burden of proof, early interventions with no possible justifications instill a precautionary approach to rapidly changing markets characterized by strong dynamic competition. This abrupt enforcement is justified in the name of swiftness, but this argument overlooks several key benefits of the traditional approach to antitrust enforcement.

First, time-consuming proceedings in antitrust can have positive consequences: During the time of proceedings, initial suspicions of monopolization or guesses about market evolutions can be confirmed or infirmed by the time judges deliver their verdict. Consequently, such judgment will better correspond to the market realities, since the antitrust analysis will be more retrospective than prospective by the time of the delivery of the decision.

Second, blanket prohibitions, such as those contained in the DMA, can have a considerable deterrent effect for companies: The opening of investigations will force the company to alter its behaviors even before legal challenges can run their course. This can result in consequences that are bad for the economy and consumers.

Third, the evolutionary process of judge-made law in the traditional antitrust approach permits companies to better anticipate and adjust their behavior according to evolving case law. Lawyers and advisors ensure that entrepreneurs quickly react to a changing legal environment.

Thus, the argument that ex post antitrust enforcement cannot work in dynamic markets such as digital markets disregard the many dynamic markets into which antitrust has historically intervened. It also ignores several key benefits of the traditional antitrust approach. Beyond these necessary concerns for the implication of ex ante rules of competition, the ex ante rules the DMA contains are concerning for two main reasons: They defy economic logic while ignoring fundamental rights.

Blanket Prohibitions as Economically Harmful

The DMA’s per se prohibitions include pro-competitive, pro-innovation practices. One seminal example is the prohibition of self-favoring. Article 5(1) prohibits self-favoring practices, although such practices are common business practices, spur competition, result in and propel innovation, and enhance consumer benefits. Moreover, irrespective of the beneficial effects of tying, bundling, and leveraging strategies on competition and innovation, the DMA bluntly prohibits such practice. The impact of this prohibition will be significant, as it will harm consumers and hundreds of thousands of businesses that operate in the gatekeepers’ ecosystem and benefit from gatekeepers’ services and products.

These prohibitions could also have a considerable chilling effect. Non-gatekeeper platforms will likely assess that such practices could be considered anticompetitive by regulators, though the DMA may not in fact impose any restrictions on them.[51] Indeed, the DMA’s per se prohibitions of allegedly “unfair” practices will trickle down to the whole economy as the DMA may inevitably be used in courts in traditional competition cases.[52]

Thus, the DMA is likely to generate numerous opportunity costs: Deterred innovation, reduced consumer welfare, and abridged consumer and business users’ choice are likely to effect the DMA’s per se prohibition rules, which do not all have economic justifications.

The DMA aims to allegedly prohibit the “egregious nature of unfair practices.”[53] In reality, the DMA will end up banning and deterring pro-competitive and pro-innovative practices such as self-preferencing, data collection, combining data, and leveraging capabilities at the expense of consumer welfare and innovation.

The precautionary logic of the DMA prevents regulators from fearing erring on the side of over-enforcement and false positives. Still, it will nevertheless be the case, with the associated costs and disincentives over-enforcement inevitably generates.

Blanket Prohibitions as Fundamentally Unethical

The DMA’s per se prohibitions are not only economically detrimental but also legally wrong. This is true for at least two reasons: (1) they violate the principle of proportionality encapsulated in EU treaties and (2) they violate the fundamental right of defendants, also encapsulated in the EU treaties. Because the DMA is a piece of legislation that ultimately must comply with EU treaties, EU judges will either declare the DMA illegal or, more realistically, trim it down to integrate proportionality considerations into the enforcement of its blanket rules.

The principle of proportionality is a foundational principle in EU law that broadly corresponds to the English common law notion of reasonableness; it comes down to the idea of justice. Disproportionate judicial sentences and rules are unjust and unacceptable. Interestingly, the principle of proportionality is also paring down to a cost-benefit analysis. A rule is unjust if the expected benefits are lower than its costs. Thus, the principle of proportionality has also become a principle of economic efficiency.[54]

As former advocate general Sir Francis Jacobs once said, “As for the principle of proportionality, there are few areas of Community law, if any at all, where that is not relevant.”[55] The EU principle of proportionality derives from the German principle of proportionality, and stands in Article 5(4) of the TEU.[56] Article 5 of this Protocol notably states, “Draft legislative acts shall take account of the need for any burden, whether financial or administrative, falling upon the Union, national governments, regional or local authorities, economic operators and citizens, to be minimized and commensurate with the objective to be achieved.” The EU courts have interpreted the principle of proportionality similar to German courts as a bundle of three cumulative rules:

1. The review of the necessity of the measures to achieve the desired objective.

2. The review of the suitability (or less-restrictive means test) of the measures for the achievement of the objective.

3. The review of the proportionality stricto sensu, whereby the burden imposed must be of proportion with the goal desired.[57]

In other words, each regulatory intervention and obligation must be strictly necessary to the goal, be the least restrictive means available to achieve that goal, and be in proportion with the goal to be completed.

It is likely that “the current version of the DMA proposal could breach the principle of proportionality.”[58] EU judges have considered that “draft legislative acts must take account of the need for any burden falling upon economic operators to be minimised and commensurate with the objective to be achieved.”[59]

When the DMA goes into effect, the blanket prohibitions applicable to designated gatekeepers will likely fail to comply with the EU principle of proportionality.[60] Current advocate general Giovanni Pitruzzella has indeed expressed concern about the compliance of the DMA’s obligations and remedies to be imposed with the EU principle of proportionality:[61]

Normally, in competition law, presumptions are rebuttable and the party can also use the efficiency defence. Doing so, it is possible to strike a balance between the need of certainty and saving time in the administrative activity (which pushes in favour of the use of presumptions) and the need to avoid false positives in antitrust enforcement and undue limitations of fundamental rights (which pushes in favour of a case-by-case approach). At a first reading, the proposal of Regulation does not allow this kind of limitations in the use of the listed presumptions. The system is more rigid than the one envisaged in the recent advice published the 8th of December 2020 by the Digital Taskforce appointed by the CMA. I wonder whether too much rigidity could hinder efficiency and introduce a disproportioned limitation on the freedom to conduct a business.[62]

The obligations and prohibitions imposed by the DMA are arguably not the least restrictive means to achieve the stated objectives. While the prohibited practices are allegedly limited to unfair practices, many prohibited practices are part of the fair competitive process.

For instance, Article 5(1)a prohibits gatekeepers from the “cross-use of personal data from the relevant core platform service in other services offered separately by the gatekeeper, including other core platform services, and vice-versa.” This would mean that, for example, Google will no longer be able to cross-use data from its Gmail service to customize its Google Pay services to compete with incumbents such as PayPal or MasterCard. It would also mean that Amazon will no longer be able to use the data on its department store to improve service for, say, its streaming video services to compete with Netflix or Hulu. In other words, the DMA’s prohibited practices are not unequivocally unfair practices but instead can be pro-competitive, pro-efficient practices.

Similarly, under Article 5(1)(f), gatekeepers will no longer be able to require consumers to register to one core platform service to use any other core platform services. For instance, Apple will be forced not to require iPhone users to use, say, Apple Pay. Similarly, Amazon will no longer be able to require an Amazon account in order to read ebooks on Kindle. These stringent prohibitions that provide no clear benefits for users impinge on the designated gatekeepers’ ability to conduct business and are even less justified when the gatekeepers are in a nondominant market position. Such impediments are not imposed on incumbents the gatekeepers try to challenge to benefit consumer welfare and innovation. Consequently, it is undeniable that “the approach followed in Articles 5 and 6 of the DMA Proposal would impinge on platforms’ fundamental right to conduct their business and their right to property.”[63]

Blanket prohibitions of potentially pro-competitive practices arguably violate the EU principle of proportionality, since the prohibitions are not narrowly tailored only to combat unfair practices, contrary to the DMA’s claims. Because the prohibited practices are not narrowly tailored to unfair practices, EU judges may trim down the DMA’s set of obligations and prohibitions to make them more reasonable (in other words, compliant with the EU principle of proportionality).

The possibility of an abridged DMA by EU judges once the DMA is adjudicated in courts is highly likely and arguably beneficial, since disproportionate prohibitions both embed a sense of economic injustice and generate economic harm. In other words, to borrow Pablo’s metaphor, the EU judges will be no Arnaud Amalric in enforcing the DMA:

To put it graphically, the DMA Proposal’s approach to proportionality (absolute prohibition coupled with the possibility of the Commission tailoring the obligations on a case-by-case basis) is reminiscent of the proportionality approach taken by Arnaud Amalric when asked how to distinguish the Cathars from the Catholics in battle: Caedite eos. Novit enim Dominus qui sunt eius (i.e., “Kill them all and let God sort them out”).[64]

This is even more likely given that the EU principle of proportionality is tantamount to a cost-benefit analysis.[65] Thus, EU judges will ensure the costs the DMA regulation are lower than the benefits expected from the DMA.

As EU judges will likely strike down some provisions of the DMA or at least apply such provisions in line with the EU principle of proportionality, the targeted gatekeepers will likely cite economic justifications to explain their practices and lead to a qualified application of these obligations in practice. Indeed, Article 16 of the Charter of the Fundamental Rights of the European Union protects the “freedom to conduct business.”[66]

The DMA’s ex ante rules of per se prohibitions are bad policy that very much frustrate the principles of fair competition and ignore the fundamental legal principles of the EU legal order. A more reasonable approach would be to apply a rule of reason where judges conduct a balancing test between rules’ positive and negative effects. Ultimately, it is likely that EU judges will inevitably adopt such an approach when adjudicating the DMA.

The DMA’s Impossible Implementation

The DMA has a positive yet unachieved goal to minimize regulatory fragmentation in the DSM. But one aspect of the DMA that deserves particular attention—and, paradoxically, appears to be overlooked—is how its rules may conflict with other EU rules. Even if designated gatekeepers willingly comply with DMA’s obligations, this will likely lead them to violate other EU obligations.

Designated gatekeepers must comply with various obligations and prohibitions applicable to digital platforms including:

▪ EU Regulation 2019/1150 of June 20, 2019, on promoting fairness and transparency for business users of online intermediation services[67]

▪ EU Regulation 2016/679 of April 27, 2016, on the protection of natural persons with regard to the processing of personal data and on the free movement of such data (General Data Protection Regulation—or “GDPR”)[68]

▪ EU Directive 2016/943 of June 8, 2016, the protection of undisclosed know-how and business information (trade secrets) against their unlawful acquisitions, use, and disclosure[69]

▪ Commission Directive 2002/77 of September 16, 2002, on competition in the markets for electronic communications networks and services[70]

▪ EU Regulation 2018/1807 of November 14, 2018, on a framework for the free flow of nonpersonal data in the European Union[71]

▪ Commission Directive 2008/63 of June 20, 2008, on competition in the markets in telecommunications terminal equipment[72]

▪ EU Directive 2019/770 of May 20, 2019, on certain aspects concerning contracts for the supply of digital content and digital services[73]

▪ EU Direction 2018/1972 of December 11, 2018, establishing the European Electronic Communications Code[74]

▪ EU Directive 2000/31 of June 8, 2000, on certain legal aspects of information society services, in particular electronic commerce, in the internal market[75]

▪ EU Directive 2002/58 of July 12, 2002, concerning the processing of personal data and the protection of privacy in the electronic communications sector[76]

▪ EU Regulation 2017/1128 of June 14, 2017, on cross-border portability of online content services in the internal market[77]

▪ EU Regulation 2019/881 of April 17, 2019, on ENISA (the European Union Agency for Cybersecurity) and information and communications technology cybersecurity certification[78]

▪ EU Directive 2013/40 of August 12, 2013, on attacks against information systems[79]

▪ EU Regulation 910/2014 of July 23, 2014, on electronic identification and trust services for electronic transactions in the internal market[80]

▪ EU Directive 98/84 of November 20, 1998, on the legal protection of services on, or consisting of, conditional access[81]

▪ EU Direction 2019/790 of April 17, 2019, on copyright and related rights in the DSM[82]

▪ The proposed Digital Services Act (DSA) was announced on December 15, 2020[83]

▪ The proposed Data Act was announced on February 23, 2022[84]

▪ The proposed European Digital Identity framework was reported on June 3, 2021[85]

▪ The proposed Cyber Resilience Act was announced on March 16, 2022[86]

▪ The proposed Artificial Intelligence Act was announced on April 21, 2021[87]

▪ The proposed Data Governance Act was presented on November 25, 2020[88]

Additionally, traditional rules of contracts, commerce, and provision of services also apply to designated gatekeepers.

The DMA will likely bring the existing complex regulatory framework into further chaos. For example, the DMA forces gatekeepers to unbundle core platform services from one another. Article 5(f) of the DMA requires gatekeepers to “refrain from requiring business users or end-users to subscribe to or register with any further core platform services … as a condition for being able to use, access, sign up for or registering with any of their core platform services identified under that Article.”[89] The DMA’s forced unbundling provisions will prevent gatekeepers from complying with Article 7 of Directive 2019/770, which requires platforms to describe the functionality and interoperability of their core platform services. This is because the DMA simultaneously allows other business users to alter the original functionality and interoperability of any given core platform service, making it functionally impossible for gatekeepers to have full knowledge of the interoperability and functionality of their services.

For example, Apple will no longer be able to require the use of its App Store (i.e., an “online intermediation service” according to the DMA’s Article 2(2)(a)) to operate its iOS operating system (I.e., an “operating system” according to the DMA’s Article 2(2)(f)). It will be forced to unbundle its iOS with its App Store for app developers to access Apple’s operating system outside of the App Store. Beyond the blatant cybersecurity risks this obligation will generate, such questionable unbundling will prevent Apple from complying with Article 7 of Directive 2019/770. How can Apple sell the functionality of its iOS should business users no longer have access to the App Store? How can Apple accurately describe the interoperability of third-party apps with its operating system if app developers can circumvent the App Store?

Additionally, the DMA lays down data portability rights for business users vis-à-vis gatekeepers. These rights may violate the very privacy rights of end users enshrined in the GDPR. Article 6(i) of the DMA states that gatekeepers must:

provide business users and third parties authorised by a business user, upon their request, free of charge, with effective, high-quality, continuous and real-time access and use of aggregated and non-aggregated data, including personal data, that is provided for or generated in the context of the use of the relevant core platform services or services offered together with or in support of the relevant core platform services by those business users and the end users engaging with the products and services provided by those business users; for personal data, provide access and use only where the data are directly connected with the use effectuated by the end user in respect of the products or services offered by the relevant business user through the relevant core platform services, and when the end user opts in to such sharing by their consent.[90]

The DMA creates a duty for the gatekeeper to port data to a business user whenever the end user has “opted in” to such sharing. But after a one-time general acceptance by the end user at the time when the end user first registers with the business user, the business user can receive any personal data from gatekeepers to an excessive extent. This data sharing would take place without the end users being fully aware that the data generated when using the gatekeepers’ services would ultimately be possessed by a large number of business users, and thereby likely conflict with GDPR rules, which outline the conditions for the end user’s consent to data processing. For instance, Article 7(4) prohibits the unnecessary processing of personal data for the performance of a contract. In conflict with this, however, the DMA gatekeeper may require gatekeepers to share data with business users beyond what is strictly necessary to provide the requested services.

At the same time, the DMA may also conflict with GDPR data portability requirements. The DMA’s data disclosure obligation relates only to the data “directly connected with the use effectuated by the end-user regarding the products or services offered by the relevant business user.” The definition of “directly connected” remains to be determined in practice by the agencies and courts; however, such restriction on the scope of data that may be transferred may lead gatekeepers to violate Article 20 of the GDPR, which regulates data portability for personal data. This article grants end users the right to port personal data “without hindrance” from the data controller. Either the gatekeeper complies with Article 6(i) of the DMA and limits data portability of personal data and risks violating Article 20 of the GDPR, or the gatekeeper provides data portability “without hindrance” as provided by Article 20 of the GDPR and potentially violates Article 6(I) of the DMA.

Article 6(i) is only one example of the many obligations the DMA contains that may conflict with EU data protection rules. Indeed, commentators agree that the compatibility between the DMA and the GDPR “remains rather obscure.”[91]

The DMA may also conflict with requirements under the DSA. For instance, Article 5 of the DSA requires online platforms to act “expeditiously to remove or to disable access to the illegal content,” or else the liability of the platform can be engaged. But Article 6(1)(k) of the DMA obliges gatekeepers to offer “fair, reasonable, and non-discriminatory general conditions of access for business users to its software application store, online search engines, and online social networking services identified in the designation services.” In other words, the gatekeepers must provide a “fair” placement to any business users.

What if the business user is an app such as Parler, which contains hate speech that is illegal under the DSA? Apple, Google, or Amazon running app stores or cloud services may remove the app to prevent unlawful content from disseminating under Article 5 of the DSA. By doing so, however, they may violate Article 6(1)(k) of the DMA, since the business user (i.e., Parler) could successfully argue that the app has been unfairly discriminated against. Similarly, websites containing possibly illegal content may successfully claim that the gatekeepers have violated the DMA’s Article 6(1)(k) by demoting or removing their websites from search engine results.

The result of these regulatory dilemmas is gatekeepers, seeking to minimize their legal risks, will adopt a more precautionary approach to innovation, ultimately harming consumers and overall economic growth.

Conclusion

The DMA is a significant piece of legislation that will revolutionize how competition occurs online—but not for the better. It will likely lead to three negative, unintended consequences.

First, the DMA is likely to increase, rather than decrease, regulatory fragmentation. EU Member States are likely to worsen regulatory fragmentation with decentralized enforcement of the DMA and additional national rules. Thus, the DMA is a missed opportunity for imposing regulatory harmonization across the EU.

Second, the DMA’s per se prohibition on practices that may be pro-competitive can hinder economic growth and harm consumer welfare. This concern is compounded by the DMA’s failure to provide sufficient economic justification for these bans likely violating EU principles of proportionality. Judges will have to grapple with the DMA’s conflict with the proportionality principle, and may ultimately curtail future DMA enforcement.

Third, the DMA will create regulatory confusion. Gatekeepers will ultimately need to decide whether to comply with the DMA or other EU regulations, including the GDPR. This confusion will further chill the ability of covered platforms to innovate and foster consumer welfare.

As the DMA nears final adoption, EU lawmakers need to improve its flaws, possibly through enforcement guidelines. At the same time, EU judges will have to determine how to square the DMA’s requirements with other rights afforded under EU law. Unless judges or lawmakers ultimately address the DMA’s fundamental flaws, European innovation and consumers will be collateral damage.

About the Author

Dr. Aurelien Portuese (@aurelienportues) is the director of The Schumpeter Project on Competition Policy at ITIF. He focuses on the interactions of competition and innovation. Portuese holds a Ph.D. in law and economics from the University of Paris II (Sorbonne), an M.Sc. in Political Economy from the London School of Economics, and an LL.M. from the University of Hamburg. Portuese’s latest book, Algorithmic Antitrust, was published in 2022 by Springer.

About ITIF

The Information Technology and Innovation Foundation (ITIF) is an independent, nonprofit, nonpartisan research and educational institute focusing on the intersection of technological innovation and public policy. Recognized by its peers in the think tank community as the global center of excellence for science and technology policy, ITIF’s mission is to formulate and promote policy solutions that accelerate innovation and boost productivity to spur growth, opportunity, and progress. For more information, visit us at www.itif.org.