A. Introduction
The numerous corporate scandals and revelations regarding failures to address harms on social media have unleashed a torrent of views on how corporations cannot moderate themselves and how their power needs to be reined in.Footnote 1 In the European Union (EU) such incidents have given rise to a strong momentum for regulation, which has already taken a more specific form in the flurry of legislative initiatives, including most prominently the upcoming Digital Services Act (DSA) and the Digital Markets Act (DMA).Footnote 2 Indeed, there is ample theoretical and mounting empirical evidence pointing to the need for constraining regulation.Footnote 3 In this vein, there is no shortage of accounts presenting the EU’s new regulatory agenda as a watershed, a nod to its ambition and novelty.Footnote 4 My goal is not to dispute these accounts. Yet change does not occur in a completely disoriented way and it is just as important to subject the patterns of continuation and consolidation to scrutiny as it is those of transformation.Footnote 5 As Sheila Jasanoff has stated, institutions “offer ready-made instruments for putting things in their places at times of uncertainty and disorder.”Footnote 6 Thus, it is apposite to analyze the current regulatory vision in the various formative contexts within which such thinking takes place in the EU. One such context concerns the collective understanding of the legally possible and desirable within the Union, that is, the context of EU law institutions.
This article analyzes how the EU attempts to govern online platforms, particularly social media, as a part of internal market construction. Internal market construction, namely Article 114 of the Treaty on the Functioning of the European Union (TFEU), lends the DSA and DMA proposals their legal basis.Footnote 7 As the Union’s jurisdictional boundaries demarcate which visions it may entertain, the shared legal understandings of “what the internal market is” and “what it needs”Footnote 8 partake in shaping what platforms are in Europe, what is thought they should be, what their potentials and attendant risks to European societies are, and how to govern these platforms. Conversely, the shared legal understandings also affect what platforms are not, inhibiting other ways to think about them.Footnote 9
The Article unfolds as follows. In Section B, I first explore how the internal market jurisdiction is conceptualized, that is, what is understood to be a rational construction of the internal market, as expressed in the reasoning of the Court of Justice of the European Union (CJEU or the Court). I find that in the CJEU jurisprudence the internal market construction is animated by a twofold aim. In internal market construction, transnational private freedom is to be constantly negotiated with public interest concerns. In Section C, I go on to map how the internal market construction, as informed by the twofold aim, has already partaken in constituting the rise of the platform economy in Europe through specific legal mechanisms. Focusing on social media as an example, I show how these mechanisms have been implicated in the emergence of platforms’ power and its attendant distributional inequalities in the first place.
In Section D, I then mirror the current “internal market rationality”Footnote 10 to the rationale for the DSA and DMA proposals. I argue that from the perspective of internal market construction, the EU’s vision as exhibited in these proposals cannot be conceived of as particularly transformative. While the proposals introduce many novel regulatory mechanisms with an equalizing potential, they evince institution-preserving qualities as well by remaining committed to both the twofold aim of internal market rationality and the central mechanisms facilitating the platform economy. Thus, the proposals risk reproducing and legitimizing various distributional inequalities and power relations in Europe. I emphasize the need to consider alternative visions of platforms in conjunction with the re-imagination of internal market construction, including its twofold aim. While I refrain from proposing normative, legal, or technical solutions ready for implementation, my hope is that the article can help expand the horizon of institutional imagination in the name of egalitarian aspirations within legal and policy discussions in Europe.Footnote 11
B. What the Internal Market Is and What It Needs
As a bread-and-butter issue for European lawyers, the EU may have recourse to any part of its legal toolbox only with proper legal competence, dictated by the principle of conferral in Article 5(2) of the Treaty on European Union (TEU).Footnote 12 Legislative intervention requires a legal basis in its Founding Treaties and thus the rules on legal jurisdiction are able to channel the direction of legal change. The Union is taken to have functional institutional design that focuses on common objectives.Footnote 13 It is expressed in the purposive formulation of legal bases, which empower the Union to achieve certain projects.Footnote 14 Therefore, to map the EU’s reasoning, it becomes paramount to explore not only the techniques of governing but also the underlying aims in the name of which such techniques are deployed. The choice of competence also influences the specific nature of competence, legal instruments that may be used, and the process including the voting rules in the Council among others.Footnote 15
The Commission has posited Article 114 TFEU as lending the Union competence to enact new platform law including the DSA and DMA.Footnote 16 Article 114(1) foresees that the Union may take action to “adopt the measures for the approximation of the provisions laid down by law, regulation, or administrative action in Member States which have as their object the establishment and functioning of the internal market.” Thus, the fact that a market economy is partly constituted by supportive legal constructions is perhaps particularly apparent in the case of European internal market.Footnote 17 Yet if we reject the idea that the internal market would have some essential meaning that could be discovered by a method and could then claim objectivity over other competing meanings, it becomes crucial to delve into how it is imagined what the internal market is and what it needs.Footnote 18 First, Article 26(2) TFEU prescribes that: “The internal market shall comprise an area without internal frontiers in which the free movement of goods, persons, services, and capital is ensured in accordance with the provisions of the Treaties.” Second, Article 3(3) TEU now stipulates that the internal market shall also, among other ideals, aim at being “a highly competitive social market economy,” seek social progress, justice and protection, and promote solidarity between generations and Member States.
Combined, these Treaty provisions appear rather indeterminate, lending credence to a host of heterodox meanings that possibly point in contradictory directions. Gareth Davies has argued that “[t]he legal bases are typically built around such open-textured concepts that almost any legislative act could be brought within a plausible interpretation of one of them.”Footnote 19 Thus, to explore the current understanding of the internal market more specifically, one must turn to institutions that are in a privileged position to interpret the provisions. One of them is the CJEU,Footnote 20 which has stressed that the choice of legal competence must be grounded in objective factors amenable to judicial review, including the aim and content of the measure.Footnote 21 The aim and content of EU platform law must sufficiently express commitment to the CJEU’s understanding of the internal market or risk annulment.Footnote 22 Therefore, the jurisprudence of the internal market legal basis appears to be a prime starting point for analyzing how internal market construction informs the governance of platforms. In the remainder of this section, I probe into how, in the case law on the internal market legal basis, the CJEU understands what the internal market is and what it “needs.”
To start with a relatively old case, in Titanium Dioxide, the Court determined that the directive, the legal basis of which was under dispute, sought to harmonize the environmental standards in the titanium dioxide industry and to improve the conditions of competition in that industry. It noted that the directive “thus pursues the twofold aim of environmental protection and improvement of the conditions of competition” which are “indissociable.”Footnote 23 The Court further accepted the Commission’s reasoning that this twofold aim could indeed be pursued within Article 100a of the Treaty Establishing the European Community,Footnote 24 a new device introduced by the Single European ActFootnote 25 and “particularly appropriate to the attainment of the internal market.”Footnote 26 Through subsequent treaty amendments, this device would eventually become the current Article 114 TFEU.
Almost ten years later, the Tobacco Advertising I case concerned a directive that included a total ban on print advertising and sponsorship of tobacco products in the internal market.Footnote 27 For the CJEU, “genuinely” improving the internal market establishment and functioning meant the removal of obstacles to free movement.Footnote 28 The removal of disparities between national rules is not an end in itself.Footnote 29 Although it is now plain that either removing obstacles to movement or appreciable distortions of competition is sufficient for competence, the two are intimately connected and mutually reinforcing.Footnote 30
Free movement refers to the Union’s fundamental freedoms—free movement of goods, persons, services, capital, and freedom of establishment across nations’ borders.Footnote 31 After brushing aside the implication that goods or services would move around on their own, it becomes clearer that the freedoms, as legal entitlements, are accorded equally to private actors seeking to cross the border of one Member State into another to advance their private projects.Footnote 32 In terms of subject production, this aim of transnational individual freedom and competition appears to disembed the subject from social settings, particularly from that of Member State polity. As Floris de Witte has argued, the emancipatory ethos of the Union seeks to uproot and liberate individuals from the exclusionary tendencies of the nation-state.Footnote 33 Fundamental freedoms and undistorted competition seek to ensure that individuals and other private actors have many transnational opportunities and choices in their pursuits. At the same time, this disembedding and the ensuing focus on the freedom of the atomistic individual tends to exclude a relational understanding of the subject, bracketing off the distributional inequalities and power relations between individuals and communities within the Union.Footnote 34 Such framing is aptly expressed in the economic term “externalities,” effects on others that are thought to be external to the transaction activity itself.
As already visible in Titanium Dioxide, the furthering of private actors’ transnational freedom and competition is by no means the only internal market need in the CJEU’s imagination. The Court has sanctioned that the Union legislature may, and even should, also pursue several public interest objectives. In Tobacco Advertising I, AG Fennelly was rather candid in stating that “the internal market is not a value-free synonym for general economic governance.”Footnote 35 Therefore, it is not necessary to interpret the internal market provisions of the TFEU “as a kind of liberal charter, entailing harmonization towards the lowest standard or even towards some sort of mean of the pre-existing national standards.”Footnote 36 The CJEU has even accepted that a public interest objective may be the “decisive factor” in the legislative choices as long as a link to fundamental freedoms or competition is found.Footnote 37
Indeed, a link may usually be found. While a public interest rationale rather than an economic one usually animates regulation,Footnote 38 the apparent tension between private freedom and public interest regulation may also be rationalized away. For instance, product bans may be framed as facilitating the free movement of other freely tradeable substitute products by helping, as in the case of a ban on seal products, to regain “consumer confidence while, at the same time, ensuring that animal welfare concerns are fully met.”Footnote 39 That way, public interest regulation can be understood as market-making as legal arrangements animated by free movement and competition.Footnote 40 In other words, public interest regulation is equally internal market construction.
It appears that the public interest objectives—especially health, safety, environmental protection, and consumer protection as listed in Article 114(3) TFEU—are imagined as feeding into the uniform EU-level standard that then fixes the level playing field for competition. Alexander Somek has argued that the Court’s understanding of the internal market competence amounts to “market holism” as opposed to “market liberalism,” the latter being rejected by AG Fennelly and the Court in Tobacco Advertising I. Market liberalism would exclude the pursuit of public interest objectives and would correspondingly militate toward lowering protective standards.Footnote 41 Generally, holism often refers to approaches where certain values are thought to pervade the totality of social intercourse. They must be integrated into practices as evaluative guidance. This can also be referred to as “the internalization of externalities.”Footnote 42
Public interest regulation makes conditions safe and products environmentally sustainable, and improves the position of the consumer in transactions. Arguably, the role of such market-intervening regulation has progressively increased with the privatization of public services from the 1990s onwards.Footnote 43 Yet public interest protection also exhibits a disembedding of social matters from local and national settings similar to the aim to liberate the individual. Public concerns are seen as rather anonymous and free-floating problems of governance to be resolved by universal expert reasoning.
During the last two decades, the Court’s relaxed interpretation of “measures for approximation” in Article 114(1) TFEU has afforded the Union power to create a supranational constellation of satellite administrative agencies orbiting the Commission and engineering the economy.Footnote 44 In matters associated with less vital interests, the Union has since the adoption of the New Approach and the Global Approach sought to create transnational markets for services certifying the compliance of products with open-ended “essential requirements” laid out in directives.Footnote 45 In addition, the European Standardisation Organisations have been entrusted with drafting Union-wide standards that entitle their adherents to the presumption of compliance with the law.Footnote 46 As Somek highlights, regulation presupposes a trusting subject that adheres to universal expert reasoning by assuming that the experts know better,Footnote 47 thus accepting intervention into their individual freedom or, in the words of the Court, regulation “circumscribing that freedom.”Footnote 48 Of course in practice, such circumspection is not uniformly accepted, as demonstrated by the legal challenges for judicial review initiated by UK corporations.Footnote 49
After Tobacco Advertising I, the Court has confirmed this twofold aim of the internal market with remarkable consistency. Despite Treaty amendments, most notably the introduction of Article 3(3) TEU, the Tobacco Advertising test is still being applied by the CJEU.Footnote 50 As stated by AG Sharpston in a 2019 case that concerned a ban on specific types of guns in the Union: “The Court’s case-law concerning tobacco products provides particularly useful guidance on the scope of Article 114 TFEU which can be applied by analogy here.”Footnote 51 The Court’s understanding may also reflect a more widely shared understanding of internal market governance in the Union, also because no such piece of internal market legislation appears to have been brought before the Court that would have invited its re-imagination.Footnote 52
The general tilt of the internal market is often argued to be, in various ways, toward “the economic,” that is, fundamental freedoms and competition.Footnote 53 Yet within judicial review case law it has also been argued contrariwise—that the internal market basis enables “covert” public interest harmonization, where legislation ostensibly introduced to further economic freedoms in fact pursues a public interest.Footnote 54 I wish to refrain from positing the abstract fundamentality of either.Footnote 55 Therefore, even though the shared legal understanding of the aims of the internal market do orient its construction across settings, my contention is that the twofold aim is nevertheless being negotiated and operationalized iteratively case by case, directing the analyst’s attention to how internal market rationality informs market construction in specific contexts.
C. Internal Market Mechanisms Constituting Online Platforms
I. The Internal Market Mechanisms of the Platform Economy
Thus far, I have explored the aims of internal market construction. I have argued that on the one hand, the internal market rationality, as expressed in the CJEU jurisprudence on internal market competence, pursues a twofold aim. On the other hand, it is connected to the facilitation of transnational freedom for private actors and a program for competition that largely brackets off concerns for distributional justice. Yet internal market rationality also seeks to infuse private activities with concern for public values through regulation informed by public and private expertise.
Now, situating online platforms within internal market construction firstly invites the question of what an online platform specifically means. In their article, Platformisation, Thomas Poell, David Nieborg, and José van Dijck define online platforms as “(re-)programmable digital infrastructures that facilitate and shape personalised interactions among end-users and complementors, organized through the systematic collection, algorithmic processing, monetisation, and circulation of data.”Footnote 56 Within economics and management studies, platforms have been understood through the organizational logic of a two-sided or multi-sided market.Footnote 57 Based on algorithmic data analysis, platforms intermediate communication and interaction between distinct groups of users on different “sides,” such as consumers, programmers, or advertisers.Footnote 58 Matching users draws upon the analysis of data that platforms produce and pool.
Julie Cohen has argued that a platform has also become the core organizational logic of a digital economy that transforms the market institution and makes it arguably more material.Footnote 59 By courting more and more users with the promise of access to information and connectivity, platforms scale up, creating “indirect network externalities” or network effects that make access increasingly valuable for everyone on any side of the platform.Footnote 60 It also becomes progressively harder for those not included to remain outside platforms “where everyone else already is.” That way, the scaling of the platform may result in reinstituting the multi-sided platform structure in place of a market.
That the economic market order is constituted partly by supportive legal constructions may appear clear enough, but one may be more circumspect regarding the relationship between legal constructions and the ordering of platforms as “multi-sided markets,” for which the material dimension may be more central. It is important to clarify that the relationship between the platform emergence and EU law mechanisms is not one-way in that supposedly autonomous technological innovation merely triggers a legal response. EU law works not just as a reactionary “coping mechanism” for the force of technological and attendant socio-economic change,Footnote 61 but also actively drives and shapes that change in the first place.Footnote 62 Online platforms and social order are mutually shaping.Footnote 63 Accordingly, internal market constructions not only constitute markets but also influenced the emergence of online platforms as multi-sided markets in Europe. Therefore, it now becomes paramount to scrutinize the specific legal mechanisms of internal market construction that have been implicated in this emergence.Footnote 64 I do not intend to present a comprehensive overview, but will delimit my analysis instead to a suite of techniques within the Union’s electronic commerce law.Footnote 65 Many other legal constructions of intellectual property law and trade secrecy, and fundamental institutions of private law such as contract and corporate law that still largely remain within Member State competence, cannot be given proper credit here.Footnote 66
To begin with EU primary law, while the free movement of goods used to serve as the leitmotif for early common market construction, for platforms the choice of fundamental freedom is that of services together with the freedom of establishment. The place of the former in the Union’s imagination gained prominence through the drive to “complete” the internal market and its attendant Single European Act.Footnote 67 Following the “X-as-a-service” business template,Footnote 68 platform firms primarily sell access to a connection based on the insights gleaned from data analytics, not data itself.Footnote 69 Despite the wide range of possible business model variations, platform firms’ overall imperatives to monetize users through data and money rents persist.Footnote 70 The Silicon Valley organizational template marries private platform start-up firms with venture capital markets oriented toward long-term returns that are expected after the platform has scaled up sufficiently.Footnote 71 Therefore, early-stage investors with ample patience play a crucial role in instituting the imperative to monetize users within the start-ups early on.
Platform firms may extract money rents from advertisers via data analysis insights, while access to a platform for end-users may be provided in exchange for data processing. They may also extract service-fee rents from buyers, sellers, or both; and they may provide two platform versions: A free standard version of the platform and another with additional features via a subscription.Footnote 72 The generously wide category of services “normally provided for remuneration” in Article 57 TFEU, and its equally lenient interpretation by the CJEU,Footnote 73 have enabled the development of various business models that the multi-sided platform structure permits. For instance, the CJEU has explicitly stated that the remuneration need not be “paid for by those for whom it is performed,” thus sanctioning the provision of a platform service to consumers without a monetary price.Footnote 74
As for the Charter of the Fundamental Rights of the European Union, its regulative function features prominently in legal scholarship, especially in the form of rights to privacy and data protection in Articles 7 and 8. At the same time, the Charter’s facilitative potential for the digital economy are somewhat less pronounced. Consider, for instance, the freedom to conduct a business in Article 16 that has helped in legitimizing, often if not always together with Articles 15 and 17, platform firms’ rent-seeking business as the exercise of a fundamental right. While the freedom may again be limited through public interest regulation,Footnote 75 in its jurisprudence on intermediary service providers, the CJEU has given weight to the service provider’s freedom of economic activity against the interests of others.Footnote 76 For instance, it has ruled that in the context of online service provision, “[t]he freedom to conduct a business includes, inter alia, the right for any business to be able to freely use, within the limits of its liability for its own acts, the economic, technical and financial resources available to it.”Footnote 77 This effectively sanctions and protects the firms’ exclusionary control over platform design and governance processes.
Moving on to secondary law, the Union had started adjusting internal market construction for the digital economy well before the Juncker Commission’s announcement of a Digital Single Market.Footnote 78 The horizontal Services DirectiveFootnote 79 was introduced after much controversy in 2006. Article 20 of the Directive lends background support for scaling platforms transnationally by obliging Member States to ensure that the recipients of a service are not placed at a disadvantage based on their residence or nationality. This resulted in prohibiting the practice of “geo-blocking” an online service.Footnote 80 Dariusz Adamski has noted that the Services Directive, along with another horizontal piece of the 2011 Consumer Rights Directive, was “primarily intended to facilitate the development of the new market environment,” only seeking “to eliminate the most rudimentary negative externalities of the digital innovations.”Footnote 81
More consequential optimization of the internal market for platforms was brought about even earlier in 2000, well before platform ascendancy. At that time, the Union introduced a specific service category to facilitate the digital economy. While defined elsewhere,Footnote 82 the category of “information society services” is operationalized in the e-Commerce Directive.Footnote 83 The directive is deeply connected to the freedom to provide services and the freedom of establishment. For the Union, the task at hand with the directive, as Graham Pearce and Nicholas Platten summarize, was “to achieve a regime in which service providers, operating within the framework of law in their home countries, would be able to offer information society services to any client elsewhere in the EU.”Footnote 84 This is epitomized most clearly in the internal market clause of Article 3 of the e-Commerce Directive, which according to AG Cruz Villalón “expresses in an instrument of secondary law a safeguard already provided for in primary law by Article 56 TFEU, and adapts it to the specific features required by the harmonisation of legislation on electronic commerce.”Footnote 85 Rather than relying on harmonization of substantive law per se, the specific mechanisms combined in the internal market clause are mutual recognition and the country-of-origin principle.Footnote 86 The approach was aligned with the one adopted earlier to avoid “regulation for regulation’s sake.”Footnote 87
These internal market techniques have enfranchised platform firms to establish themselves in propitious habitats within the Union, and most of the US-founded platform firms are now based in Ireland or Luxembourg. While the combined critique of internal market mechanisms and their interpretation by the CJEU has often been voiced in terms of regulatory competition,Footnote 88 these techniques tend to channel economic benefits to Member States that firms have selected as their hosts even without the race-to-the-bottom of public interest protection. The social consequences of transnational economic activity, however, often remain local and thus scattered around the peripheries of the Union.Footnote 89 Similarly, the country-of-origin principle serves as an intellectual device for centralizing public power within “the coordinated field” of law delineated in Article 2(h) of the e-Commerce Directive.Footnote 90 It channels front-line authority to the Member States hosting platform firm establishment, also legitimizing the fact that in most nations platform firms lack any representatives who could be made easily accountable for local consequences for local authorities based on local evaluation criteria.
Article 4(1) of the e-Commerce Directive prohibits the prior authorization of information society services even by the country of establishment. As a result, the classification as information society service, and “intermediary service” as its sub-category, entitles the provider to disregard the pre-existing regulatory frameworks for underlying services altogether, a theme well-mapped in scholarship from various angles.Footnote 91 Through the last decade, the CJEU has slowly but surely affirmed the placement of various platforms, including eBay,Footnote 92 Google AdWords,Footnote 93 Airbnb,Footnote 94 FacebookFootnote 95 and YouTubeFootnote 96 within these categories. A notable exception has been Uber, whose designation as a transport service instead of an information society service dislocated it from the scope of the Services Directive per Article 2(2)(d), and even from the free movement of services in Article 56 TFEU.Footnote 97
II. Platforms’ Socio-Political Implications: The Case of Social Media
Another special benefit of providing an information society service for platform firms is the provisional immunity against civil or criminal liability for harm in Section 4 of the e-Commerce Directive. A lot of effort has already been put into clarifying the conditions of this immunity, and I do not intend to reproduce those accounts here.Footnote 98 Instead, I would like to point out that even though the arguments for the exemption in contemporary debates are often framed by freedom of expression, around the turn of the millennium freedom to provide services was avowedly coupled with free expression. Recital 9 of the e-Commerce Directive notes that “[t]he free movement of information society services can in many cases be a specific reflection in Community law of a more general principle, namely freedom of expression.”Footnote 99
Understanding the liability exemption as facilitating both free expression and the free movement of services illustrates another crucial aspect of the platform economy, namely the profound interconnectedness of platforms’ economic and socio-political dimensions. Power ensuing from control over the platforms strikingly stems from the putative understanding of “the economic.”Footnote 100 The largest platforms in particular cross the public-private divide rather spectacularly, implicating public values and converting platform firms into “public actors without public values.”Footnote 101 Consequently, the co-production of platforms and legal institutions entails that the internal market mechanisms facilitating the platform economy are also implicated in the platforms’ socio-political effects. The manifestations of these effects are most appropriately analyzed separately for each platform type. Even so, platforms’ influence on public values is perhaps particularly conspicuous in social media. Social media platforms transform the conditions under which people communicate and access information online, changing the conditions of public deliberation.Footnote 102 Accordingly, internal market construction has facilitated not only the emergence of social media platforms, but also the transformation of the public sphere in Europe.
Social media platforms have enabled people to circumvent the homogenizing mass media whose social mechanisms of organizing and selecting information were seen as exclusionary for those that did not identify with the dominant media narratives, especially various minorities. These social media communities serve the same needs that other communities do.Footnote 103 Yet social media does not restrict the availability of communities by territory, as users may also select from transnational ones, affording individuals rather free play with group identification.Footnote 104 As the costs of adding new users to a platform are practically non-existent, and platform use requires comparatively little digital aptitude, social media could include and give everyone the opportunity to voice their concerns in a formally equal manner. All of these affordances have had emancipatory potential.Footnote 105 This potential also appears to exhibit a certain similarity to that of the EU with its subject freed from the constraints of one’s location and lifted to the transnational plane of the internal market.Footnote 106
Yet following the up-scaling of social media, predominantly from California and spearheaded by Facebook’s connecting the world ethos,Footnote 107 the sheer volume of information has made its ordering “an extra-ehuman task” requiring automated systems of sorting and recommending.Footnote 108 Scaling also affected an unprecedented centralization of control over the information organization within platform firms. Due to the abundance of information on platforms, mere access to a platform is a necessary but insufficient condition for gaining visibility and recognition.Footnote 109 Marion Fourcade and Fleur Johns have argued that within machine learning systems “the benefits of inclusion now depend less on access itself, and more on one’s performance within each system and according to its rules.”Footnote 110
On social media, visibility as a pre-condition for recognition becomes a scarce resource that must be continuously competed for,Footnote 111 creating a form of never-ending competition for the indicia of recognition in the network.Footnote 112 At a minimum, visibility requires ongoing participation—posting, sharing, liking—or else the user risks falling into invisibility.Footnote 113 At best, successful activation of the network may trigger platform algorithms’ further amplification of one’s views. In particular, various cultural workers have become highly dependent on social media visibility for their livelihood.Footnote 114 The gamification of social participation also keeps the flows of platform data running and the metric of monthly active users high. This ensures the capital markets’ enduring preference for platform firms, maintaining their market capitalization, which currently amounts to an immense concentration of wealth.Footnote 115
It is also at this level of performance within the platform, rather than access, where the persistent social inequalities between users seep in to reproduce distributive patterns and undermine social media’s original emancipatory promise. Social media profiles usually encourage the convergence of one’s online and offline identities,Footnote 116 while the response metrics of aggregated subscribers and likes make one’s constantly fluctuating position in the visibility competition clearly visible.Footnote 117 Thus, those who are short of time and money are arguably worse off “not only because that makes them less desirable in the real world, but also because they cannot afford the effort and expense needed to overcome their disadvantage in the online world.”Footnote 118
Platform firms’ imperative to monetize users exacerbates these effects through offerings of preferential placement for those with sufficient resources.Footnote 119 Targeted adverts empower marketers to find “good matches” for their messages, which are “perceived as natural because they fit well with how things already are.”Footnote 120 Such control may result in intriguing offerings and valuable discounts, making ads that are more relevant a cause for celebration for many. At the same time, the control also enables acting on various automatically created categories of the disadvantaged which, while “low value” for some organizations, are specifically sought after by others.Footnote 121 Thus, misinformation may be channelled to those categorized as “interested in pseudoscience,”Footnote 122 online casino adverts to those with gambling problems, and predatory loan offerings to those already heavily in debt. Similar looping effects that circulate pre-existing disadvantages can be associated with current recommender systems that suggest new information. As algorithms cannot be trained on future data, the past is presumed to be a relevant indicator of future value.Footnote 123 On top of the platform mechanisms, “a whole industry of social media derivatives” has emerged to serve the more affluent users through profile optimization, reputation management, and purchasable interaction.Footnote 124
Effectively, social media platforms may have distributed attention and recognition perhaps less to oppressed minorities but rather to the same quarters that were already in a position to have ample public outlets for their voice. Yet platforms have also empowered other previously marginalized but at times well-resourced interests of vicious hatred, racism and conspiracy that have come to enjoy extraordinary stickiness in the online public sphere.Footnote 125 As platform interfaces do not enable seeing others’ information feeds, comparing one’s own information delivery to that of others becomes hard, thus directing concern away from others’ relative disadvantage and toward the convenience of one’s personal service.
D. The DSA and DMA as Internal Market Construction
I. Internal Market Rationality in the DSA and DMA Proposals
After analyzing the emergence of platforms in the internal market and some of its attendant inequalities in the case of social media platforms, I will now delve into the rationale for the new EU laws that are set to target online platforms more specifically than the earlier laws on platforms. My intention is to scrutinize the patterns of (dis)continuity in relation to previous internal market construction. While there is already a somewhat thick corpus of more specific legislation that could also be subsumed under the heading of platform law,Footnote 126 I focus here only on the DSA and DMA proposals because of their more horizontal character and relative importance within the Union’s agenda. To be sure, much more could be said about the content of these proposals. My aim here is not to provide an all-encompassing analysis of the Articles’ myriad contours, which are bound to be amended during the legislative process, but to focus on the animating reasoning running through the proposals. Such reasoning arguably remains more stable as the specificities of legal proposals are agreed upon, perhaps partly precisely because more fundamental rethinking may be expected to require a different legal basis, which, in turn, may be understood to be unavailable.Footnote 127
The DSA is set to update the e-Commerce Directive, and it centers heavily on platforms. Even so, the Commission affirmed that: “This proposed Regulation is without prejudice to the e-Commerce Directive, and builds on the provisions laid down therein, notably on the internal market principle set out in Article 3.”Footnote 128 Re-imagining internal market foundations was “discarded at an early stage” because “[t]he evaluation of the e-Commerce Directive and all other available evidence shows that the single market principle has been instrumental for the development of digital services in Europe. This principle increased legal certainty and reduced compliance costs significantly, which is crucial for smaller services in particular.”Footnote 129 Similarly, the Commission affirms that “[t]he liability regime set in the e-Commerce Directive for online intermediaries is considered a cornerstone for allowing online intermediaries to emerge in the 2000s” while also fostering freedom of expression.Footnote 130 Under the fundamental rights impact assessment, the Commission hastens to ensure that “[n]one of the measures in either one of the options should jeopardise the protection of trade secrets or proprietary products of online platforms.”Footnote 131
Building on these foundations, the DSA seeks to both harmonize “the conditions for innovative cross-border digital services to develop in the Union, while maintaining a safe online environment.”Footnote 132 This twofold aim can only be achieved at the EU level.Footnote 133 Indeed, reliance on the mechanisms of mutual recognition and country-of-origin principle tends to summon the specter of fragmentation calling for unification. Thus, the underlying logic of the proposal is neatly aligned with internal market rationality—transnational freedom to provide services combined with public interest regulation to mitigate the societal consequences of such freedom. Undeniably, the emphasis is now on the latter aim. A safe online environment is envisioned to be brought about by fostering “responsible and diligent behavior by providers of intermediary services.”Footnote 134
In that regulatory vein, many of the DSA provisions seek to have the online platform firms, especially the largest ones, to internalize online safety standards in their services.Footnote 135 For instance, Articles 3–9 aim at careful modification of the tenuous relationship between the liability exemption and platform responsibility for illegal information, slightly in favor of the latter. More structural concerns, now dubbed “systemic risks,” particularly concerning illegal information, threats to fundamental rights, and manipulation, should also be taken into account by very large platform operators, as provided for in Articles 26–27 of the proposal.Footnote 136 Which specific risks and how they should be taken care of is a task delegated to the corps of corporate bureaucrats, in particular the foreseen compliance officers and independent consultant auditors.Footnote 137 Yet as Ioannis Kampourakis has noted of such open-ended requirements, with much interpretative leeway for various professional communities there may be a “substantial space for regulated firms to decide the process by which they might achieve the outcomes sought.”Footnote 138 Moreover, protective regulation will likely end up being market-making also in the sense of consolidating the downstream markets on content moderation, impact assessments and algorithmic auditing services.Footnote 139
To be sure, public oversight provides a possibility to intervene if firms remain irresponsible and things go awry. On one hand, the proposal again commits itself to maintaining the country-of-origin principle, as changes to “the requirement for the country of establishment to supervise services would inherently undermine the development of digital services in Europe, allowing only the very large players to scale across the single market.”Footnote 140 On the other hand, the proposal lifts the supervision of very large online platforms away from platforms’ host states and centralizes it within the Commission itself.Footnote 141
At the same time, users are thought to need more individual agency within platforms. For instance, at a minimum, very large platforms must provide the individual with “at least one option which is not based on profiling” and more control in determining the kind of information one encounters on a platform.Footnote 142 In addition, Article 24 entitles them to more information on the delivery of adverts. In sum, the reasoning in the DSA proposal exhibits the perceived need to intensify the interventionist logic largely in the name of public interests, but with some individual empowerment added. Yet the envisioned regulatory framework is largely meant to be imposed on top of the various facilitative internal market mechanisms, some of which I have outlined in the previous section. These mechanisms are kept largely intact, demonstrating commitment to the twofold aim of internal market rationality in its entirety.
The DMA proposal foregrounds the aim of undistorted competition in digital services where platforms now dominate, branded here as “gatekeepers” in the area of “core platform services.” According to the proposal: “A few large platforms increasingly act as gateways or gatekeepers between business users and end users and enjoy an entrenched and durable position.”Footnote 143 Consequently, the proposal seeks to create an internal market for digital services “by ensuring that markets across the Union where gatekeepers are present are contestable and fair. This should promote innovation, high quality of digital products and services, fair and competitive prices, and free choice for users in the digital sector.”Footnote 144 Corporate dominance in itself is not the targeted harm but rather dominant firms’ certain business practices, which are seen as unfair. At the same time, the Commission is candid about its wish of seeing new (European) platforms emerge, as the DMA “will foster the emergence of alternative platforms, which could deliver quality innovative products and services at affordable prices.”Footnote 145
The DMA proposal consistently stresses that more contestable digital markets and the curtailment of unfair practices would be instrumental in lending individual users more choice in pursuing their personal interests.Footnote 146 Recital 32 further notes that the obligations of gatekeepers are needed “to address the risk of harmful effects of unfair practices imposed by gatekeepers, to the benefit of the business environment in the services concerned, to the benefit of users and ultimately to the benefit of society as a whole.”Footnote 147 The increased individual choice is identified as being in the interest of society.
Both the mandatory obligations of gatekeepers in Article 5, and those whose application is at the Commission’s discretion in Article 6, aim at enhancing businesses’ ability to offer more choice in the service market and, correspondingly, individuals’ ability to choose among the options.Footnote 148 As in the DSA, the interventionist regulatory techniques are to be deployed in the DMA proposal as well. The techniques listed in Articles 5 and 6 anticipate deep intervention into the current logics of platform business.Footnote 149 For instance, the foreseen interoperability mandate in Article 6(h) of the DMA proposal would potentially burrow deep into the platforms’ technical protocols to enable the transfer of data between platforms.Footnote 150 Yet in the DMA this is not so much in the name of mitigating social consequences, but rather to further the other aim animating internal market rationality, undistorted competition and individual empowerment. From the competition law perspective such increased emphasis on the “ex ante approach” of regulation may amount to considerable innovation,Footnote 151 but from the perspective of the internal market, as we have seen, regulation in the name of many different values is nothing new.
II. Distributive Patterns of the DSA and DMA Proposals
I will now pin down the allocative effects of internal market rationality between various actors. First, there is the case that internal market freedoms systematically favor the mobile, that is, those with the means and motivation for transnational activity. Accordingly, the internal market freedoms are structurally biased toward more movable capital,Footnote 152 and the nature of the digital economy has enabled capital to become extraordinarily mobile indeed. Second, and more generally, the financial sector is among the recipients of benefits as shareholders in platform firms because the envisioned European alternative platforms would likely rely on the same financial model of patient venture capital as the Silicon Valley firms to scale up and challenge the dominant players.
Having said that, the actors to whom history has accorded the upper hand—the US-based platform behemoths—are the most prepared for undistorted competition in the digital service market. While the preserved internal market entitlements may help small businesses to scale up transnationally, the idealizing luster of European start-up entrepreneurs conceals the fact that through the law’s universalizing logic these same entitlements are granted to Meta, Alphabet and Amazon. It may be that the incumbent firms are able to make the most of them.
The states continuing to benefit the most from the internal market mechanisms—investments and tax revenue—are the ones that have a relatively strong digital industry, generally in the European North, or those that have been able to invite the establishment of the US-founded platform behemoths, that is, Ireland and Luxembourg. Those that have less to gain are the Eastern and Southern Member States and their small businesses. Similarly, the country-of-origin principle continues to channel power primarily to the public authorities of host states while disempowering those of others. In addition, open-ended obligations of public interests entrust bureaucracies within the Commission, platform firms, and formally independent consultancies with epistemic authority to interpret and decide the meaning and distributional effects of amelioration and its trade-offs.Footnote 153
This is not to claim that the incumbents’ position is set to persist, nor to reject the regulatory measures’ destabilizing potential. Regulation targets the largest platform firms the most, and likely evens out certain structural asymmetries peculiar to the digital economy. My point is that there is no equal starting position to which one could fall back to start market competition “anew,” but the actors benefiting from historical accumulation do get a head start. Thus, when a framework of formally equal undistorted competition and transnational freedom is imposed on social reality characterized by pervasive inequalities, many distributional patterns are likely to be reproduced. This casts doubt on the delivery of competition’s supposedly equalizing effects. For those that lack the possibilities to invest time and resources in starting a platform firm, say in Greece, or that fail to appeal to the whims and expectations of the venture capital market, there is less reason to feel sanguine about entitlements to fair competition against established players.
Finally, among individual social media users, those that likely benefit the most from the increasing agency within and among platforms are still the ones that already benefit from those services the most, that is, the more well-off. Again, my point is not to plainly dismiss the equalizing potential of regulation. For instance, removing illegal information more scrupulously may free up the space for marginalized voices. Similarly, while giving users more individual agency has some merit, it also responsibilizes disadvantaged users to opt out of feedback loops as the more well-off users may continue to enjoy the more rosy effects of targeting. At the same time, the internal market mechanisms help maintain the control over platform configurations within the platform firms and in line with their imperative of user monetization.
In sum, both the DSA and DMA evince rather than question the animating twofold aim of current internal market rationality. They seek to offer individuals more freedom within and across online platform services. Yet the proposals also aim at mitigating the social externalities of platform businesses. Firms are to internalize the externalities of platform business by incorporating values such as online safety, fundamental rights and fairness into their operations. The goal is to have the platforms permeated with public interests while existing internal market mechanisms are preserved. While containing many novel regulatory devices with the potential to alleviate some inequalities, the proposals nevertheless risk reproducing and legitimizing many distributional patterns and power relations within the digital economy and the online public sphere. They are primarily instances of what Roberto Unger has called “progressive pessimistic reformism” as “the pursuit of programmatic goals, such as more economic competition or greater equality of practical opportunity and cultural voice, within the limits imposed by the established institutional order.”Footnote 154
E. Conclusion
This article has explored various patterns of continuation and change in how the EU attempts to govern social media platforms in particular as a part of internal market construction. I argued that the construction of the EU internal market is generally informed by a twofold aim. On the one hand, the goal is to ensure that individuals and other private actors may pursue a life that one sees as fitting across borders under the conditions of undistorted competition. On the other hand, the internal market also affords concern for the societal harms of private behavior through the imposition of mitigating regulation informed by public and private expertise. I then explored how specific mechanisms of internal market construction have participated in the emergence of social media platform power, and its attendant inequalities in Europe.
After mapping the internal market construction prior to the Union’s recent platform law agenda, I turned to examine the rationale for the DSA and DMA proposals, arguably the EU’s most important pieces of upcoming platform law. I argued that despite numerous regulatory innovations with potentially equalizing distributional effects, the proposals remain committed to both the twofold aim of current internal market rationality and many of the already existing mechanisms of the internal market. Therefore, from the perspective of internal market construction, these platform law proposals evince at least as much continuation and consolidation as they do change. Accordingly, they also risk reproducing various unequal distributional patterns and power relations in Europe.
My hope is that the analysis has conveyed the suggestion that counter-conducting projects aiming to resist platform power and harms in Europe should rethink and disrupt the existing aims and mechanisms of the current internal market rationality. This should extend even to the point of asking whether the current aims are “the values we should have, in view of everything that we know about the world.”Footnote 155 While it is important to continue exploring and experimenting with regulatory techniques that could “tame the giants,” there should also be room for more intensified questioning so that the existing internal market mechanisms as informed by the current internal market rationality, which have helped the current platforms rise in the first place, would be brought under serious scrutiny and re-imagination as well. To be frank, this kind of re-imagination should begin by sidetracking, at least to some extent, consumer choice and competition within or across platforms as its orienting virtues, taking the various unequal distributional patterns and socially embedded subject as its points of departure instead. Such re-imagination might also help drive and shape the emergence of radically different platforms for Europe.
Profound social change is possible in a piecemeal way if the direction of change persists.Footnote 156 One could re-frame many current efforts of resistance as intermediate episodic changes toward deeper social transformation. For instance, the attempts at platform interoperability, digital tax,Footnote 157 and changes to corporate lawFootnote 158 could all be re-framed as intermediate steps toward a profound alternative vision of the digital economy. Nevertheless, envisioning such alternatives would likely also require extending the horizon of imagination beyond the currently dominant aims of transnational individual freedom and market competition.
Open access
