25 November 2022

Protecting Media Content on Social Media Platforms

The European Media Freedom Act’s Biased Approach

On 4 October 2022, the Council approved the final version of the Digital Services Act (DSA). The DSA is, no doubt, a remarkable and unique proposal in the field of online platform regulation. As horizontal legislation, the DSA establishes a general framework for the provision of online services after the adoption of a set of more specific norms, in particular the Copyright Directive, the Audiovisual Media Services Directive, or the Regulation on Terrorist Content Online, among others.

The flow of legislative proposals does not stop here, though. On 16 September 2022 the European Commission released a proposal for a Regulation establishing a common framework for media services in the internal market, also known as the European Media Freedom Act (EMFA). This is a very special proposal, as it aims at tackling at the EU level fundamental issues connected to the exercise of the right to freedom of expression by media actors and media organizations. So far, the prevailing legal interpretation referred these issues to the respective constitutional, legal and regulatory provisions at the national level.

The EMFA proposal includes safeguards against political interference in editorial decisions and against surveillance. It also tackles the issues of the independence and stable funding of public service media, as well as the transparency of media ownership and of the allocation of state advertising. A key tool introduced by the EMFA is the increased regulatory cooperation and convergence through cross-border coordination tools and EU-level opinions and guidelines.

Beyond these general principles, the EMFA also includes a series of provisions targeting online services and establishing additional obligations and regulatory powers in this field. This post will focus on these specific areas, particularly referring to new obligations on the presence of media content on very large online platforms (VLOPs).

This post will present how these provisions not only create serious problems of interpretation, and force online service providers to engage in extremely complex assessments, but also establish arbitrary and unjustified differentiations between protected categories of speech.

What is media?

In its definition of ‘media’, the objective of the legislator appears to have been to enact a limited and relatively certain definition of media services.

The EMFA unjustifiably embraces a limited definition of media services which is in fact neither aligned with applicable international and regional human rights standards nor justified from a communication or a journalistic perspective. This definition may moreover be considered discriminatory, as it has the consequence of excluding certain forms of media and journalistic activity from important protections contemplated in the draft, including vis-à-vis interferences in editorial freedom deriving from detention, sanction, interception, and surveillance or forceful disclosure of sources.

Article 1 of the draft EMFA defines media as a service where the principal purpose consists in “providing programmes or press publications to the general public, by any means, in order to inform, entertain or educate, under the editorial responsibility of a media service provider”. The notion of ‘programme’ is defined by the proposal as “a set of moving images or sounds constituting an individual item, irrespective of its length, within a schedule or a catalogue established by a media service provider”, whereas press publications need to be understood as “a collection composed mainly of literary works of a journalistic nature, but which can also include other works or other subject matter”.

This last definition is taken from Article 2.4 of the Copyright Directive. Therefore, the draft EMFA focuses on a ‘traditional’ notion of press publications, i.e., those targeted by news aggregators or media monitoring services, and which are currently facing problems in licensing the online use of their publications.

However, the notion of media services is much broader, as established by the United Nations Human Rights Committee in its General Comment number 34. Hence, the Committee of Ministers of the Council of Europe has also recommended to “adopt a new, broad notion of media which encompasses all actors involved in the production and dissemination, to potentially large numbers of people, of content”.

Despite the ‘traditional’ notion of press publications, the EMFA confuses concepts, best encapsulated in Recital 8. It affirms that in the digital environment “providers of video-sharing platforms [i.e., services such as YouTube] or very large online platforms [for example, Facebook Watch] may fall under the definition of media service provider”, due to the role they play in the “content organization, including by automated means or algorithms”. This takes the legislator to state that “in the increasingly convergent media environment, some providers of video-sharing platforms or very large online platforms have started to exercise editorial control over a section or sections of their services. Therefore, such an entity could be qualified both as a video-sharing platform provider or a very large online platform provider and as a media service provider”.

If we were to take this language literally, it would introduce a complete change from the current conceptual framework established by the eCommerce Directive, the case law of the Court of Justice and even the recently adopted DSA. Content moderation practices could be considered a form of exercise of editorial responsibility, and therefore at least certain types of online services would become ‘traditional’ media services. However, this Recital looks more like an aspirational statement or perhaps an invitation for further amendments in this direction than an actual legal determination.

In any case, some video sharing platforms’ (VSP) users’ exercise of editorial control over a section or sections of these services does not mean that “an entity could be qualified both as a video-sharing platform provider or a very large online platform provider and as a media service provider”. EU law regulates services, not entities. When an audiovisual service is provided via a VSP platform, this specific activity is regulated under the Audiovisual Media Services Directive, and it obviously needs to be separated, legally and technically, from the service provided by the platform acting as a mere intermediary.

Privileged treatment of content of media service providers on very large online platforms

A new issue included in the EMFA refers to the moderation of ‘media content’ by online platforms. The most relevant legal systems in the world (including the European Union) incentivise content moderation by platforms by ensuring that they are not penalized for good faith measures against illegal or other forms of inappropriate content. Legal provisions, including the DSA, protect the capacity of platforms to decide how to organize, prioritize, demote or simply eliminate content to different extents.

The EMFA seems to attempt to connect the exercise of editorial responsibility by media content providers with the enjoyment of specific conditions for online content moderation from the side of online platforms. However, this option appears poorly justified so far and based on very vague and problematic criteria.

According to Article 17 EMFA, VLOPs shall provide a functionality allowing media service providers to declare their status according to the definition included in the draft. Media service providers must prove they are “editorially independent from Member States and third countries”, and that they are “subject to regulatory requirements for the exercise of editorial responsibility in one or more Member States, or adhere[…] to a co-regulatory or self-regulatory mechanism governing editorial standards”.

Provided that these cumulative requirements are met, media service providers are granted two main privileges:

  1. When a VLOP decides to suspend the provision of its services to a media service provider on the grounds that its published content is incompatible with the VLOP’s terms and conditions, it must communicate to the concerned media service provider a statement of reasons “prior to the suspension taking effect”. In addition, complaints by media service providers must be “processed and decided upon with priority and without undue delay”.
  2. Where a media service provider “considers” that a VLOP frequently restricts or suspends the provision of its services without sufficient grounds, the VLOP shall “engage in a meaningful and effective dialogue with the media service provider, upon its request, in good faith with a view to finding an amicable solution for terminating unjustified restrictions or suspensions and avoiding them in the future”.

These provisions are problematic based on a series of legal considerations.

Designated media services

The previous section has already elaborated on the draft EMFA’s very limited definition of media service providers, and the reasons for this choice. The arbitrariness and discriminatory nature of this limitation becomes particularly clear in light of Article 17. Is it reasonable to accept, for example, that a social media post by a commercial or a public broadcaster may receive a more cautious treatment, when it comes to possible restrictions or suspensions, than the content posted by a big human rights organization reporting about possible crimes against humanity?

Editorial independence from member states and third countries 

The first criterion and concept of editorial independence is not defined by law, including the EMFA. Moreover, from both a theoretical and a practical perspective, factors that can affect editorial independence are extremely diverse, depending on the conditions of the concrete media space under analysis, as well as the media’s public or private (commercial) nature.

Regarding public service media, editorial independence is usually subject to the fulfilment of several complex indicators. Several European institutions have offered guidance, for example, the European Broadcasting Union has issued a series of Governance Principles and the Council of Europe has adopted a significant number of standards. At the EU-level, some relevant indicators can be found in the Protocol annexed to the Treaty of Amsterdam as well as the Communication from the Commission on the application of State aid rules to public service broadcasting. That being said, even within the limited EU context, assessing the actual independence of public media is becoming progressively complex and controversial in certain countries, as a recent publication by the European Audiovisual Observatory shows.

Regarding private/commercial media, actual editorial independence not only relates to legal and political factors, but also to other circumstances, including financial support, business and revenue models, and vertical or horizontal integration. These factors can be found in the recent Recommendation of the Committee of Ministers of the Council of Europe to member States on principles for media and communication governance.

In sum, assessing the editorial independence of media services in the EU is an extremely complex task, given that the editorial independence of both public and private media companies will always be relative. It remains unanswered what degree of independence is sufficient for the purposes of the EMFA. It seems therefore that the EMFA contemplates a requirement almost impossible to assess in a proper and consistent manner by VLOPs when implementing the provisions included in Article 17.

Subjection to regulatory requirements or adherence to co-regulatory and self-regulatory mechanisms

The second criterion for media service providers to benefit from the regime contemplated in Article 17 is that they must be subjected to “regulatory requirements for the exercise of editorial responsibility in one or more Member States”. This vague requirement is to be assessed, again, by VLOPs upon the declaration by media service providers. Defining whether a complex and diverse national legal framework is conducive to editorial responsibility would require an in-depth analysis from a communications, legal and economic perspective. Considering the complex nature of such assessment, it would also be reasonable to expect different conclusions being adopted by different VLOPs applying Article 17. Therefore, the lack of legal certainty is complete in this area.

The same concerns also apply to the requirement of adherence to self-regulatory or co-regulatory mechanisms that are “widely recognized” in one or more Member States. For example, the Media Pluralism Monitor released in 2021 by the Centre for Media Pluralism and Media Freedom concludes that the freedom of journalists and editors to make decisions without inference (be it from the owners of the publication or external political pressure) is at risk in 26 out of the 32 European countries analyzed, and at high risk in 14 countries. Moreover, only six countries are considered to have effective journalistic self-regulation: Belgium, Denmark, Estonia, Germany, the Netherlands, and Sweden. Once again, considering the poor situation of self-regulation and co-regulation in most EU countries, fairly assessing when a system is “widely recognized and accepted in the relevant media sector” becomes a very sensitive task. In addition, the fact that a self-regulatory system is widely accepted does not guarantee its independence or quality.

Nature of the privileges granted to media service providers

If media services providers meet the above conditions, they can look forward to two kinds of privileges.

Firstly, before the suspension of the provision of services, the VLOP must provide a prior statement of reasons. Since possible complaints or appeals by media companies must be “processed and decided upon with priority and without undue delay”, this provision seems to grant something close to a must-carry privilege or at least limit a takedown’s duration, until a final decision is adopted by the online service provider. Once again, this represents a clear difference of treatment vis-à-vis other media content, which would not enjoy such protection.

This discussion has some connections to the broader debate about a so-called (and finally not accepted) media exemption during the elaboration of the DSA. The main argument in favor was that the media sector is legally liable for what they publish, therefore, it should not be subject to the editorial control of online platforms that unilaterally decide their terms and conditions. This position mixes two different elements: the editorial responsibility that publications hold when it comes to respect for applicable legislation as well as professional and ethical standards on the one hand, and the role of online platforms in the establishment and enforcement content policies (stimulated by different stakeholders, including governments and regulators) to avoid harm, promote civility and protect users when engaging in conversations and mutual interactions, on the other. Compliance with one aspect does not necessarily guarantee alignment with the other.

Secondly, and as already mentioned, where a media service provider “considers” that a VLOP frequently restricts or suspends the provision of its services without sufficient grounds, the VLOP shall “engage in a meaningful and effective dialogue […] to find[…] an amicable solution for terminating unjustified restrictions or suspensions and avoiding them in the future”. No details are given regarding the nature and scope of such amicable solutions. The provision also seems to presume that frequent restrictions or suspensions would be unjustified per se and therefore need to be avoided by platforms.

The must-carry proposals currently under discussion in the United States

In the United States discussions about regulatory-imposed privileged treatment of “special” speakers have triggered political discussions and interesting litigation. In these cases, platforms may be facing an actual obligation to leave certain categories of content up, thus losing the capacity to moderate content in these areas. These provisions differ from the lighter touch that can be found in article 17 EMFA.

Attempts in the United States particularly emerged from Republican ranks and were triggered by President Trump’s suspension and the broader political claim that platforms tend to restrict “conservative” views. In 2021, the Governor of Florida signed the Act SB 7072 on Social Media Platforms, requiring that companies use the same criteria across their platforms when deciding to take down a post or remove an account, and barring them from removing the account of any “journalistic enterprise” or political candidate in Florida. The U.S. Court of Appeals for the 11th Circuit considered that the law violated social media companies’ First Amendment right to decide what speech is expressed on their platforms. There is now a pending petition before the Supreme Court to reinstate the law. A similar law was also adopted in Texas, prohibiting large social media companies from banning users’ posts based on their political viewpoints. In this case, the 5th U.S. Circuit Court of Appeals upheld the law, thus dismissing for the first time in US case law the argument (presented by the plaintiffs as in the case of Florida) that Internet companies have a First Amendment right to curate content posted on their platforms.

Beyond the mentioned First Amendment argument, authors like Daphne Keller have criticized this kind of laws as they appear to be based on the principle of taking away private companies’ control over speech and impose government-defined rules instead. In particular, such laws would actually subdivide lawful speech into legally favored and disfavored categories. Therefore, according to this author, platforms are pushed to make judgments subject to new, state-defined rules in the absence of a capable and independent adjudicator. Neither courts nor government agencies are equipped to oversee a deluge of disputes deriving from the vagueness of the mentioned legal provisions, much less do so while providing constitutionally adequate process. The same holds in Europe. A possible option would of course consist of empowering a national media regulator. However, such an entity would not be able to resolve every and each individual dispute, which effectively leaves platforms as the final arbiters of any new state-imposed speech rules. Thereby, the circle closes to the EMFA, and the problems outlined above inherent to an approach which foregrounds platforms’ assessment.


The proposal of a directly applicable legal instrument to establish a series of basic protections within the area of media freedom represents a remarkable step forward by the EU institutions in guaranteeing a common fundamental rights framework for all citizens of the Union.

This being said, the proposed EMFA must be improved in order to become an instrument fully aligned and consistent with existing international and regional human rights instruments. In addition, the EMFA must also adapt the regime governing the treatment of media actors and content by VLOPs.