Abstract: A number of platform economy services are composite insofar as they consist of an element provided by electronic means and another that is not provided by electronic means. Determining whether platforms provide only the online intermediation service or an overall service comprising the underlying physical service as well is instrumental in ascertaining which EU law provisions apply to these platforms. The aim of this Insight is to explore the legal classification of composite platform economy services and its policy implications, in order to propose a way forward at the moment when the Commission is drafting a proposal for a new Digital Services Act (DSA).
Keywords: Digital Single Market – Regulation of the platform economy – composite services – illegal content – online platform liability – Digital Services Act.
“Isn’t Amazon’s technology simply a tool to sell goods?”, Judge Jan L. Passer of the General Court asked during the hearing in tax ruling case T-318/18 Amazon EU and Amazon.com v. Commission. Whether the major players of the digital economy are mere intermediaries that make the offer of goods and offline services more accessible thanks to online technology or operators that provide both the online and the underlying offline services themselves is the issue at the heart of a fascinating on-going debate whose outcome will be of great significance for platform regulation. This question has been brought up more than once before the Court by national jurisdictions struggling with the classification of platform services under the existing, and allegedly out-dated, legal framework.
Since the adoption of Directive 2000/31, also known as the e‑Commerce Directive, and Directive 2006/123, the Services Directive, markets have changed dramatically. Innovative platforms have emerged, bringing about healthy competition and flexible ways to generate revenue, raising high hopes for growth. However, in the face of indisputable market failures, such as threats to public health and safety as well as housing shortages, many national and local authorities across the EU have decided to take action, amid uncertainty as regards the legal classification of these new services.
Specifically, the provisions of Directive 2000/31 do not settle the question whether the coordinated field thereof covers services that are only partially provided by electronic means, to which AG Szpunar refers as “composite” or “mixed” services, in line with settled case law regarding services composed of several connected supplies in the fields of taxation and procurement. This category encompasses not only collaborative economy services, such as those provided by Uber and Airbnb, but also e‑commerce services, insofar as they are usually linked to the delivery of goods. Of course, it could be considered that the online and the offline supplies should be considered separately. However, there might be good reasons to consider a single, comprehensive service. This is true, for instance, where the supply “comprises a single service from an economic point of view” or where partial liberalisation is not desirable.
Thus, two questions arise when analysing composite platform economy services: must the intermediation (online) service provided be considered separately from the underlying (physical) service? If not, then both types of services form part of an overall service provided by these platforms: in that context, what should be considered the main element of this comprehensive service? The Court has already examined the intermediation services provided by eBay, Uber, and Airbnb, and provided guidance as to the way these questions must be answered. Still, as the Commission puts it, “[w]ith 20 years of case law interpreting the e-Commerce Directive, sometimes in diverging ways, the legal notions on the services covered and their respective responsibilities lack precision”.
Against this background and pending additional cases, the aim of this Insight is to explore the legal framework applicable to composite platform economy services to better understand how they are currently regulated and how the EU legislature could make this regulation more efficient.
Section I first briefly recalls the main EU law provisions applicable to composite platform economy services to illustrate that the classification of such services has significant legal consequences for both platforms and regulators.
Section II then analyses the Court’s landmark judgments to determine against which standard we should assess composite platform economy services. It argues that the Court has somewhat clarified how such services fit into existing legal categories, but that it still does not have a robust solution to this regulatory conundrum, at least not one that is beyond dispute.
Finally, section III explores the potential policy implications of the current legal framework, in light of the Court’s case law, and proposes a way forward, in the context of the forthcoming Digital Services Act (DSA).
II. The applicable legal framework
Arts 49, 56 and 57 TFEU lay down the principles of freedom of establishment and freedom to provide services. These provisions play a residual role in relation to the existing directives governing the field of services, except when it comes to taxation, inter alia, which is usually explicitly excluded from the scope of these directives. In this regard, the Court has yet to clarify if ‘administrative requirements necessary for the enforcement of tax laws’ also fall outside of the scopes of such directives, as the Commission has already suggested regarding Directive 2006/123. The judgment in pending case C-723/19 Airbnb Ireland and Airbnb Payments UK, concerning obligations for platforms to collect and remit the tourist tax as well as data to Italian public authorities, may provide useful guidance on this matter.
Directive 2006/123 and Directive 2000/31 are instrumental with regard to the regulation of platform services. The former establishes general provisions facilitating the free movement of all services, with the exception of a ‘long and broad’ list of activities and matters. The latter creates a legal framework to ensure the free movement of information society services (ISS), i.e. “any service normally provided for remuneration, at a distance, by electronic means and at the individual request of a recipient of services”, pursuant to Art. 1, para. 1, let. b), of Directive 2015/1535. In case of conflict between the provisions of these directives, precedence is to be given to Directive 2000/31, which is a lex specialis in relation to Directive 2006/123. The former is a stronger “market access opener” than the latter, as the following paragraphs demonstrate.
The two directives contain conflicting provisions as regards authorisation schemes. Under Arts 9 and 10 of Directive 2006/123, Member States may make access to a service activity subject to such a scheme, provided that strict conditions are fulfilled, while under Art. 4, para. 1, Directive 2000/31, subjecting the activity of ISS providers (ISPs) to “prior authorisation or any other requirement having equivalent effect” is prohibited. However, only Directive 2006/123 also governs other requirements relative to the establishment of providers through the black and grey lists provided for in Arts 14 and 15 thereof.
Both directives also provide for the freedom to provide services. Setting aside the debate on the distinction between the country-of-origin principle (CoOP) and the freedom to provide services, there is no denying that the provisions of Art.16, para. 1, Directive 2006/123 and that of Art. 3, para. 4, let. a), Directive 2000/31 are very similar. Pursuant to the former, “Member States shall not make access to or exercise of a service activity in their territory subject to compliance with any requirements” which are discriminatory or cannot be justified by reasons of public policy, public security, public health or the protection of the environment. Under the latter, Member States may only restrict the freedom to provide ISS from another Member State by measures that are justified by reasons of public policy, public health, public security or the protection of consumers.
The CoOP however truly materialises in Art. 3, para. 4, let. b), Directive 2000/31, imposing on Member States the obligation to notify measures restricting the freedom to provide services of operators established in another Member State to the Commission and other Member States. Although Directive 2006/123 also contains such an obligation, it does not require that “before taking the measures in question”, the Member State has “asked the Member State [of establishment] to take measures and the latter did not take such measures, or they were inadequate” as Art. 3, para. 4, let. b), Directive 2000/31 does, with the aim of preventing a Member State from impinging on the competence of the Member State where the provider is established. Further, only the provisions of Directive 2000/31 are complemented by the obligation to notify rules on ISS provided for in Directive 2015/1535, bearing in mind that the Court has already held that Member States’ failure to notify a measure under one directive or the other renders this measure unenforceable against individuals in criminal proceedings and in disputes between individuals.
Another unique element of Directive 2000/31 that has underpinned the development of the Internet in the EU is the principle that ISPs should not be held liable for the content that they transmit, store or host, as long as they act in a strictly passive manner. Art. 14, para. 1, of that directive exempts the host ISP from liability where it satisfies one of the two conditions listed in that provision, that is to say, not having knowledge of the “illegal activity or information”, or acting “expeditiously to remove or to disable access to that information” as soon as it becomes aware of it. The Court has already specified that only ISPs that play a neutral role may benefit from this provision. Further, pursuant to Art. 15, para. 1, of the same directive, Member States must not impose a general obligation on providers, when supplying the services covered by Art. 14, to monitor the information that they transmit or store, or a general obligation to actively seek facts or circumstances indicating illegal activity. Such a prohibition does not concern obligations imposed “in a specific case”. As a result, national courts may order platforms, such as Facebook, to remove some information when the relevant content is identical or equivalent to content which has previously been ruled unlawful, or to block access to that information worldwide, provided that certain conditions are met.
In light of the above, it is clear that ISPs benefit from far more favourable provisions than the providers of services that fall within the scope of Directive 2006/123 or relevant Treaty provisions, but outside that of Directive 2000/31. The other side of the coin is that Member States are particularly restricted in their efforts to regulate ISS. Thus, it is clear that the issue of the legal classification of composite platform economy services is by no means neutral for companies and regulators.
III. The Court’s case law on composite platform economy services
Over the past twenty years, the Court has provided guidance as to legal classification of the services provided by e-commerce, ride-hailing and short-term accommodation rental platforms. These categories of platforms will be examined in turn.
III.1. E-commerce platforms
The surge of e-commerce constituted the first wave of the ongoing digital revolution. In the course of a single generation, Amazon grew from a fledgling online bookseller to the world’s most valuable brand. Directive 2000/31 was then adopted to (de)regulate the services offered by such e-commerce platforms, prompting the Court to tackle some of the questions the directive had left open.
In Ker-Optika, the Court first examined whether the fact that goods purchased online are subsequently delivered to the buyer makes the online sale of these goods fall outside the scope of Directive 2000/31. Although the case concerned sales through a traditional retailer’s website, it is instructive for third-party platform services. The Court indeed found that the coordinated field of Directive 2000/31 covers national provisions prohibiting the online offer and the conclusion of sales contracts by electronic means but, by virtue of Art. 2, let. h, sub-let. ii), thereof, does not cover requirements applicable to the delivery of goods, which fall within the scope of Treaty provisions on the free movement of goods. The Court then considered another underlying offline service, namely the requirement that a doctor had physically examined the customer prior to the sale. In this regard, it adopted a more complex approach, which partially prefigures the methodology it later followed to analyse collaborative economy services. The Court assessed whether the medical examination at issue was inseparable from the online sale to determine if the existence of the former may influence the legal classification of the latter. It answered in the negative noting that, inter alia, such an examination could be carried out independently.
The Court has not applied Ker Optika to e-commerce platform services yet. In the landmark L’Oréal case, it however clarified that an Internet service facilitating relations between sellers and buyers of goods, such as that provided by eBay, is, in principle, an ISS. While the fact that some e‑commerce platforms provide an increasing number of physical services could eventually lead the Court to reach a different conclusion, it has yet to make a step in that direction. Indeed, in the recent trade mark case Coty, although AG Campos Sánchez-Bordona had contended that Amazon did not operate like a neutral e-commerce platform under its “Fulfillment by Amazon” scheme, given that Amazon takes care of all operations on behalf of retailers, the Court carefully stuck to the facts presented by the referring court, leaving the question open for future proceedings.
III.2. Ride-hailing platforms
As e-commerce platforms were starting to generate tremendous revenue, collaborative platforms emerged. After Uber went international in December 2011, with a launch in Paris, it quickly enjoyed commercial success, prompting harsh criticism by taxi companies and trade associations, which complained it did not meet the regulatory requirements applicable to the taxi industry. The Court analysed the classification of the UberPop service in Asociación Profesional Elite Taxi (hereinafter “Elite Taxi”) and Uber France. The legal rationale behind both judgments is the same: the intermediation service provided by Uber is, in principle, a separate service from the urban transport services at issue, but Uber offers the latter, which it renders accessible through its application and whose general operation it organises for the benefit of users.
In this regard, the Court adopted a two-pronged approach. First, Uber provides an application without which “(i) […] drivers would not provide transport services and (ii) persons who wish to make an urban journey would not use these services”. Second, Uber “exercises decisive influence over the conditions under which that service is provided”, inter alia, by determining “at least the maximum fare”, receiving that amount from the client before paying the driver, and exercising “a certain control over the quality of the vehicles, the drivers and their conduct, which can, in some circumstances, result in their exclusion”.
Consequently, the Court held that Uber’s intermediation service must be regarded as “forming an integral part of an overall service whose main component is a transport service” and “inherently linked to a transport service”. The Court then stated that “accordingly, [it] must be classified as ‘a service in the field of transport’” and added that this classification is confirmed by its case law regarding transport services. It finally specified that this meant that ‘it is for the Member States to regulate the conditions under which [the UberPop service is] to be provided’.
The first criterion seemingly draws on the Court’s case law. The “inherently linked” formula can be traced back to Itevelesa, to which the Court referred to confirm its finding. Further, in essence, the Court said that the intermediation service provided by Uber was indispensable to the exercise of the main activity of transport, just like in Itevelesa. This criterion is also consistent with Ker-Optika. Indeed, the Court ultimately seeks to determine if the online service may be separated from the offline service at stake by, inter alia, assessing whether the latter can be provided separately from the former. However, besides the argument that Uber’s application is not actually indispensable to the services provided by non-professional drivers, the very relevance of this criterion is debatable (see section IV).
In any case, the Court laid out a second, and arguably stronger, criterion, which appears best suited to bring about some predictability for platform economy services. In its observations in Elite Taxi, the Commission had warned the Court against ruling that any intermediation service linked to a transport service falls outside the scope of Directive 2006/123, contending that this would penalise too many services, including travel agencies and ride-hailing platforms. In its communication on the collaborative economy, it had proposed a stricter approach, according to which price control and ownership of key assets was instrumental in determining whether a collaborative platform controlled the underlying service. On the one hand, the Court took up the idea that control, and specifically price control matters. On the other hand, it dismissed the fact that Uber owns no vehicles, whose relevance for the assessment of collaborative economy services had been previously challenged in the literature.
Last, the Court concluded that Uber’s intermediation service must be regarded as forming an integral part of an overall service whose main component is a transport service. Although it can be argued that this is a mere consequence of the reasoning analysed above, it could be considered that this, in fact, constitutes a third criterion, as Airbnb Ireland suggests (see section II.3.).
This case was abundantly commented. Some argue that the Court was driven by political economy considerations. Yet, many acknowledge this case set a new standard against which collaborative services should be analysed.
III.3. Short-term accommodation rental platforms
After Uber, Airbnb soon came into the spotlight as the next troublemaker of the collaborative economy. As the platform was accused of hollowing out European city centres and competing unfairly with hotels, the Court ruled that its services must be regarded as an ISS in Airbnb Ireland.
In the first place, the Court examined the service provided by Airbnb. First, it held that the service is separable from the “accommodation service” provided thanks to the platform, and does not form an integral part of an overall service whose main component is an accommodation service. The Court indeed considered that Airbnb’s service does not solely aim at providing an immediate accommodation service, but that the “essential feature of the electronic platform” lies in “the creation of a list” for the benefit of both the hosts and the persons looking to rent a place, a list which cannot be regarded as “merely ancillary” to an overall accommodation service “because of its importance”. The Court added that the service is in no way indispensable to the provision of accommodation services, since guests and hosts have many other channels at their disposal. Finally, it noted that Airbnb does not set or cap rent prices.
Second, the Court explained that the other services offered by Airbnb do not call into question that finding, since they are ancillary in nature and do not substantially modify the characteristics of the main service. Third, the Court refuted the arguments of the hotel industry and the French government, arguing that, unlike Uber, Airbnb does not exercise decisive influence over the accommodation services, since it neither controls rent prices nor does it select the offers put up on its platform.
In the second place, the Court clarified that the obligation to notify restrictions under Directive 2000/31 also concerns measures enacted before the entry into force of the directive and sanctioned the unenforceability of un-notified measures.
The Court’s analysis of the main service provided by Airbnb is divided in three parts, in each of which the Court contrasted this service with UberPop, whether explicitly or not. In the first part, it found that Airbnb’s intermediation service was not “merely ancillary to an overall service”, while it had implicitly held, in Elite Taxi, that Uber’s intermediation service was merely ancillary to the transport service at issue. Admittedly, Airbnb’s listing does allow guests to select hosts, and vice versa, which the Uber application does not. Still, as the authors have already argued, “the connection stage” is neither “self-standing” nor “the main supply”, but “merely preparatory” in both instances.
In the second part of its analysis, the Court essentially applied the first criterion set out in Elite Taxi and said that Airbnb is not indispensable to the provision of accommodation services, despite the fact that, similar to Uber, Airbnb is arguably a market maker that has expanded the market for residential accommodation and created a new supply of short-term rentals that would not exist without the platform. Bearing in mind that AG Szpunar had argued in his Opinion that this criterion should constitute only ‘an indication that a service provided by electronic means forms an inseparable whole with a service having material content’, it is unsettling that the Court relied on it so extensively.
The third part of the Court’s analysis is reminiscent of the second criterion laid out in the Uber case. Like in Elite Taxi, the Court dismissed the fact that the platform does not own the key assets used to provide the underlying service, confirming that this factor should not be given too much importance. Unlike in Elite Taxi, the Court focused solely on price control. This minimalist approach has the merit of offering a robust, albeit simple, criterion to distinguish between the services offered by Uber and Airbnb, which cannot be overstated with regards to fast changing services. It nevertheless appears overly simplistic (see below).
The Court then examined whether the existence of other services called into question the finding that Airbnb’s ISS is separable from the underlying accommodation service. It strictly followed the Commission, according to which the fact that collaborative platforms provide services that are “ancillary to the core [ISS] offered by the platform”, such as payment facilities, insurance coverage, and user rating or review mechanisms, in itself, does not “constitute proof of influence and control as regards the underlying service” This part of the judgment is very much in line with Ker-Optika. Regrettably, instead of clearly referring to this case, the Court chose to rely on fiscal cases whose connection with the legal question at issue seems, at best, uncertain and indirect, and whose interpretation sits unwell with the rest of the judgment.
The Court never clarified why the complementary services do not allow Airbnb to exercise decisive influence on the accommodation service. It merely specified that Airbnb neither “determine[s], directly or indirectly, the rental price charged” nor selects the hosts or the accommodation. It concluded that the complementary services at issue did not ‘provide evidence for the same level of control found by the Court’ in Elite Taxi. This part of the judgment, albeit crucial, is so short that it is hardly convincing. It ignores some of the very elements at the heart of Elite Taxi, such as the management of payments as well as the ability to control the quality of services and exclude service providers, even though the AG had rightly pointed out that Airbnb can actually “suspend a listing, cancel a reservation, or indeed prohibit access to its platform”. It therefore leaves the question open whether quality control is capable of rendering an ISS inseparable from the underlying offline service when the price is not fixed by the platform.
In the end, the Court arguably upheld most of the framework set out in Elite Taxi and Uber France, but cherry-picked the facts of the case to conclude that Airbnb provides an ISS.
III.4. Conclusion on the Court’s case law
It can be inferred from the aforementioned judgments that, when confronted with a composite service made up of both online and offline supplies, the Court adopts two different approaches. On the one hand, when the underlying offline service at issue consists in delivering goods cross-border, it simply applies the provisions applicable, in principle, to each of the supplies. On the other hand, when the underlying offline service at stake does not relate to the cross-border delivery of goods, the Court further proceeds to ascertain whether the online supply is separable from the offline supply. The test that the Court applies in order to do so seemingly depends on the specific nature of the offline service concerned.
When confronted with an offline service that constitutes a pre-condition to access an online service, the Court relies on the mere fact that the offline service may be provided independently from the online service (see Ker-Optika). Incidentally, while there is a possibility that the Court would not confirm this jurisprudence after its recent judgments, there are indications that it may still do so. Indeed, AG Øe recently analysed whether offline advertising preceding an online sale was separable from such an act, by relying solely on Ker-Optika. Further, in Dobersberger, the Court itself adopted an approach in line with Ker-Optika to determine whether on-board services, cleaning services or the provision of food and drink on trains are inherently linked to the service of rail passenger transport.
When confronted with online intermediation services that aim at connecting clients with the service providers of an underlying physical service, the Court relies on at least two criteria (see Elite Taxi, Uber France and Airbnb Ireland). It examines (i) whether the intermediation provider creates the offer and the supply, and (ii) whether it exercises decisive influence on the conditions under which the underlying service is provided. Amongst the relevant factors for the analysis of the second criterion, price control emerges as a key determinant. Arguably, the Court also considers (iii) whether the examined online service is susceptible of forming part of an overall service whose main component is the offline service at issue or, on the contrary, cannot be regarded as merely ancillary to such an overall service.
However, the many inconsistencies between Elite Taxi, Uber France and Airbnb Ireland do not allow for the identification of a precise legal standard. A number of significant questions indeed remain: are there two or three relevant criteria? Are these criteria cumulative? What are the determining factors when assessing if the criteria are fulfilled? In addition, one may wonder how much Uber’s unique characteristics influenced the Court’s judgements. First, “in the specific context of urban passenger transport”, which Member States are competent to regulate for now, an “incomplete – or simply apparent – liberalisation [would have] create[d] legal uncertainty, […] encouraging infringements of the law”, as AG Szpunar warned in Elite Taxi. Second, the control Uber exercises on its drivers is so strict that it may be indicative that these drivers’ so-called independent status is purely notional.
Future cases will hopefully provide additional clarity. The fact that most prominent composite services of the platform economy differ from the services examined in Elite Taxi, Uber France and Airbnb Ireland may force the Court to deliver more information on the relative importance of the criteria and of the factual elements used in these cases. For instance, BlaBlaCar strictly caps the price of the underlying transport service to comply with the French legal framework regarding carpooling, but has no control over the other conditions under which the underlying transport service is provided. Similarly, the Romanian mobile application Star Taxi may fix the price paid by the user, but it exercises close to no control on the quality of the vehicles, the drivers or their conduct. The opposite is true for the “Airbnb Plus” scheme, under which Airbnb exercises stringent control over the hosts, while it does not set or cap prices. This scheme only accepts hosts who agree to “a home visit with a third-party inspector” and who can secure continuous conformity to Airbnb’s hospitality standards, high ratings, as well as specific booking acceptance and cancellation rates. Were the Court to examine these services, it would need to weigh quality control against price control.
Other collaborative economy services may well be controlled by the platform, but they most likely did not create a new supply. After Elite Taxi, it may be considered that Uber, Uber Eats and Deliveroo exercise decisive influence on the respective underlying services. However, the professional drivers and riders providing the latter certainly have “a number of other, sometimes long-standing, channels at their disposal”. If the Court were to hold that these services are transport services, as it has been argued, it may need to acknowledge the indicative nature of the first criterion of the Uber test, which it has refrained from doing up until now.
The considerations above clearly show that the existing case law does not allow for a definite conclusion as regards the legal classification of the main platforms’ composite services.
The only element that is now beyond dispute is the fact that, until the EU legislature adopts common rules for UberPop-type services, Member States may regulate such services, in conformity with the general rules of the FEU Treaty. It is also settled that Member States may regulate the underlying offline service of platforms such as Amazon or Airbnb, provided that they comply with applicable EU law provisions. In this regard, the judgement in joint cases C-724/18 and C-727/18 Cali Apartments is expected to provide a thorough interpretation of the provisions of Directive 2006/123 relating to the freedom of establishment of providers. In the opinion of AG Bobek, the objective of tackling a shortage of long-term housing constitutes an overriding reason relating to the public interest capable of justifying an authorisation scheme imposed on Airbnb hosts in Paris. The Court will most likely follow his approach and the indications it might give as regards the proportionality of this scheme will be of great significance for the regulation of rental platforms. Still, even if the Court were to largely preserve Member State powers to regulate the underlying service, platforms that provide ISS would still be able to rely on the liability “safe harbour” set out in Directive 2000/31 to avoid any involvement in such regulation.
IV. The policy implications
The current legal framework appears suboptimal, insofar as it is not conducive to a sufficient degree of platform involvement in the regulation of third-party content or services across the EU. An overhaul may be needed.
IV.1. The shortcomings of the current legal framework
The liability exemption of Directive 2000/31 is accused of allowing, or even encouraging, platforms to refrain from taking part in any regulatory effort. The test laid out by the Court in the aforementioned judgments further provides platforms with the wrong incentives. It is hardly desirable that companies may lose the benefit of the favourable provisions of Directive 2000/31 by virtue of the mere facts that they create a new supply and fix the price of the underlying service. Indeed, there is no reason to provide a disincentive to the development of cross-border activities and the offer of new services. Further, platforms, such as BlaBlaCar, that use price control to exclude profit-seeking behaviour from their marketplace should not be penalised. More generally, the case law may encourage platforms to loosen the control they exercise over underlying services, thereby creating another so-called “good Samaritan paradox”, which keeps optimal regulation at bay.
This is all the more regrettable given that, although platforms offer “the promise of greater control”, this does not translate into practice. Theoretically, the ubiquitous control they exercise over economic agents enables them to “play an active role in correcting market failures that are traditionally addressed through regulation”. Yet, there is mounting evidence that major players have “lost control of their massive platforms – or decline to control them”. In this regard, the latest inquiry of the European Consumer Organisation shows that two-thirds of 250 products bought from online marketplaces fail safety tests. In addition, experience shows that platforms are often restrictive and selective in granting access to their user data, and reluctant to cooperate with public authorities when it comes to preventing breaches. In this last regard, the mayors of numerous EU cities complained that short-term accommodation rental platforms “claim that they are willing to cooperate with the authorities, […] [but] don’t or only do so on a voluntary basis”. 
For this reason, although commendable, the Commission’s actions to foster self-regulation, convince Alibaba, Amazon, eBay and Rakuten, Cdiscount and Allegro to pledge to remove dangerous products from their marketplaces faster, and persuade Airbnb, Booking, Expedia and Tripadvisor to provide Eurostat with data on short-stay accommodation rentals are insufficient. As for existing binding legal acts, the rather timid Platform to Business (P2B) regulation excludes all platform services that do not classify as ISS, preventing a consistent approach to the responsibilities of platforms supplying composite services at EU-level. Finally, as the Parliament argues, “the need to go beyond the existing regulatory framework is clearly demonstrated by the fragmented approach of Member States […] [and] by the lack of enforcement and cooperation between Member States”.
IV.2. A way forward
For the above reasons, it is of the utmost importance that the proposal for a new Digital Services Act (DSA), due at the beginning of 2021, compels platforms to leverage the data they are amassing to contribute to the enforcement of sector-specific regulations.
There are several, not mutually exclusive options. New obligations may require platforms to share relevant data with EU and/or national public authorities. They may also demand that platforms enforce certain regulations themselves, reporting to a regulatory body. For e-commerce platforms, this could consist in further cooperation with regulators and obligations to improve compliance of products sold on their marketplaces. For short-term accommodation rental platforms, this could translate to obligations either to transmit user data to public authorities, or to automatically limit rentals to a given number of nights per calendar year and block listings that do not include a registration number, possibly in accordance with the applicable (sub-)national regulations. Thus, in the framework of the DSA, tackling illegal content online could not only mean taking down hate speech on social media platforms, but also blocking offers of unsafe products or accommodations that do not comply with applicable regulations. Finally, to take multi-homing into consideration, the forthcoming Data Act announced by the Commission could set up a legal framework for data sharing among platforms.
Alternatively or complementarily, larger powers should be left in the hands of Member States, even if this would lead to further fragmentation of the Digital Single Market (DSM). In this regard, the list of public-interest requirements provided for in Art. 3, para. 4, let. a), Directive 2000/31 could be complemented to cover additional policy fields, especially where it is desirable to implement rules at a more local level.
Last, ride-hailing, as well as food ordering and delivery platforms, whose services do not classify as ISS, should also take their fair share of responsibilities. Within the ambit of the competences laid out in the Treaty, it would be advisable that forthcoming legal acts do not turn a blind eye on these platforms, which provide some of the most prominent services of the platform economy.
Up until now, the issue of third-party hate speech and fake news has taken centre stage in the current debate on online platforms. The case of unsafe products sold via e-commerce platforms is also heavily debated. On the contrary, collaborative economy services are seemingly left aside. Yet, only a comprehensive legal framework that finally addresses the negative externalities they create will put a stop to the current undesirable surge in uncoordinated national initiatives that do not always comply with EU law.
* Référendaire, Autorité de la concurrence, firstname.lastname@example.org.
** Senior Academic Assistant, College of Europe, PhD candidate, UCLouvain, email@example.com. The views expressed are solely those of the Authors and should not in any circumstances be regarded as an official position of the French Competition Authority or the French Government. The Authors would like to thank the Editor and the reviewers for their time and valuable remarks. All websites referred to in the following footnotes were accessed on 24 April 2020.
 European Parliament, Draft Report with recommendations to the Commission on Digital Services Act: Improving the functioning of the Single Market, 24 April 2020, 2020/2018(INL), www.europarl.europa.eu, p. 3; see also L. Bellulo, Réflexions dans la Perspective du New Digital Services Act européen, in Digital New Deal, 2020, www.thedigitalnewdeal.org, p. 13.
 Directive 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on certain legal aspects of information society services, in particular electronic commerce, in the Internal Market (“Directive on electronic commerce”).
 Directive 2006/123/EC of the European Parliament and of the Council of 12 December 2006 on services in the internal market.
 D. Adamski, Lost on the Digital Platform: Europe’s Legal Travails with the Digital Single Market, in Common Market Law Review, 2018, p. 721.
 B.G. Edelman, D. Geradin, Efficiencies and regulatory shortcuts: how should we regulate companies like Airbnb and Uber?, in Stanford Technology Law Review, 2016, p. 309.
 Opinion of AG Szpunar: delivered on 30 April 2019, case C-390/18, Airbnb Ireland, para. 44; delivered on 11 May 2017, case C‑434/15, Asociación Profesional Elite Taxi, para. 28; delivered on 4 July 2017, case C-320/16, Uber France, para. 15.
 E.g. Court of Justice, judgment of 10 November 2016, case C-432/15, Baštová, para. 77; see also, for an overview of the case law, Opinion of AG Kokott delivered on 4 May 2006, case C-251/05, Talacre Beach Caravan Sales, para. 27.
 E.g. Court of Justice, judgment of 11 December 2014, case C-113/13, Azienda sanitaria locale n. 5 «Spezzino» and Others, para. 40.
 According to the Commission, the term “collaborative economy” refers to business models where activities are facilitated by collaborative platforms that create an open marketplace for the temporary usage of goods or services often provided by private individuals; see Communication COM(2016) 356 final of 2 June 2016 from the Commission, A European Agenda for the Collaborative Economy, p. 3. For issues relative to the legal challenges specifically posed by the collaborative economy platforms, see V. Hatzopoulos, The Collaborative Economy and EU Law, London: Hart, 2018.
 M.Y. Schaub, Why Uber Is an Information Society Service, in Journal of European Consumer and Market Law, 2018, p. 109.
 Court of Justice, judgment of 25 February 1999, case C-349/96, CPP, para. 29.
 Opinion of AG Szpunar delivered on 11 May 2017, case C-434/15, Elite Taxi, para. 66; see also D. Adamski, Lost on the digital platform, cit., p. 745.
 Court of Justice, judgment of 12 July 2011, case C-324/09, L’Oréal.
 Court of Justice: order of 27 October 2016, case C-526/15, Uber Belgium; judgment of 20 December 2017, case C-434/15, Elite Taxi; order of the President of the Court of 12 April 2018, case C-371/17, Uber; Uber France, cit.
 Airbnb Ireland, cit.
 Working document SWD(2020) 54 of 10 March 2020 of the European Commission, Business Journey on the Single Market: Practical Obstacles and Barriers, p. 30.
 Regarding the specificity of the case law on fiscal measures, see opinion of AG Kokott delivered on 12 September 2019, case C-482/18, Google Ireland, paras 36 and 37; see also A. Chapuis-Doppler, V. Delhomme, Non-Discrimination and free Movement in a Member State to Member State Fiscal Dispute: Case C-591/17 Austria v. Germany, in Maastricht Journal of European and Comparative Law, 2019, p. 855.
 European Commission, Handbook on Implementation of the Services Directive, p. 13.
 C. Barnard, The Substantive Law of the EU, Oxford: Oxford University Press, 2016, p. 418.
 Directive (EU) 2015/1535 of the European Parliament and of the Council of 9 September 2015 laying down a procedure for the provision of information in the field of technical regulations and of rules on Information Society services.
 Opinion of AG Szpunar, Elite Taxi, cit., para. 92; see also Art. 3, para. 1, Directive 2006/123.
 V. Hatzopoulos, S. Roma, Caring or Sharing, in Common Market Law Review, 2017, p. 97.
 Court of Justice, judgment of 4 July 2019, case C-393/17, Kirschstein, paras 66-82.
 C. Barnard, The Substantive Law, cit., p. 361-362.
 Ibid., p. 421.
 For an overview of the debate a debate between the Court’s AG regarding the nature of the list in Art. 16, para. 2, see opinion of AG Wahl delivered on 8 May 2018, case C-33/17, Čepelnik, para. 71.
 Airbnb Ireland, cit., paras 84-85.
 Ibid., para. 95.
 Although it is unclear how the two work together after Airbnb Ireland, AG Øe has proposed an interesting interpretation in his Opinion delivered on 27 February 2020, case C-649/18, A, paras 117-119.
 Court of Justice: judgment of 30 April 1996, case C-194/94, CIA Security International, para. 54; judgment of 27 October 2016, case C-613/14 James Elliott Construction, para. 64; judgment of 4 February 2016, case C-336/14, Ince, para. 84; Airbnb Ireland, cit., para. 98.
 Communication COM(2015) 192 final of 6 May 2015 from the Commission, A Digital Single Market Strategy for Europe, p. 12; see also J.B. Nordemann, Liability of Online Service Providers for Copyrighted Content – Regulatory Action Needed?, 2018, Study for the European Parliament PE 614.207.
 Court of Justice, judgment of 3 october 2019, case C-18/18, Glawischnig-Piesczek, para. 23.
 Court of Justice, judgment of 23 March 2010, joint cases C-236/08 to C-238/08, Google France and Google, para. 114; L’Oréal, cit., para. 116.
 Glawischnig-Piesczek, cit., paras 31-53.
 Court of Justice, judgment of 2 December 2010, case C-108/09, Ker-Optika, paras 28-30 and 46.
 Annex I of Directive 2015/1535 specifies that such service is not an ISS since it is not provided “at a distance”.
 Ker-Optika, cit., paras 34 and 37.
 L’Oréal, cit., para. 109; see also L. Edwards, Law, Policy and the Internet, London: Hart, 2008, pp. 278-279.
 Opinion of AG Campos Sánchez-Bordona delivered on 28 November 2019, case C-567/18, Coty Germany, paras 55-62.
 S. Jacques, CJEU Rules that an Online Marketplace Storing and Dispatching Infringing Products, Without Itself Selling These Items on the Platform, does not Infringe Trademark, in EU Law Live, 6 April 2020, eulawlive.com.
 D. Geradin, For a Facts-Based Analysis of Uber's Activities in the EU: Addressing Some Misconceptions, in TILEC Discussion Paper, 2017, pp. 2, 4 and 5.
 Elite Taxi, cit., para. 34.
 Ibid., para. 38.
 Ibid., para. 39.
 Ibid., para. 41.
 Ibid., paras 40-47.
 Ibid., para. 41.
 Court of Justice: judgment of 15 October 2015, case C-168/14 Grupo Itevelesa and Others, paras 47-50; opinion 2/15 of 16 May 2017, para. 61; see also D. Adamski, Lost on the Digital Platform, cit., p. 743.
 Written Observations of the Commission, Elite Taxi, cit., para. 64.
 Ibid., para. 62.
 Communication COM(2016) 356 final, cit., p. 6.
 M. Finck, Distinguishing Internet Platforms from Transport Services: Elite Taxi v. Uber Spain, in Common Market Law Review, 2018, p. 1630.
 G. Smorto, Critical Assessment of European Agenda for the Collaborative Economy, 2017, Study for the European Parliament PE 595.361, p. 18.
 For another perspective, see M.Y. Schaub, Why Uber is an Information Society Service, cit.
 D. Adamski, Lost on the Digital Platform, cit., p. 742.
 C. Busch, The Sharing Economy at the CJEU: Does Airbnb Pass the “Uber Test”?, in Journal of European Consumer and Market Law, 2018, p. 172.
 Airbnb Ireland, cit., para. 53; Due to poor translation, the English version of the judgment suggests the opposite.
 Ibid., para. 57.
 Ibid., paras 53-56.
 Airbnb provides hosts with an offer format, an optional photography service and a rating system, offers hosts guarantee against damage and optional civil liability insurance, and collects the rents and transfers it to the host only 24 hours after the guest checks in, thus giving the guest assurance that the property exists and the host a guarantee of payment.
 Airbnb Ireland, cit., paras 58-64.
 Ibid., paras 65-68.
 Ibid., paras 88-98.
 Ibid., para. 54.
 Opinion of AG Szpunar, Elite Taxi, cit., paras 64-65.
 Airbnb Ireland, cit., para. 55.
 C. Busch, The Sharing Economy at the CJEU, cit., p. 173.
 Opinion of AG Szpunar, Airbnb Ireland, cit., para. 65.
 Airbnb Ireland, cit., para. 56.
 Ibid., paras 58-64.
 Communication COM(2016) 356 final, cit., pp. 6-7.
 At Airbnb Ireland, cit., para. 64, the Court merely refers to the paragraph of the AG’s Opinion where the latter specifies what he infers from Ker-Optika.
 Airbnb Ireland, cit., para. 68.
 A. Chapuis-Doppler, V. Delhomme, A Regulatory Conundrum in the Platform Economy, cit.
 Airbnb Ireland, cit., paras 65-68 (emphasis added).
 Ibid., para. 66.
 Opinion of AG Szpunar, Airbnb Ireland, cit., para. 55.
 Opinion of AG Øe, A, cit., paras 48-49.
 Court of Justice, judgment of 19 December 2019, case C-16/18, Dobersberger, para. 26.
 Beyond the factors identified below, see D. Adamski, Lost on the Digital Platform, cit., p. 742.
 Airbnb Ireland, cit., para. 55.
 M. Finck, Distinguishing Internet Platforms from Transport Services, cit., p. 1636.
 Opinion of AG Szpunar, Elite Taxi, cit., para. 66; see also D. Adamski, Lost on the Digital Platform, cit., p. 744.
 French Court of Cassation, judgment of 4 March 2020, no. 19-13.316; see also UK Employment Appeal Tribunal, judgment of 10 November 2017, no. 2202550/2015. However, AG Szpunar contended that the decisive influence Uber exercises on its drivers does not mean that these drivers must necessarily be regarded as its employees, see Elite Taxi, cit., para. 54. Further, the Court recently held that delivery drivers of UK’s Yodel are not workers under EU law, see Court of Justice, order of 22 April 2020, case C-692/19, Yodel Delivery Network.
 French Court of Cassation, judgment of 12 March 2013, no. 11-21908.
 This issue is at the heart of pending case: Court of justice, case C-62/19, Star Taxi App.
 Although UberPop has been discontinued across the EU, Uber still offers several other ride-hailing services.
 Airbnb Ireland, cit., para. 66; for a similar argument regarding BlaBlaCar, see L. Van Acker, C-390/18 - The CJEU Finally Clears the Air(bnb) Regarding Information Society Services, 2020, in Journal of European Consumer and Market Law, p. 77.
 D. Adamski, Lost on the Digital Platform, cit., p. 741.
 J.B. Nordemann, Liability of Online Service Providers for Copyrighted Content, cit., p. 10.
 Opinion of AG Szpunar, Airbnb Ireland, cit., paras 62-63.
 G. Smorto, Critical Assessment of European Agenda for the Collaborative Economy, cit., p. 18.
 B.G. Edelman, D. Geradin, Efficiencies and Regulatory Shortcuts, cit., p. 325-326.
 G. Smorto, The Sharing Economy as a Means to Urban Commoning, in Comparative Law Review, 2016, p. 7.
 K. Frenken, J. Schorb, Putting the Sharing Economy into Perspective, in Environmental Innovation and Societal Transitions, 2017, p. 3.
 Communication COM(2017) 555 final of 28 September 2017 from the Commission, Tackling Illegal Content Online. Towards an Enhanced Responsibility of Online Platforms; Commission Recommendation (EU) 2018/334 of 1 March 2018 on measures to effectively tackle illegal content online.
 European Commission, Press Release IP/20/194 of 5 March 2020, Commission Reaches Agreement with Collaborative Economy Platforms to Publish key data on Tourism Accommodation.
 Draft Report on Digital Services Act, cit., p. 6.
 Regulation (EU) 2019/1150 of the European Parliament and of the Council of 20 June 2019 on promoting fairness and transparency for business users of online intermediation services.
 This loophole was yet discussed during the negotiation process; see T. Madiega, Fairness and Transparency for Business Users of Online Services, 2019, European Parliament Briefing PE 625.134, p. 8.
 Draft Report on Digital Services Act, cit., p. 3-4.
 For other ideas on how to further regulate platforms, see L. Bellulo, Réflexions dans la Perspective du «New Digital Services Act» européen, in Digital New Deal, 2020, p. 13, www.thedigitalnewdeal.org; see also European Law Institute, Model Rules on Online Platforms, www.europeanlawinstitute.eu; see also M. Inglese, The Collaborative Economy Legal Conundrum: A Way Forward Through Harmonization, 2018, in Legal Issues of Economic Integration, p. 388.
 The Parliament may take the view that a central regulatory authority should be established, but that such authority should prioritise cooperation between Member States, in close cooperation with a network of independent National Enforcement Bodies (NEBs). See Draft Report on Digital Services Act (2020), cit., p. 7.
 The Parliament may call on the Commission to remedy the current legal loophole which allows suppliers established outside of the Union to sell products online to European consumers which do not comply with Union rules on safety and consumer protection, without being sanctioned or liable for their actions. It may also explore expanding the “Product Safety Pledge” and making some of those commitments mandatory. See Draft Report on Digital Services Act, cit., pp. 11, 16.
 The Parliament may call on the Commission to clarify what falls within the remit of the notion of “illegal content”. See Draft Report on Digital Services Act, cit., p. 7.
 Communication COM(2020) 66 final of 19 February 2020 from the Commission, A European Strategy for Data, p. 13.
 Opinion of AG Bobek delivered on 2 April 2020, joined cases C-724/18 and C-727/18, Cali Apartments, para. 136.
 P. Van Cleynenbreugel, Will Deliveroo and Uber Be Captured by the Proposed EU Platform Regulation?, cit.
 The Parliament’s draft report on the DSA only states that this instrument should “clarify the legal regime applicable to professional and non-professional services in all sectors, including activities related to transport services and short-term rentals, where clarification is needed”. See Draft Report on Digital Services Act, cit., p. 10.