Uber picLorna Woods, Professor of Internet Law at the University of Essex and a Senior Associate Research Fellow at the Information Law and Policy Centre, analyses the Opinion of the Advocate-General in a CJEU case concerning the status of Uber’s taxi services. The post first appeared on Steve Peers’ blog, EU Law Analysis

Case C-434/15 Asociación Profesional Elite Taxi v. Uber Systems Spain SL, 11 May 2017

This case is the first before the Court of Justice specifically on the sharing economy and the extent to which coordination via platform should be treated as removing unnecessary red-tape, or as seeking to avoid regulation in the public interest (in the form of concerns about passenger safety) as well as permitting unfair competition.  While the Commission seems in favour of the (unequal) sharing economy, Advocate-General Szpunar sees the position a little differently.


Spanish law envisaged that taxi firms, and transport intermediaries, should hold a licence.  Asociación Profesional Elite Taxi (APET) sought to challenge the use of Uberpop, an app which allows non-professional private drivers to transport passengers using the drivers’ own cars, where neither Uber nor the drivers have the requisite licences.  Passenger users download the app from Uber and provide their bank details to Uber.  On receiving a request for a car, the app notifies drivers and calculates the fare (this latter based on distance but also demand for taxi services at that time).  The payment is made to Uber, which deducts a percentage and then pays the remainder to the driver.  To use the app as a driver, an individual must comply with Uber’s terms and conditions. APET sought a cease and desist order and a prohibition of future similar behaviour on the basis of unfair competition.  Uber resisted APET’s claims on the basis that it was not providing transport/taxi services but was rather a digital intermediary.


The national court referred the questions of how to classify Uber’s services to the Court of Justice. The answer would affect which EU derived legal regime would be applied to Uber, with the corollary that the State’s freedom to impose licensing requirements would correspondingly vary depending on which regime was held to be applicable.  In essence, the question was whether Uber fell within the provisions of the e-Commerce Directive (Directive 2000/31) as an information society service provider, or whether the Services Directive (Directive 2006/123) or the TFEU itself applied in this context.  Here, there is a distinction between a service general and a service in the field of transportation.


The first phase of the opinion comprises some general remarks about the significance of the ruling and the impact of different types of competence on the outcome.  The Advocate-General also assumed that the respondent in the case should be the Dutch company (Uber BV), which operates the app in the EU, rather than the Spanish company, Uber Spain, which is responsible for advertising.

The Advocate-General then moved on to consider the scope of the e-Commerce Directive, specifically the meaning of ‘information society services’ as defined in Article 2(a) of that Directive by reference to Article 1(2) of Directive 98/34 (the Directive on notifying new technical barriers to trade). Under Article 1(2), an information society must be a service provided for remuneration, at a distance, by electronic means and at the individual request of a recipient. In the view of the Advocate-General, the questions of whether there is a service provided for remuneration and at individual request appeared unproblematic [para 27], but questions arose as to the test of whether the service is provided at a distance by electronic means.  In the eyes of the Advocate General the problem related to the fact that what was in issue was a ‘composite service’ [para 28].

The Advocate-General emphasised that the definition concerned services “‘entirely transmitted, conveyed and received by wire, by radio, by optical means or by other electromagnetic means'” [para 29, quoting 2nd indent of 2nd subparagraph of Article 1(2), Directive 98/34, emphasis in Opinion].  So, services not delivered by electronic means did not fall within the scope of the e-Commerce Directive; services which were incidental to such services would likewise not be liberalised by the e-Commerce Directive. Assuming that they did would undermine the perceived effectiveness of EU law [para 31]. Thus:

‘… an interpretation of the notion of information society services which brings online activities with no self-standing economic value within its scope would be ineffective in terms of the attainment of the objective pursued by Directive 2000/31.’ [para 32]

The Advocate-General suggested that a composite service would be treated as an information society service in two circumstances:

  • where the two elements could be seen as economically independent of one another they would be treated separately for regulatory purposes – the electronic element likely falling with the eCommerce Directive; and
  • where the service provided was substantially or predominantly provided by electronic means.

A common example of the first case would be a three party situation where an intermediary service provider facilitates a transaction between a user and an independent service provider/seller.  While the intermediary provides added value, the trader here pursues an independent business.  In a two party situation – where the intermediary provider is also the provider of a service not provided by electronic means, the two elements cannot be seen as separable; rather, they ‘form an inseparable whole’ [para 35].

In that instance, it will be necessary to see if the composite service falls within the second category; that is, whether the bundle falls within the e-Commerce Directive or outside it. To determine the answer to this second question, the key element is where the economic value lies. So where the main component is performed online that service should be classified as an information society service (assuming the other elements of the test are met); conversely, where it is not then the service does not fall within the e-Commerce Directive. The Advocate-General, citing Ker-Optika (Case C-108/09) suggests that this test would be satisfied in the case of online sales (via the seller’s own website).  Delivery of goods is ‘simply the performance of a contractual obligation’ [para 36].

Applying these tests here, the Advocate-General noted that Uber provided more than a matching service of passengers to taxi drivers. It sets down the essential characteristics of the service to be provided (e.g. quality and age of vehicle; drivers to have licences and no criminal record), it informs drivers where and when there are likely to be a high volume of trips and/or preferential fares, and it sets the prices.  Uber maintains indirect control over drivers through its ratings function.  Thus, in the view of the Advocate-General,

‘… Uber exerts control over all the relevant aspects of an urban transport service ….  (…) Uber therefore controls the economically significant aspects of the transport service offered through its platform’ [para 51].

While the Advocate-General sought to distinguish this case from the cases concerning whether drivers are employees of Uber, on the basis of this indirect control the Advocate-General concluded that:

‘Uber’s activity comprises a single supply of transport in a vehicle located and booked by means of the smartphone application and this service is provided, from an economic standpoint, [citations omitted] by Uber or on its behalf’ [para 53].

The Advocate-General sought to distinguish the activity of Uber from intermediary services on the basis that Uber drivers do not carry out an independent activity. Instead, their activity exists solely because of the existence of the platform.  By contrast, flight or hotel booking systems are separate from the independent services operated by the hotels and airlines and for whom the websites are just one mechanism of advertising their services. Furthermore, it is the hotels and airlines which control the prices and the conditions on which their services are offered. Finally, a choice is offered to the user between hotels/airlines.

In opposition to the Commission’s views on the sharing economy, the Advocate-General did not think that the fact that Uber did not own the cars was determinative. Uber is more than a ‘mere taxi booking application’ [para 64].  Because of the extent of the innovation on the transport sector caused by the way the apps links drivers and passenger and the conditions on which it does this, ‘it is undoubtedly the supply of transport which is the main supply and which gives the services the economic meaning’ [para 64].  The supply of connection services is ancillary to this.

Having determined that Uber’s services do not fall within the e-Commerce Directive, the Advocate-General considered the Services Directive: Article 2(2)(d) specifies that the Services Directive does not apply to transport services.  The Advocate-General confirmed that Uber’s taxi services were transport services in the context of Article 2(2)(d) Services Directive as recital 21 refers to ‘urban transport [and] taxis’ [cited para 68].  The service likewise falls within the exception to the Treaty rules on free movement of services (Article 58(1) TFEU) and therefore subject to the specific transport sector rules in Article 90 TFEU et seq.

The Advocate-General concluded by considering the position should Uber’s app be deemed to fall within the e-Commerce Directive.  He noted that Member States would then be limited in terms of the conditions that they could apply to such a service; drivers however would still be subject to any relevant national regulation. The Advocate-General argued that Uber nonetheless could be penalised for unfair competition as ‘it is responsible not only for the supply whereby passengers and drivers are connected with one another, but also for the activity of those drivers’ [para 86], whether or not the booking app were seen to be separate from the transportation service or not.  The Advocate-General therefore proposed that the e-Commerce Directive

‘does not preclude requirements relating to the activity of transport in the strict sense being established in national law or the imposition of penalties on Uber for failing to comply with those requirements, including by means of an injunction ordering it to discontinue the service’ [para 88].


This opinion will be grabbing the headline news because of the headline fact that – shock, horror – Uber is a taxi company.  The reasoning used is worth a little more attention, because our understanding of that reasoning, if the Court of Justice follows the same lines, will affect any wider ramifications for the ‘sharing’ economy more generally.  So while the Advocate-General starts his opinion by suggesting that the subject matter of the case is ‘narrow’ (para 2), the repercussions are potentially a little broader.

A recognition of this fact can be inferred by the approach of the Advocate-General to the question of scope of the e-Commerce Directive and the insistence that the e-Commerce Directive regulates services that are entirely delivered by electronic means – the emphasis is that of the Advocate-General (para 29), with the result that it cannot be said

‘that any trade-related online activity, be it merely incidental, secondary or preparatory in nature, which is not economically independent is, per se, an information society service (para 37)’.

The opinion reiterates this point when considering the app as an information society service (see below). The significance of this is that the e-Commerce Directive cannot be used to avoid regulation of the main service, at least to the extent that such regulation is not a barrier to trade unacceptable to EU law more generally and as exemplified by the Services Directive, just because some aspect of the business is online.

This leads to a second general point: the Opinion is noteworthy for the way it manoeuvred the circumstances of the case around to the twin obstacles to national regulation of the e-Commerce Directive and the Services Directive, especially given that the Commission, in its Communication on the Collaborative Economy (COM(2016)356) seem to view the Services Directive in particular providing a basis for such services.

The tests identified by the Advocate-General will not seem unusual – talking about whether the different elements of a composite service are severable or preponderantly one thing or another can be seen elsewhere, for example in the case of goods and services (an example of which is given by Advocate-General Szpunar in his reference to Ker-Optika, para 36), or even questions of competence.

A couple of points can be made here.  The first is that in some of the factors to be taken into account in determining independence, the Advocate-General comes close to eliding the question of what is the nature of the service with the question of who is providing it. Many of the factors the Advocate-General considered reflect the Commission’s Communication from last year. Of potentially more consequence is the impact of the Opinion when considering the question of whether the service would exist without the app as a factor.  This can be seen by contrasting the position of Airbnb with Uber.  If we look at control, Uber is caught because of its control over key aspects such as access to passengers, and price (detailed in paras 43-51); on these considerations Airbnb which does not set price might not be caught. However, if the test is that ‘the activity exists solely because of the platform’ (para 56), for Airbnb as for Uber, the answer might well be ‘yes’.  Of course, Uber and Airbnb are different in that transport does not fall within the Services Directive but rather is dealt with by specific provisions within the TFEU.  Nonetheless, the recitals to the Services Directive specify that:

it does not apply to requirements, such as road traffic rules, rules concerning the development or use of land, town and country planning, building standards (Rec 9),

some of which may affect the running of hotels or B&Bs. Further, the directive does not apply to taxation (Article 2(3)), so for example, taxes on short term lets may be unaffected by the Services Directive (though dealt with under the TFEU).

A further point that is worthy of note is the consequence of seeing Uber and the drivers as providing separate services.  The requirement to have a licence to provide a connection service in the context of transportation would fall within the scope of the eCommerce Directive and would be caught by the prohibition on authorisations set out in Article 2(h)(i); the Advocate-General was of the opinion that it was unlikely that any restriction on the electronic aspect of the service could be justified by considerations of public interest.  The prohibition does not, however, extend to the regulation of transport services – they are not provided by electronic means.  So the regulation of taxi services remains possible and, as noted, Uber remains responsible for the drivers’ activities (para 86).  Arguing Uber’s activity as a whole should benefit from the liberalising principles in the eCommerce Directive would run the risk of undercutting any form of regulation ‘because all traders are currently in a position to offer services by electronic means….’ (para 87).  Could a similar argument be put forward in other sectors of the sharing economy?  In principle, yes, but presumably only where the platform has exerted Uber-like control over the actual provision of the services.

Even when the Grand Chamber Court of Justice has handed down its ruling, this will not be the end of the Uber saga.  Currently pending before the Court is a reference from France concerning the imposition of penalties on Uber for running an unlicensed taxi service: Case C-320/16 Criminal Proceedings against Uber France, which raises the question of whether France should have notified its rules as a technical regulation under Directive 98/34.

Barnard & Peers: Chapter 14