Thursday, 21 December 2017

Why Uber isn’t Appy: the ECJ defines the difference between transport and digital services

Lorna Woods, Professor of Internet and Media Law, University of Essex

The Court of Justice has followed its own established tradition and given us a significant judgment just shortly before a holiday break, perhaps in the hope that any emotionally driven (negative) response will have gone away by the time every one has got back to work. Last year we had Tele2/Watson (discussed here); this year it is Uber. The Court, in a relatively short judgment, seems to have followed its Advocate General (discussed here) - at least in outcome - to find that Uber provides a transportation service that may be regulated by the relevant Member State. While Uber may suggest at least in media reports that this ruling is not that significant to its business, the ruling may be the subject of some scrutiny by legal advisors to intermediation services to understand precisely the point at which such services are no longer to be considered information society services (within the eCommerce Directive) and what they might be considered to be instead.

To understand the significance of the judgment at this point, it is necessary to have some awareness of the patchwork of EU legislation potentially covering these services. The eCommerce Directive provides the framework for online businesses and famously provides the safe harbour provisions for those providing some forms of intermediation services: mere conduits; providers of caching services; and hosts. This directive also provides for regulation in place of establishment within the field of the directive, and limits the sorts of regulation that may be allowed. Specifically licensing for the service is not permitted.

The Services Directive provides a similar approach, but certain sectors are excluded (gambling as well as transport, for example) and may therefore be subject to more detailed regulatory systems. Broadcasting (audiovisual media services) are subject to a separate regime too, which specifies certain rules about the content that may be provided on such services.

In sum, the categorisation of a service into a particular category affects the extent to which those services may be subject to regulation. The providers of electronic intermediation services have to date benefitted from a regulatory approach which not only limits regulation but in some instances protects them also from liability in the interests of developing the market in such services.

What we see with Uber, and other intermediation services, is not just the connection of two individuals or entities who want to contract with one another but the creation of that circumstance as well as the control over the terms on which that business is carried out. Essentially, the question before the Court was whether such behaviour was just about providing technical services (an app) or whether the control was such that the intermediary had gone beyond intermediation and at what point that boundary was overstepped. 

The Court here confirmed that, in principle, a service providing information from one party to another would fall within the eCommerce Directive as an information society service (though it did not provide the detailed analysis of the Advocate General); taxis are by contrast transportation services. The service provided by Uber was more than "an intermediation service consisting of connecting, by means of a smartphone application, a non-professional driver using his or her own vehicle with a person who wishes to make an urban journey" (para 37).

From the Court’s perspective, it seems that there are three elements – though their interrelationship is not clear:

That the services be integral to the provision of other services (para 40);

The market creation aspect (para 39) – that is the service providers would not provide the service without the platform’s intervention (in the Court’s words the service providers were ‘led’ to do this); and

The decisive influence by the platform over the conditions under which that service is provided (para 39).

Given the structure of the Court’s judgment (and its use of the word ‘thus’ at the beginning of para 40), it could be argued that the over-arching concern is that identified at (1), but that the issues identified at (2) and (3) help identify a services integral nature. Whether, however, both market creation and control of conditions are necessary.  The Court did not make this point express but merely listed them as factors in this instance; in this there is a similarity to the Opinion of the Advocate General. 

As previously noted, this issue may have consequences for other platforms in the sharing economy and whether they are seen as part of the service that they facilitate. In essence we are seeking to place these platforms on a scale between bulletin board, client management system and agency.

The Court did not expressly consider a point which was central to the reasoning of the Advocate General – that is, the question of whether the two elements could be seen as economically independent of one another (in which case, they should be viewed separately).  One response would be to note that the Court has talked of the electronic communications aspect being integral to the overall service. The terrain described by points 2 and 3 above however is not exactly that of economic interdependence, though the Court suggests as the Advocate General argued that the economic activity is taking place only because of the role played by the platform. 

Does this then suggest that there is a difference between service which amateurs are encouraged to supply and those supplied by those professionally so engaged? This distinction may not make much sense in the context of the gig economy and piecework systems of what is de facto employment.

By contrast to the Advocate General, the Court – after concluding that neither the eCommerce Directive nor the Services Directive was applicable to Uber’s app – took the view that the app was part of a service in the field of transport since that concept “includes not only transport services in themselves but also any service inherently linked to any physical act of moving persons or goods from one place to another by means of transport” (para 41). 

Consequently, the Court avoids having to follow the reasoning of the Advocate General to find Uber responsible for the taxi drivers, instead going straight to the point that Member States may regulate the circumstances in which such intermediation takes place (para 47). This seems a more logical consequence of the reasoning than the somewhat artificial distinction that the Advocate General made between app and taxi service. It does mean, however, that those providing apps to facilitate the provision of services may find themselves having to look to the regulatory framework for that particular service, rather than the regulatory framework for electronic communications.

Barnard & Peers: chapter 14

Photo credit: The Independent


  1. In my reading, I respectfully disagree with the title. The ECJ did not define the difference between transport and digital services and the author mentions that clearly in the text of the article. The Court only qualified Uber's services. There is no real theory of precedent harmoniously featuring in ECJ's practice. Given the diversity of matters and difference of cultural backgrounds of the cases that make it to the court that would not even be a desirable practice. The patchwork concept applies equally to ECJ caselaw. Nevertheless, making more of that is said in the Court rulings could be a dangerous path.

    1. Not really - as Professor Woods said "this issue may have consequences for other platforms in the sharing economy..."