European Commission v Google: In search of a theory of abuse

The European Commission’s failure to secure acceptable commitments from Google in its search engine investigation has raised questions about the effectiveness of EU competition law to tackle abuse of dominance in digital markets. 

The five-year investigation remains in the ?public eye, but on successive occasions where ?a resolution has seemed imminent, this has not materialised. The Commission’s statement of objections (the objections), which was sent to Google in April 2015, suggested a new direction. However, recent statements from Google’s owner, Alphabet, indicate that the company is set for a long battle with the Commission, which is likely ?to test the outer limits of the EU law on abuse of dominance under article 102 of the Treaty on the Functioning of the European Union (TFEU).

Protracted battle

The focus of the Commission’s antitrust concern is the prominent display, within Google’s web search results, of links to Google’s own specialised web search services (e.g. Google Shopping) relative to links to competing specialised web search services. 

The issue of the formal objections to Google marks a key juncture in the case. It is true that, ?even before this development, the case had already established itself as one of the époque-making cases in EU competition law. It involves a company that only a few years previously was one of the market challengers. In common with the Commission’s investigation into Microsoft over tying and interoperability, which culminated in ?an infringement decision in 2004, it is another ?case where both the Commission and US antitrust authorities have examined similar issues. 

Google has now provided a response to the objections, consisting of more than 100 pages. Although the response is not available publicly, ?the company strongly refutes the Commission’s case. In an interview reported in the Wall Street Journal in October 2015, EU Commissioner Margrethe Vestager said that the Commission is analysing Google’s response but will ‘take some time’ to conclude its case.

Theories of harm

While the case was proceeding as a potential commitments case, it was perhaps to be expected that the Commission would seek to pursue more novel ‘theories of harm’. Now that the Commission is one step closer to an infringement decision, this raises the more fundamental question of whether the theories of competitive harm that it is advancing have a sound legal basis.

The redacted documents released by the Commission when it issued its objections are interesting for the apparent simplicity in terms of the theory of harm that appears to be at the crux ?of this case: whether a dominant firm is entitled to discriminate in favour of its own services. 

The Commission evidently believes that there could be article 102 liability in such instance, ?but on closer inspection the conclusion that a dominant company may not discriminate in favour of its own services does not automatically follow. A consideration of some of the putative theories of harm that seem to be driving the Commission’s thinking reveals that the Google case cannot be neatly characterised within the existing legal orthodoxy.

The theory is controversial for two main reasons. First, it is clear that a dominant company may compete on the merits and is entitled to differentiate itself from its competitors, provided that this is not based on ‘methods different from those which condition normal competition’ ?(Case 85/87 Hoffmann-La-Roche v Commission). This implies that a dominant company may, in >> ?>> principle, compete on marketing elements such as displaying responsive search results, even those that favour its own services. It can be asked why Google cannot show what it considers to be its own directly responsive results, since that is precisely what a search engine does and is a core value proposition. Search engines will compete on the basis of their own offering by showing exactly what they consider to be responsive to ?a user query. 

Second, any obligation on a dominant company to deal with its competitors has traditionally been confined to the situation ?where the firm controls access to facilities that ?are essential to compete. Whether the theory ?of harm is characterised as one of the denial of access to an essential facility or the discriminatory grant of access to a distribution platform by ?a vertically integrated firm, a unifying theme appears to be that of anticompetitive foreclosure. 

However, the Commission has not articulated ?its theory of foreclosure, and it is unclear to what extent access to Google’s platform is an essential facility. Moreover, even where access to an essential facility has been mandated in previous cases, such access need not be on identical terms to that granted to the dominant firm itself, provided that access allows for the provision of a commercially viable service (Sealink/ B&I Holyhead: Interim Measures [1992] 5 CMLR 255).

The key issue for competition is what Google should or should not be permitted to do in terms of differentiating itself. Putting it another way: should third parties be entitled to an equal position in Google’s search results? Even if that is accepted, how is that to be achieved in a way that allows consumers to make an informed choice, without destroying Google’s and other parties’ incentives to innovate? These are the issues at the heart of the Google case.

Infringement decision

Whatever route the case takes – whether ?towards a settlement or an infringement decision – Commissioner Vestager has made plain her view that the case ‘may eventually be tested in court’. Should the Commission proceed to a formal finding of infringement, or even a settlement, ?it cannot expect to be immune from litigation given the vigour with which all sides have pursued their case so far.

A key difference between the commitments route under article 9 of Regulation 1/2003 (apparently on the back burner) and an infringement decision under article 7 is that ?the Commission will usually impose a penalty ?in the latter case. If the Commission issues an infringement decision, Google could, in principle, be fined up to 10 per cent of its worldwide turnover (i.e. up to $6bn). The actual level of ?the fine imposed will be based on a number of factors, including the gravity of the infringement. ?The Commission will typically take as a starting point the relevant sales in the market concerned by the infringement. The highest fine imposed on a single company for a breach of EU competition law was on Intel in 2008 and amounted to just over €1bn, which fell some way short of 10 per cent of Intel’s worldwide turnover. However, an infringement decision and fines of that level are by no means inevitable, and Vestager has hinted at the possibility of a consensual remedy.

The Commission’s focus on Google (and Microsoft before it) has prompted the often recurring question of whether US companies ?are receiving rather more scrutiny under EU competition law than their European rivals. 

The search engine case is not the only abuse ?of dominance investigation that the Commission is pursuing against Google. The Commission is also investigating Google’s practices in relation ?to its Android operating system, alleging that ?it is limiting the development of alternatives by requiring smartphone and tablet manufacturers to pre-install its own applications. 

Another case relates to the effect of exclusivity arrangements that are allegedly preventing advertisers from moving their online advertising campaigns to rivals. A separate case concerns complaints about Google scraping copyright information from other websites. Yet another case relates to the alleged favourable treatment of Google’s other specialised search services, including maps and travel. These cases are all ?at different stages, although they are apparently not being considered by the Commission with ?the same priority as the main search engine case.

The announcement of the objections has ?been hailed as a win for many complainants, including rival search engines. The commitments offered so far have not appeased complainants, but an infringement decision may not give ?them complete satisfaction either. This is ?because a prohibition decision is just that: ?it will say what is not permitted but it is not ?as capable of prescribing changes to market conduct as commitments. 

Anyone who is expecting the case to be resolved speedily will be disappointed, as the Commission will want to be assured that any decision is robust in case of judicial challenge. ?It is continuing with a theory of harm – the duty ?of a dominant company not to prefer its own operations – that is superficially appealing but fraught with legal challenges. This is a novel theory not previously seen in EU law. It has implications for both the new economy and the more traditional brick and mortar industries.

Suzanne Rab is a barrister practising from Serle Courtwww.serlecourt.co.uk