Sélecteur de langues
Autres langues disponibles: aucune
Vice President of the European Commission responsible for Competition Policy
The Google antitrust case: what is at stake?
European Parliament hearing
Brussels, 1 October 2013
Honourable Members of Parliament,
Ladies and Gentlemen,
I would like to thank the organisers, Mr Tremosa and Mr Schwab for having invited me.
It is a good opportunity for me to inform the Parliament about the latest facts on our investigation into Google's business practices, before you discuss it this morning with the participants in your hearing.
My remarks will also be available to the media. It is important that everyone is well-informed, given the great interest in this antitrust case.
We formally opened the Google investigation in November 2010.
After an in-depth analysis of the complaints, a careful consideration of many concerns transmitted to us informally, and a lot of exchanges of views with Google and other stakeholders, in March 2013 we formally informed Google of our preliminary conclusions in our Preliminary Assessment (PA), a step used in so-called "Article 9" procedures – I refer here to Article 9 of the EU Antitrust Regulation – to explain the Commission's concerns.
The PA clearly explained the four types of business practices by Google that we consider as problematic under EU antitrust rules:
1) The favourable treatment, within Google’s web search results, of links to Google’s own specialised web search services as compared to links to competing specialised web search services (for instance, services allowing users to search for specific categories of information such as restaurants, hotels or products).
Users are generally not aware of this promotion of Google's offer within the search results. In addition, competitors' results that could well be more relevant for the user are significantly less visible and even sometimes not directly visible, so they become difficult for the user to find. Our concern is that this practice unduly diverts traffic away from Google's competitors towards Google's own services.
2) The use by Google, without consent, of original content from third party web sites in its own specialised web search services. This may reduce competitors' incentives to invest in the creation of original content.
3) Conditions on publishers preventing them from displaying search advertisements from Google's competitors on their websites; and
4) Contractual restrictions on advertisers preventing them from porting and managing their search advertising campaigns across Google's and competing search advertising platforms.
These four business practices are all liable to harm consumers as they will likely have less choice of innovative services.
On each of the four, we believe there is a strong case for action under our antitrust rules.
Let me clarify that other antitrust cases that are being investigated by the Commission also involve Google but relate to different types of competition concerns.
One is about some of the standard essential patents owned by Google’s Motorola unit in the telecommunications sector and the way in which Google Motorola sought an injunction against a willing licensee; this investigation is well-advanced and an Oral Hearing took place yesterday.
Another investigation which is at a preliminary stage concerns the allegations received about some aspects of the Android ecosystem.
And of course other issues voiced about Google are not within my remit as commissioner for competition: for instance, the current privacy disputes with national regulators and litigations on copyright.
But even if this is not a case about all of Google's activities, the case I am talking about today is complex in many ways. And for several reasons.
It is the first time we are investigating the online search market. We are dealing with many formal complaints, but a lot of informal concerns arrived on my desk as well. They were complemented by submissions of other parties during our market investigations.
The industry we are looking at is also particularly fast moving, because online search itself is constantly evolving. Since we started the investigation, the way search results are presented and the kind of services provided have changed many times.
Moreover, the initial complaints focused on static devices, but in the meantime mobile devices have become more and more relevant, which led not only to the introduction of new search services but also to changes in the way users access and interact with existing ones.
As I said earlier, public opinion and stakeholders pay attention to this case more than to any other. This is one of the reasons I accepted your invitation for today's event, something that is quite unusual in a case that is still being investigated.
As you can understand, I cannot disclose aspects protected by confidentiality, nor advance conclusions in an on-going antitrust investigation, but I will try to give you as much information as possible within these limits.
Let me make some remarks, before I tell you where we are in our investigation.
As you know, antitrust enforcement is about consumer welfare, innovation and choice, not about protecting competitors.
My responsibility when enforcing the antitrust rules in this case is to make sure that Internet users are provided with choice, so they can decide between services based on their merits, and to preserve incentives to innovate across the board, so that users can benefit from new or better services tomorrow.
Antitrust is not an adequate instrument to impose on Google a specific algorithm or to prevent Google from improving its services if it wishes to do so. Nor, as a competition authority, can the Commission act in this case as a regulator for all the issues arising in the online world or raised by stakeholders regarding Google.
Against this background, in the context of our investigation, since the adoption of our Preliminary Assessment we have been negotiating with Google in order to find a potential solution through the Article 9 route, which would mean a Decision making legally binding Commitments which resolve our concerns.
I sought comments on a first proposal by Google through a market test launched in April of this year.
The feedback received from the market test was very negative. Therefore, I asked Google on 9 July to improve significantly its proposals.
Google has now improved the commitments it has offered. We have negotiated improvements until yesterday.
Although I cannot describe the details, I can tell you that the new proposal more appropriately addresses the need for any commitments to be able to cover future developments. Therefore, the new proposal relates to queries entered in Google in whatever form - whether they are typed or spoken – and irrespective of the entry point or the device.
One of the most significant improvements relates to the vertical search concern, which was the point that received the strongest critical comments during the market test.
Many respondents during the market test said that in this Google proposal the links to rivals that would be displayed for certain categories of specialised search services were not visible enough.
In my opinion, the new proposal makes these links significantly more visible. A larger space of the Google search result page is dedicated to them. Rivals have the possibility to display their logo next to the link, and there will be a dynamic text associated to each rival link to better inform the user of its content.
Market test respondents also contested the organization of the proposed auction to determine the rival links that would be displayed on Google's search results page for the most commercial categories of specialised search services.
The new proposal foresees an auction mechanism which includes the option to bid for each specific query. This is important to also ensure that smaller specialized search operators can be displayed.
As regards the second concern included in the Preliminary Assessment – the use of content produced by others – Google has improved the granularity of the opt-out that is offered to third party web sites. It also tightens the provision that ensures that Google cannot retaliate against web sites that make use of the opt-out.
With regard to the third concern, Google has committed to no longer include in its agreements with publishers any provisions or impose any unwritten obligations that would require publishers to source their requirements for online search advertisements exclusively from Google in relation to queries from EEA users. The new proposal improves the safeguards against possible circumventions.
As regards the fourth concern, Google has offered to cease to impose any written or unwritten obligations that will prevent advertisers from porting and managing search advertising campaigns across Google's services and competing services. The new proposal again provides stronger guarantees against circumvention.
Finally, in a commitment that in my view deserves careful attention, an independent monitoring trustee would be put in place to provide assistance to the Commission in ensuring that the principles outlined in Google's proposals would be implemented in practice.
We have reached a key moment in this case.
Following the first market test, I had serious doubts whether it was possible to continue the route towards a Commitment decision. I expressed my opinion to Google and in public.
Now, with the significant improvements on the table, I think we have the possibility to work again and seek to find an effective solution based on a decision under Article 9 of the Antitrust Regulation.
Now, what are the next steps?
Google has committed to support its new proposals with empirical data to show their impact.
At the same time, we will work with Google during the next weeks to finalize the precise drafting of the proposed commitment text.
As a next step, I will seek feedback on the improved commitments proposal from complainants and other relevant market participants. To that end, we will send information requests, on the basis of the EU Antitrust Regulation 1/2003, on the improvements that are being proposed.
We know the general positions of the complainants and other stakeholders. What we need now is to receive concrete technical elements on the effectiveness of the proposed package in order to conclude whether this new proposal is satisfactory from a competition point of view.
If our investigation on this improved proposal is satisfactory, I will continue the Commitments route and end up with a formal decision next Spring. Otherwise, I will be forced to turn to a procedure under Article 7 of the Antitrust Regulation: this would mean sending a Statement of Objections to Google in the coming months, to which Google could formally respond in writing and during an Oral Hearing.
"Prohibition" and "commitment" decisions have different advantages and disadvantages.
Generally, Article 7 decisions to prohibit anticompetitive practices and impose fines have a strong deterrent effect and a precedent value. They are generally preferable when we have to sanction past behaviour or when there is no remedy available to solve the competition problem other than an order to "cease and desist".
As to Article 9 decisions to accept commitments proposed by a company making them legally binding, we can use them to solve competition concerns more quickly and concretely, with an immediate impact on the market and a forward-looking vision of how the market should function.
In the past, the Commission sometimes had recourse to prohibition decisions to address anticompetitive conduct, for instance in the case against Microsoft regarding server interoperability information and Media Player, or in the Intel case. Other decisions, for example on Microsoft Internet Explorer or in the recent E-books case involving Apple and five publishers, have relied on commitments made legally binding.
Of course, if a company does not comply with commitments we can impose a fine for the breach itself, without having to prove an infringement of the Treaty's antitrust rules. A fine of this kind was imposed on Microsoft earlier this year for an amount of €561 million.
In the case I am talking about today, and for the reasons I briefly described when I referred to Google’s latest proposals, I think that the settlement route remains the best choice. I hope the answers we will receive to our questions will confirm this.
European users want undistorted competition and choice in online search and search advertising.
They want it now and, if possible, deserve it now, and not after many years of litigation.
Thank you for your attention.