Other available languages: none
[Check Against Delivery]
Vice President of the European Commission responsible for Competition Policy
Looking back at five years of competition enforcement in the EU
Global Antitrust Enforcement Symposium (Georgetown)
Washington, 10 September 2014
Ladies and Gentlemen,
I want to thank Sharis Pozen and Larry Center for inviting me again to share my views with such a distinguished audience, which this year includes William Baer and Edith Ramirez. I look forward to the conversation Sharis and I will have after this presentation.
This is the last time I am speaking to you as European Commissioner for Competition. So, I thought this would be the perfect opportunity for me to take stock of the work done over the past five years, of the challenges facing competition authorities, and of how we in the EU have used and developed the instruments to tackle them.
Let me start with a general comment on my perspective on the fundamental role of competition authorities, looking back at my experience during these years.
There’s no such thing as an ideal world where all economic actors see the advantage of open markets and competition on the merits, and where there is no need for efficient and independent systems of competition enforcement.
That world does not exist. We know that the reality is very different: competition authorities and courts are needed to preserve open and competitive markets.
Whatever the system put in place in each jurisdiction, one of the main challenges is to find the right balance between its ability to prevent and sanction anticompetitive practices in the public interest and the respect of due process, which is not only essential to ensure that sound and impartial decisions are taken, but is also inherent in our democracies.
Based on these principles, we have in the EU and in most EU Member States administrative systems whereby public authorities – namely the European Commission and the National Competition authorities – take decisions, sometimes imposing fines or legally binding commitments, which can then be appealed before the relevant Courts.
While I have a lot of respect for other systems stemming from different traditions, my own experience tells me that the EU system delivers effective and independent enforcement and guarantees due process. The system, anchored in the rule of law, secures the respect of the rights of parties at all stages of the procedure and ensures a high standard of fairness and impartiality.
I remember that, at the beginning of my mandate, when the entry into force of the Lisbon Treaty implied the increased relevance of the jurisprudence of the European Court of Human Rights in the case-law of the EU Court of Justice, some stakeholders were questioning whether administrative competition enforcement systems – and in particular the combination of investigation and decisions in the same body – were in themselves compatible with fundamental rights.
During these years this issue has been raised before the European Court of Human Rights and the EU Court of Justice. We now have a string of very recent case law confirming that the EU institutional framework complies with fundamental rights requirements.
By the way, contrary to some opinions I’ve heard, the European Court of Justice also provide a thorough review of our cartel and antitrust decisions, including the amount of the fines imposed.
Over and above these requirements, the Commission has, on my initiative, taken a range of measures early on to enhance openness and transparency in the relationships between companies and DG Competition during the proceedings.
In 2011 we approved new antitrust best practices to give parties more opportunities to interact with DG Competition and to explain their arguments at an early stage. We have also strengthened the role of the Hearing Officers, who are the guardians of the parties’ procedural rights and report to me and to the College of Commissioners before we adopt decisions.
Of course, at the same time we must make sure that our procedural safeguards are not abused and our investigations remain efficient.
Beyond the fundamental rights of defence companies are entitled to, competition authorities must strive to keep their enforcement and policies clear and predictable – whatever the systems’ architectures.
Preventing unnecessary delays will be a challenge for the future – of course, without throwing the baby out with the bathwater.
Another important issue is to guarantee that there is no discrimination between companies regardless of their national origin.
And we must aim at providing the best possible guidance to facilitate compliance among businesses.
To this end, since the beginning of my mandate I introduced new antitrust rules and guidelines to cover a number of issues such as distribution, standards, exchanges of information, and so on. These rules are designed to keep enforcement in touch with the realities in today’s markets.
More recently, updated rules on R&D agreements have also been adopted to make sure that these agreements are used to promote, rather than stifle, innovation.
Similarly, our rules regarding fines are transparent and clear, thanks to the Guidelines adopted in 2006.
This brings me to cartel enforcement, which has been one of the main priorities during my mandate.
Over these years, our practice has shown two main trends: a higher proportion of global cartels – which calls for strong international cooperation – and more cartels detected in sectors other than the traditional industrial sectors – from finance to energy trading.
As part of our efforts to adapt to these and other developments in the markets, which often call for a swifter resolution of cases, we have introduced cartel settlements in the EU.
The first case was settled in 2010. Four years later, I can tell you that the new procedure has already had a positive impact on our practice.
Many companies today prefer to settle to limit the damage to their reputation, clean up their operations, and go back to business as quickly as possible.
So far, the European Commission has settled 14 cartel cases in a variety of sectors, involving 65 undertakings and imposing fines for a total of € 4.1 billion.
According to our legal framework, settlements offer quicker procedures, a 10% reduction in the amount of the fines, and shorter decisions.
It's important to understand that not all cases are suitable for settlement. The Commission can use the procedure at its discretion and if companies do not play by the rules, we can revert to the normal procedure at any time.
As a matter of fact, in the Smart Card Chips decision adopted a week ago, we started settlement discussions and then reverted to the ordinary procedure for lack of progress.
In principle, in contrast to the practice here in the US, a settlement is concluded with all parties to a cartel at the same time. But we also had a few “hybrid” cases where, because some of the companies involved ultimately decided not to settle, we nevertheless took a settlement decision while reverting to the normal procedure for the remaining parties.
The best known case is our investigation in the Euribor cartel in which, following the settlement decision taken in 2013 vis à vis some of the banks involved, normal proceedings are ongoing vis-à-vis JPM, Crédit Agricole and HSBC.
Although these hybrid cases will likely remain the exception, keeping this possibility open allows us to use the settlement procedure without being held hostage to the decisions of the companies that prefer not to settle.
Now, let me say a few words on our fining policy.
Our general approach is to deter through fines and use our leniency programme to encourage companies to cooperate with us to uncover and investigate cartels.
Our leniency policy offers incentives to the companies that help us to detect such infringements. A large proportion of our cartel investigations result from participants who approach us to reveal illegal practices.
As to our fines, they are set following our fining guidelines. Over the last years, the EU Courts have confirmed all the main features of the guidelines; including on fundamental principles and rights, gravity, duration, entry fee, and deterrence multiplier.
Since 2010, the Commission has meted out fines for a total of about €8.8 billion against companies that had formed cartels. These include €1.71 billion in our financial decisions of last December concerning interest-rate derivatives, and almost €1 billion in the recent automotive ball bearings settlement.
I sometimes observe some contradictions in the comments I hear in Europe on our fining policy. Some say that they are too high; more recently, others have said that they are too low and that setting up a cartel is still a bargain for companies.
I do not agree with either view. Each case is analysed and sanctioned in its own merit, while keeping in mind the need for overall deterrence.
In light of the figures I just mentioned and the risk of a tarnished reputation, our fining policy tells companies not to get involved in cartels because the negative consequences far offset the expected gains.
I can formulate my message in a clearer way: If you did get involved, get out of it as fast as you can. And when you’re out, don’t do it again.
Regarding fines, I still expect two major developments in the future.
One is further convergence on fining principles within the EU. National Competition authorities enforcing EU competition rules still apply national guidelines for fines but the basic principles should be the same.
For example, parental liability has evolved in my decisions over the last few years. However, in some EU countries parent companies cannot be held liable.
Also, the turnover used to determine the maximum amount may vary in some countries. Differences of this type can lead to situations where the threat of fines in some Member States does not achieve the desired deterrent effect.
In light of this, I recently called for more convergence among EU National Competition authorities based on our ten-year experience of decentralisation.
This means aligning the fining principles but also ensuring that all NCAs have sufficient resources, the necessary autonomy, and the same fundamental investigative and decision-making powers.
It will be up to the next Commission to decide on the most appropriate initiatives to be taken.
The other development I expect is linked to private enforcement, thanks to new EU legislation on antitrust damages actions.
As you know, under my tenure the Commission proposed a Directive on antitrust damages actions to facilitate compensation of victims throughout the EU.
The European Parliament and the European Council of Ministers have reached political agreement on the text before the summer and formal adoption is set to take place in the coming weeks.
Once the new Damages Directive becomes law in the EU, one can expect more actions brought by customers who seek compensation for the harm they suffered.
The new legislation will democratise enforcement and empower the victims of antitrust infringements to receive effective compensation, because damages actions will be more readily available to victims throughout the EU.
The directive will also enforce the principle that all victims have the same right to compensation in the EU, which is excellent news especially for SMEs and consumers.
In addition, the damages directive will fine-tune the interaction between private and public enforcement – that is, between court actions and the work of national and EU agencies – by allowing victims to wait for and rely on the decisions of the Commission or the national authorities.
Finally, the directive will preserve public enforcement and a company’s incentives to cooperate with competition authorities through leniency programmes and settlement procedures.
All in all, this new EU law will ensure that private enforcement complements and relies on a strong public enforcement, first and foremost as regards cartels but also as regards other anticompetitive practices such as abuses of dominance.
Ladies and Gentlemen,
I will now turn to the domain we call in Europe “abuse of dominance”.
The laws that apply in our respective jurisdictions do not overlap completely. For various reasons, we make a distinction between “exploitative” and “exclusionary” abuses.
This distinction derives from article 102 of the EU Treaty and the case-law.
Unlike the Sherman Act, our Treaty does not prohibit the acquisition of dominance. It only applies to abuses of already existing dominance, hence the need to exert additional control over “exploitative” practices of dominant companies.
Still, this field of EU competition policy is bound to evolve in light of legal and market developments.
Recently, the EU General Court – which is the Court of first instance – has issued important rulings in the field of abuses of dominance. Let me say a few words on them.
On the 12th of June, the General Court upheld a €1.06 billion fine the Commission had imposed on Intel back in 2009 for abusing its dominant position in the microprocessor market.
The Court found that Intel had engaged in two types of abusive behaviour.
- First, Intel gave wholly or partially hidden rebates to computer manufacturers on condition that they bought all, or almost all, their CPUs from Intel.
- Second, Intel made direct payments to manufacturers to stop or delay the launch of products containing competitors’ CPUs and to limit the sales channels available to these products.
The Court has reviewed the facts of the case extensively and set clear-cut principles for this type of exclusionary conduct.
The judgment sets out the test for the assessment of exclusivity rebates. In essence, the Court held that these practices are abusive as such, unless they are justified by economic efficiencies.
The Court thus made it clear that the Commission is not bound by law to always establish the economic effects of certain abuses, especially when they clearly do not constitute competition on the merits.
However, outside the context of this type of behaviour which is abusive by its nature, in a range of cases the EU Courts have assessed the potential effects of abuses before finding that the dominant company infringed our rules.
This is, for example, how the Court of Justice dealt with predatory pricing or margin squeeze practices by a dominant company.
By confirming the fine imposed by the Commission on Intel, the Court also suggested that serious abusive behaviour on the market should be sanctioned as harshly as collusive behaviour. So, this judgment will likely have implications on our current policy on rebates and exclusive-dealing arrangements, and the type of economic analysis to be conducted. We may need more time before we can reach more definitive conclusions.
More recently, in July, the General Court gave another important judgment in the Greek Lignite case as regards abuses by State-owned companies. In this case, the State-owned energy supplier had been given exclusive access to national energy sources, thus necessarily excluding potential competitors.
Essentially the Court signalled that granting certain exclusive rights to such companies can be “automatically” abusive, irrespective of the actual behaviour of the company.
This could be an important development for competition enforcement in the future as regards this kind of State-owned undertakings.
Second, beyond legal developments, economic developments in our digital age also raise new issues when dealing with abuses of dominance.
In digital markets, successful players can emerge as leaders very quickly. Over time these companies can build large platforms, both locally and globally. This is often the result of innovation and smart business strategies, which are central to economic growth and should be encouraged.
However, competition concerns may arise when leading market positions transform companies into gatekeepers to consumers for other market players. This is particularly so in industries in which network effects reinforce and perpetuate market dominance.
In the digital sectors, network effects can be closely linked with big data – the access to big data is becoming a major barrier to entry.
Digital platforms also tend to operate as two-sided platforms, with users on one side and advertisers on the other. The more users you have, the more valuable the platform becomes to advertisers. And the more advertising revenues are earned, the more investments can be made to attract additional users.
These patterns are important to explain why digital platforms tend to become dominant. As I said before, achieving a dominant position does not in itself constitute a competition problem in the EU.
Competition policy is about ensuring that the competitive process remains healthy so as to foster and allow innovative companies to grow. And this is where our Google search investigation fits in.
The case is not about Google’s dominance in search. The case is about the alleged abuse of that dominance with the aim to exclude competitors from other markets where Google faces competition from large or small innovative rivals.
In this particular case, one of the main issues relates to Google using its dominant position in search to enter other markets – the so-called vertical search services – where it faces more competition.
By giving its own services preferential treatment, Google could leverage its dominant position in search to weaken competitors in vertical search markets in a way that would harm consumers.
There would be a disincentive for current and future competitors to invest in vertical services as these would be at a competitive disadvantage vis-à-vis the prominently displayed Google services.
This would be all the more damaging for users since they are generally not aware of the preferential treatment Google gives its own services.
Other concerns in the investigation relate to the use by Google of third-party content as well as contractual elements of AdSense and AdWords.
Over the last two years, we have explored with Google possible remedies in order to solve our competition concerns and have asked for significant feedback from the market.
We have received new elements and data from complainants concerning the effectiveness of the latest remedies package. In accordance with the EU antitrust procedures, we had asked the complainants to react to the Google proposal. Some of their arguments seem to be solid enough to be taken into consideration. Therefore, we now need to see if Google can propose solutions to these issues by improving its offer.
Crafting effective remedies is a major challenge in “digital” antitrust cases. These are dynamic, fast-evolving markets.
In the case of Google, the result page for a given keyword may differ from one day to the next. This is the result of a product being constantly optimized and is in itself not an issue. But it also increases the demands on competition authorities that need to decide on effective remedies in an ever-changing environment and ensure their compliance over time.
Finally, we have also seen in recent years an increasing interaction between competition and intellectual property rights, especially in the context of standards and, more specifically, in the so-called “smartphone wars”.
Last April, the Commission adopted two decisions involving the smartphone manufacturers Samsung and Motorola.
The decisions establish that a company can always use injunctions to fight back patent infringements. However, such injunctions may be abusive when the holder of a standard-essential patent has given a commitment to licence it on FRAND terms and when licensees are willing to respect them.
These principles strike a good balance between the interests of patent holders and those of the companies that need those patents to produce their devices.
The former should be fairly remunerated for the use of their intellectual property. The latter should get access to standard-essential technology without the threat of anti-competitive injunctions.
I hope our decisions will bring clarity to the market and help avoid the protracted patent disputes we have seen in the smartphone industry.
These decisions also illustrate the two different types of decisions that the EU Commission can take when it decides to pursue a case.
The first is to formally find an infringement under a “prohibition decision”, potentially with fines. Alternatively, it may take a “commitment” decision, which allows companies to make legally binding commitments to alleviate our concerns.
The latter decisions represent slightly more than half of our decisions since May 2004 when they were introduced. They are essentially aimed at opening up markets more quickly and effectively for the future, whether in the energy sector, in payment services, or in digital industries.
But not all cases can be solved with commitments, either because nothing effective is proposed or because it is better to set a strong precedent, as we did recently with several “pay for delay” cases in the pharmaceutical industry.
Let me conclude my presentation with a short overview of merger control in the EU.
Globalisation is a reality that can be particularly observed in this field. But this does not mean that the definition of the relevant market, when it comes to the analysis of a particular merger case, will always conclude that the market is global.
First, our practice shows that, in many sectors, markets have broadened from being purely national to becoming at least EU-wide. This is a direct effect of the EU internal market being effective in some sectors.
Once industrial customers or final consumers tend to procure on a European rather than a purely national basis, we take that reality into account in our assessment.
We see for instance that merger decisions involving purely national markets now represent only 13% of our decisions, compared to 20% ten years ago.
More significantly, in more than 60% of the cases in which an investigation is launched – i.e. cases which do not fall under a simplified process and are thus potentially problematic – markets were considered at least EU-wide if not global, compared to 48% ten years ago.
Of course, there are variations from industry to industry or among products within a sector, depending on various elements. Still, this shows that our practice evolves with the realities of the markets.
But in some sectors there is no such evolution in Europe; unfortunately, some markets still remain national because of regulatory and economic barriers.
A well-known example is the EU telecoms industry, where operators, despite being global companies, still compete on a national basis.
European consumers only have access to mobile operators present in their respective Member States. When we drive from Strasbourg to Brussels, we have to switch providers three times – and incur roaming fees.
This is because spectrum allocation and telecoms regulations in Europe remain very much national affairs, thus leading to a fragmented European market along national borders.
Existing barriers in the internal market prevent EU companies from competing on a European scale and reach out to a customer base of half a billion potential users. This is what makes the difference with other large telecom markets outside Europe.
To bring Europe’s telecom industry into the 21st century we would need a genuine Single Market for telecommunications.
In an integrated EU market, the definition of relevant markets would automatically change. The Commission would assess the impact of most mergers on the whole of the EU and not on individual national markets, which would no longer exist. Until then, we must rely on this reality.
The rules we have been using in the EU since 2004 to assess mergers have worked well, but there’s always room for improvement. At the end of last year, we simplified the process for unproblematic mergers to make the rules even more business-friendly.
As a result, 70% of the transactions we review already follow a simplified process, in which less information is required from companies and no investigation is needed. This allows us to focus on the most important cases.
In addition, in July we published proposals that would allow the Commission to deal with the acquisition of non-controlling minority shareholdings and make referral procedures from national competition authorities to the EU simpler and faster.
At present, mergers that do not reach a certain threshold need to be notified in various EU countries, which mean that companies have to deal with different processes and sustain higher transaction costs. I believe that in such cases, which are not referred to the Commission, more convergence among national systems would be needed in any event.
Our proposals are currently in public consultation. On the basis of the comments we receive, the next Commission will decide whether legislative initiatives need to be launched.
Ladies and Gentlemen,
Since the start of my mandate, the EU has gone through a period of serious economic difficulties. These have been difficult years for Europe’s companies, for our workforce, and for consumers.
In this juncture, the recurrent risk has been to yield to protectionist and populist views regarding the role of competition authorities and their alleged interference in the way markets should function.
We have resisted with great determination repeated calls for weaker enforcement on the assumption that laxer control would help companies weather the storm.
I know that all our sister agencies in the Member States and the leading competition authorities around the world have similar stories to tell. Our mutual cooperation has become stronger in these difficult years for competition enforcement.
Competition policy is a pillar of Europe’s internal market and the internal market is the single largest asset Europe has to create value and jobs. It is our decisive competitive advantage.
Open markets, even business conditions, and fair implementation of competition laws are also prerequisites for global markets to thrive. It is with concern that I read reports in the press of some jurisdictions using competition enforcement for different ends.
It has never been more vital than in these years – marked by a severe recession with serious economic and social implications in Europe and in other world regions – to preserve the best possible competition conditions and keep markets open, vibrant, and innovation-friendly.