Chillin’ Competition Conference, Brussels, 9 December 2019

'Please check against delivery'


Ladies and gentlemen

It’s a very great pleasure to be here with you today.

This is an exciting moment – just a few days since the European Commission made a new start, under the leadership of its first woman President.

These new starts can often be a little bittersweet. The joy of starting new challenges can be mixed with the sadness of leaving good friends and colleagues behind. So I feel very lucky to be back here today, with the promise of another five years in which I’ll be able to work closely with this wonderful community of competition experts. 

Competition enforcement in a new world

But I don’t see the five years ahead of us as just an opportunity for more of the same. You can never step in the same river twice.

Business these days is increasingly global. That means huge new opportunities for European companies – but it also means they have a lot more at stake. And globalisation can offer a better deal for consumers – but only if competition keeps working the way it should.

Meanwhile, digitisation is also creating new opportunities. New companies, with millions or even billions of users, have emerged from nowhere in the last decade or two. And even that is just one small part of a transformation that’s affecting every part of our economy.

These changes are causing an earthquake in our markets; they’ve helped some companies to grow very large, while others have found it hard to compete. And so it’s no surprise that in the last few years, questions about the future of competition enforcement have moved right into the mainstream of Europe’s debates.

That discussion has put some important issues on the table. There have been valuable questions about how to keep competition working in a world where you sometimes depend on data, to be able to compete. We’ve seen interesting discussions about who exactly global businesses have as their competitors..

It’s true that, at times, the debate can be a bit contradictory. We hear, at the same time – and sometimes from the same direction – that competition enforcement is both too strict and too lenient. On one hand, we’re asked to do more, or act faster, to stop big tech companies driving out competition. On the other, we’re told that by taking firm action in defence of European consumers and businesses, we’re stopping some companies in Europe from reaching the scale they need to be world leaders.

Keeping the rulebook up to date

But that doesn’t take away from the importance of checking whether our way of enforcing the rules is still right for this new world. The challenges we’re facing, at the start of this new decade, mean that we need to look again at the tools we use to enforce the competition rules.

That’s not to say that we need fundamental changes to the rules themselves. It’s true that our world is changing. But human nature doesn’t change so fast. The fundamental motives, like greed and fear, that tempt companies to harm competition are the same as they’ve ever been – and the competition rules are still well designed to deal with them.

But even when a musician knows and trusts her instrument, she still takes the time to tune it before a concert. And that’s what we need to do right now – to make sure that everything’s in tune, so we can give our best performance in the years to come.

We already started that process in the last Commission’s term of office. We’re reviewing the antitrust rules on vertical agreements between the different levels of the supply chain – and we have valuable input from the consultation we ran in the spring. We’ve just launched a review of the rules that deal with horizontal agreements between competitors – so it’s clear when rival companies can work together to produce better results for consumers. We’re also carrying out a “fitness check”, to evaluate whether our state aid rules are fit for the future.

Market definition

But there’s one other thing that isn’t yet part of that work – but which keeps coming up, in discussions of competition policy. And that is the very first step that we take in competition cases – defining the market.

At the end of 1997, the Commission set out the principles for doing that, in a Notice on Market Definition. That notice still applies today. But of course, a lot has changed in the world since then.

Changes like globalisation and digitisation mean that many markets work rather differently from the way they did, 22 years ago. In that time, we’ve also developed and refined the techniques we can use to define the boundaries of a market and the kinds of evidence we use. And experts – like the three special advisers who prepared a report on competition policy in the digital era last spring – have drawn our attention to new challenges with market definition.

So the time has come to review the Market Definition Notice. We want to be sure that the guidance it gives is accurate and up to date, and sets out a clear and consistent approach to both antitrust and merger cases across different industries, in a way that’s easily accessible. And we want to give everyone a chance to set out their views in detail during the review process.

As we set out on this journey, we mustn’t allow the term “market definition” to create the wrong idea about what we really do. That term can give the impression that competition enforcers make choices about what we think the market ought to be. We do not have a magic wand, which allows to "define" what are the boundaries of any given market.

What we do when "defining" markets is to study the complex interplay between product characteristics, customer and supplier behaviour, firms' substitution decisions and regulation. And on this basis we try to answer the question: "Can customers turn to someone else if prices go up, quality down or innovation stops?"

Defining markets isn’t like agreeing the border between two countries, by drawing a line on a map. It’s more like charting a coastline. The shape is already there – our job is just to measure it as accurately as we can.  And nothing we do will change the shape of that coastline itself

The changing scope of geographical markets

But it does develop as time passes. And we can see some general trends in the way that markets are developing.

The achievements of the single market, in knocking down the barriers that once divided Europe’s national markets from each other, have made it much more likely that customers will be willing to turn to suppliers anywhere in the single market. The growing ease of global trade, together with the power of digitisation, can make it easier for customers to benefit from truly global markets.

And this can have an effect on the geographic scope of markets. In 2002 and 2003, for example, we found geographic markets that could be as wide as the single market – or wider – in about half of our merger decisions.  By 2017 and 2018, that figure was close to two-thirds. And we can find, when we return to a market after many years, that we need to update our ideas about where that market’s boundaries lie.

In 1991, Alcatel bought Telettra – an Italian company that made, among other things, telecoms equipment. And the Commission’s decision found that the relevant market for this equipment was still national – because, among other things, there were different technical standards in different European countries. But the markets then opened up. Standards for telecoms equipment were harmonised, first within Europe and then across the world. And so in later decisions, we found that these markets covered at least the single market - if not the world.

So there are genuine trends in the way markets are developing. But our job is not to deal in generalisations.  Our job is to understand the options that consumers really have, by looking in detail at how specific markets work.

A product like cement, for instance, costs so much to transport that there’s often no point trying to buy from a plant that’s much more than a couple of hundred kilometres away. So the market that customers can rely on is very small. And the fact that cement companies are global businesses, with rivalries in many countries, doesn’t change that reality at all.

So it doesn’t make much sense to lay down a general rule for every market – to say, for example, that all markets have now become global. Instead, we need to continue to look at each market on its own merits. Of course, it’s important to know whether the way we do that assessment is appropriate. That’s why, in 2015, we asked two independent competition economists, Bruce Lyons and Amelia Fletcher, to take a look at how we define geographic markets.

The main message of their report was that we’re on the right track. They believe that our way of defining geographic markets is generally in line with the latest economic thinking, and the practice of other leading competition enforcers around the world. At the same time, they also offered some suggestions on how we could do things better.

For example, they suggested that we could make it clearer that market definition is not an end in itself. It’s a valuable starting point for our work; but it doesn’t mean we ignore competition that might come from outside the market.

Earlier this year, for example, when Tata Steel proposed to merge with ThyssenKrupp, we looked at whether imports from other countries – including China – might keep prices down for steel customers in the European market.

In that case, it turned out that those imports didn’t have the quality and reliability they’d have needed, to be a genuine alternative to steel made in Europe. But the principle remains, that we can and we do look at what’s happening outside the geographic market. And I completely agree with the two economists, that we should be clear about this.

Our review of the Market Definition Notice will give us a chance to take stock of this and a range of other issues – and to explore ways to update and improve the way we deal with geographic market definition.

The changing nature of product markets

But the challenge of market definition isn’t only about geographic markets. The changes that we’re going through – especially digitisation – are also creating new challenges for defining product markets – for working out which products consumers are willing to substitute for each other.

One thing about being competition commissioner for five years is that you pick up some handy pieces of jargon. One of my favourites is the “SSNIP test” – this idea that you can see where the boundaries of a market lie, by seeing which products people would switch to if the price for the one they’re using goes up a little. The trouble is that a test like this, which is linked to changing prices, can’t be used with a product that consumers use for free.

In our case involving Google’s Android mobile operating system, for instance, we had to deal with the fact that Android is accessible for free – so there’s no price that we could use for a SSNIP test. So instead, we asked ourselves what would happen if Google reduced the quality of Android a bit. And we found that this wouldn’t make consumers switch to Apple. Without a certain expertise, they might not register the lower quality; and if they did, then the costs of switching to Apple would discourage them from switching – costs like buying a new smartphone, downloading new apps and transferring their data in situations where providers don’t ensure portability. And because of this, we concluded that Google’s dominant position for licensing Android to smartphone makers wasn’t affected by the existence of Apple’s iOS.

Anddigitisation also raises some more challenging questions for the way we define product markets.

Very often, we find that big digital businesses don’t just provide one or two kinds of service. They’re often active in a whole range of different areas, providing consumers with an ecosystem of services, that are all designed to work together well. And as the special advisers pointed out in their report, it can be difficult for consumers to switch from one ecosystem to another. So there may be times when we also need to look at the way that these ecosystems can leave consumers locked in.

And questions like this will be among the things we’ll be able to look at, as we work on reviewing the Market Definition Notice.


So we have a whole series of interesting questions ahead of us, as we launch this review. We’ll need to draw on ideas and experiences that come from many different angles – from public authorities and consumer groups, businesses and individuals.

And we’ll also need you, Europe’s competition experts. Because no one is better at understanding how competition works, or at spotting the difficulties that competition rules need to be able to deal with. In the last five years, the debates and discussions that we’ve had have helped to make competition enforcement work even better. And I count on you for more of the same rich debate, in the five years to come.

Thank you.