Oral evidence - Online Platforms and the EU Digital Single Market

Members present

Lord Whitty (Chairman)

Lord Aberdare

Baroness Donaghy

Lord Freeman

Lord German

Lord Green of Hurstpierpoint

Lord Mawson

Lord Rees of Ludlow

Lord Wei

_______________________

Examination of Witnesses

Daniel Gordon, Senior Director, Markets, Competition and Markets Authority, Alex Chisholm,[1] Chief Executive, Competition and Markets Authority, and Nelson Jung, Director, Mergers Group, Competition and Markets Authority

Q43 The Chairman: Thank you very much for seeing us. We have just been through one session, in which various industrial representatives gave us their views on the market, on whether there should be further regulation, and on whether current regulations were being effectively enforced. We are looking at online platforms. The Commission has indicated that it thinks there may be a problem in relation to online platforms. First, is it possible to provide a useful definition of online platforms and their different forms? Secondly, is it possible to provide a definition of what market domination or the abuse of data through market domination might mean in this rather new area where the market is not very clearly defined and where price discrimination is not evident, because by and large most things are free at the point of use? In other words, how do the traditional competition authorities approach this area?

Alex Chisholm: Thank you very much, Chairman. We are delighted to have been given a second chance to come here. You have already heard from my colleague Jason Freeman on the consumer side. We plan to talk more about competition today.

I will briefly introduce my colleagues. On my right is Daniel Gordon, a competition economist who runs our markets programme; and on my left is Nelson Jung, a competition lawyer who works in our mergers area. I think you have also seen our written evidence, which we stand by. I will try not to repeat any of the points made in that.

Just to jump ahead on your questions, the first thing you mentioned is that you have been focusing very much on problems associated with online platforms. It is also healthy to keep in clear focus the opportunities and benefits that come from the use of these platforms. If one was to do a vox pop asking people about the most exciting things they have in their lives today, the answers would be things associated with the use of new technology and the platforms and content that they are able to access through that.

In terms of the wider economy, the improved choice that you get from using online platforms, the improved connectivity and the ease of making comparisons between prices are all very positive, generally speaking, for the performance of the economy and the accountability of firms to consumers. We also see a terrific amount of innovation in this area. It is absolutely right and wise for the Committee to focus on the potential downsides and problems, but I hope that you will also keep in clear sight the benefits that are coming.

It is also fair to say that a lot of the harms that have been identified at this stage are theoretical harms; they are about things that might go wrong in the future. When pressed, most people are unable to come up with very much evidence of things that are not working out right at the moment. These have certainly been the findings from our work.

To turn to your specific questions, you asked first about a useful definition of online platforms. There are certainly some common elements within them. In particular, they are, broadly speaking, trying to facilitate and connect different groups of users, and obviously doing that in an online space. More than that, that definition does not get us very far, because there are lots of other ways to connect different groups of users in the physical world: for example, supermarkets and cafés are also opportunities for manufacturers and ordinary customers to connect. So in looking at online platforms, people have discerned the quality of it being a two-sided market—you are selling on both sides of that market. Again, we do not see that as being especially unique. In fact, overall, the more we thought about digital markets and platforms, the more diversity we have seen in the business models that are out there and the kinds of policy issues that they give rise to. For each online platform, what their customer proposition is, who their competitors are, how they are adding value, what their money-making model is, pricing and quality issues, privacy and security will vary according to the type of platform. We are very struck by the variety. For that reason, we are doubtful as to whether there is a single definition that opens up a very powerful analytical framework to say that anything within that definition needs to be treated in this way, as opposed to anything outside the definition, which can escape that treatment.

Certainly from my perspective, we try to be very evidence-based and operate on a case-by-case basis. We have not found it useful to say that within online problems there is a whole class of economic problems that we can usefully group together. So we focus on digital markets because they are a very important area of the economy, with great potential for consumers, for productivity and growth, and other things that matter. But we have not tried to say that online platforms are all in a special class of their own and can be treated in the same way. We see diversity there.

To begin to tackle your second question, it is a big one and I will call on my colleagues to help me. First, you asked where dominance comes from. Dominance does not come from size alone. The ability to be able to add customers and get to a certain level of market share does not itself lead to a presumption that you have done a bad thing. Indeed, often it means that you have done a very good thing and that you have satisfied those customers very well. From a competition analysis point of view, big is not bad. We look at what potential it gives companies to be able to act independently. Therefore, to attempt to give you a definition of market dominance, I would say that it is more about one’s ability to act without reference to one’s competitors or customers; to be able to dictate terms, so to speak. That is quite a high standard for a firm to be able to reach, and obviously there have been cases where market dominance has been found. But certainly in the European competition arena it is fact-based and based on patterns of behaviour and evidence of being able to act without those constraints that I spoke of.

In terms of where potential sources of dominance could lie in relation to online platforms, one very important feature, which you have highlighted in your own work and in your questions to us, is that these are two-sided markets. As such, they exhibit network effects, meaning that the more users use them, the more valuable they become. Potentially, this can create barriers to entry. It can also lead to a situation where in some markets there is a tipping phenomenon, whereby more and more people use it until it seems almost pointless to use any other platform because there is so much value in that. If, for example, your ability as a seller to be able to reach very large numbers of potential purchasers is so great on one platform, or, in a social media context, the chance to be able to connect to other people, why would you consider other platforms? There is that risk of the markets concentrating excessively in that way and, from the consumer perspective, of people saying, “I don’t need to go anywhere else. I can just ‘single home’ ”, which is the jargon for using one platform for all one’s needs. We have to be very alert to that, because if people ever felt that one company were providing all their needs in this very large category, it would mean that that company would be able to act without restraint. We are probably not in that situation at the moment. There is quite a lot of dynamic competition, which suggests that the barriers to entry are not as high as is sometimes felt. If you think about what it takes to provide a digital business, there has probably never been a better time to get hold of the capital that you need to do that, the technology and the people with relevant skills. All those elements are available, and the evidence to date from the past several years is that some of these firms get a strong market position for a time but do not manage to hold on to them. Certainly we have seen cases of that, such as Bebo, which you may remember from a few years back. It no longer has a strong market position, and there are other examples like that.

We should not assume that there is something inherent in online platforms that leads inexorably to dominance, but we should be mindful that there is the potential in that regard. That is why it is very important as a competition authority that we are watchful and work very hard to try to understand what is going on in the relevant markets that we study.

Q44 Lord Mawson: We have heard that there is fierce competition between online platforms, but we remain sceptical that the biggest ones could be dislodged at any moment. To what extent is there dynamic competition in the biggest platforms’ core markets?

Alex Chisholm: We see a lot of evidence of dynamic competition, in that the companies that are out in the market are continuing to innovate a lot. Also, new companies have come up in the last few years and achieved high market share and, indeed, high financial values in the markets. We also see a tendency whereby firms might start in one particular area and then move across from that area to be able to compete more effectively across the piece. Let us say that firms start in a core area, such as search, social media, book and music sales, and then look to extend their offer to other potential customers and suppliers through wider e-commerce opportunities. That has meant that these firms, which started off relatively insulated from each other and with their own particular specialisation, are now increasingly bumping into each other, that is to say competing with each other.

A particular example is entertainment content, which I am sure you will find over the course of this inquiry is an increasingly important ingredient for many of these online platforms. Many of them are using entertainment content as a way to differentiate themselves and create some increasingly distinctive offer to their customers and to make their customers stay with them for longer, so it is a device for retention. That has meant that content that you might have associated historically with broadcasters and telecommunications businesses or publishers is now very often a big feature of these e-commerce offers, and I think you will see more and more of that. That means that these large digital companies are directly competing with each other for eyeballs as well as for users and purchasers, and it is important to recognise that they provide a strong restraint on each other and that if you are running, say, digital giant A, probably the thing that keeps you awake at night the most is what digital giant B might do next week, as well, obviously, as what your customers think about your offer.

In short, we see a lot of potential for, and evidence of, dynamic competition, and in looking at whether that relatively benign story can get interrupted we then have to look very carefully at the conditions in a particular market and ask whether there is evidence that they are continuing to grow because they are continuing to innovate and to do a good job for customers, or whether they are trying to hold on to their position by putting in place artificial restrictions on suppliers to prevent them moving elsewhere or trying to lock their customers into them so they are unable to switch their business elsewhere. Those are the types of problems where we need to be alert as regulators to intervene if we need to, but I do not think we should assume that that is an automatic outcome.

Lord Mawson: Say, for example, that there are very large platforms buying up small and very innovative things that might fundamentally challenge it. What research would you do? What access would you have to see whether these very innovative products are being lost from sight?

Alex Chisholm: From a merger control point of view, there have been cases where we have been able to assess that. Nelson, do you want to talk about Facebook and WhatsApp?

Nelson Jung: Yes, sure. There have been a few cases. One that we looked at as the OFT, the CMA’s predecessor, was Google Waze, which in a sense exhibits exactly the sort of concerns that one might have about one digital giant potentially sweeping up smaller competitors, which at that stage do not generate a lot of revenue but might still impact on the marketplace overall as a result of innovative business models that they bring to the table. One what we call theory of harm that we looked at there was the extent to which that merger would result in reducing Google’s incentives to innovate in that space. Ultimately, the evidence that we looked for was what the internal strategic business documents told us, what plans those companies had, how important Waze was from Google’s perspective at that point in time in terms of developing its own products’ enhanced features for consumers. That was a case that we felt comfortable clearing at the time, but that is indeed one of the common theories of harm that we look at in the context of these digital platforms. I think it is fair to say that the UK merger control regime is particularly helpful compared with other merger control regimes. The EU merger regulation, for instance, looks at mergers only if the parties to the transaction reach very high turnover thresholds. That is not the case in the UK, where we can also examine transactions just on the basis of what we call the share of supply test: one of these smaller innovators might not yet generate large revenues but still have a share of supply that would add, in this case, to what Google had already established at that point in time. It enables us to look at those mergers and assess whether there is harm that goes above and beyond pricing and also looks at quality and innovation.

Alex Chisholm: That can also be the case at the European level. When Facebook took over WhatsApp, that went through scrutiny not only in the US but in Europe and other parts of the world, and that was a company with a lot of users but at the time very low revenues and low profits. Certainly the analysis looked at whether that could diminish competition in the future, and that is a standard part of competition assessment.

Q45 Lord Rees of Ludlow: Platforms tend to take a long-term view and to prioritise building up their user base rather than short-term profits—

Alex Chisholm: Yes.

Lord Rees of Ludlow: —and that might mean, therefore, that they establish complete dominance before there is any intervention, which might then be too late. Is that a concern?

Alex Chisholm: It is a potential concern, but we have to look at the counterfactual. If you were so worried about what the end game might be that you intervened at an earlier stage in order to avert that potential harm, would that be a cost-free choice? The answer, I think, is that it is not. It is very likely that if you act in advance of harm you are going to risk reducing consumer welfare and chilling innovation.

Furthermore, the solution—we would say the remedy—that you would put in place would prevent other things from happening which you will therefore not see by that process. When the market is in a very dynamic phase, new things are happening the whole time and consumers are continuing to express themselves, sometimes in quite surprising ways. To interrupt that process brings risks with it.

I will turn to my colleague, Daniel Gordon, to expand my answer a little.

Daniel Gordon: I think there is a distinction to be drawn between growth by acquisition, which Nelson has just talked about in the context of merger control; and organic growth. If we are looking at organic growth, companies, by exploiting network benefits, economies of scale and economies of scope, will be growing on a proposition that serves consumers, and to intervene in the course of such a proposition runs the major risk not only of preventing those potential benefits from reaching consumers from further growth but of dampening the longer-term incentives to innovate, which are key to dynamic competition in this market. So, generally, we pause for a long time before thinking of intervening at that point.

A more common phenomenon that occurs in online markets is that because of these economies of scale and networks they can tend towards concentration, so you look more at competition for the market as a mechanism more than competition within the market. It is the competition to replace that is the dynamic incentive. We are therefore most concerned about entrenched positions of the sort we mentioned earlier, such as whether there are entry barriers that cannot be surmounted. Those are the sorts of things that you would focus on. Early intervention is dangerous.

On the question of whether that intervention is left too late while you focus on those entry barriers, competition law is well positioned in respect of that, because not only is it ex post but it comes with serious potential fines that are intended and act as a deterrent. Companies are well aware of that. They know that if they get to that position, they really do run those risks.

The Chairman: Lord Green, your question is on related matters.

Q46 Lord Green of Hurstpierpoint: I want to pursue the point a bit further. If the network effects and economies of scale are powerful, resulting in what might be described as a natural monopoly in some of these activities, there would obviously, theoretically, be the risk of customers being disadvantaged, even when the service is provided for free. That could be, for example, by degrading the quality of service over time once a company is in that monopoly supplier position. What is the answer to this? Is it to wait and see and test by the normal competition considerations that you have outlined? Or is there a case, as has been suggested to us in some of our other hearings, for requiring at least the major platforms that are deemed to be in this position to be more transparent about their operating strategies and procedures? Is that a means of making it clear to customers and competition authorities what is going on, and thus alerting you earlier in the game to the potential risks?

Alex Chisholm: Thank you very much. First, you talked about building up a natural monopoly. Many of us think about natural monopolies based on typical analogue-type situations. Take, for example, a gas distribution network, for which the idea of a rival set of pipes being built over the same ground would be inherently implausible, or similarly a port, providing access to a small island. I think that when one speaks about online platforms, a natural monopoly is not quite as complete as that.

You also mentioned network effects and economies of scale. Although the network effects are pronounced, they are not complete because many users are free to go elsewhere. For example, when using a search engine, if you decide, even for a moment, that you have had enough and want to use another one, it is very easy to do that with just one click—you are not bound to it. That is on the demand side. On the supply side, although there are undoubtedly economies of scale, it is also quite easy to scale-up a digital business, much easier than it would have been in the analogue or physical world. It is not as though you have to build a new semiconductor plant or something like that. We have seen that with the success of businesses such as WhatsApp. In the space of only a few months it has been able to acquire 200 employees, and build a business valued at 17 billion. It is quite unlike the physical world that came before it. That is my first point.

Secondly, I very much agree with you, Lord Green, about transparency being very important, both for maintaining consumer trust and building public understanding, and for giving regulators, including us, a real understanding about the way in which the businesses operate.

Lord Green of Hurstpierpoint: If I may press this just slightly, this is not just about transparency in pricing structures and the way in which they impose, if that is the right term, on suppliers using their platforms. It is about their operating and business model, and the way in which they use the network effects that accrue to them because of their position to make the next acquisition, or whatever it might be.

Alex Chisholm: Yes, I think that is exactly right. An example from our own work at the CMA is when we were dealing with payday loans. Some of the payday loan websites made it look as though you were going to a place where you would get the best value loan—in other words, the lowest cost loan—that could exist. But actually, when you entered your details and your requirements, those were auctioned off to the highest bidder on the other side to have the chance to sell you this very expensive loan. Following the intervention of the CMA, that practice has stopped and it is much clearer that price comparison websites are doing the job they are supposed to be doing, as opposed to being lead-generator sites. That is an example of where the transparency of the business model to the user is very important in a market like that.

Nelson Jung: Can I just make one legal observation on the framework as a whole, because your question is also about the extent to which a degradation of quality could raise concerns—and it could. We have looked at our legal framework and the tools that we have. We have Article 102 of the Treaty on the Functioning of the European Union, which talks about the abuse of a dominant position. That is a very flexible tool in that it is a non-exhaustive list of different types of abuses that we can challenge by virtue of using that tool. Recently, there was a case in which making misleading statements in front of national patent authorities resulting in an extension of the patent was deemed to be an abuse of a dominant position. Of course, that is not in the treaty, but regulators such as us or the European Commission can use that provision to tackle that harm. The reason why it is important to look at harm specifically on a case-by-case basis is that although these network effects may give rise to harm, they may also give rise to beneficial effects for consumers. We had a newspaper case involving the traditional two-sided market or platform—advertisers on one side and readers on the other. In that case, the existence of these network effects was a strong contributing factor to us clearing that newspaper merger. We did not think it was plausible that the newspaper would chose to downgrade the quality of its editorial contributions because it would lose readers, as a result of which it would struggle to gain advertisers. That is also important to bear in mind and is therefore behind our drive to ensure that we look at the specific evidence for each case.

Lord Rees of Ludlow: Can I infer that you gentlemen have no sympathy with Peter Thiel’s argument in a Wall Street Journal article that a near monopoly is better, because then the company can make enough profits to do more innovation than it could if it was in a more competitive market?

Daniel Gordon: There are two famously competing schools of thought on this that, if we drill down into them, are not inconsistent. One is that what a monopolist has to lose is greater, so they have a greater incentive to innovate to maintain their position. That is about the ability to appropriate the benefits of that innovation. That points in one direction. The other is that competition spurs on innovation because you want to separate yourself from the pack, so you will win and gain the greater benefit. There have always been these competing thoughts. As I said, when you drill down into them, they are totally consistent, but you have to look at the evidence and the empirics. They generally show that on most measures, higher levels of competition, which are always a hard thing to measure, correlate with a greater degree of innovation. It can tail off—

Lord Rees of Ludlow: Greater in the sector, not just generally.

Daniel Gordon: Generally, competition and innovation tend to go hand in hand.

Alex Chisholm: There is some evidence in digital markets of the famous upturned U. It suggests that at both ends—in other words, with too much competition and too little competition—you do not get it right and you do not get innovation. That is an argument that gives some academic support to some of what Thiel has said, but I think that some of his other arguments go too far.

Q47 Lord Freeman: We have been told that there has been a race almost to the bottom in the sense that the very large platforms have been seeking to add as much information and as many customers as possible. To protect the concept of a sensible degree of competition, is there an argument for intervention in setting standards for how much information can be collected from individuals or companies?

Alex Chisholm: First, there is a clear argument for that. Data protection is, in all jurisdictions, a recognised area for regulation. It is necessary for both consumers and companies to know the rules of the road. So, yes, it is uncontroversial to say that. Having said that, the precise levels of privacy require very careful judgment, and continue to evolve in our view. There is no single standard of privacy. Having spent time in Europe, I know that different countries seem to have different concerns about privacy. I have just come back from Germany, where they are very concerned about privacy in a different way from the way the debate works in this country. We also see a lot of variety in consumer views about privacy. Certainly regulators, policymakers and, I am sure, concerned parliamentarians tend to think that consumers need good protection for their privacy. Yet if you look at the behaviour of consumers online, very often when given a choice between a bit more privacy and a bit more convenience, it is convenience that is chosen. That is also the case when it comes to security. A consumer might say ‘I could have a bit more security or a bit more convenience. I will take the convenience, thanks very much.’ That is a bit of a caution to all of us when trying to capture precisely a privacy and data protection framework that would command wide assent on behalf of consumers, who, after all, we are trying to serve. Many of those consumers themselves might place those protections in a slightly different place from where our first instincts would be.

That is one of the reasons why, in the work the CMA did on the commercial use of consumer data, we came across a number of paradoxes, such as that although there was quite a lot of consumer concern about what happened to the data, their behaviour did not show very much concern. I think that is the reality of where we are in these markets. There are also many tools that you can use to protect yourself with—browser settings, privacy dashboards, opt-out tools, ad-blocking software and what have you—but relatively few consumers have explored that potential, suggesting that it is not a very large concern for people. Our view is that we want there to be clarity regarding the regulatory requirements, because obviously that is important for trade and confidence. We would also like companies to take as much responsibility as possible for keeping their consumers onside—in their terms and conditions, privacy policies and cookie notices—not just in a formal, legalistic way, but in an active conversation with consumers to the effect, “This is what we are doing. Is this okay with you?”. That is the way to think about this—as an ongoing dialogue.

We also think it is important that consumers and firms should share in the benefits of consumer data. Consumer data is sort of like a currency in digital markets, and if all the benefits are going to the supply side, it seems a very asymmetrical situation. We want consumers to have a lot of clarity from transparency and control over what happens to their data and its collection. We see some signs that companies are competing to some extent on how best to reassure consumers about taking care of their privacy concerns. I do not think there is a very clear trend of a race to the bottom. We want to encourage a higher level of transparency and more options for control to encourage the opposite of that—a race to the top, with firms competing with each other about who can best maintain the trust of consumers in the online space. The fact that that trust, when broken, is very hard to reclaim is a significant incentive for the big firms to get that right, and to continue to get it right and preserve that consumer trust.

Daniel Gordon: Perhaps I may add to that. The importance of trust in these markets is absolutely critical, particularly given that we have wholly new products coming online. We need to separate out genuine issues of trust, in the sense that the consumer is not getting what they bargained for and expect, from just the sheer, rapid progress and the confusion and mistrust that can come naturally as a result of that. We need to be very clear about the real trust issues as opposed to perceived trust issues. Sometimes our assumptions as an authority and as public servants might not be totally congruent with the real issues. We need to test ourselves on that.

The Chairman: On another issue, Lord German.

Q48 Lord German: Thank you. I want to raise the issue of price parity clauses, and I have several questions. We have received lots of evidence, and people have been asking questions, about the anti-competitive nature of these clauses, and some people have said, “Thank goodness we have the CMA, which can deal with it”. What conditions do you think you should examine in order to identify whether price parity clauses are anti-competitive? What benchmarks would you set? Given that Germany and France have just outlawed certain aspects of price parity clauses and the fact that the platforms are across different nations, do you think that this matter is dealt with more effectively at a European level, or should it be dealt with, and can it be dealt with effectively, within the UK?

Alex Chisholm: I will give the short answer and Nelson can expand on it, if that is okay. First, price parity clauses are potentially anti-competitive—we recognise that. Secondly, the CMA has been very active in tackling these in specific situations where we have seen harm arising. We do that both in our own right, where it applies in the UK to British markets, but also in co-operation with other authorities in Europe where necessary. The short answer is that we are ‘on it’, and I will ask Nelson to give the longer version.

Nelson Jung: We have been looking at these provisions for some time now, the reason being that in certain circumstances they threaten to undermine some of the benefits that the internet has brought about for consumers: that is, the ability, at low cost, to search and shop around to find a better deal. Although these provisions are not new—they have been around for decades—we have found them to be increasingly prevalent in online settings. We looked at whether we should be concerned about that and the circumstances in which it is the case. By way of background, these price parity provisions are sometimes called retail most favoured nation clauses. That is because the online platform imposing that particular provision on the supplier effectively obliges the supplier to ensure that no other online platform, and indeed that supplier itself through its own sales channel, is able to offer that good or service at a lower cost than that offered by the entity imposing that provision is offering it at the retail level. In one sense, it ossifies the retail price at a given level and eliminates price competition.

Why would we be concerned about them? One reason is that we thought they might soften competition between those online platforms. One example would be that they might disincentivise online platforms from competing over the commission fees they charge those suppliers. They might therefore also facilitate collusion between those online platforms. They might create a focal point on the fees or commission levels that they charge suppliers—ultimately, potentially again, resulting in artificially inflated retail prices for consumers. We also think that they are able to restrict entry by potentially more innovative online platforms that cannot compete on price where that is an important parameter of competition. They might also facilitate collusion between suppliers because, of course, the supplier’s ability to vary retail prices across different sales channels is limited if not eradicated as a whole. Lastly—and this applies specifically to what we consider to be wide retail most favoured nation clauses, which apply not just to the supplier’s direct sales channel but to other online platforms—there is an element here of horizontal collusion between them. I have heard people talk about them almost as if, in one sense, they were exhibiting almost cartel-type behaviour that could give rise to concerns.

From our perspective, when we look at the market conditions, it is important to assess each market and each online platform in its own right, based on the evidence available. We would look, for instance, at the extent to which that online platform exercises market power. We would want to see whether, when there is not one large platform, there might still be a network of these types of arrangements that ultimately cover a wide proportion of the market in question. We would want to examine the extent to which consumers are still able to ‘multi-home’, to use the jargon—to switch to other platforms—or whether there are other ways in which that has been made more difficult, for instance by virtue of contractual restrictions.

We also want to see how important price is as a competitive parameter in the market that we are looking at. I can give you a few examples of where we have looked at this more specifically. On private motor insurance, for instance, we ultimately concluded that these wide MFNs were harmful, so we ordered them to be banned. We saw no justification for the restrictions that that would impose on the other online platforms, and, ultimately, precisely the harms that I outlined earlier manifested themselves in those wide MFN settings. However, on the narrow MFNs—where the provisions relate just to the extent to which the supplier is able to compete on price with the entity imposing the retail MFN provision—we found no evidence that they would be sufficiently harmful for us to ban them. That is another example of why we think it is particularly important to look at each market in its own right. Another case is Amazon Marketplace. We took enforcement action a couple of years ago and Amazon unilaterally withdrew these provisions shortly after we commenced our formal investigation. Of course, we are now also contributing to investigations that take place at EU level. You will have seen that the European Commission is now investigating these types of provisions, as it is formally investigating Amazon’s e-book distribution arrangements. We also previously looked at e-books when Apple entered the e-book space. We worked very closely with the European Commission and ultimately assisted it in achieving a commitments outcome that was pursued with the Department of Justice and the European Commission.

Lord German: So your answer to the question whether this matter is dealt with more effectively at a European level or at a UK level is basically, “We’ll work with our colleagues”.

Alex Chisholm: It depends. The private motor insurance example was a purely UK market investigation. The way in which the online motor insurance market works is different in different countries in Europe, as is motor insurance itself. That was very much a UK-focused case and its resolution, by orders that came into effect in April this year, was purely in the UK. Amazon obviously has a number of Europe-wide policies. It is very active in a number of markets in Europe. This is not the only English-language market in Europe, because of Ireland in particular, but the restrictions that the Commission is concerned with—and we share its concerns—about the contracts with publishers apply in the German-language market as well as in some other markets in Europe. If there is a Community dimension and the policy has been applied in a number of different countries in Europe, there is a strong argument for that to be taken at the European Commission level, so we agree that it is correct that the Commission has taken that case.

Q49 Lord Wei: A number of platforms have been using different methods to lock in consumers, whether through getting consumers to put all their photos or playlists online, or purchased content with proprietary formats, such as books that are purchased or rented. Should you be working with the Commission to reduce this kind of lock-in and the costs associated with switching platforms, whether for consumers or suppliers, which in practice make it quite difficult for people to go to another competitor within a given vertical?

Alex Chisholm: Thank you very much for the question. It is a good question and one that we ask ourselves quite a lot in the particular cases that we deal with. I have a few reactions to it. First, although we use the language of “locking in” consumers, in many cases it is consumer demand that causes services to be provided in that way. You give the example of all photographs being available in one place. Sometimes people want that, as they find it very useful. That is also true with some social media. The convenience factor and demand factor for consumers need to be taken into the equation.

Also, overall within digital systems you can distinguish between open and closed systems. Although people typically think that open is good and closed is bad, when we have done a closer economic analysis of some of those things we have found that there are some good aspects of so-called closed ecosystems. There is a difficult [2] contrast with the Apple ecosystem, which is relatively closed compared to the Android system. The more open system has the advantage that it is easier for others to innovate on and no one is limited in the way they are in the Apple system. On the other hand, the ease and security of interoperation within the Apple system seems to bring benefits to consumers. We did a joint study on this with the French competition authority and found that, overall, open and closed systems each had benefits.

Thinking more about your question, I suppose that every business in the land is keen to hold on to its customers. They try to do that by adding more valuable things to that customer. Again, we need to think quite carefully about trying to prohibit that behaviour. When we think about the particular data types that the digital platforms hold about us, we need to ask whether it is information that they hold about us uniquely or whether it is information that other digital platforms would also have about us. That obviously makes a difference to the competition between platforms. And is it certainly ours, as consumers, or is it shared between us and the service providers? If, as a kind of thought experiment, you think about getting your phone record from your service provider, you would not think twice about your right to be able to ask for and get it. You would think, “It’s my data and I should be able to get it”. That, I am sure, is the case with your bank account details—you would feel that you had the right to ask your bank for the record of your bank transactions, because they are your transactions and your data.

The example of an online marketplace, where you have built up an online reputation, is not quite so straightforward, because your online reputation and profile are not just your data; that data reflects a lot of other user feedback on you that has been generated solely through that marketplace and by others, not just you, so the sense that that is your data to be able to take somewhere else becomes more arguable. These few points are about saying that we should not simply say, “It’s my data and I should be able to take it anywhere else as an absolute rule”. We need to look at whether it is causing problems in the particular circumstance. If it is, there is the potential to intervene as a competition authority and, indeed, to uphold any consumer rights that are not being protected, given that we are a joint competition and consumer agency. Again, we also need to look at the phenomenon that I described earlier whereby online marketplaces are very susceptible to consumer reactions where they lose trust. If it was felt that an online marketplace was abusing its privilege by putting a lot of artificial constraints in place, making it very difficult for consumers to exercise their freedom to shop elsewhere, that would obviously be very damaging for its reputation and it would suffer a consequence in the market.

The Chairman: We have a couple of questions relating to the regulators themselves.

Q50 Lord Aberdare: One of the messages for me from today’s session has been that in this very fast-moving, dynamic and innovative market, there is a preference for an ex-post competition-based approach to management, rather than ex-ante regulation. There are two questions about that. Do you feel that you have all the tools and resources that you need to be able to do that job fully, and/or is there a role for a degree of ex-ante regulation to supplement or underpin that?

Alex Chisholm: Thank you very much. First, when people debate online platforms, there are often lots of different issues involved. Those include privacy, data protection, intellectual property, and competition and consumers. We are really only speaking to the competition and consumer aspects. As I said in my speech to the Bundesnetzagentur last week, and I have said it elsewhere as well, we have a preference for ex-post analysis—after the event—in dealing with the economic analysis. I absolutely accept that certain fundamental rights should be protected up front in relation to things such as privacy and data protection. We should, if you like, be able to take that for granted as the regulatory framework, and within that be able to treat firms as they behaved, based on that ex-post analysis.

On the tools that we have, probably the cautious answer is, “I think so”. As we have talked about, with some of the digital cases that we have already in our portfolio, we feel that with our competition and consumer powers we have quite wide-ranging tools to be able to deal with problems where they emerge. We also have, pretty uniquely in the UK, the market investigation regime. That proved very useful when dealing with the issues that Nelson mentioned earlier relating to private motor insurance.

But it is also very important to recognise that, as you say, the market is very fast-changing and we need to up our game and continue to update our practice to be able to deal with the challenges ahead. One big way in which we try to do that is by engaging, as you have probably realised today, very actively with what is happening in these markets to stay up to date with commercial practices, consumer concerns, the best academic literature and what is happening in other countries around the world. We publish a lot of research studies in this area.

We also need to make sure that the competition regime can react quickly enough to situations. We do not want situations in which we have very long-running cases, for many years, getting tied up in litigation, while in the meantime the harm is being done. We need to be ready, where justified, to take speedy action through so-called interim measures or in securing voluntary behavioural changes through commitments, where necessary.

We also need to be able to update our thinking. Some traditional anti-trust concepts, such as essential facilities, probably do not work so well when thinking about these digital markets, as I tried to explain earlier. We also need to take the opportunity to recognise that online markets are very data rich. Therefore, it is much more practical to test a remedy in that market to see whether, by presenting information in this way, it will deal with the problem that we are trying to deal with than it would have been in some physical markets. That gives us great potentiality to be able to update our practice.

Two other areas are particularly important when making sure that competition regimes are fit for purpose for the challenges of digital markets. The first issue is that in the large ecosystems, many of the people participating in the market, and who might themselves suffer competition abuse, are so dependent on one of the platforms that they might not want to come to us with that complaint for fear of retaliation. They may think by the time we built up our case, gathered the evidence and perhaps gone to court, they would have suffered that retaliation effect. So we need to give some thought as to whether there is a need for wholesale [3] protections for complainants in that situation.

The other area, which I think has already been brought up, is the need for international joint working in these areas. It is clearly not satisfactory to have dozens of different parallel investigations for the same underlying activity. Nor is it good in the interests of the internal market to have different solutions being developed in different countries in Europe.

The Chairman: Thanks very much. I am going to have to make this the last question.

Q51 Baroness Donaghy: There have been suggestions from some tech firms and some politicians in the United States that competition enforcement at EU level is sometimes used to pursue other aims. One respondent to our call for evidence suggested that because the Competition Commissioner is part of the college of Commissioners, she is subject to political pressure and collegiality. Do you share these concerns, and is a sharper separation of powers required?

Alex Chisholm: I should say first that I am not here to speak for the European Commissioner. Having said that, neither I nor my colleagues have any reason to doubt her objectivity or independence, which her record gives strong testament to. I watched the evidence that she gave to the European Parliament hearing at the time of her appointment. She was very firm then on her understanding of the need for the Competition Commissioner to exercise independent judgment on cases. She has taken a number of opportunities to reassert that since then, most recently in a speech just two weeks ago, emphasising that obviously as a commissioner, she needs to look at the overall objectives for the European Commission but that independence of decision-making in cases was “non-negotiable”.

I notice that when the announcement was made by the European Commissioner about the role of the various vice-presidents, there was a footnote. As you know, in European matters, footnotes are always tremendously important. This footnote asserts that there is an exception in that role for competition cases, where the Competition Commissioner’s judgment does not require the process of working through the vice-presidents on cases. It is very important to have that independence, because consumers have the right to expect decision-makers to do things that are in the interests of wider consumers and do not reflect national or industrial preferences.

I also think that we need to recognise that as the competition authority we have terrific powers to fine firms and potentially even to lock people up. Therefore, companies need to have confidence that what is being decided has been done in an objective, thorough, evidence-based fashion and that there is no secret agenda. All the evidence that I have seen to date suggests that the European Commission understands and accepts that. Indeed, in their review of Regulation 1/2003, which is the fundamental ordinance governing the way in which the European competition network works, they have gone out of their way to try to explore ways to further guarantee the independence of competition authorities to preserve their confidence [4].

The Chairman: Are all the national competition authorities on the same page in this arena?

Alex Chisholm: I would say so. I used to be a telecoms regulator and I have seen, in both telecoms and competition, that it is a tremendously joined-up system across Europe. That is particularly true in competition, because identical law is applied right through the European community. On that basis, the European Commission and national competition authorities have to apply that law. There is a sense of a collegiate purpose. We are constantly sharing cases with each other, trying to decide who should handle which case but also developing our process within that. It is very consistent and I am not particularly concerned about that. Indeed, the approach to competition and consumers that has been followed over the last several decades in Europe has been pretty robust and very much in the interests of consumers.

The Chairman: Thank you very much. I will have to draw this session to a close. You have been very forthcoming. Thanks for that, and for your written evidence. If there is anything else you think we should know, please either say so now or write to us. Meanwhile, watch this space. We are going to Brussels next week, where we will get the view not only of the Commission but of the French and German authorities.

Alex Chisholm: Thank you very much.

The Chairman: Thank you very much indeed.

[1] Richard Feasey, Specialist Advisor to the Committee on this inquiry is acquainted with Alex Chisholm.