 Good afternoon to all of you. It's very nice to see you all again after the summer's break and a very warm welcome to today's webinar on enforcing the Digital Markets Act. My name is Joyce O'Connor and I chair the digital group here at the IIEA. I'm delighted to introduce our distinguished speaker today, Thomas Kremler. A very warm welcome to you Thomas and thank you very much for being with us today and for taking time out of your busy schedule and we look forward to your presentation. Thomas Kremler is head of the unit of e-commerce and the data economy in the European Commission's directorate general for competition. He was previously head of the digital service market task force responsible for the e-commerce sector inquiry. Thomas also served as deputy head of the unit for antitrust in the information industries, internet and consumers, electronic sectors. Before joining the European Commission, he worked as an agent representing the Austrian government before the European courts. Thomas presentation will take around 20 to 25 minutes and then I will go to you the audience for questions and answers and you can use the Q&A function as you know at the bottom of your screen. As is usual, today's presentation and Q&A is on the record and I'd appreciate very much if you give your name and affiliation when you're asking questions and thank you very much for doing that. Please join our discussion on Twitter using the handle at IIEA. There is a lot of interest and in and discussion about the Digital Markets Act. Today's webinar is very timely as Thomas Kremler is talking to about us about enforcing the Digital Markets Act. Thomas will discuss its main features and what the European Commission hopes to achieve with the DMA. He will also elaborate on how the Commission is preparing for the enforcement of the DMA and its strategy to ensure that the DMA is effectively enforced and how it will prioritize the different areas. He will also discuss the challenges the Commission expects to face in relation to enforcement and how it plans to address such challenges. Finally, Thomas will discuss how third party stakeholders in the digital economy can contribute to successful enforcement of the DMA. Thomas is one of the two antitrust officials to head up the new EU directorate set up to enforce the Digital Markets Act. He is highly placed to give leadership in this area. We are privileged to have you here today, Thomas, at such a key time. So it's over to you and we look forward to your presentation. Thanks very much, Joyce, for this intro. Now the expectations are very high. I will hope I can fulfill all the high expectations that you set. You said is very correct. This is a very timely moment to talk about the Digital Markets Act. It has been signed, co-signed by the co-legislators, the parliament and the council last week. That means that it will be published in the official journal in the next couple of days and will then enter into force around mid-October and then the deadlines will start. Before looking into the concrete enforcement challenges and the concrete enforcement issues that lie before us, not only us as the regulator, but also those who will be the beneficiaries of the DMA and those who will be at the receiving end, if I may say so, of the DMA, the potential gatekeepers. I want to take a look back actually to the genesis of the DMA. So I want to talk a bit about the why. How did we actually arrive where we are now? Why do we have the DMA in the first place? Then a bit about the how. How is the DMA going to work? What are its main features? And then at the end a bit about the when. How is this going to unfold? So let me first start with the why. So why did we get there? That needs a bit of looking back into let's say 20 years of enforcement into unilateral conduct cases in the EU. I think there's many lessons that we can draw from that. We have started enforcement in platform markets back in the 2000s with the Microsoft cases where we looked into tying and interoperability issues. This has then gone on in the Google shopping cases, which has been confirmed by the court recently into issues of self, so called self-preferencing. So where a platform actually favors its own products on the platform to the detriment of rivals. We have in the Intel and Qualcomm cases looked into exclusivity rebates along running saga, which is still ongoing after 20 years and bouncing up and back in courts, especially when it comes to the Intel case. And you've just recently won, if I may say so, the Android decision in court where the court looked into Google's behavior in relation to mobile, its mobile ecosystem. Now, concretely, we are turning in terms of concrete enforcement issues we are turning to look into data related cases more and more. We have an Amazon case pending where we're just on the antitrust side where we're just consulting on potential commitments. So Amazon would basically silo separate its data sets between the marketplace and the retail. We have a retail business, we have a Facebook case that is about data use and we have a Google case about ad tech, which is also very prominent in terms of data use. All of this to basically set the scene to draw the lessons of what lessons can actually be drawn from all of these cases and from all of the experience that we have in the EU with enforcement of antitrust rules and tech markets. I think one of the lessons to be drawn that the case by case approach that we have followed in the last 20 years has definitely an advantage as it allows for very, very deep diving into the specific markets, getting to the bottom of the market realities. That's very important in the antitrust world and it's important that we get it right. But what it leads to, there's a nice long German word in my native tongue, it's very good for what is called in German Einzelfeiergerechtigkeit, which translates into individual justice in each and every case. So antitrust is very good at that, at achieving that. It's also good if remedies need to be tailored to a very specific conduct or very specific abuse, but there's also drawbacks at that. Now I'm coming to the DMA. The drawbacks are that these procedures take very long. That sometimes when we finish an antitrust case, the market has actually moved on and the remedies that we have imposed might actually come too late in some of the markets. For instance, because a market has dipped and a dominant position can no longer be challenged because of that. The remedies that we can impose in antitrust cases as it is an abuse control system as you know in the European Union needs to be focused on the concrete abuse. So we cannot go beyond the abuse that limits us in the kinds of remedies that we can basically impose on companies. So antitrust is really a difficult tool and it's not impossible, I would say, but it's difficult to address with the current antitrust tools that we have, systemic issues. For instance, if a market is going to tip, not because certain behavior of a company, but because of the market structure, think about network effects. If a market is prone to direct or indirect network effects and people go to a platform because other people are there, that's more market structure type of thing that's not sometimes not related to conduct. That's very difficult to tackle with an abuse control system. Another example is leveraging. If a company is very strong in one market and then leverages this market power in another market. That's, as we know now from from the Google shopping judgment in and of itself, not prohibited or not abusive, but it can be tackled under antitrust. The problem that arises when this behavior covers not only one market but five markets or more. In terms of an antitrust case you'd have to basically investigate all these five markets in parallel that have to establish that there are foreclosure effects. This is time consuming. This is not really reflecting the market reality of ecosystems and development strategies where companies, gatekeeping companies can leverage from, from one strong market into many different markets. It's very hard to deal with that under the current, the current antitrust tools. Also data access issues. They're really hard to tackle under under the current antitrust tools because the threshold to give access to data is relatively high in the EU. So all of this has led to a lot of policy policy reflections in the world basically not just in the European Union many reports have been written in the UK in the US in Japan in Australia. And all of those reports come to the conclusion that antitrust as it stands needs to be complemented by ex ante rules which are basically prescribing or imposing obligations on companies with substantial market power, strategic market status, bottleneck power, or you name it as many names that have been found for these companies, the DMA uses the name gatekeeper to impose on these companies, actually ex ante, so not ex post after the conduct has happened but upfront before conduct has happened obligations that they have to comply with. So I think there's consensus around the world now that this complimentary powers are actually needed. Now you can have two models of how to implement that. And one model is the DMA model which is largely I would say a legislative driven model, but the legislative has taken the big decisions, the big trade offs have been made by the the legislator has basically struck the balance between flexibility, efficiency, and legal certainty, and has made this to an innovation I shouldn't forget, and where the legislator has made this trade offs already in formulating the specific obligations that are going to be imposed on gatekeeping compromise. And another model is is a more regulated driven model where the legislator just works with broad brush clauses and basically entrust the regulator to come up with the right solutions. Both of these models have advantages and disadvantages the US deliberately chosen the first model for two reasons for constitutional reasons, because in the you obviously we work with the with the principle of conferral of powers. So do you only has the competences conferred to it by the treaties and the treaties in this case the legal base that we used is an internal market legal base, which means that we have to rely on the internal market rules for the DMA purposes. And an even more important reason is that we have opted for the system because of the drawbacks of the antitrust enforcement system the case by case system where we have to analyze conduct specific conduct and the basis of a case by case analysis which turned out to be difficult in this market so we have opted in the you for a legislator driven model. What does this now mean in practice. That means, first of all, which is quite important that there's no efficiency defense in the DMA. So you cannot justify your conduct as a gatekeeping company by saying wait, wait a moment what I'm doing here is actually very efficient pro competitive doesn't lead to the harm that you think it would lead to and therefore this conduct in a specific circumstances of my case is positive. That's positive for consumers. That is not a defense that is possible under the DMA, because the assumption is that the legislator has basically already when designing the rules, make these tradeoffs. So if the legislator has come to the conclusion, for instance, that it is on balance better that an app store provider opens up its system and provides a second app store. The assumption is that the legislator has basically weighed the innovation incentives for this app store provider against innovation incentives for the app developers and has basically made this tradeoff already in the law. Therefore, an efficiency defense is not necessary. So the DMA, as I said is an ex ante regulatory tool. It's not to competition law, but it has goals, which are partly similar to competition law, but not totally overlapping. So the two goals of the DMA are contestability, ensuring contestability of markets where gatekeepers are present and fairness. I will find a description of what is meant by contestability in recital 32 of the DMA. It's basically about the ability of companies that are reliant on gatekeepers to overcome barriers to entry and expansion so that they can also enter into the market with the gatekeepers present in very short words. So obviously the recital is a bit longer and more sophisticated but that's a summary. So fairness is a concept that is described as the imbalance between rights and obligations between the gatekeeping companies and their business users. So the contestability concept is obviously not alien to competition law. It comes from lowering barriers to entry, but it goes wider. So it's not for us to have non-conduct based remedies, which the DMA basically provides for. For instance, as I said before, access remedies irrespective of any conduct that the company has been carrying out, the DMA provides for, for instance, for data portability. So users who are able to take out the data from the gatekeeping platform and move them to competing platform that should ensure contestability of this market, markets, I should say. When it comes to fairness, fairness is actually a very old concept. You find it in competition law, but you find it in many laws that are actually older than competition law in the European Union in many unfair trading laws that date back to the 1900s. So fairness is not something alien to our legal systems and fairness between business users and bigger companies with more market power is not something totally alien to what we know from different laws. There is obviously a relationship between what competition law and antitrust law wants to achieve in terms of contestability and also in terms of fairness, but the DMA goes beyond the DMA. Tackles actually issues that are related to market structure that are independent of conduct of a specific company. So quickly, and I guess many of you are very familiar with the DMA so I'm not going to dwell on that very much. This is about the how the legislature has implemented that the legislature has basically defined a number of so-called platform services that includes storage engines, online intermediation services, messaging services, and if a company fulfills certain thresholds, which you find in the DMA, then this company will be designated for a specific service on the basis of quantitative criteria as a gatekeeper that will have to comply with all the obligations that DMA provides for. So if a company can rebut this presumption that the figures would actually lead it to have a gatekeeper status, but only in very, very limited circumstances that's on purpose, the legislature has said, if a company fulfills certain business user and end users thresholds is 45 million and 10,000 45 million end users 10,000 business users, then a company is deemed to be a gatekeeper for the specific service. If it's not deemed to be a gatekeeper or doesn't fall within the thresholds, the commission still has the possibility to qualitatively assess the position of this company and come to the conclusion that a company is a gatekeeper. As I said, I'm not going to bore you with all of these details on the DMA. I'm sure you're all familiar with it. You want to learn, I guess, more how it's going to be implemented in practice. So I think the law is less interesting, I would think. So the obligations therefore I'm not going to go into them very much. They're all set out in the articles five, six and seven. I think what is going to be really interesting is how they're going to be interpreted, and what are the challenges and all that we talk about that. I mean, let's acknowledge it. The DMA is uncharted territory for both us as regulator and for the potential gatekeeper but also for third parties who have an interest in in the enforcement of the DMA. So it's all, I don't want to say trial and error because that sounds as if we are doing we obviously have quite have our objectives in mind and the contestability and fairness objectives are very firmly set in the law, but it's testing. And this implies a lot of discussions with the gatekeepers who have to explain their business models to us who have to, and also third parties who have to explain the market realities to us so that we can apply the law in a way that makes sense in the market. One of the big challenges I think for a regulator in this kind of an enforcement system is the big informational symmetry between what we know as regulators about the market and what the gatekeepers know about their products about the functioning of the market. That's very difficult, at least in the first stage when you're on a very steep learning curve to bridge. So how can you bridge that I think there's a number of ways to bridge that first of all, we want to start with a collaborative enforcement approach as a starting point. So we're inviting the gatekeeping companies to explain to us how their business models look like what we would want to see in terms of compliance and build your open to a dialogue to find solutions which are compliant with the law at the same time, but also make sense from a business reality perspective for the gatekeeping companies. But if that doesn't work, then you'd also have to bring in the guns if I can say so. That means there is plenty of very tough enforcement possibilities that the DMB provides for. So we can go for periodic penalty payments relatively high fines as a last resort even structural remedies if a company doesn't want to comply with the DMA. But in the end, I think the starting point is that we want to try a collaborative method. You can also call it the carrot and stick approach if you want to and we will first try to discuss with the gatekeepers and only in a second stage resort to enforcement if necessary. So that's the first thing to bridge the informational cemetery. The second thing is bring in third parties bring in third parties so that we basically have, as I will call it a reality check on the markets and there's quite some tools in the DMA where we can bring in third parties formally speaking but there's also obviously the possibility to talk to parties informally about potential complaints potential compliance issues. We really want to have an open door policy where we bring in the parties early early on to help us in determining whether a gatekeeper is actually compliant with the obligations. And then we need to ramp up our own knowledge. We will do this in house. We will also do this with the help of external experts. So a lot of enforcement in the DMA is actually front loaded that means that a lot will happen in the in the first phase when gatekeepers will have to be designated under the formal quantitative designation process which is going to happen over the summer next year and then gatekeepers after designation of six months to comply and then compliance efforts and compliance will really get hot if I may say so beginning of 2024. This is when gatekeepers will have to comply with all of the obligations in one go, which for many of the gatekeepers already means that they have to prepare now in order to be ready for for the 2024 date. How in practice will the DMA interact with competition law. That's also a question that is often raised. So are we going to stop antitrust enforcement in digital markets. That's certainly not the plan at all. What obviously doesn't make a lot of sense is to replicate DMA procedures with parallel antitrust proceedings. That's not something that we're planning on, but the DMA is a very, or has a very limited set of obligations. It has its limits. It doesn't go beyond the obligations which are in the act or does an updating mechanism, but there are limits to what can be done under the DMA. So in essence, antitrust enforcement will fill the gaps between what is regulated under the DMA and what is not regulated under the DMA. So the DMA deals with, I may say, so the most egregious behavior of gatekeeping companies. And if new behavior, new conduct comes up or conduct which isn't fully covered by the DMA then antitrust enforcement can kick in. And as in the past, there's obviously the two instruments will run in parallel. That's not new. That is something that's done in telecommunications law and the Court of Justice has recently confirmed in a case that concerns the postal sector that there needs to be between the sectorial regulation which the DMA is basically and general antitrust that needs to be coordination between the two tools and the regulators that work on the two tools in order to avoid the issues of double jeopardy. And in the end, I will have to be very clear coordination between those who enforce the DMA, which is the Commission, but also the National Competition Authority, which go on enforcing their national competition laws, but also data protection regulators. Because as we know the DMA refers a lot to concepts of the GDPR and obviously these need to be interpreted uniformly. Just recently yesterday, there was an opinion by Advocate General Rantos in the German Meta case where he reiterated that it is very important that the data protection offices and authorities and the competition authorities work together on the DMA, not in this context, but here it was a competition on GDPR, but obviously this can be extrapolated to the DMA. Looking at the time, I think I'm going to conclude here and leave it to questions. What I wanted to say is obviously the DMA is not just a challenge for the regulator, it's also a challenge for the beneficiaries of the obligations. So we invite very much a collaborative discussion with them as well. As I said, the doors are open and also for the gatekeepers, we are very much already engaging with many of the gatekeepers as we speak going forward, I think this has to intensify. To make this work, I think it needs a common effort and it's in the interest, I think of everybody involved to make it work because if it doesn't work, there's the risk that there will be even more backlash. I think that I think that's pretty simple. This is just we need to make this work collectively. And I think the best way to do that is to try this in a collaborative way. If that doesn't work, I mean the commission is obviously ready to use all the tools available under the instrument to make gatekeeping companies comply with the law. Thank you very much.