Sarah Cardell and Will Hayter.

Q1 The Chair: This is the Communications and Digital Select Committee. Today, we have a one-off accountability session, if I can put it that way, with the Competition and Markets Authority. I am pleased to welcome again Sarah Cardell, its chief executive, and Will Hayter, who is the leader of the Digital Markets Unit.

As I think is widely known, this committee has been very interested in the digital markets, competition and consumer legislation that has now been enacted—we will cover quite a bit of that in the course of the questions that we want to ask today—but we are also interested in that legislation and the activity of the CMA in the context of other inquiries that we have held since the last time you were before us. We have had a news inquiry, and we are just completing and will soon publish an inquiry into scaling up in both AI and the creative tech industries. Some of our questions may be influenced and informed by some of our work in that area.

To get us going, I know that you issued a statement this morning with an update from the CMA on its plans for implementing the new legislation. We are grateful that you put that out but slightly disappointed that there was not a bit more information. We are keen to know from you, in providing a brief update on where you are, when we are going to hear what the first topics of investigation will be.

The other thing on which we are concerned to get some assurance from you at this stage today is the breadth of these investigations because, as I am sure you will know, there is some concern from some stakeholders that the investigations you will do—I am assuming this even though you are not saying it—will perhaps be on digital advertising and mobile ecosystems. I see that you are not even flinching as I say that to you—oh, no, I got a bit of a smile. There is a concern around whether they will be sufficiently broad so that we, or you, do not spend a lot of time revisiting some of this stuff. Anything you can say to give us some assurance in that area will be really helpful to kick us off.

Sarah Cardell: Thank you so much, Chair and members of the committee, for the opportunity to speak to you again this afternoon. As you say, this is a timely opportunity for us to come back to the committee and provide a fuller update on the work that we are doing. Obviously, we have had many discussions over the years in terms of the development of the legislation. Perhaps I could contextualise briefly the announcement that we made today; I can then talk through it in a bit more detail.

As you say, the legislation has been passed. The digital markets competition regime aspect of the legislation came into force with effect from 1 January. Just to put it on the record, we also published the final version of our guidance, which sets out in full how we intend to implement and operate the new regime; it was published just before Christmas, having received the Secretary of State’s sign-off. That leaves us, in effect, good to go with the implementation of the new regime.

It is perhaps important, again, to refresh and remind the committee of the context and the ambition of the digital markets competition regime. We will, I am sure, get into this in a lot of detail this afternoon, but we really do see this new regime as a critical opportunity for the UK to create and enhance opportunities for effective competition in order to drive growth, investment and innovation not just across the UK tech sector—clearly, that goes without saying—but across the whole of the UK economy, given how critical digital markets are to the many consumers and businesses who rely on those products and services; in the case of businesses, they do so to bring their own products and services to market. We think, at this moment of focus more broadly on the importance of growth for the UK, that this legislation and this regime are critical components of that.

We have been clear throughout that the way in which we intend to implement the new regime will be characterised by an approach that is participative, proportionate and predictable. We will seek to move with pace but also to balance that against appropriate process. This focus on predictability, in particular, is what underpins the announcement that we have made today. Will and I will unpack this a bit more as we go through the afternoon, I am sure, but we have obviously been thinking very carefully about how to plan and sequence this first phase of SMS designation investigations.

We wanted to provide at the outset an announcement that framed the expectations of activity over the first six months of the new regime. What we did not want to do was have a drip feed of occasional information where we launch one investigation and everybody is wondering, “Well, what’s next? What’s coming next and when?”, as well as—as you asked—“What will the scope of that investigation be?” So what we have done in today’s announcement is set out our expectation that we will launch SMS designation investigations in relation to two separate areas of digital activity during the course of January. I will come back to your question on scope.

We then expect to launch a further designation investigation in relation to a third separate area of digital activity, most likely towards the end of the first six months of the regime. The reasons for that sequencing are twofold. One is the focus on proportionality. We are particularly mindful as we launch investigations—particularly because, to your question, those investigations are likely to be fairly broad in scope—that they will have a significant impact on not just the prospective designation firms but the many interested stakeholders. Of course, the participative nature of the regime means that we expect to have a lot of engagement, so we want to make sure that that engagement can be timely but is not overly burdensome on the stakeholders. We think that it is important to have a degree of sequencing for that reason; we also think that it is important to do this so that we can manage our own internal resources and timelines accordingly. So, those are the reasons for having a degree of sequencing there.

On your point about scope—Will might get into some of that in a bit more detail—what we cannot do, as you anticipated, is give full details on what activities will be covered by those investigations. I should say at the outset that we take incredibly seriously our obligations to provide as full a disclosure as we can to this committee. We will certainly—this goes without saying, I hope—provide updates to the committee as soon as those individual investigations are launched so that you have a full description of what is covered by them. We would also be more than happy to come back once those investigations are under way and give a fuller account of—and have a fuller discussion on—some of the questions that we may not be able to answer fully today.

On the question of scope, it is fair to say that we would expect to take a reasonably broad approach when we look at the overall scoping of the digital activities for those first designation investigations. The other thing we can say is that, because we want to be able to move as quickly as possible through the process for these first designations, you will recall that, as we launch those investigations, the legislation enables us to investigate then designate a firm as having SMS status in relation to a particular digital activity. Once a firm has been designated, we can impose conduct requirements and introduce pro-competition interventions. What we expect to do with these investigations is in parallel to investigating whether or not a firm should be designated. We will also be considering and consulting on potential conduct requirements so that, we hope, we are in a position to introduce at least an initial set of conduct requirements in the event that we conclude we should be designating. So we do not want to have an entirely staggered process, if that makes sense.

As I say, that is the expectation. We will provide specific updates as we announce those first two areas of investigation; that will be during the course of January. At that point, we will set out in full the scope of the investigation and the activity in respect of which we are contemplating designation, but we will also set out initial areas of potential concern and potential action to address those concerns by way of potential conduct requirements. We can get into more detail on that.

As a sort of closing opener, if I may, I did just want to come back to the engagement that we have had with the committee over the last couple of years. I am mindful that both the Chair and a number of members of the committee may shortly be coming to the end of their terms.

I just wanted to say, on behalf of the CMA, that we are very grateful for the close interest and very constructive scrutiny that we have experienced from the committee over the last few years. It is much appreciated that we have the opportunity together to discuss the fact that this legislation is now in force.

Q2 The Chair: Good. Thank you very much for that and for your comments. On this specific matter, I have a couple of follow-up questions on some of the things that you have said. Have the topics for investigation been decided or are they subject to some kind of approval process? I am just trying to get to why it has not been possible for you to identify the topics today. What is behind that?

Sarah Cardell: We need as a practical matter to sequence the launch of our first two investigations. There is the very practical point that these will be wide-ranging and extensive investigations. If we launched two investigations on the same day, they would in essence be running to the same nine-month timetable. This would mean that we had formal milestones hitting at exactly the same point in time.

Let us think about some of the internal governance that we have, for example. You will, I am sure, recall from the legislation that we have a Digital Markets Board Committee, which is the decision-maker as we go through the investigation process. We want to stagger those investigations by a couple of weeks. You should expect to see a gap of a week or two weeks between the two announcements; this is just so that we have a bit of airspace as we are running those as two substantial, parallel investigations.

The Chair: But it has been decided and you know what you will be doing, basically.

Sarah Cardell: We have now. We had to take that decision through our board, but our board’s decision has now gone through.

The Chair: Mr Hayter, it looks like you want to add something. Do not feel that you have to because we will come back to preparedness in detail a bit later on; you just looked like you were about to say something.

Will Hayter: It was only to observe that, as well as our internal management and governance and so on, it is important to bear in mind that, in particular, some of the smaller stakeholders out there may potentially have an interest in both of these two investigations. So we also do not want to have a situation where the same deadlines are falling on small, perhaps less well-resourced, stakeholders at the same time.

Q3 The Chair: That is a fair point—understood. I am sure that, in the course of the questions that we will proceed through this afternoon, we will return to some of this in more detail. Can I stick to the more strategic level, as it were, before we get to some of the detail? I want to ask about the challenges that you have been set as a regulator, along with other regulators, in terms of being more pro-growth and not just minimising risk and that sort of thing; I am talking in general terms there.

I want to understand how the CMA is responding to this challenge that has been set. The importance of this legislation has become very evident to us, particularly through the lens of the inquiry that we are just completing on scale-ups; we have been looking at AI and creative tech. Its importance not just to those sectors but certainly in that context, in terms of the ability to operate in a world that is very much controlled and dominated by a small number of very powerful firms, became evident to us in that inquiry. This legislation should not be seen as anti-growth, although it obviously has an impact on the big tech firms. I would just like to hear a bit more from you about how you are responding to that at the CMA.

Sarah Cardell: Maybe if I can say a few words from a broad CMA perspective, then drill down a little into this particular regime specifically. From my and the CMA’s perspective, the starting point is that our job is to promote competition and protect consumers. That is the essence of our statutory function. We have a number of different ways in which we do that. Effective competition is a critical driver to sustained and productive economic growth. It is well understood and well evidenced that markets where competition is stronger are typically markets where we see higher levels of investment and innovation, as well as higher levels of productivity. Obviously, all those things are key ingredients to driving growth.

I think that there is a very high degree of alignment between the purpose and the statutory functions of the CMA to promote competition; that leads to a resulting outcome that is a pro-growth outcome. Then, there are obviously lots of choices that we can make about which particular issues we prioritise, the work that we do and how we work, in order to make sure also that they are all driving in the same direction in terms of supporting growth. But the essence of our mandate being to promote competition is one that is entirely consistent with and supportive of that focus on growth.

Again, this is a very timely conversation. Next week, we will lay before Parliament the CMA’s draft annual plan for the coming year, 2025-26. You will see in that draft annual plan an absolutely clear and central focus. We are framing it in terms of our job to promote competition and protect consumers to drive growth, opportunity and prosperity for all because it is important also to view growth as a means to an end—that is, what is that economic growth ultimately delivering in the opportunity and prosperity that it brings to the UK economy but also to the people and businesses who live, work and thrive in that economy? This has been central to our strategy since I took up the role of CEO; Marcus Bokkerink took up the role of chair a couple of years ago. You will see it having, I think, a renewed and central focus in the annual plan that we will be consulting on from next week. Again, I am very happy to answer more questions on that and have a follow-up conversation on it as well.

In that plan, we set out a number of ways in which we will really deliver against that mandate in the coming year. Outside of the context of digital markets, for example, we have a broader markets function; we will really drill down and focus on those areas where we can take action to unlock barriers to investment innovation—for example, in critical infrastructure markets and those core enabling markets that are so central to driving growth. An example from last year was the market study that we did into housebuilding. So that is our broader mandate.

We also have a really important role: to provide advice to Government on an ongoing basis. Towards the end of last year, we published our response to the Government’s industrial strategy Green Paper, which set out our thoughts on how we could contribute to that growth focus and industrial strategy, which is very much focused on growth. We have within the CMA our microeconomics unit, which is focused on providing economic analysis. We have a growth programme of activity for that unit, which drills down into economic analysis around the key drivers of and blockers to growth. So there is a lot happening across the different areas of our work; again, I am very happy to provide more detail on that.

An absolutely core element for the coming year will be how we go about implementing the digital markets regime. Let me come on to that more directly. As you alluded to in your question, I believe that the new digital markets competition regime provides quite a unique opportunity for the UK to make sure that we are opening up digital markets in a way that maximises the opportunities for innovation, investment and growth for all participants in the market. I do not think that it should be a choice between the big players and the small players. I think that this is genuinely a regime that can harness, as well as continue to sustain and encourage, the critical activity of the very largest players, which bring very important investment and innovation to the UK—we absolutely do not want to see any sort of chilling of that activity—but what the legislation, as we intend to implement it, should achieve is to bring alongside that important activity from the largest firms a levelling-up opportunity, by which I mean an opportunity to create a level playing field where smaller companies can innovate and thrive.

On your question about scaling up—this is a conversation that we have had; I am sure that we will talk about this more—many of the smaller players in the market want that opportunity to scale up but there are a number of different barriers for them. One barrier can be the dependency that they may have on a larger digital ecosystem, with the need to get access to key inputs.

We have also done quite a lot of work in the AI foundation model space. We talk in that review about the importance of access to critical inputs; data; compute, obviously; technical expertise; and capability. It is also about making sure that, where you have smaller players who are perhaps looking to bring in a new innovation at one level of the supply chain, they are not disadvantaged by the incumbency position of a larger player who is able to leverage and extend their position of market power in order to close off those opportunities to smaller challengers. That is very much at the heart of the regime and a key challenge for us, as we are now taking this forward and prioritising which areas to look at and, importantly, which areas and issues to address through interventions. That focus on opening up opportunities for, in particular, UK businesses to grow and thrive in these digital markets is absolutely key. Having a UK regime in which to do this, rather than being reliant on a regime that is more international in scope, is a unique opportunity for us.

There is a huge opportunity there. By design, that is enhanced because of the targeted and bespoke nature of the digital markets competition regime that we have in the UK. This is not a regime based on broad-brush, blanket rules that apply by default across the sector. It is one where we have to designate an individual firm in relation to an individual activity, then design bespoke and targeted interventions for particular harms. All of this means that, through the work we do, informed by all of the engagement that we have had and will continue to have, we can target those interventions with maximum confidence that we will be opening up opportunities, rather than chilling investment and innovation. That is sometimes a risk from alternative approaches, which have more of a blanket, uniform application with a one-size-fits-all set of rules.

The final element that I very much want to emphasise is the participative—I have already used that term several times—approach that we will take, which is at the heart of this regime. It is about constant iterative engagement with the very largest firms and a whole host of different stakeholders that have an interest in how these markets develop. This means that our interventions will be incredibly well informed and that we will operate on a no-surprises basis; that will, and should, foster the right environment where all businesses, large and small, and their investors can have confidence that this regime is designed and is being implemented in practice in a way that seeks to create and maximise those opportunities for growth.

The Chair: Some of my colleagues may want to pick up on or pursue some of this in more detail as we proceed. Do you feel that you are getting support from the current Government and Ministers for that balanced approach? We can understand and see the challenge in terms of the investment opportunity from some of the large players into the UK, as well as the pressure that may be being brought to bear on the way in which this legislation is being implemented. Are you getting the right level of support from the Government and Ministers for hitting that sweet spot, as you have just described?

Sarah Cardell: We have had good discussions exactly about that balance. When we think about it in terms of creating those opportunities for all, this should be a win-win situation where we can harness the benefits of the very largest firms and the investment and innovation that they bring, as well as creating opportunities for smaller firms in the UK. That is a conversation that we have had with officials and Ministers. This is a very positive opportunity for the UK. It is quite reasonable for there now to be an expectation around how we, the CMA, are going to deliver this; that bit is now on us, in terms of demonstrating how we will deliver this in practice over the coming months, but there is a good understanding around that ambition.

Q4 The Chair: The final thing that I want to talk to you about before I hand over is mergers and acquisitions. This is an area where, as you have said, there will be a change in approach from the CMA. This has certainly been raised with us at different points over the years, certainly in the context of the inquiry on scale-ups.

One of the things put to us by somebody from one of the VCs—I forget which—was that the CMA can tend to be quite interventionist, which is problematic, on what they called internal roll-ups; basically, those are British companies buying other British companies, which is of course a way for some of these smaller firms to scale up. There is concern there that there might be a deterrent from the CMA’s approach to M&A as regards some of these firms being able to scale. That is sometimes aligned with concerns that a firm will not want to scale—or at what point they would do so—if they are at a certain size but are then going to be subject to the kind of attention that scale may bring from the CMA. What has been the change that you are adopting, and what should people expect to see as a result of that?

Sarah Cardell: It is an excellent question. We have absolutely had similar conversations. We have really stepped up over the last few months our engagement more broadly with businesses and investors. Interestingly, you referred to a comment from a VC. We have had a couple of round tables over the last couple of months with groups of both venture capitalists and private equity investors to discuss many of these sorts of themes. Some of this sits in perception rather than reality in the sense that, when you look at the hard numbers in terms of the deals that we actually look at—particularly the handful of deals that are ultimately subject to an in-depth investigation or prohibited—you are talking about single-digit figures every year. The vast majority of deals, including deals involving smaller companies being bought out by larger ones, will pass without any review whatever or, if any, a very quick review.

That said, we recognise that there is this perception of a chilling effect. There is a sense of, “Is this opportunity for exit closed off because of a perception that the UK merger control regime is more interventionist or more intrusive?” We are working hard to address that perception, as well as the reality on the ground, because that can have a chilling effect, even by itself.

As I say, part of that is about engagement. We have had good discussions. A good example of that is a conversation that we had with venture capitalists where we were talking about the fact that we might look at a merger and be concerned enough to ask, “Is this start-up the one that could provide that future competitive challenge? Is it being taken out of the market?” Those are rightly the questions we ask ourselves. However, are we also aware that the buyout might free up that entrepreneurial capital and capability to go off and launch the next innovation or initiative that might create that opportunity and can be replicated? It is about making sure that we are not inadvertently reinforcing the chilling effect that closes off that innovation or diverts some of that activity outside the UK. This is the sort of thing where, in undertaking our kind of competitive analysis, we need to make sure that we are taking full account of those different dynamics as we look at the potential effects of a transaction.

Other things that have come through from some of our conversations concern the way in which companies engage with the CMA, or the way in which the CMA engages with the companies—particularly for smaller companies that may not have much experience of engaging with the CMA or may not be able to afford well-paid advisers who can help them through the process. The thought of the process alone can be quite a deterrent; certainly, being subject to a lengthy process can be so. We have a non-executive on our board who has been through the experience. It took a year out of his commercial life; that is a year taken away from running a business that is trying to grow and scale.

We are also incredibly mindful about making sure that, as businesses come within our purview, we try to tailor the experience accordingly and move as quickly as we can. In a speech that I gave just before Christmas, I talked about four Ps: proportionality, predictability, pace and process. Those are really important, particularly in a merger control context where that uncertainty alone can kill a company. We have heard that many times over. So it is not about a radical change of approach. It is about us taking a much more rounded, informed approach in terms of how we look at particular deals: understanding the context, understanding the ecosystem and understanding what that vibrant investment scale-up ecosystem needs.

I would say, though, that merger control is only one part of that. As we were just discussing, actually our role and responsibility to keep markets open, to make sure we are breaking down those barriers to competition, are equally important for those scale-up opportunities. We have also heard loud and clear that some of the other barriers to scaling up that sit outside the competition landscape in terms of listing rules and access to capital can have an equal, if not greater, deterrent effect. So, in our engagement with investors and with smaller businesses we are having parallel conversations where we want to act on the areas of concern but also want to reassure that we are their champions to create, support and enhance those opportunities as well.

Q5 Baroness Harding of Winscombe: From high strategy, maybe we can get a bit more operational now. I think the Digital Markets Unit has been operating in shadow form since April 2021. Could you just set out what you see as the main operational challenges of moving from shadow operation into full delivery and what you are doing to address them?

Will Hayter: Operation is where I come in, I hope. I tend to think about this question in terms of four categories. You have the tools, the structures and processes, knowledge, and people. When I think about tools, I particularly mean the Act to start with—obviously it is a good moment to have had that come into force—but also our guidance, which, as Sarah said we completed and had signed off by the Secretary of State, and it was published just before Christmas. So those are the key tools without which none of this could get going.

When I say structures and processes, I am talking about things such as the establishment of the Digital Markets Board Committee and the other governance structures within the CMA which Sarah also touched on, but also all the broader plans for how we will actually run these SMS investigations, and other pieces such as the memoranda of understanding with other regulators, which we published also just before Christmas. So, there is a whole set of things around structures and processes. Knowledge is probably the one which, if you like, has taken the biggest investment in terms of numbers of people over time, because that is where all of our catalogue of previous and existing cases in relation to digital markets comes in. As you know, we have for a number of years been building the understanding that will be necessary to be able to start and then proceed with these investigations in a way that is, I hope, effective.

That goes right back to the earlier studies such as the one on digital advertising and on mobile ecosystems, but now through into the various competition enforcement projects that we have had running; the foundation models project, where AI developments clearly inform a number of investigations that we might do; and consumer enforcement work as well. So, there is a huge body of knowledge that we are looking to build on as we now get this regime up and running.

Then, really importantly, people: we have—again, for a number of years—been building up the skills that we need in order to be able to deliver effectively, both in the Digital Markets Unit team but also strengthening our body of legal and economic advisers. Then, critically, as we have talked about before, there is our data unit, which is the real tech specialists, data engineers, data scientists and behavioural scientists. That is a genuinely strong group which we are proud to have been building since 2019. Now, if you include the important strategic business and financial analysis experts that we also have, that total group number is about 100. So, it is important to get all these people in place and get them set up into project teams ready to get the first investigations up and running.

Baroness Harding of Winscombe: Looking forward, what are the biggest concerns that you have operationally?

Will Hayter: There is plenty to work through. We are confident in our preparations and plans but should be expecting to learn and to adapt, as I think one would expect with any new framework. I know that one of your previous evidence sessions, I think back in September, raised a number of challenges and pitfalls to look out for. We have had those in mind as well: things such as needing to consider the balance between competition and other considerations, and, as someone raised, the issues around privacy and security. Those are fair questions to raise, and we have been thinking about them hard throughout the course of our mobile browsers market investigation and drawing on both our in-house and external expertise.

Another challenge raised was the risk of being snarled up in lots of process challenges. As Sarah said earlier, we think it is important to set up the right process which enables us to draw in the evidence and think about that hard, but also to proceed with due pace, and giving due regard to the new duty of expedition that was also introduced in the Act that will apply to the new regime. But there is also then a question about how the firms that engage in this will choose to do so. We have set out our stall in terms of the participative approach and wanting to act in a collaborative, open and transparent way, but that will need the firms to play by that mantra as well.

Another one that I think did not come up in your previous session but which I want to highlight is the risk of regulatory capture. All competition authorities worry about market failure, but of course regulatory failure could be a risk too, and regulatory capture could be an example of that. That is where we have to think really carefully about how we, again, go about this participative approach. A participative approach does not just mean cosiness with the potential or designated SMS firms; it means drawing on a real range of inputs from all participants in any of these ecosystems: so, the candidate or actual designated firm itself, its competitors but also its customers, partners, advertisers on its platforms, consumer groups, and broader civil society. We need to be drawing on all those different inputs. I have been working hard through our existing work and I am looking to do so even more under the new regime to make sure that it is as easy as possible for all those groups to provide input.

Baroness Harding of Winscombe: Thank you. There was quite a lot in what you said, Will. First, just to get it on the record, I did not hear you say that you had trouble recruiting tech expertise. Is it fair for me to take that conclusion away?

Will Hayter: It would be wrong for me to say that it is entirely straightforward. It is quite clear that the salaries available in the private sector are significantly higher in some cases than those that are able to be paid in the Civil Service. I should issue that caution. But we are happy with the offer we are able to make in terms of the public service ethos that we offer, the impact and interest in the work, a very positive environment in terms of diversity, well-being, flexibility, different working patterns and so forth, which has proved able to support, again, a very strong body of tech experts. But it is not impervious to approaches from outside.

Baroness Harding of Winscombe: You were just talking about the risks of regulatory capture and how you make sure that you get input from all sorts of directions. As you know, in the past this committee has been concerned with how you make sure that the voices of much smaller organisations are heard in specific inquiries. Could you give us an update on how you will make sure that that happens?

Will Hayter: Yes, that is a very important area. As I said, that participative approach is about being open and collaborative, and so on, but that means open to everyone, including all those groups I touched on. But it is important to recognise that being open does not just mean passively sitting back and waiting for people to come and talk to us. We need to be mindful of the fact that, particularly at the smaller end, many stakeholders may not have regulatory affairs teams or lawyers waiting to go off and talk to competition authorities, so we really need to make it as easy as possible. We have been doing that and piloting it, if you like, under our existing work and are looking to increase that as we go forward under the new framework.

To take a couple of examples, recently on the guidance consultation we did a few things to try to make that as easy as possible. So, as well as issuing the guidance draft for consultation, which was the legal requirement, we also issued a sort of overview summary—a more plain English version and much shorter—to try to make that much more accessible, recognising that although the long version is not terribly long, it is still probably of an order that is most likely to be read by the legal advisory community in full and not that many other people. So, we are trying to create products that are more accessible for a broader audience.

We also used a mix of in-person and virtual round tables to get input in a time-efficient way, and in a way that is accessible to people all around the country and beyond. We recently put in place a new consultation portal to give a variety of routes for people to provide their inputs without having to necessarily draft a lengthy consultation response. So there are a few things we have done on the guidance piece. Also, our privacy sandbox process, as you will remember, is a set of commitments given to us by Google under the Competition Act. But, again, we have been trying to pilot some of the approaches we would want to take here. For example, we have had a good number of industry round tables to seek out industry conferences and so on that are particularly attended by smaller groups. In any of these sectors, there are particular trade groups that cater to the smaller end of firms. We have tried to engage with them fully. Also on that project, the testing and trialling that we did included Google in this case but certainly included ad tech providers much more at the smaller end of the spectrum. We would want to replicate that again in the new framework.

Those are things that we have been doing thus far. As we look ahead, there are a number of things that we are trying to do. One is this phasing of the kick-off of the first investigations, as we touched on earlier, which is, again, partly about making sure that those organisations at the smaller end do not get hit with multiple consultation deadlines all at the same time, as well as the statutory investigation notice that will describe the scope of the investigations. At the start, we are issuing invitations to comment, and there are broader narrative documents explaining the kinds of things that we think we ought to focus on—picking up on the Chair’s point earlier—describing the breadth of issues, opening up for input on that in plain English, hopefully.

We will be doing an awful lot of meetings and round tables, in person and virtually, trying to cater to different kinds of audiences. When we think about the process of considering setting conduct requirements, for example, we will very much work hard to seek out the views of smaller firms, recognising that we are certainly likely to hear from the bigger ones anyway.

Q6 Baroness Harding of Winscombe: In your framework of tools, structures, knowledge and people, you did not reference accountability frameworks. That is an area certainly this committee has looked at during the whole period in which I have been on it. How will you ensure not just that you are acting accountably but are seen to be doing so? What is the way in which we should think about the accountability framework?

Sarah Cardell: I might pick that up first but do jump in, Will. It should be multifaceted. Specifically in relation to parliamentary accountability—we have said this before but it bears repeating—we fully recognise that we have been entrusted with a significant and substantial set of powers through the new regime. We take that responsibility seriously. It is critical that we have regular accountability through this committee and others in terms of reporting back. We are literally on the cusp of getting into the action, and it will be quite different from our accountability on merger control, for example, where you have a one-off review into a particular deal; you move on. This is going to be an iterative regime. Just as our engagement will be iterative in the way that Will has outlined, so too should our accountability be in that sense. Obviously, it is for this committee and others to decide but I expect and hope that we will have regular opportunities to appear before parliamentary committees to be able to discharge that element of accountability. There is also a set of responsibilities on us in terms of our transparency and reporting.

As an organisation, the CMA operates with high standards of transparency and openness. We consider that to be part of our DNA in how we operate. But as you will see through the coming weeks, we intend to exemplify that in openness and transparency as we set out what we are doing and why we are doing it. In our own reporting and transparency, you will see that come through. When it comes formally, for example, to our annual report, we have already said that we will include specific reporting on numbers of investigations and all the detail on where we are introducing conduct requirements and the rationale for that—particularly regarding Will’s point on that coming through and much of that wider public consultation and engagement exercise that we are doing. As we have said all along with this process, we also remain open to ideas and discussions about ways in which to further enhance that accountability and reporting. We will want to develop that together as the regime is put into effect. But we take it very seriously.

Q7 Baroness Harding of Winscombe: You mentioned co-operation earlier. I was pleased to hear you describe competition and growth being one and the same, not an either/or. I am a bit confused as to what the Regulatory Innovation Office is going to be doing and how it will interact with your new powers, if at all. Can you help me to understand that better?

Sarah Cardell: I am not sure whether I have a position on that. We have no detailed insights into that at the moment. Obviously, we have through the DRCF our own self-created, if you like, version of maximising that regulatory co-operation across the regulators with the closest and most overlapping interest in issues across digital markets. I think we are we are waiting to hear from DSIT more detail on the RIO.

Baroness Harding of Winscombe: Should we be worried about the growth in the number of different organisations with overlapping responsibilities for co-ordinating regulation?

Sarah Cardell: Inevitably it is a complex landscape. That goes without saying. I would not want to pre-empt or judge your own collective perspective on things but, from the CMA’s perspective, it is essential that we engage closely with agencies domestically—there is obviously the international element to it, too, but just domestically, it is critical that we engage closely, as we have done, I think, over the past few years, both through the DRCF and bilaterally. We have worked closely with the ICO, for example, in relation to privacy matters. We work closely with Ofcom in relation to news and publishing. There are no hard and fast boundaries in so many of the issues that impact on digital markets. It is inherent that there is a requirement to have close and co-operative working. We see that in the legislation. We have an obligation to consult a number of the regulators. Both Ofcom and the FCA have the ability to make references to the CMA. I should have thought it is absolutely worth continuing to monitor it and watch closely how well that is working.

The Chair: I have a couple of colleagues with supplementaries, first Lord Young and then Lord Dunlop.

Q8 Lord Young of Norwood Green: Will, you got to the people bit at the end and I am going to address this question to both of you because it is such an important area. There is going to be huge competition for skills and how you advise the Government on creating those skills is going to run right through the industry, education, et cetera. I want to get a feeling for how you see this and your strategy towards it.

Will Hayter: That is a really important question. There is a big set of challenges. Any of these new developments need a lot of investment in people in order to understand and be able to work with them. We have been fortunate with some decent decision-making in the past to try and invest in new areas like the data unit that I mentioned. It has been possible to do that, partly due to being of sufficient scale. It is a bit more difficult for smaller organisations who cannot perhaps invest in that kind of substantial team. It is also important to make sure that you do not just bring in those external experts or newer types of skills but really integrate them with the existing people in the organisation. We have worked hard to do that. In our case, we have settled down effectively now but when it first started, there was a bit of a learning process for the new teams and skills coming in, as regards working with the more familiar lawyers, economists and policy delivery people in the organisation. It is something we have been working on hard but I would not underestimate the challenges involved in building those capabilities, keeping them up to date, retaining the people, as I was touching on earlier, and—

Lord Young of Norwood Green: On that last point, you have to have a retention strategy as well.

Will Hayter: Absolutely; CMA-wide we have worked very hard on that to try to make sure that we have a positive working environment. Again, going back to what I said earlier, it is about making sure that people are motivated by the interest of the work and the positive impact that it can have. I think that people really feel that strongly in the CMA and believe in the purpose that the organisation has set out. We need to keep on reinforcing that and making sure that it is as positive an environment as possible.

Lord Young of Norwood Green: Is there anything you want to add to that on the skills front?

Sarah Cardell: Maybe just a couple of quick points. On your retention point, I fully agree. Perhaps also going back to Baroness Harding’s question, one of our biggest challenges will be retaining the skills and capability that we develop, and the people working on these regimes in the first year will be gold dust, frankly, to others. We have done fantastically in recruiting the talent that we have got; retaining that talent will be equally important, if not more so, over the next couple of years.

On a broader cross-government perspective, with something like the DRCF, one of the real benefits has been pooling our knowledge and expertise, not just in terms of individual opportunities to work together on particular issues but in trying to leverage a sort of “greater than the sum of the parts” sense. Horizon scanning is a good example of the activity that we can carry out within DRCF, where there is the need to invest in knowing what is coming down the track in terms of digital activities and innovations, without necessarily needing to identify whether something raises a competition issue, a privacy issue or an online safety issue. Rather, it is about that solid understanding of the future horizon. That is the sort of thing where we really can pool that capability and knowledge, and it will also become increasingly important, as Will said, to find ways to share that appropriately with smaller organisations which perhaps are not able to kind of leverage the capability that we have been fortunate enough to be able to build.

Again, it is beyond the remit of the CMA but, from a government perspective, when you are thinking about AI skills and capability, there are pockets. Again, we have been fortunate in the teams that we have built up primarily to drive our sort of front-line work in understanding developments in digital markets. They are the very same teams that we are deploying to help us as an organisation think about how we are using AI in the work that we do. They are at the forefront of our digital transformation efforts in how we are operating as an organisation. Many of them are wearing two hats for us. From a cross-government perspective, trying to leverage those skills internally and then thinking about sharing those learnings, not reinventing the wheel, is hugely important.

Q9 Lord Dunlop: Following on from Lord Young’s question, could you give us a sort of sense of where you are getting your external tech expertise from? It would be quite interesting to understand that.

Will Hayter: A number of sources, is the answer. At one level it is the actual people we recruit, who, as you would expect, come from quite a range of different backgrounds, some from elsewhere in government and others from outside—a real range of sources. But it is really important then that we complement the people in the organisation with input from outside as well. Two years ago we appointed a group of specific digital experts—again, a range of individuals with world-leading expertise in their fields—to advise on a variety of projects, and we have used those across our AI work, in some other work in advertising, a whole range of different projects, and on a lot of the thinking on the guidance and policy design, and so forth. So that is another source.

We also recognise that it will almost certainly be necessary on specific projects to pull in particular expertise, because you can get to a position where the expertise needed is so specific to a particular area that it does not necessarily come from a standing group of experts. Just as an example, we have procured external expertise on our mobile browsers investigation, specifically on the privacy and security issues which are so important there. That is exactly the same kind of approach that we would expect to look to draw on as we go through these investigations. Again, it is really important that we have the internal people who can engage in these issues, but sometimes you really need that hard-edged expertise that might come from a very specific place externally as well.

Lord Dunlop: Are you recruiting from tech firms themselves? Are you getting poachers who are turning gamekeeper?

Will Hayter: Sometimes, yes. Obviously, you get into this interesting question about salary differentials and so on, which is just a fact of life. But yes, it has been interesting that we have certainly had a good number of approaches in recruitment campaigns from people from the very biggest tech firms, for sure.

The Lord Bishop of Leeds: By definition, if people are attracted by the notion of public service, they are likely to be shorter term than people who are building a career, are they not? So, if you are getting in tech experts on lower salaries with that incentive—presumably to enhance their CV as well; it gives them broader experience—you then run into problems later, potentially with turnover.

Sarah Cardell: I think we have to accept a degree of rotation in and out—that is inevitable. To be honest, if you can draw in some of that talent and have the benefit of it for a few years, as long as we are growing that talent pool, that is fine. One of the things that we are doing a lot of broadly across the CMA is looking hard at how we build those career development programmes. A lot of the talent that we have brought in has been at the more junior end, so you have people who have good technical capability but they are actually looking for a career development opportunity, and we need to continue—that needs to be a sort of constantly refreshing source of talent.

Will Hayter: But we see, and we should expect to continue to see, a bit of a mix. As you say, you might get some of those people who come in for a couple of years to add to a CV and then they go elsewhere. We also have people who are in the organisation for longer who are perhaps more attached to that public service ethos.

The Chair: We must move on—we still have quite a bit of ground to cover.

Q10 Baroness Wheatcroft: I want to broaden the field rather away from tech expertise and ask you about regulatory expertise. In particular, it is now more than 18 months since the EU regime came into play. What lessons have you learned, and are you convinced that the very different approach that the UK is adopting is still the right one?

Will Hayter: The headline answer is that we have learned and are continuing to learn an awful lot about and from the implementation of the Digital Markets Act in the EU, and that is both from our counterparts at the Commission but also from all the many firms that are experiencing it from different angles. There is an important opportunity for us to keep learning, as we do. I think it is probably not the right thing for me to say yea or nay as to whether it is definitely still the right approach versus the DMA, because they are at different stages and we need to see how they work out. It is important to note the differences but also some of the similarities—the problem that was identified and some of the subject matter, both of which we will end up thinking about. But the differences are quite clear in the process, and that is in terms of the designation, as I think you know, and therefore the likely scope, particularly early on in our regime, and then differences in the remedies.

In particular, as you know, in the Digital Markets Act you have this list of obligations and prohibitions in the law up front which apply equally to all the seven gatekeepers across the 24 gatekeeper core platform service combinations that the EU has designated, whereas we will be going about this very targeted, tailored, bespoke nine-month investigation and thinking about targeted, tailored, bespoke conduct requirements.

Baroness Wheatcroft: But what you have seen already is that the big tech companies are going to use every trick in the book to try to avoid total compliance with what they are asked to do. Are you sure that the approach that the UK is adopting of asking nicely, persuading and negotiating, will get the results?

Will Hayter: It is important to emphasise up front that there is an element, as you say, in the participative approach of, “We really do want to work collaboratively together to try to identify solutions and avoid surprises”, and all that, which we have emphasised several times. Yet there is a hard backstop, which is potentially quite substantial fines if the firms in question do not engage in that process in the way that we would hope they do. That is important.

I often think of that quote about speaking softly and carrying a big stick. In this context, you need both for it to work, but, again, we have definitely learned an awful lot from the DMA. The helpful thing in our framework, as with so much of it, is the flexibility inherent in it. When you think about setting conduct requirements, for example, we are not given the rule up front in the law and that is that for the long term; the allowable categories of conduct requirement then allow us to consider the right level of tailoring specification for any conduct requirements. In doing that specification and consideration, as well as the work that we will do with the SMS candidate firms and others, we will be thinking exactly about learning from what we have seen in the compliance process in the EU, including the compliance investigations that have been done over there and the outcomes of those.

Baroness Wheatcroft: Have you, albeit privately—although publicly too, I hope—got in your head some sort of target time within which you will, in effect, lose patience with the wriggling that is inevitable from the big tech companies?

Will Hayter: I might hesitate to put a schedule on it. The investigations could be into several firms. Once we have got a little further down the road, we will, I think, learn a lot from not just the DMA but from the different approaches taken by different firms. We will want to get to a position where it is clear that what we think is the case is indeed true: the better solution for all concerned is us working constructively together to try to resolve things, rather than the “walking backwards slowly” approach that I think you are describing.

Sarah Cardell: I want to reinforce that. There are choices that will need to be made all round about what direction this regime will take. Our starting point is that, absolutely, we would like, we expect and we hope to pursue a participative approach; this is the conversation that we have both had with a number of firms. I do not think that anybody would expect each and every potential issue to be resolved in an amicable fashion, but let us take a sensible approach where we can filter down and pick those issues that we are capable of resolving fairly quickly and those where it becomes a more substantive issue.

As Will said, it will be quite interesting. We will potentially have some designation investigations where it is a single digital activity with more than one firm and, therefore, you might end up with a different approach to interventions depending on the approach that the firm in question chooses to take. Obviously, this is not a one-shop game: there will potentially be ongoing engagement over time and a number of different designations in different areas of activity. We will learn from the experience that we have had in previous investigations what approach has worked or has not worked, as well as how that informs our approach going forward. I am not sure that I want to compare this to a chess game but it is a complicated set of engagements where I very much hope—because I think that it is genuinely in the best interests of the UK economy—that we can resolve many things in a constructive way. In contrast to some of the other regimes, perhaps, this regime is designed to enable us to do that, but it does require that shared ambition on both sides.

Q11 Baroness Wheatcroft: If you are confronted, as the EU has been, with companies saying, “Well, I just don’t think we’ll bring this new product to the UK”, for instance, what will your reaction to that be? Do you think that there could be a detriment to the UK consumer if that happens?

Will Hayter: We certainly want to see UK consumers benefiting from all the latest and greatest technologies that these markets can offer. Some of those aspects that I was just talking about give us good confidence that we should not see those sorts of issues arising in the same way.

That is apparent when you go back to some of the things that the firms have said at the point when they have described their respective rollout approaches. For example, when Apple described its approach to rolling out Apple Intelligence, it talked about regulatory uncertainties, in particular those arising from the interoperability requirement in the DMA; that is the framework, and it is quite a broad requirement. It said, “We’re just not sure how this applies”. Similarly, when Meta made some announcements about its AI products, it talked more broadly about regulatory uncertainties in the EU.

Now, it is not for us to comment on the predictability or certainty—or otherwise—of the EU regime, but it is really on us to make sure that we offer as much certainty and predictability as is possible across any of these instances. I just cannot see any scenario where we would end up with a rule that looked anything like the EU’s interoperability requirement because any kind of requirement will have involved such a degree of engagement through our participative approach with the firm in question, as well as with others around that ecosystem, before it is anywhere close to being put in place.

Baroness Wheatcroft: That is very reassuring. Given what we have seen in the EU and all the work you have already been doing—you said that you have built up this huge body of expertise and knowledge—what do you think is the extent of the harm already being experienced in the UK in terms of the issues that you are now empowered to deal with?

Will Hayter: It is an important question. We talk in broad terms about the effects for people, businesses and the economy; some of that is about people. It could be about high prices, reduced choice for businesses or the lack of a level playing field. If you are pushing me or us to try to put a number on that—

Baroness Wheatcroft: Consider yourself pushed.

Will Hayter: You might need to watch this space on that front. We have put numbers on aspects of this in the past, but I should say that some of them are from investigations or studies that are now a couple of years out of date. We talked about the market study that we did on the cost of digital advertising, for example. The number was £500 per household in the UK, I think; that is just the cost of advertising. We then talked about a number in terms of excess profits; at that stage, it was £2.4 billion, looking across those markets. That is data from 2018-19 so, clearly, the numbers would be different now.

It is also not always a straightforward calculation. It is important to note that many of the most worrying harms are likely to be less easy to put a number on because they are really all about forgone innovation from new offerings being cut off from being brought to market. It is truly difficult to put a number on forgone future opportunities but, where possible, we try to understand the scale of the harm—we will continue to do that as we go through these investigations. But, as I say, the important and perhaps bigger prize is future innovation, with the unknown and wonderful new services that might emerge on the back of us being able, I hope, to pull away some barriers to that innovation.

Baroness Wheatcroft: I have just one more question. Given that, do you think it would be right for people to sense what appears to be a lack of urgency in the timetable that you are working to?

Will Hayter: We certainly feel urgency; we have felt that ever since the Digital Markets Unit was set up in shadow form in 2021. We have always, in this committee and elsewhere, stated our intention to hit the ground running. That is exactly what we are doing with today’s announcement. The regime came into force six days ago and our board met yesterday. We are doing what we said we would do: getting on with the job in hand.

Baroness Wheatcroft: What about the nine months?

Will Hayter: Nine months is a limit on the length of the investigation. Obviously, we want to do any investigation as quickly as we possibly can, but it is important to recognise that what we set out is our ambition, as allowed for in the Act, to consider the first set of conduct requirements in parallel with the SMS investigation. As you will appreciate, given the complexity of the issues, that may well take a good deal of those nine months.

Sarah Cardell: I just want to reinforce that. Again, I come back to one of our earlier comments. When these investigations are announced, you will see that they are likely broad and complex in scope, as Will said. Because we have a sense of urgency, we think it is important that we move and put ourselves in a position—obviously we are not pre-empting the designation investigation—to be able to introduce at least an initial set of conduct requirements immediately upon designation, if that is where the investigation ends up.

On the other point, we were talking earlier about the potential challenges and the risks of litigation and derailment. The slowest outcome will be for us to seek to move too quickly, to be challenged in court, and to have to go back and do it again. Although that absolutely does not mean that we should be taking an overly cautious approach, it means that we need to strike the right balance, because we want to make sure that the action that we take is robust and defensible.

Q12 Lord Hall of Birkenhead: Our recent report on news included a number of recommendations for you—can I ask you about a couple of them? The first one was that we asked what steps you would take to investigate tech firms that are leveraging their dominance to secure anti-competitive advantages in obtaining data for gen AI training. We suggested that this “should be an immediate priority given the pace of market developments and impacts on news media business models”.

Will Hayter: We certainly registered the recommendation in a very clear report and will be responding formally soon to that. It is important to state where we are in terms of the status of our thinking on this issue. As you know from previous conversations, we have been thinking about these broad sets of issues right back through to 2019-20, through the digital advertising study, where we heard lots of concerns from publishers about their relationships with, in particular, Google and Facebook—as it was at the time—and about transparency, control, fair payment for content, and so forth. The advice that we worked on with Ofcom and published in 2022 was part of that same strand of work. We have absolutely been continuing to talk to publishers and all participants in that market since, and of course the issues have evolved somewhat to focus more on AI and the points you raised in your report about possible concerns on connections between search indexing and scraping data for training models or for inference, grounding, and so forth.

It is important to recognise that we have not formally gathered evidence on those issues more recently. If this ends up being the subject of one of the early investigations, we would absolutely expect to go into that more formal information-gathering stage. But at the moment we are, I suppose, in a similar stage to you in that we have been talking to the same publishers that you heard evidence from last year, those same sorts of issues have been raising their heads, and we definitely want to be getting to grips with those.

Lord Hall of Birkenhead: The answer may be the same for this, but we recommended that you provide detailed guidance on how you would deal with this issue. Is that part of what you are thinking through before you say anything on this?

Will Hayter: Exactly. Again, this is one of these ones where, once the early investigations are out in the wild, as it were, I will be very happy to come back and talk more about the specifics.

Q13 Lord Hall of Birkenhead: We also made recommendations about the Digital Regulation Cooperation Forum, which, when it comes to news, I think struck us as being phenomenally important in working through how you as a group of regulators deal with issues around news. We thought that it would be good to establish a workstream to look at areas of regulatory crossover that affects the new sector. What progress has been made? What do you make of that?

Will Hayter: We are always grateful for the committee’s support for and interest in the work of the DRCF, something we have had a number of conversations about over the years. A number of news issues relate to existing DRCF work: for example, the workshop we did in August on AI transparency. As I have said already, it has also been a big focus of lots of engagement between us and Ofcom specifically, including that formal advice a couple of years ago. It is important to recognise that not all DRCF work is four-way; some of it is two-way. Indeed, the very successful programme of work the CMA has had with the ICO is part of a DRCF work programme but it is really something for the two of us in particular, because it is specifically about competition and data protection. It may be that news fits a bit more in that bracket. You would want to get the formal answer from the ICO, but I do not think news is such a focus for the FCA, for example, whereas, given this new framework and Ofcom’s media responsibilities, it is clearly squarely for the two of us to be thinking about.

I should just add that the regulatory co-operation provisions in the DMCC Act complement what we have already been doing under the DRCF. Those memoranda of understanding that I mentioned provide for and require engagement by us with the relevant regulator where what we might be doing touches on their remit, and obviously that would be the case here if we were going to be engaging in issues that relate to news. It is a minimum requirement in the Act, and we have already been going beyond that and would want to keep doing so. But I just note that that is an important part of the structure as well.

Lord Hall of Birkenhead: What is your sense of the importance of issues around news to you and Ofcom in terms of the other priorities you have?

Will Hayter: I cannot speak for Ofcom but, from our point of view, it is clear that the news industry has a broader importance than just its economic impact given the sort of broader public interest function that it has and its importance for citizens, not just consumers. But the flip side of that is what you also recognised in your report, which I was pleased to see, which was the need for a broader policy effort to think about these issues. Action to address problems with competition may be part of a solution but is unlikely to be the whole solution given some wider systemic issues.

Lord Hall of Birkenhead: Just to help me on something very specific which you mentioned earlier: on web crawlers scraping journalistic content, is that you or Ofcom, or again, would you want to have a look at it with others?

Will Hayter: That is something that we have already been talking to Ofcom about as part of this ongoing dialogue on these broad issues, and I suspect that we could both end up conceivably digging into it further through our respective remits, but joining up. As I said, right now, our understanding is broadly where yours is in terms of the engagement we have done, because the conversations with the publishers that you have also had have raised that as an issue. If and when the time arises and the right investigation is under way, we would certainly want to be getting into that.

Lord Hall of Birkenhead: Just one final thing: at the very beginning you mentioned horizon scanning. Is that an area again where you would come together with other regulators to say, for example, “We want to have a look at AI-powered search”, which, bluntly, is going to change the nature of the journalism which is done by search tools. Would you want to look at that with others or would you do it yourselves?

Will Hayter: Both, actually. With horizon scanning, by its nature we cover lots of topics and we do a lot of that work ourselves.

Lord Hall of Birkenhead: In that specific case, though.

Will Hayter: Yes, I mean exactly those kinds of developments: what is going on with foundation models more broadly, and then how is that being applied in search? We were already thinking about that in our foundation models work. Search has one of a number of deployments of foundation models alongside other sectors as well. That is definitely something that we would want to consider if we were to look at the search market, for example.

Lord Hall of Birkenhead: You would consider it—you are not actually looking at it at the moment, from what I gather.

Will Hayter: As I said, we do not have an open investigation with formal evidence gathering and process but we have lots of behind-the-scenes discussion going on.

The Chair: That is a neat segue to our final group of questions, where Lord Knight is going to lead us on new and emerging technologies.

Q14 Lord Knight of Weymouth: Thank you and apologies: I had to step out for an Oral Question earlier on. I am interested in how the CMA is keeping abreast of emerging technologies, including AI, and the way in which those tech companies can leverage that dominance into adjacent markets: for example, if you were a significant search provider, could you then get into data and text mining in order to fuel your AI products? How effectively will emerging technologies be captured under the new digital markets competition regime?

Sarah Cardell: On the broader point again, perhaps jumping back a little bit to the previous conversation as well, in terms of our own horizon scanning and tracking of developments in AI and foundation models, this has been front of mind for us at the CMA for the last 18 months or so. We started work back in the early part of 2023. Again, the benefit of having the Data and Technology Insight function capability that we have built up internally is that we have some experts who are really good at tracking what is happening across the markets, engaging very constructively with a wide set of different stakeholders. We put out a couple of reports back in 2023 and early in 2024, which were absolutely designed to shine an early light on the opportunities in this space as well as the potential risks for competition and consumer protection.

In those reports, we set out what we considered guiding principles to help ensure that as these markets evolved rapidly, they did so in a way that maximised the opportunity for open and effective competition. We talked about access to key inputs upstream. We talked about diversity in terms of seeing a different range of models coming through. We talked about choice for downstream users, both business users and consumers, not just in terms of an initial choice but, critically, in the ability to move between different models over time. We talked about fair dealing, which is capturing the essence of making sure that you do not have that kind of anti-competitive leveraging and self-preferencing. Then we talked about accountability and transparency along the supply chain. Those are principles that we first developed in the latter part of 2023 but they still hold good for 2025.

In a way, that very much foreshadowed the new regime because that principled-based approach is very much at the heart of the way in which we would want to see the regime operating. It goes back a little to some of the earlier questions, because we were talking about some of the potential benefits of the UK regime compared to the EU regime. My personal view is that one of the advantages of the way in which the UK regime has been designed is that it is truly forward looking and gives us that flexibility to keep pace with market evolutions. We obviously know that these digital markets are so fast moving that what we absolutely need to guard against is taking a sort of static, backward-looking view of issues and not being able, through the process of designation and then designing interventions, to make sure that we can take account of those evolutions in relation to AI, as in other areas.

In the report that we put out back in April 2024, we talked about how we could see not necessarily emerging concerns but potential areas of concern—the possibility that these markets could evolve in a way whereby those sorts of scenarios could arise, where you could see greater entrenchment, extension of market power and foreclosure of the kind of innovation needed. Again, this comes back to the theme of a lot of this afternoon’s discussion: making sure that we keep open those opportunities. Many of the UK innovators in this space are not necessarily going to produce one of the groundbreaking foundation models but they may well bring to market some more bespoke modelling or evolution, or build on those models in a way in which it enables them to offer an innovative new product or service. That is absolutely the kind of opportunity that we want to keep open with this regime.

So, I am confident in the way in which the UK regime has been designed from the get-go. As to Lord Hall’s question, we will consciously be able to take account of those evolutions as we see them. But the benefit of this regime is that it is not a one-shot set of rules. We designate, we may impose an initial set of conduct requirements, and if we see market developments moving in a particular direction, obviously we can go back and introduce additional or amended conduct requirements accordingly.

I hope the regime will be working well and we will be able to do so in an engaged way, so it should not be a sudden surprise to the companies. I hope that over time we can have those conversations to guard against the need for that further intervention. That is why the founding principles become so important. But that leaves us well placed as regards the combination of the way in which the regime is designed and the capability that we have built up to act on those sorts of potential concerns.

Lord Knight of Weymouth: How do you deal with what I perceive as a tension around that? We saw with the development of generative AI how it is hugely expensive, requiring huge amounts of energy and computing power. It is therefore only really available to a few very wealthy, established players. You do not want to constrain that innovation but inevitably you are then going to end up in an anti-competitive place because these huge players have the innovation and you have got to somehow try and unlock that for everybody else.

Sarah Cardell: Our job is obviously to avoid that inevitability. We need to recognise that there are inherent features in these markets. As you say, the computing power required is phenomenal, and that requires huge financial capital. One of the consequences of that, for example, is that you see lots of partnerships arising. Those are not inherently problematic; in many cases, they provide a route to market for smaller players. But, to take that as an example, we want to have comfort and potentially use principles, as we have done already, in future, and use conduct requirements to safeguard against enabling the incumbent players to leverage and foreclose opportunities. It does not enable them to get a better opportunity to take some capability and leverage it downstream or in an adjacent market that a smaller partner has the ability to switch to if they decide they want to move to another partner over time. So, we need to make sure that we are working with those big players to build in that flexibility, which, coming back to my earlier point, enables us as an economy to harness the benefits of that investment, which is critical, but to do that in a way that keeps open those opportunities. That is where the principles of diversity, choice and access on fair and reasonable terms are absolutely critical.

Q15 Lord Knight of Weymouth: Are there lessons from what you have observed with the development of AI that you would then want to apply as the next wave of technology, perhaps quantum, comes? Google is some way from a commercial application of Willow but it is an exciting, powerful, innovative technology. But it is hugely expensive for them to develop, and who else is going to be able to do it?

Sarah Cardell: Almost certainly, yes. We will need to be continually looking for those parallels. The benefit, once we have the regime in place, is that a lot of those principles are transferable. Once you can see how those principles and that approach are being deployed in a particular market and you see market evolutions coming through—there will be differences, of course, but a lot of those foundational principles are about access, about closing off or keeping open opportunity, and making sure that you are not building up closed ecosystems that are then impenetrable to others or locked in a first mover advantage—a lot of that replicates across.

Lord Knight of Weymouth: Finally, do you see a need for legislation specifically on AI, and if so, what would you like to see in it?

Will Hayter: Helpfully, in terms of market power issues, we do not think there is a need for specific AI legislation because this new piece of legislation is sufficiently flexible, does not contain the words “AI” or indeed any other technology, but is couched in broader economic concepts that allow us to think about any market power issues that might be either in AI markets or enabled by AI in other markets.

So, in our area, no. I know there is a broader question about other types of AI legislation and regulation. As we understand it, what is being contemplated is more tightly focused as regards the critical safety end of things. That seems to follow a somewhat similar philosophical approach to the one that underpins this work, in that it is intended to identify where the most acute risks might be and target those quite specifically without a sort of broad-blanket approach and so on to keep this mantra going of supporting growth and investment, not shedding innovation and so forth.

However, we are ready to work with the Government as much as they would like in terms of their thinking in that area and as the AI opportunities action plan comes out, thereby continuing the themes and the work that we have been doing on issues such as using AI expertise to interrogate bidding data in procurement and so forth. That is all covered in this area as well.

Q16 The Chair: I just have one very small question, which goes back to the staffing issue. Did you in the end get released from the constraints of Treasury pay settlements and were you brought more into line with the FCA and Ofcom in terms of the DMU? That was something we put in a request to the Treasury on your behalf at one point.

Sarah Cardell: The short answer is no, in the sense that we are broadly subject to the normal Civil Service pay rules.

The Chair: So you are not on a par with the FCA and Ofcom in that area?

Sarah Cardell: No.

The Chair: But it does not sound like it has been a barrier to you. Have they given you a reason as to why not?

Sarah Cardell: I am hesitating because I am trying to think back to a specific conversation. As part of our broader conversations, our approach to pay has been and continues to be that we fall within the Civil Service pay rules. I am not sure that we would want to end up in a situation where we had different pay rules for some members of staff in the CMA because that could become quite organisationally complex for us. At the moment, for all the reasons that Will has said, we have managed. More generally, the pay difference vis-à-vis other agencies such as the FCA and Ofcom, as well as compared to the private sector, is something for us to keep an eye on.

The Chair: Okay. I noted that you are going to respond formally to our news inquiry and the points in there that are relevant to you; we look forward to that very much. That just leaves me to thank you very much for all your answers to our questions today. It has been very informative. I hope that it has been in line with the way you described your hope and expectation for the session at the beginning of our meeting today.

Sarah Cardell: It has, thank you.

The Chair: Thank you again for your kind comments about the committee. I am sure that, under new leadership, it will continue to be constructive in its approach to yourselves and to other regulators.