Introduction
Thank you for the opportunity to open today's conference.
Since taking over as Chief Executive of the CMA, I have been having conversations across the UK and internationally about how the CMA can best help people, businesses and the UK economy through our role in promoting competitive markets and tackling unfair behaviour.
Some of my most informative and thought-provoking conversations have been with businesses and investors. I have heard first-hand how crucial competitive markets are to enable talented British entrepreneurs to challenge incumbents, innovate and grow. And I have heard how Britain's open and competitive markets make us an attractive destination for overseas investors, providing capital which helps our economy to grow productively.
But I have also heard questions, including in the field of merger control. Is the CMA too interventionist in its approach to mergers? Has it over-stepped its jurisdictional reach by reviewing and blocking global deals with limited impact on the UK? Has it acted with insufficient regard for consistent outcomes internationally? And, as a consequence, is there a risk that the benefits to consumers - and, indeed, businesses - from robust merger control are outweighed by greater uncertainty and unpredictability?
For the reasons I will explain today, I believe these concerns are misplaced. But they deserve a direct and open response. Having spent many years of my previous professional life advising on the merger control risks of major M&A activity, I understand the desire for clarity and transparency, particularly in the context of the CMA's expanded role post-Brexit. And I think the CMA can and should be taking the opportunity to explain the approach that we take to assessing mergers and how that approach is evolving over time. We do this through individual case decisions and formal guidance, but we should also be ready to have a wider discussion with businesses, investors and their advisors. And we should be open to feedback on our approach.
Today I will set out my reflections on 5 aspects of UK merger control:
- First, I will briefly recap the fundamental benefits and purpose of merger control.
- Second, I will address the concerns that the CMA has taken an overly interventionist approach to merger control, including the question of alleged jurisdictional overreach as well as our approach to identifying and remedying competition concerns.
- Third, I will reflect on our approach to international alignment.
- Fourth, I will draw out some key aspects of the UK regime that underpin our commitment to an objective and evidence-based merger control assessment.
- Finally, I will return to the themes of certainty and predictability and offer some closing reflections.
I do not expect every practitioner or business person to agree with every decision that the CMA takes. But the clearer we can be about the UK's approach to merger control, particularly in our expanded role post-Brexit, the more we help ensure competitive businesses can innovate and grow both organically and through acquisitions in a way that benefits people, businesses and the whole UK economy.
Purpose and benefits of merger control
First of all, I want to take a step back and remind ourselves why merger control matters.
The purpose of our UK merger control regime is to prevent mergers between businesses that substantially weaken competition.[1]
Competition matters because it helps to ensure that markets deliver benefits to people, including through lower prices and higher quality goods and services, as well as enabling businesses to enter new markets and grow. CMA estimates show that during the past 3 financial years, the merger regime saved consumers more than £2 billion - around a third of the CMA's overall impact.[2]
Of course, the vast majority of acquisitions will not impede competition, and some may well enhance it. In a competitive market, M&A activity can enable firms to accelerate innovation, broaden their product offering and reduce their prices as well as enabling a new competitor to grow more quickly to challenge incumbents. And competition for corporate control can also improve governance and hasten the spread of innovation in management.
But in a minority of cases, mergers can provide a route to create or consolidate market power in a way that is not readily reversed. This may occur most obviously where 2 direct competitors merge. But it may also result, for example, from the acquisition of a potential future competitor or a business active in an upstream or adjacent market. The CMA's duty, when discharging its statutory merger control responsibilities, is to consider the potential impact of such an acquisition on competition.
Mergers assessments are, by definition, forward-looking. The CMA has to judge how market conditions will develop absent the merger and weigh that against the likely competitive impact of the deal. This carries the risk of blocking a pro-competitive deal or allowing an anti-competitive one. Neither outcome is cost free but the risks from the latter should not be understated.[3] The CMA's postbag is full of complaints about competition and consumer problems arising in markets that have become too concentrated. Even if some of these problems can be tackled retrospectively using other tools, that takes longer, competes for priority with other matters, and the harm that arises in the meantime can never be undone.
Is the CMA overly interventionist in its approach to merger control?
So, is it correct that the CMA is intervening too often, or in the wrong cases?
There have been 2 elements to this claim: first, that we have been too expansive in relation to jurisdiction; and second, that having obtained jurisdiction, we have been overly interventionist in blocking deals. I'll take each in turn.
Has the CMA exceeded its jurisdictional reach?
Turning first to the question of jurisdiction.
The starting point is to remind ourselves that the role of the CMA is to protect UK consumers and businesses from anti-competitive mergers. UK merger control gives the CMA jurisdiction to review a transaction either where the UK turnover of the company being acquired exceeds £70 million or where the merging companies will together supply more than 25 per cent of a particular good or service in the UK or a substantial part of it (and where the merger results in an increment to that share of supply).
Following Brexit, the CMA has become more actively involved in the consideration of a wider range of global deals: we review international transactions that would have previously been reviewed by the European Commission where they meet our jurisdictional thresholds and we consider that they may harm consumers or businesses in the UK.
And the deals over which we have jurisdiction are not limited to mergers involving a UK company or ones which principally impact the UK. Sometimes, a deal between 2 non-UK companies that has its 'centre of gravity' elsewhere can still have a material effect on the UK market. The jurisdictional tests in the UK legislation clearly contemplate this broader reach.
In many cases, large international deals will comfortably meet the UK turnover threshold. But some global deals may come within our jurisdiction because they satisfy the share of supply test. This test is not new - it is a long-standing feature of the UK regime. By design it carries a degree of flexibility. And it goes hand in hand with the voluntary nature of the UK regime which can accommodate jurisdictional tests that aren't "bright line" in nature. But our application of the share of supply test is not arbitrary: we have provided a detailed explanation of our approach in published guidance and the approach we have adopted has been upheld when challenged.[4]
Additionally, there are important mechanisms that parties can use to help manage any concerns they have about whether the UK has jurisdiction to review their deal.
This includes the option of engaging our Mergers Intelligence Committee (MIC) - essentially, a quick clearing house for an early view on jurisdiction and whether we think a deal needs to be called in for a phase 1 review. Some time ago, we instituted a practice where parties could submit a short briefing paper to us and receive a quick response that enables most deals we look at in the MIC to proceed without any further inquiry from the CMA quickly and easily - a core strength of our voluntary regime. While we have seen a very substantial increase in the number of briefing papers submitted to MIC, this has not led to an increase in the number of cases called in by MIC which has been on average around 13 cases in each of the past 5 years.
So, on jurisdiction, I do not think that the CMA has exceeded the jurisdictional mandate set down in legislation and while our approach rightly has a degree of flex, to ensure that we can properly assess the impact of deals that may be detrimental for UK consumers and businesses, it is not arbitrary or unpredictable. The CMA only looks at deals where there is an impact on UK markets - and if parties are unsure whether there is a UK nexus, there is a clear path open to merging parties to achieve certainty.
Is the CMA overly interventionist in blocking deals?
Turning now to the second question, is the CMA overly interventionist in its substantive review of mergers?
When this question is posed, it is often framed with reference to one or more of the following considerations:
- Has our frequency of intervention increased?
- Have we shifted the goalposts in terms of when we intervene?
- Are we too quick to dismiss behavioural or other remedies?
In addressing these points, it is important to look at the nature of the cases where the CMA has intervened to block a merger (or where a deal has been abandoned in the face of detailed merger control scrutiny).
My first observation is that the vast majority of competition concerns we identify in merger cases continue to involve existing horizontal overlaps - traditionally viewed as the most obvious form of competition concern that a merger can raise.[5]
These deals typically involve significant market concentration or close competitors. The fact that these types of deals raise problematic competition concerns should not be a surprise to parties or their advisers.
There has been no material shift in our approach to mergers involving orthodox horizontal overlaps and any year on year changes in the number of these transactions being blocked will principally reflect the volume of those types of deals coming to us for review.
That leaves the smaller number of cases which are blocked on the basis of what some might term more "novel" concerns, including threats to dynamic or potential competition and innovation, or relatively more complex theories of harm across supply chains or where there is an accumulation of market power across vertical or adjacent markets.
It is true that a greater number of these deals are being scrutinised - both in the UK and elsewhere.
Partly that reflects an increase in the number of mergers in more dynamic and rapidly evolving markets including, but not limited to, digital and technology markets.
But it also reflects a conscious decision by merger control authorities around the world, including the CMA, to adapt our approach in light of evidence of historic under-enforcement in these areas.
For the CMA, this has been a relatively long and well-signalled evolution. In 2019, the Furman and LEAR reports were issued, both of which identified historic under-enforcement in respect of acquisitions by the major digital platforms.[6]
At the CMA we recognised the need to respond to this evidence of under-enforcement, to adapt our analytical approach and evidence gathering to address the types of competition concerns that might result from these kinds of acquisitions and to ensure that the inevitable element of uncertainty that comes with a forward-looking merger assessment in a rapidly evolving market is not in itself a bar to effective merger control.
In 2021 we updated our Merger Assessment Guidelines to reflect these developments, including a section on potential and dynamic competition, and we signalled our willingness to take decisions about mergers applying a balance of probabilities test based on the evidence, analysis and information that was available to us, even if it came with some inherent uncertainty.
It is the case, then, that over recent years there has been a clear evolution in the CMA's practice in this area of merger control. For example, where the evidence supports it, we've been ready to consider a more competitive counterfactual - to take account of the likely strengthening of the target as a future competitor. We've increased our focus on dynamic competition - a core concern in Meta/Giphy, where our substantive approach was considered - and upheld - by the Competition Appeal Tribunal. We have been ready to consider whether deals might lead to reduced rates of innovation or slower product development. And in some cases, 'flywheel' or ecosystem theories of harm have been considered. There's also been a clear level of interest - both from merging parties and, consequently, from us and from other competition authorities - in the role played by data held by merging companies.
Reflecting on our approach today, I am clear that the CMA will continue to give careful scrutiny to acquisitions in the digital, tech or other rapidly evolving sectors, where those deals involve a party that already has market power in a relevant or related market. And that applies to our consideration both of whether we have jurisdiction to review the deal and to our substantive assessment if we do. We do that to ensure that UK consumers and businesses are protected from the potential adverse impact of any such deal. But each case will be considered objectively and assessed on its individual merits. And it's important to emphasise that it is still only a handful of such cases that are referred to an in-depth investigation, and even fewer that are ultimately prohibited.[7]
And whilst the issues arising in those types of deals will undoubtedly continue to evolve as the markets themselves do, we are committed to providing as much transparency as we can about our approach, including through carefully reasoned decisions, regularly updated guidance, speeches and wider engagement.
A further view that is sometimes expressed is that the CMA is resistant to behavioural remedies or that we are harder to persuade on remedies generally than other competition authorities.
The starting point with remedies is to remind ourselves what they are for. Under the UK statutory scheme, remedies aren't a bargain or a settlement - they must provide a comprehensive and effective remedy to the identified competition concern.
It is important to emphasise that the CMA's approach to remedies is highly transparent. We have published guidance setting out our general approach and, in individual cases, parties have many opportunities to engage with the CMA on remedies. Our remedies process is open and extensive, including the opportunity to avoid an in-depth investigation through offering undertakings in lieu, publication of a remedies notice, sharing a remedies working paper with the merging parties, and a dedicated remedies hearing with independent decision makers in phase 2. Our process is also flexible, and parties can approach the CMA at an early stage to have exploratory discussions about remedies.
Turning to the question of whether we are too quick to dismiss behavioural or other remedies - it is certainly the case that we will give careful consideration to any remedy proposal that is put to us - but we will also apply clear criteria in that consideration.
When we consider a structural remedy, we are not seeking simply to replicate the number of players that were active in the market prior to the merger, but to replace the competitive constraint that would be lost as a result of the merger. This means, for example, that we must carefully consider both the proposed divestment business and the proposed purchaser to ensure not simply that the business can be sold, but that it can be sold to someone who will operate it as an effective competitor.
On behavioural remedies, we have been clear that the bar to show that such a remedy can address a competition concern is high, for 2 reasons.
- First, a behavioural remedy does not typically address the underlying causes of the competition concerns identified by the CMA - instead it seeks to manage market outcomes by controlling the behaviour of the merged entity for a period of time. As such, it may involve arrangements that the parties would not enter into under normal commercial circumstances. And once the remedy expires, the underlying causes of the concern may remain and then have a significant effect on competition.
- Second, even where a behavioural remedy might in theory be capable of restoring competitive outcomes in a market, there are many practical challenges to designing an effective behavioural remedy.
These concerns are not theoretical - they are supported by case study research.[8] One particular challenge is that a remedy can become ineffective or be circumvented as market conditions change, particularly in a sector where technology or business models are changing quickly. Another is that the monitoring and implementation of behavioural remedies can be very challenging, particularly given that there are information asymmetries between suppliers and customers.
Some of these difficulties can be mitigated, of course, by the use of mechanisms such as monitoring trustees or fast-track arbitration. But if those mechanisms aren't effective, the remedy will fail to work properly, or the CMA may end up in the role of a referee in a market for an extended period of time. It is not clear that a remedy that led to that sort of outcome could be viewed as an intervention that has served to prevent the lessening of competition, compared to the position absent the merger.
It is important to emphasise, though, that this does not mean that the CMA approaches remedies with a closed mind. On the contrary, our statutory scheme requires us to consider whether a remedy is first effective and then proportionate: we will not impose a remedy that is more onerous than required to effectively address the competition concern. The CMA's guidance, and our decisional practice, shows that there are circumstances, albeit limited, in which a behavioural remedy may be accepted - typically where structural remedies such as divestment aren't feasible, where the harm arising from the merger will be short-lived, or where substantial relevant customer benefits will be preserved[9]. But we want to be clear with parties at the outset about the standards for accepting different remedies.
Conclusion on intervention
So what is my conclusion on the perceived 'interventionist stance' of the CMA?
First, I do not believe the CMA is over-reaching on jurisdiction. We exercise jurisdiction in line with the statute and to ensure that deals which could adversely impact UK businesses and consumers are properly scrutinised.
Neither do I think that the CMA is overly interventionist in blocking deals. Our approach to reviewing standard, static horizontal mergers hasn't changed. We have stepped up our scrutiny of mergers in dynamic and evolving markets, including where the competition concerns relate to dynamic competition or the potential extension of market power across adjacent or vertical markets. And we have taken steps to ensure that the existence of some uncertainty about the future evolution of a market doesn't itself create a presumption in favour of clearance. But we still apply the same legal test weighing all available evidence and analysis to determine whether, on the balance of probabilities, the merger will harm competition. That is the question we are legally required to answer.
Finally, I don't consider that the CMA is overly interventionist in our approach to remedies. We have set out clearly the basis on which we will accept remedies and the difficulties that we often see with a behavioural solution. But each case is considered on its own merits.
That said, as mentioned at the outset, it is critical that we are as clear as we can be about our approach to these issues. And it is understandable if the CMA is perceived to have become more activist in recent years, given the evolution in the nature of the deals that we are tending to review.
The importance of international alignment
My third reflection relates to the question of whether and to what extent the CMA should pursue international alignment in merger control outcomes.
In the post-Brexit environment, not only have we been reviewing international mergers to a much greater extent than previously, we are increasingly doing so in parallel with other agencies.
The CMA has always worked closely with other competition authorities around the world including the FTC and DOJ in the United States and the European Commission. Following Brexit, the scope, scale and intensity of that cooperation has increased, and today we have a close working relationship with a range of merger control authorities globally[10].
We recognise that in global mergers, coordinating on merger review and remedies, where applicable, is a good idea - it's efficient for both merging parties and authorities. We strive to achieve this as far as possible. But consistency is not an end in itself. The CMA does not seek, or seek to avoid, divergent outcomes but will seek the outcome guided by the evidence that is right for UK consumers and businesses.
Inevitably there will be a few cases where our positions diverge. Our experience has been that this arises in a very small number of cases - typically driven by specific differences in the regimes or in the evidence base across jurisdictions. But divergence is very much the exception, not the rule.
And on questions of process, where we can be flexible, we are open to that, and our experience is that other agencies tend to approach these questions in the same way. While there's no "one size fits all" approach to international alignment, our procedural guidance was updated, in anticipation of our increased role in global mergers post-Brexit, to explain in detail the various levers available to support multi-jurisdictional coordination.
Procedural safeguards - our commitment to an objective and evidence-based assessment
Finally, I want to talk about the important procedural safeguards that are built into our system. Some commentators have suggested that the UK system can be less transparent than other jurisdictions and that we are not subject to effective judicial scrutiny. I would politely disagree.
The UK system (in common with many other jurisdictions but in contrast to the US, for example) provides for administrative decision-making, so it is the CMA taking decisions rather than a court - but our decisions are of course then subject to judicial review.
Because we are decision-makers, not prosecutors, we must weigh up both sides of the argument, and set out an objective and balanced view of the evidence we have gathered and that underpins our ultimate decision. We consult extensively and engage with parties to understand their points, and to give them a fair opportunity to put those points effectively - both in relation to the identification of potential competition concerns and the consideration of possible remedies.
In both our phase 1 and phase 2 processes, the UK model allows parties to make their points directly to the decision-makers, who are required to take their decisions independently and in circumstances that minimise the risks of confirmation bias or groupthink. While the 2 phases are distinct processes, a common thread is that, at each stage, we are challenging and testing the emerging case, both internally and by considering the points being put to us by the parties.
A key feature of the UK merger control regime is the independence of the phase 2 group drawn from the CMA Panel. This independence is hard-wired in the statutory regime and deeply embedded in long-standing practice. The Panel itself reflects a strong and deep bench of expertise drawn from a variety of relevant professions, with first-hand experience of commercial life. If you are interested to understand more about that process, and the way in which independence of decision-making is reflected in the Panel's approach, I commend to you a recent speech by Martin Coleman, our Panel chair, in which he sets out these points in detail. [11]
The final and important safeguard is that our decisions can be challenged before the Competition Appeal Tribunal. The CAT is a specialist appeals body which holds the CMA accountable for the legality and rationality of our merger decisions, and the fairness of our decision making process.
As a result of this institutional design, the UK has a strong and well-deserved reputation for high-quality public decision-making and regulatory independence.
Final reflections
To conclude, I began by describing some of the questions I've been hearing about the CMA's merger control work, and in particular, the impression that is sometimes given that the CMA is overly interventionist and even unpredictable in its approach.
It is certainly not surprising that, compared with 5 years ago, the CMA is perceived as a more active and visible merger control authority. Brexit has meant that we are looking at a greater number of international deals than previously and our evolved approach to assessing mergers in rapidly developing digital and technology sectors in particular has meant that we give deals in these sectors careful scrutiny, especially where they involve established incumbents. That means we will apply theories of harm that match the competitive dynamics of the markets that we are reviewing and we will take a measured approach when weighing the impact of uncertainty on a forward looking assessment.
But I do not agree that we are excessively interventionist or that the UK regime has become unpredictable as a result. Indeed, the changes that have occurred are driven by evidence about gaps or the ways we can improve merger control - to make it more effective and, ultimately, more predictable.
Looking ahead through 2023 and beyond, I am confident that UK merger control is well-placed to maintain competitive markets in the UK for the benefit of people, businesses and the wider economy, helping to meet the UK's ambitions for innovation and investment, and offering a high degree of predictability and certainty. That includes:
- certainty as to our processes;
- predictability about the approach we taken to jurisdiction and the types of global deals we will review;
- clarity about our analytical framework, which reflects widely acknowledged norms of best practice in competition analysis, as well as our best view on how to ensure that merger control works well even in cases where the company is young or the market is nascent, or where the threat is to innovation or future competition;
- commitment to an objective and case by case assessment, which is shaped by the evidence and which rests on the bedrock of our independent decision-making, including the role played by the UK's unique Panel system.
When you look at our decisions, our published guidance and the clarity we can provide through wider engagement with businesses and their advisors, I am confident that for the vast majority of contemplated deals, businesses should have enough clarity to properly assess the antitrust risks that they're likely to face.
That leaves a small minority of deals in the "grey zone", for example where markets are evolving or market participants have significantly different views about how competition works in practice, where it is harder to predict in advance how a merger control investigation might play out. But those are precisely the cases that require a thorough and detailed examination of the evidence to ensure that the interests of UK consumers and businesses are protected. And in those - as in all - cases, we will always be ready to set out our thinking and to listen hard to the arguments being made by all parties before reaching our final, objectively determined decision.
Importantly, outside individual cases, we want to do all that we can to 'join the dots' for all those with an interest in our work, to explain our approach and engage in an open discussion about how to improve the regime. The shared objective is to maintain and strengthen a merger regime that achieves its core objective of protecting consumers, businesses and the wider UK economy, while also being a system that enables deals to come forward quickly and efficiently in a way that supports healthy M&A and investment.
[1] The CMA's merger control is solely concerned with impacts on competition. There is a separate regime for vetting deals based on concerns about national security, administered by the Department of Business and Trade (formerly BEIS).
[2] £2,029.7 million in total over 3 years, giving an average of £676.6 million per year, of a total annual impact of £2.264 billion (30%) (CMA Impact Assessment 2021 to 2022)
[3] The risk of underestimating the impact of errors arising from failing to intervene (sometimes termed 'type 2 errors') has been a particular focus in merger control. See, in the context of digital mergers, the discussion in section 1.5 of the LEAR report.
[4] 1345/4/12/20 Sabre Corporation v Competition and Markets Authority.
[5] In our last reporting year (2012/22), the CMA found a substantial lessening of competition, either at phase 1 or at phase 2, in 20 cases. 17 cases involved horizontal unilateral effects based on existing overlaps.
[6] The Furman Report was issued in March 2019; the LEAR Report was issued in May 2019.
[7] The CMA has referred only 3 transactions involving one of the major digital companies for an in-depth phase 2 review: Amazon/Deliveroo (which was cleared in 2020); Meta/Giphy (which was prohibited in 2022); and Microsoft/Activision (which remains under review). Many other transactions have been cleared unconditionally following a phase 1 review, including Google/Looker, Facebook/Kustomer and Microsoft/Nuance; in addition, other digital cases have been considered by the CMA's MIC team and have not been called in for review.
[8] See, for example the case study research on behavioural merger remedy evaluations published by the CMA in 2019.
[9] We've used behavioural remedies in relation to rail franchise mergers, for example, where the problem is both time-limited and where there is a body that can oversee implementation effectively: see First Rail/West Coast (2019), Abellio East Midlands/East Midlands (2019) and FirstGroup/South Western (2017), each of which involved a price control remedy accepted at phase 1 in lieu of a reference to phase 2.
[10] We have cooperated both with the European Commission and the FTC/DOJ in recent cases, for example, S&P/IHS which was cleared with remedies in all 3 jurisdictions, Nvidia/Arm, which was abandoned last year, Cargotec/Konecranes where we collaborated extensively with the DOJ, ACCC, the European Commission and others, and a large number of cases that have been cleared unconditionally by all the regulators that considered them including SMD/Xilinx, Astrazeneca/Alexion and Microsoft/Nuance.
