The role of the CMA panel in decision making: Merger enforcement and reform
Richard Feasey, panel member, discusses his experience at the Competition and Markets Authority (CMA).
This speech was delivered to the Law Council of Australia’s Competition and Consumer (ACCC) Annual Conference, virtually in London.
Thank you to Linda, Jacqueline and the Law Society for the invitation to speak in what must be challenging times for you all. I will speak about my experience at the CMA Panel, which is a feature of the UK competition regime that is not, so far as I know, replicated elsewhere in the world.
I must emphasise that I am speaking in personal capacity, drawing my experience of a specific set of merger cases and other work. Others may have had different experiences. The CMA publishes detailed guidance on how the merger process works in the UK on its website, for those that are interested.
Some of you know me but most won’t. I worked in the private sector in telecommunications for 25 years, over half as the Policy Director for Vodafone plc. In the decade before I joined Vodafone had used cross-border M&A to move from being a UK business spun out of a defence contractor to become the largest wireless company in the world. I was involved in various mergers, including the first Vodafone merger with Hutchison in Australia in 2009.
I left corporate life in 2013. In 2017 I joined the Panel of the CMA. I have done various other advisory roles in the public and private sector since then, and had some involvement on the VHA merger with TPG.
I’d like to start by saying a bit about my reasons for joining the CMA Panel. In 2002 I led a team dealing with an appeal on the merits of a decision taken by the UK telecoms regulator, Oftel. Regulatory appeals in those days were made to the CMA’s predecessor, the Competition Commission. Our expectations were quite low. We feared the costs in terms of outside advisers and senior executive time would be significant and that the Commission would end up supporting a fellow regulator. The advisers had dealt with the Competition Commission before, but nobody at Vodafone had.
The appeal was heard by a Panel of members, chaired by an academic, Paul Geroski. One of the other members had recently retired as Finance Director of a business of similar scale to Vodafone’s UK business. During the hearings we were asked difficult questions about the way we set prices and ran the UK business. Some we realised we didn’t but should have known the answers to. When we went back we found that some of the answers also proved commercially valuable.
That appeal ended up being a score draw in financial terms for Vodafone. But we all came away feeling we’d been given a tough but fair hearing by people who had no particular axe to grind. The hearing with the Competition Commission had felt more like a difficult Board meeting than a day in court. I was left thinking being a panel member at the Competition Commission might be something to think about in the future.
The role played by Panel members has changed a little, but not much, since 2002. It was also the year the British Government made 2 important changes to the competition regime. The first was to formally require the Competition Commission to stop using a broad ‘public interest’ test to assess mergers and move to the ‘competition effects’ or SLC type test we see today in the UK and Australia.
Secondly, after 2002 Ministers no longer decided which mergers to refer to the Competition Commission and no longer had the final say on whether to block or remedy if the Competition Commission recommended doing this. Instead, there was a 2 stage process with the Office of Fair Trading undertaking the first phase assessment of the merger and the Competition Commission the second phase if the OFT referred it on to them.
In 2014 the Office of Fair Trading and the Competition Commission were then merged into the Competition and Markets Authority, or CMA, of which I am a Panel member today.
Throughout this evolution the Competition Commission and then the CMA have retained the approach, which had actually started back in 1949, of having decisions taken by independent members rather than by a professional civil servant. Today, the CMA Panel consists of 33 members. Each serves one term of up to 8 years, with no possibility of renewal. The number of members has varied between 30 and 45 in recent years.
There is a Chair of the Panel who ensures that it functions properly and that groups of Panel members (a group consists of at least 3 and normally between 3 and 5) are allocated to mergers and to other work. These allocations are made on the basis of availability of members, the mix of expertise and experience required and the need to balance the workload. Each merger or sector enquiry will be chaired by one of 6 Inquiry Chairs, of whom I am one.
Inquiry Chairs receive a monthly salary whilst members are paid a (relatively low) hourly rate for attending meetings and reading papers. The need to avoid conflicts also limits income from advisory or other work which some Panel members might otherwise expect to earn outside the CMA. A consequence of this is that most Panel members are approaching retirement rather than being at earlier stages of their careers when financial concerns loom larger.
The composition of the Panel varies over time. We currently have approximately 7 members who were formerly senior partners in law firms or judges, 4 or 5 partners from economic or other advisory firms, 12 members with significant commercial experience in a variety of businesses, 3 or 4 former regulators or senior civil servants, 4 with expertise in consumer advocacy or social enterprises and 2 academics. I would argue that the Panel provides the UK competition regime with access to a remarkably deep pool of experience and expertise at very modest expense.
Let me next explain how the Panel has operated whilst I have been a member and how it operates today.
The CMA’s permanent staff, who report to the CEO and of whom there are currently about 850, will undertake the first phase of the merger assessment. This takes 40 working days after a merger is formally notified or called in. At the end of this, the phase one decisionmaker, who is a senior member of the CMA permanent staff, will decide whether there is a ‘realistic prospect’ of an SLC. If there is (and the merging parties decline to offer undertakings that would remedy those concerns) then the merger is referred to phase 2. The phase one process therefore functions as a filter to identify a sub-set of mergers which look likely to be problematic and create the pipeline of work for the Panel.
The size of the pipeline informs the size of the Panel that’s needed. The CMA has undertaken an average of 50-60 phase one reviews every year in recent years, which is about 10% of all the mergers identified by the CMA. Around 15% of those phase one mergers were on average referred to phase 2, which leaves the Panel to assess about 10 mergers a year, with 3 or more members allocated to each inquiry. That require a Panel of around 30.
The second phase normally takes 24 weeks but is extendable by a further 8 weeks, which occurs about half the time. This means the Panel group that is appointed can investigate issues in much greater detail than might have been possible in the first phase, including issues that might not have been addressed in phase one at all.
The parties get to make new, often more substantive, written submissions which the Panel members will read and to attend hearings with the Panel group. Importantly, the phase 2 inquiry also has time to undertake its own primary research, often in the form of large consumer surveys, rather than rely only on surveys submitted by the parties.
The Panel group is supported a team of staff drawn from the CMA, some of whom may have worked on the phase one review and will bring their knowledge with them, but others, including the Project Director that leads the staff team, won’t. The phase one staff decisionmaker is not involved in the second phase at all.
Although the Panel members have more time to do their work, we are also held to a different and higher evidential standard than the phase one decisionmaker if we wish to find an SLC - a ‘balance of probabilities’ or 50%+1 basis rather than a ‘realistic prospect’. The phase 2 assessment is exhaustive and Panel members are expected to be engaged on an almost daily basis. The final merger decision typically runs to over 300 pages with another 100 pages of appendices and provisional or interim findings are not much shorter.
Crucially, although the role is part-time, a Panel member is not an advisory role in the way that a conventional Non-Executive would provide advice to a Board. Panel members are the decision makers in the inquiries they undertake and are expected to read almost all of the materials produced by the staff team and a good part of those provided by other parties.
The 24 weeks includes time to consider any remedies. After the enquiry group has issued its decision, the same members may be reconstituted as a remedies group to oversee the implementation of remedies like divestitures.
I like the statutory deadlines in the UK regime. They impose a discipline on the group and staff at the CMA and on the parties and their advisers, and they help create momentum in the process. It means that all phase 2 mergers adhere to the same timetables and processes even if decisions are being made by different groups of members.
Despite concerns in the UK, including from the Government, about the length of time taken to reach decisions in our 2 stage review process, the data suggests that UK cases that involve both phase one and phase 2 reviews are on a par with the US at around 12 months and about 3 months faster, on average, than the European Commission. It is longer than the informal public merger reviews or authorisations undertaken by the ACCC, but about 3 months shorter if the ACCC’s decision is contested in the Federal Court, as in TPG / Vodafone or Pacific Horizon / Aurizon.
The outcomes of the CMA’s phase 2 process vary significantly year to year even if the underlying regime remains the same. For example, in 2015 to 2016 the CMA cleared 8 out of 12 phase 2 mergers without remedies and prohibited none. In 2020 to 2021, we cleared 1 out of 12. I have been a member of 6 merger reviews since joining the CMA in 2017, of which one was cleared unconditionally, 2 with remedies, 2 were blocked and one is ongoing.
There are number of features of the Panel system that I want to discuss briefly. The first is the rationale for having the Panel system in the first place.
I suspect the answer to this question has changed over time. In the days when mergers were assessed not on competition but on wider ‘public interest’ grounds, there was a case for saying that determining where that interest lay was something best left to a group of people drawn from a range of different backgrounds in society rather than a professional civil servant. It has long been common for difficult questions of public policy to be referred by UK Ministers to independent commissions, panels or other ‘public interest’ bodies populated by these kinds of groups.
It might be argued that the shift to a competition effects or SLC approach makes the case for having a generalist decision maker is less compelling than it once was. The economic and legal issues involved in assessing mergers these days are complex and seem to become more esoteric every year. Keeping up with them is a full time job for professionals.
I am not sure there is much to this argument. Firstly, many Panel members have significant expertise in competition law or competition economics or both, even if it is not always up to date. Other Panel members bring other experiences and skills - in IP or commercial law, in M&A and corporate finance, other forms of regulation or consumer interests – which can be equally valuable. To the extent there are any gaps in the Panel’s understanding of the leading edges of competition law or economics, the professional CMA staff can and do provide advice and support.
Crucially, whilst issues undoubtedly arise that require specialist legal or economic advice, a merger assessment is a multi-disciplinary exercise not a specialist one. It requires people with skills and confidence in interrogating evidence, asking good questions and drawing sensible conclusions. These skills are often acquired through experience but they can be acquired working in many different environments.
They are not the exclusive preserve of lawyers, economists or judges. Specialists often take too blinkered a view of the issue, or have too much confidence in their own discipline or their own results. Technical debates can obscure, rather than illuminate, where the consumer interest lies. My own experience is that decision making at the CMA is improved by having diverse perspectives and experiences around the table, which is also the case for many other types of decision making.
The other argument for having the Panel today follows from the decision to merge the 2 stage process into the CMA after 2014. Prior to that, the OFT and the Competition Commission each had their own staff, buildings and organisational cultures, which created a clear institutional boundary between the first phase of the merger review and the second phase. Removing that institutional boundary has unlocked some important efficiencies for the CMA - which was why the change was made – but it has removed some of the safeguards against confirmation bias that came from having an hard separation between the 2 phases of the process. The Panel’s role is to preserve that boundary.
This is achieved by having the Panel be part of the CMA – in institutional terms the CMA consists of both the Panel and the Board – but at the same time operating at some remove from it.
For example, Panel members are appointed by Ministers not by the CMA Board or the Chief Executive. Most of us had not worked at the CMA before joining the Panel and don’t expect to do so when our fixed term finishes. Panel members don’t manage CMA staff or report to, or take instructions from, them. We are not involved at all in the phase one merger review process and have nothing invested in the decision of whether or not to refer to phase 2.
These features of the Panel system ensure that the phase 2 review provides a completely ‘fresh pair of eyes’. The figures I quoted earlier show that phase 2 groups do come to different conclusions on an SLC from the phase one staff decision maker (albeit we are also applying different evidential thresholds).
The Panel system also means that the decision making function is diffused across the Panel as whole, rather than being held by a single individual or a small Board. Each merger is assessed by a different group of panel Members, and neither the merging parties, the phase one staff team or the members themselves know in advance what the composition of that group will be.
Nobody suggests that the allocation of members to groups is done in order to secure particular outcomes, or that Inquiry Chairs or the Chair of the Panel can determine what the group as a whole, who may never have worked together before, will end up deciding. Nor does anybody suggest that Panel groups can be ‘captured’ by either the CMA’s staff or by the parties to the merger. Repeat interactions between parties and the same Panel members are rare. All this ensures that members come to cases and deal with parties without preconceptions.
Dividing the workflow amongst a 30 person Panel system also allows for far more intensive scrutiny of each merger by the independent members than non-executives on a Board could ever hope to provide. No non-executive, however diligent they were, could process the thousands of pages of materials or attend the meetings that Panel members do. They might receive summaries, challenge decisions and advise or question executives, but they are neither expected nor in a position to make the decision themselves in the way that Panel members are.
Let me also say something about the interrelationship between the Panel system and the grounds for appealing second phase merger decisions in the UK. Decisions by the Panel groups can and sometimes are appealed to the Competition Appeals Tribunal. The CAT is the specialist competition court in the UK, similar to the ACT in Australia. It is more legalistic than the Panel in the sense that cases are always chaired by a senior lawyer and the President of the CAT is appointed by the UK’s chief legal officer. It applies different standards to different types of appeal, but all appeals of decisions taken by the CMA Panel, including phase 2 mergers, are limited to judicial review.
I’ll make four brief points about this:
-
First, I have already said that I believe that the nature of the issues means that substantive decisions in merger cases are likely to be better taken by multi-disciplinary groups of Panel members than by a judge
-
Second, I think evidence can be more effectively obtained by having business executives speak directly to a Panel group, straightforwardly and in their own words, than have a process that is too heavily intermediated by lawyers. A prosecutorial or adversarial system is quite alien to the way in which most executives conduct their day to day business. Unlike a cartel case, business executives are not committing a crime when they decide to acquire another company and I don’t think it helps anybody to have a process which treats them as if they were
-
Third, I think having a second stage performed by an independent body like the Panel but subject to judicial review is one way to achieve an equilibrium in which the interests and rights of the different participants are seen to be fairly balanced. In the UK I have not heard many demands from business or advisers for Panel decisions should be subject to a full merits review. Nor have I seen arguments as to why they should be. However, this also depends on the phase 2 process being fair and transparent. The Panel group publishes an Issues letter, extensive working papers on different aspects of the case, and comprehensive provisional findings to allow parties to understand the group’s thinking at different stages of the process and to provide an opportunity to make representatives in light of this
-
Fourth, although the Panel system may require some additional investment upfront in terms of running a 2 phase process and in recruiting and then supporting over 30 independent decision makers, it avoids the costs a full merits review by the courts might otherwise consume. I have never seen anyone undertake a comparison of the end to end total costs of the CMA merger regime against the comparable cost of the merger regimes in the US or EU but my guess is that the UK regime would compare well.
I also think there are some other, less tangible, benefits to the Panel system. One goes back to my experience in 2002 because I think the impressions which members of the business community form when they engage with competition authorities like the CMA or ACCC matter. The regime needs to command the trust and support of the business community if it is to be sustainable.
So, having people on the Panel who have previously been in the same position as the executives who now find themselves in front of the CMA helps overcome the preconception that regulators are run by people who have no real experience of business or of how markets actually function in real life. Having a process that doesn’t alienate business people also helps.
Experience of how businesses and advisory firms operate helps Panel members do their work in other ways. I know from my own experience that synergies in mergers can be real and significant. But I also know they can be reverse engineered to support a decision which has already been made by the CEO on other grounds or to advance the interests of one division or subsidiary of a company competing against others for capital or other resources.
Experience of writing or reviewing documents and knowledge of how they inform decisions being made inside businesses can be very helpful when assessing internal documents in merger cases. I also remember that whilst there will generally be good reasons to merge from a commercial point of view, business executives tend, like the rest of us, to confuse what it is their private or corporate interest with what is in the public interest. I have done it myself.
I should tell you that the British Government is currently consulting on changes to the UK competition regime, including aspects of the Panel’s work. I won’t cover all the issues being considered. Some are intended to make it easier for parties and the CMA to use undertakings to resolve competition concerns earlier in the process, or to reduce the number of occasions in which phase 2 reviews are extended beyond 24 weeks.
One interesting proposal would reduce the size of the Panel – currently just over 30 – to some smaller number but with each member then devoting more of their time to the role. The Government says it hopes this would produce more predictability and consistency in decision making, speed up the process and allow the CMA to pay a salary sufficient to attract people – perhaps younger people at an earlier stage in their career – who might not be able to afford to be members under the current arrangements. I expect this will prompt an interesting debate about how much time Panel members would then be expected to devote to each merger and what implications this might have for the role they are expected to perform and potentially for other aspects of the phase 2 process as well.
I should also mention that Panel members perform other functions at the CMA, although being decision makers on mergers represents the bulk of the work. I have no personal experience of Case Decision Groups, which involve some Panel members sitting alongside executive staff of the CMA to consider abuse of dominance or cartel cases. CDCs are established to make decisions on enforcement cases after a statement of objections has been issued by the CMA staff. They perform a similar function to a phase 2 merger review, but cases generally last much longer. CDCs are not required to include Panel members and do not, to my knowledge, ever consist only of Panel members as phase 2 merger groups do.
The Panel is also the decision maker when the CMA hears certain regulatory appeals, such as appeals of price control decisions that have been taken by other economic regulators. These average about 2 a year.
The role of Panel members in market investigations is slightly different from mergers. As with mergers, the CMA has a 2 stage regime in which the staff undertake the first stage, called a market study, at the direction of the CMA Board. Market studies can conclude with non-binding recommendations or guidance, lead to enforcement actions using other competition powers, or accept undertakings from businesses but, just as mergers which can only be prohibited by the Panel, binding remedies can only be imposed after a market investigation by a Panel group.
Market investigations are less common than either market studies or second phase merger reviews, recently averaging about one a year. This means individual Panel members may only do one market investigation or CDC, or none to all, during their term at the CMA. In many respects, the same interactions with parties and weighing of evidence is undertaken in regulatory appeals or market investigations as in a merger assessment and many of the points I have made about the Panel system for mergers should apply to these other activities.
Finally, the CMA is currently establishing an Office of the Internal Market, which is a post-Brexit arrangement to ensure the free flow of goods and services between the different regions of the United Kingdom. Panel members are currently being recruited for this. The other significant new development, which is the Government’s proposals to legislate for a new Digital Markets Unit to regulate digital platforms with ‘strategic market status’, does not currently appear to envisage any role for the Panel.
There seems to be an active debate in both the UK and Australia at the moment about what changes or improvements might be made to the respective competition regimes. There is always room for improvement and I am not for one moment suggesting the UK’s Panel system is perfect, that it would be invented if it didn’t already exist or that is necessarily something to be replicated elsewhere.
On the other hand, the Panel has survived since the late 1940s and it continues to perform a key decision making role in the UK competition regime, in the ways and for the reasons that I have tried to explain to you today.
Thank you very much.