The past year has seen unprecedented political intervention in the work of antitrust authorities.
In the US, President Donald Trump dismissed two commissioners at the Federal Trade Commission in defiance of Supreme Court precedent while a former senior official at the Department of Justice alleged that politically connected lobbyists had been able to influence case decision-making. In the UK, the chairman of the Competition and Markets Authority was forced to resign by ministers who thought the CMA was insufficiently focused on the government’s economic growth agenda. And in Europe, there have been multiple signs of Commission President Ursula von der Leyen seeking to exercise greater control over DG COMP, while heads of government and CEOs have called for EU competition rules to be changed to allow the creation of European champions.
But what should competition authorities do in the face of this interference? At a time of political turbulence and populism, simply demanding that politicians respect precedent and convention is unlikely to do much good. Those who wish to protect the independence of antitrust agencies must argue their case from first principles and be ready to answer some deeper questions. To whom, if anyone, should competition authorities be accountable? Why? And how?
These questions matter because antitrust agencies are not merely enforcers of a clear set of statutes and a well-established body of case law. They exercise discretion and they create policy. The people who get appointed to lead these authorities, and the agenda they choose to pursue, have a big impact on enforcement outcomes.
This is not an accident. When the bill to create the Federal Trade Commission was going through the US Congress in 1914, it became clear to the administration of Woodrow Wilson that it would be hard to win legislative support for banning lots of individual specified forms of “unfair competition”. The easier way to get the bill passed would be “to prohibit unfair methods of competition and to leave it to whomever was administering the law to determine whether a method in a particular case was unfair and harmful or not.” This has created the potential for conflict with politicians over the years. For example in the 1970s congressional committees wanted an activist FTC and that is what they got, with the FTC issuing rules governing the conduct of entire industries. But those industries subsequently engaged in furious lobbying, after which congressional committees changed their minds, criticising the FTC as a “run away” agency and cutting its budget.i
We see a similar issue in merger control. The UK CMA is empowered to block mergers that are likely to lead to a “substantial lessening of competition” but the term is not defined in statute or in guidance from central government. In setting out their approach, agency heads can sometimes give the impression that there is a binary choice between being for or against competition. Take this from Sarah Cardell: “The economic evidence is overwhelming that competitive markets – not monopolies – provide the best environment in which to encourage investment, innovation and productivity.” (I quote this not as a criticism of Sarah. It was me who drafted those words.) Or this in the past week from Hiroo Iwanawi, Secretary-General of the Japanese Fair Trade Commission: “The JFTC has embraced a simple message: no competition, no growth.” But in deciding whether to clear or block a merger an antitrust authority is rarely making a straightforward choice on whether to favour competition. It is making a judgment call over how much competition and how much enforcement is needed, and in doing so it acknowledges there may be trade-offs – for example with efficiency gains foregone – and that there can be error costs if it gets its decision wrong.
We can see this in a range of sectors. In recent years, several authorities have reviewed 4:3 mergers in the telecoms sector. But why should the line be drawn there rather than at 3:2 or – if competition is such a good thing – why not at 5:4? I worked on the Sainsburys/Asda merger which came just as the CMA was toughening its approach to merger control and the agency blocked that deal based on a materially lower threshold for prohibition, through its GUPPI decision rule, than in previous retail mergers. And this is even before we get to big tech, where my arrival at the CMA happened to coincide with the publication of the Furman report, which called for more active enforcement in digital markets.
It is not as if these shifts reflected a changing academic consensus which could easily be incorporated into enforcement. There were, and are, live debates amongst the experts themselves. On the one hand, evidence has been put forward in the US, the EU and the UK that competition may be weakening across the economy on a range of indicators and that a tougher approach to enforcement might be required. Meanwhile, other academics have cautioned that some of these indicators might not actually point to a widespread decline in competition at all and that there is no case for a toughening of enforcement.
Deciding the enforcement stance of a competition authority is therefore a significant policy choice, so who should have the power to make that choice?
The traditional view is that it should be the leaders and staff of the authority itself because they are the ones with the expertise, rather than politicians who are at risk of capture by business, and that any necessary correction should be provided by the courts. However, I think that this is increasingly problematic.
First, if the experts disagree on major questions of antitrust policy, what mandate does an individual pro- or anti-enforcement head of agency have for the policy stance that they pursue? In the US, the agency heads and judges who took a less aggressive approach to enforcement from the 1970s onwards did not generally express scepticism about enforcement in their confirmation hearings so their views were not tested by those approving their appointment. In the UK, the CMA took a tougher enforcement stance under Andrea Coscelli but the chief executive of the CMA is not subject to any pre-appointment hearing at all. Putting policy choices in the hands of antitrust agencies themselves means that enforcement may be shaped by their leaders’ personal policy preferences, and these preferences might not be publicly known in advance.
Second, it is not only politicians who are at risk of capture. Until very recently a bigger concern among academics and campaigners in the United States and Europe was that the antitrust agencies were the ones being captured by business, partly because they increasingly rely on technical economic evidence provided by the merging companies themselves and partly because of a revolving door between the agencies and private practice. Indeed, in the US, it was President Biden and his political appointees who criticised the “retreat” by the antitrust agencies over the previous four decades and who pushed for more robust competition enforcement on behalf of consumers. This is the opposite of the dynamic we currently have in the UK and, if the CMA softens its enforcement stance, we may well see it in future on this side of the Atlantic too.
Third, competition policy increasingly involves trade-offs which are difficult for an unelected agency to make. For example, when I started out in economic regulation I remember being taught that independent agencies should not get involved in distributional questions. But there is a growing literature on the links between market concentration and economic inequality (for example here, here, here and here) so a government which wants to reduce economic inequality might well want to push its competition authority for more robust enforcement. On the other side of the ledger, there are policy trade-offs which might legitimately make an elected government want to soften enforcement in some areas. An increasingly important one is with foreign policy. When the UK’s independent competition regime was established by my old boss Gordon Brown, the main targets of enforcement were domestic incumbents while the beneficiaries were international companies seeking to enter the British market, so a robust competition policy was either neutral or positive for our foreign policy. But recently the main targets of enforcement appeared to be big American companies, and this came just as the UK’s alliance with the US is coming under the most strain since the end of the second world war. In that context, it is impossible for the elected government to ignore the geopolitical implications of the CMA’s enforcement.
Fourth, the courts are not well placed to adjudicate on all these issues. They can judge whether an antitrust agency is acting within the bounds of its authority or is sufficiently protecting the rights of the companies involved; and if they conclude it is not doing so they can impose a correction e.g. as in the landmark Airtours case in the EU. But if the issue is how to decide between competing public policy priorities, the courts have even less of a mandate than the antitrust authorities to resolve this because they are even further removed from the democratic process. Elected politicians have to be involved.
A common retort is that if legitimate wider policy issues are raised by a merger case, it is better to have a narrow route for governments to intervene on carefully defined “public interests” grounds e.g. as under the EU Merger Regulation and the UK Enterprise Act. Governments could even create a new ground for intervention in a live case, as the UK government did for example in Lloyds/HBOS in 2008. However, I think there are two problems with this.
First, if politicians are going to start intervening more in individual merger cases, this will encourage exactly the kind of corporate lobbying that competition authorities and lawyers all want to avoid. The prospect of greater ad hoc political intervention is also hardly going to make the regime more predictable. A far better approach is that, where a government is concerned that antitrust enforcement could come into conflict with other policy priorities, it should give some overall policy direction to the agency (as in the recently published strategic steer in the UK) and then leave it to the agency to take its individual case decisions independently.
Second, if an antitrust agency pushes too hard for politicians to keep their tanks off its lawn, it risks an even greater political correction. The period between the 1980s and 2000s saw a trend toward greater independence for competition authorities and a greater focus on objective economic analysis in decision-making but it can be easy to forget how historically unusual that period was and therefore how fragile that settlement remains. In the UK, Conservative governments played a key role in setting up independent regulators and giving them the status of “non-ministerial government departments” with a high degree of insulation from political interference. However, as I have written elsewhere, this tended not to be out of a commitment to independence in principle, but because regulatory independence was a tool to achieve another policy objective at the time, usually around winning investor confidence. If investors no longer demand regulatory independence, support for it on the political right is likely to weaken, and conservatives are likely to return to their traditional prioritisation of parliamentary sovereignty and the belief that those exercising power under the royal prerogative or under statute should be properly accountable to parliament, through ministers.
On the centre-left, it was the Labour governments of Tony Blair and Gordon Brown – who I served as a special adviser – that entrenched competition authority independence in the UK. I have written about this in more detail here, including how Brown was strongly influenced by his experience of giving operational independence to the Bank of England. Bank independence became a symbol of New Labour’s commitment to economic credibility and it could then be applied in other areas, including to the competition authorities and regulators. The process was eased by the fact that for most of that period interest rates were low and utility bills were falling, so the independent institutions appeared to be delivering. By contrast, Keir Starmer and Rachel Reeves took office at a time when rates and bills were rising and when many of the regulators – from Ofwat to the FCA as well as the CMA – were mired in controversy and attacked for being bad for business and investment. The inheritance and perspectives of today’s Labour ministers are very different from those of their predecessors.
Sarah Cardell and Doug Gurr have taken flak for how they have adjusted to this new environment – for example with a range of process changes and a more accommodating approach to merger remedies – but what other approach realistically was there? If the CMA had ploughed on as the “world’s policeman on mergers” it could have won plaudits from neo-Brandeisians internationally but it would have led to continued tension with the government on live cases, undermining confidence in the CMA’s independent case decision-making, and if the government felt its broader policy concerns were not being addressed it would have resorted to legislation to emasculate or even abolish the CMA. Going out in a blaze of glory would hardly have been in the long-run interests of UK consumers. Agency heads cannot avoid paying attention to the political context in which they operate and grappling with questions around democratic accountability.
How, then, should these issues be resolved? I think there is a role for governments, for the agencies and for the courts.
Governments can set a broad policy direction for competition authorities on how they should prioritise their work and on how they should manage policy trade-offs. In the US, Assistant Attorney General Gail Slater has set out a philosophy of “America First antitrust”, which seeks to combine vigorous enforcement of the cross-economy antitrust laws with deregulation of sector-specific rules, all in the name of promoting economic liberty. In the UK, the Labour government has set out a new strategic steer which tells the CMA to focus on cases where it can have the biggest impact for UK consumers; to take into account what other authorities are doing when it reviews international deals; to take a proportionate approach to remedying competition problems that it identifies; and not to deter investment, either by its substantive decisions or by the signals it sends about what it is like dealing with UK regulators. One can agree or disagree with the content of these steers but they strike me as legitimate ways to align antitrust enforcement with the broader economic priorities of the elected government.
But a line must be drawn when it comes to the nature of political involvement in individual cases. As Bill Kovacic has argued recently, “There has to be an accountability regime but you can’t have political leadership telling the agency to punish enemies, reward friends and otherwise simply be party to a negotiation between top political leadership and individual business interests. Once you do that, confidence in the entire system of government tends to erode.” That is why the allegations about lobbyists influencing case outcomes at the DoJ are so damaging. In the UK, ministers have pledged to protect the CMA’s independent decision-making and observers will be watching closely to see if this is followed in practice.
This leads to the question of whether it is realistic to draw such a line. Even if politicians insist that they only want to issue a general policy steer, is there not a risk that having done so it will open the floodgates to lobbying and will politicians really be able to resist the temptation to intervene in individual cases? But I think this ignores the fact that, in the UK, the increase in lobbying began not when the UK government removed the CMA chairman but much earlier when the CMA began taking a tougher stance on mergers (see here, here, here and here). There were suspicions that ministers were intervening behind the scenes on Microsoft/Activision and the AI partnerships. The lobbying floodgates were already open and the question became what to do about it. Furthermore, I think it is mistaken to assume that ministers have a desire to intervene in controversial merger cases. Doing so generally brings them flak from all sides while distracting them from their policy agenda. In my experience, at least in the UK, they would much prefer to be able to rebuff calls for intervention by saying that any relevant policy trade-offs had already been taken into account and that any outstanding concerns were ones that the CMA or the courts could address. The problem is that in the past few years Conservative and then Labour ministers found it increasingly hard to make that argument because, as explained above and here, there did appear to be significant unresolved policy trade-offs.
There is also more that antitrust authorities themselves can do to strike the right balance between being more accountable to their stakeholders and discouraging illegitimate political interference. A starting point may be to be more modest about the goals of antitrust and what it can deliver. If heads of antitrust agencies say that their purpose is to “protect democracy” from corporate power or that they are part of the solution to a “growing discontent about capitalism”, they are likely to attract interest from politicians – including from those whose preferred enforcement stance may be very different from their own. Antitrust agencies can be players in solving major societal issues, or they can be technical experts doing a job from which politicians should steer clear, but it is very hard to be both.
Agencies can also be much clearer about the policy agenda they are pursuing and how they are taking into account the policies or strategic steer of the elected government. The Annual Plan which the CMA is required to lay before Parliament at the start of each year would be the ideal place to do this, but historically it has tended to function more as a summary of work that is currently in train than as a genuine guide to changes in policy (for example the agency’s toughening of merger control from 2018 was not announced or explained in the Annual Plan), while the Annual Report published at the end of the year contained not a single key performance indicator. If an agency does not report at the start of the year on what it intends to achieve, and at the end of the year on whether it has been successful, it can be quite hard to hold it to account. Under Sarah Cardell, the CMA is doing more to explain its policy stance publicly, for example at Chatham House and in other speeches, and as a next step it should encourage an open conversation about how its success can be judged.
This transparency should extend to an agency’s dealings with the government and with the legislature. Agencies should actively seek to engage with the media and parliamentary committees about the policy choices they are making; and those lobbying the agency should be warned that what they say will be made public, reducing the risk that ministers, advisers or lobbyists are able to impact policy or decision-making privately.
For this to work, parliamentary committees need to be more effective in their scrutiny. After the CMA chairman and chief executive gave evidence to the House of Commons Business Select Committee on their work in June 2019, they were not called again for another four years, until May 2023 in the heat of the controversy over Microsoft/Activision. If Commons committees are too busy to provide in-depth scrutiny of competition policy absent a live and controversial case, the House of Lords may be more able to provide focused attention, for example through the Industry and Regulators Committee. Furthermore, pre-appointment hearings should be extended to include the nominees for chief executives of regulators as well as their chairmen, given the extent to which chief executives are able to influence the direction that an antitrust agency takes.
It is true that greater legislative scrutiny of competition authorities also creates opportunities for vested interests to influence the debate. But the cross-party nature of parliamentary committees, and the accountability of MPs to their constituents, means they can be responsive to a broad range of influences. Indeed, the Commons Business and Trade Committee under the chairmanship of Liam Byrne has called for tougher enforcement by the CMA, for example in consumer protection law. Their work also tends to be conducted transparently, with oral evidence taken in public and written evidence being published. Where the committee has taken evidence on a live deal, in particular Vodafone/Three, it has forced supporters and opponents of the deal to make their case in a non-technical way to a non-specialist audience, and to respond directly to each other’s arguments, which I think is good for public discourse on competition policy.
All this still leaves an important role for the courts in holding agencies to account. They are far better placed than politicians to review an agency’s interpretation of the law and the facts of a case, given their expertise, so where politicians hear complaints from companies on this score they should direct them to the relevant judicial process. They are also more likely than politicians to take an unbiased approach to complaints from companies that they have been treated unfairly during the investigation process. But what they cannot do is judge whether an agency is following the right policy stance or striking the right balance between competing policy priorities. That is where democratic accountability is more important.
Finally, if governments and antitrust authorities do not take these steps to increase transparency and accountability, what will happen? If the policy stance of antitrust authorities is set in private it makes for an unpredictable regulatory environment, which is hardly good for business decision-making, investment and growth. And if elected governments cannot address policy trade-offs at the level of an agency’s policy-setting process, they will be forced to do so in individual live cases, either privately or publicly, which is bad for an agency’s independent decision-making. The resulting uncertainty may be good news for those of us who advise companies on these issues but it is not good for the competition regime as a whole or for the consumers it is supposed to serve.
In conclusion, antitrust agencies across countries and over time have made very different choices about the enforcement stance to pursue. These choices can be left to the agencies themselves on the grounds of their superior expertise, but very rarely are agency heads appointed with a mandate to pursue a particular approach, and so left to their own devices there is a risk that they will pursue their own personal agenda, or the agency’s institutional interests of accumulating more power and resources, or that they will be captured by the corporate users of the system. Courts can provide a corrective where an agency misreads the facts or the law, or where it treats parties unfairly, but they are not well placed to adjudicate on an agency’s overall policy stance or on how it balances competing policy priorities. The involvement of democratic institutions is therefore a necessary part of the competition policy-making process and this can be done in a way that provides a clear steer to agencies and enhances the democratic legitimacy of their work while protecting their decision-making on individual cases from lobbying and interference.
i Wilson, J. Q., 1989. Bureaucracy: What Government Agencies Do and Why They Do It. New York: Basic Books, pp. 82-83
