Andrea Coscelli: Ahead of the curve – Bannerman Competition Lecture

photograph: Andrea Coscelli

It is an honour to be joining you all, virtually, to give the Bannerman Competition Lecture.

In his time, Ron Bannerman was a pioneer and a forward-thinker. It seems self-evident now that the business cartels he confronted caused harm to Australian consumers and the economy. But his work over 2 decades to cut back at what he later described as the ‘astonishing web of restrictions’ [footnote 1] across Australian industry was conducted in the face of considerable resistance and antagonism. His commitment and persistence were crucial to the development of the strong culture of fair and open competition that now prevails in Australia.

The ACCC has continued that legacy of leading from the front. Under the Chairmanship of Rod Sims, it has done pioneering work on digital markets, and competition and consumer protection more generally, that has significantly influenced the CMA’s own work.

How competition authorities stay ahead of the curve; how we deploy our current toolkit; and the additional tools we might need, in order to deliver for consumers are the themes of my talk today.

Operating, as we do, predominantly at the level of individual firms and transactions, it is easy to lose sight of the wider importance of this. Robust and timely intervention to cut back at that web of anticompetitive agreements and practices, and to stop anticompetitive mergers, carries wider economic benefits: stronger productivity, lower inequality. These things are important at the best of times. As our economies begin to recover from the devastating consequences of the coronavirus pandemic, they matter more than ever. [footnote 2]

Staying ahead of the curve as markets and business practices evolve and change

So how do we stay ahead of the curve as markets and business practices evolve and change? Some might say that we know the answer already.

That, despite quite different starting points, there has been a convergence of international opinion about how to deliver well-functioning, competitive markets.

To some extent, this is true. 3 principles have been particularly influential over the last 20 to 30 years.

  1. The first is that economic analysis should sit at the heart of assessments of whether particular conduct, or a particular merger, should be prevented. And the focal point of that analysis should be the effects on consumer welfare. This is reflected in the institutional design and statutory objectives of many of the competition regimes established during the 1990s and 2000s, including the UK’s. [footnote 3]

  2. The second principle is that we should be cautious in the face of uncertainty. Over-enforcement, it is argued by some, can have a ‘chilling’ effect, deterring investment and innovation that could benefit consumers. [footnote 4]

  3. The third principle is that competitive markets – supported by competition authorities working under those first 2 principles – are preferable to regulation: or as Ron Bannerman himself put it, ‘competition itself is an important regulator’. [footnote 5]

I think it is now widely agreed that the consensus around these principles is showing signs of fracture. There are calls for competition authorities to look beyond consumer welfare and take other objectives into account. [footnote 6] There are arguments that we have been overcautious and too ‘narrow’ in our assessments. And the case is increasingly made that many markets, especially in the digital sphere, are under-regulated. [footnote 7]

There are a number of possible reasons for this shift in thinking. One of the most important explanations, to my mind, goes back to my theme: that as the digital revolution has transformed markets, competition authorities have struggled to stay ahead of the curve. The largest digital platforms, it is argued, have attained unassailable positions of dominance right under our noses. The ‘web of restrictions’ is reasserting itself, this time in new guises, such as algorithmic collusion, anticompetitive self-preferencing, and acquisitions of nascent rivals. And we are struggling to address these concerns in a timely and effective way.

I’d like to say a few words on where the CMA stands on each of these principles. I will say at the outset that I think they remain relevant. And individually, none of them is necessarily wrong. But in combination, they have led to a number of legitimate concerns given the evidence in front of us that we have fallen behind the curve in dealing with the competition issues arising from the growth of digital platforms.

Externally, we need to look at whether we are empowered by the legislation that we apply to act sufficiently swiftly and effectively in the face of developments and behaviour that stand to harm consumers. In the UK, at least, we think there is scope for improvement, and we have said it publicly over the last couple of years. [footnote 8]

And internally, we need to look at whether we are taking the necessary risks; and whether the lessons from decisions made yesterday – and especially decisions not to intervene – are bearing, as they should, on decisions made today.

For the CMA, this challenge to adapt and develop is given extra impetus now that having left the EU, we have acquired responsibility for major antitrust and mergers cases – many of which will be in the digital space – that were previously reserved to the European Commission.

The importance of economic analysis

Turning to that first principle, that our decisions should be based on economic analysis of price, quality, and innovation that benefit consumers in both the short and long run. Economic analysis sits at the heart of our decisions, and that will remain so. And I think it is right we should stay focused on the consumer interest; or in other words, what they stand to gain or lose. This is our current statutory duty.

If, as some suggest, we ask competition authorities to focus also on other objectives – the protection of workers, for instance; the promotion of small businesses; or the advancement of industrial strategy objectives – then we must face up to the question: what happens when those objectives come into conflict? What do we do when a merger might harm consumers but save some jobs? These are not easy questions – and, many would contend, not appropriate questions – for independent competition authorities, staffed by unelected officials, to answer.

From a practical point of view, the system risks becoming unpredictable and hard to administer. As Bill Kovacic – who, among many other roles, serves as a non-executive director on the CMA’s Board – puts it:

You need to explain the exchange rate [between the different objectives], and explain the hierarchy of values. What gets weighted, and by how much, in each case? [footnote 9]

It is also important to reiterate that we believe that consumer welfare should be interpreted broadly. Non-price factors, such as quality and innovation, may be the primary focus of the analysis in certain cases. Similarly, in a merger between 2 firms in the digital space, we may well be concerned with the privacy implications or the impact on innovation. Even an issue like the security and resilience of supply chains – a concern that has intensified as a result of the pandemic – is something that could form part of our consideration, to the extent that it affects consumers over time. So, I do not think competition policy would be well served by shifting its focus away from consumers, or by broadening it to include other considerations. However, I think it is important to ask whether we are doing all we can to deliver on the terms we have been set.

Here, I think there is scope for improvement. By way of example, the evidence we receive in mergers and antitrust cases comes overwhelmingly from the businesses concerned and their advisers. Naturally, this evidence goes to great lengths – often through highly complex economic analysis – to make the case that the conduct or merger in question is not harmful to competition; and that, in fact, it stands to benefit consumers. And naturally, the CMA looks at this analysis with due scepticism.

But the consumer voice in this process is largely silent. And, if anything, the silence is still more deafening during litigation, when our decisions are appealed against. For 2 reasons, we need to think about how that consumer voice can be amplified.

  • first, amplifying the consumer voice would help to entrench the legitimacy of the competition regime, and to address reasonable concerns that it is systematically biased against the consumers whose interests it purports to represent.
  • second, at a more practical level, it can enrich our evidence base and improve our analysis.

Caution and uncertainty

What about the second principal – that we should be cautious in the face of uncertainty?

On that, I would make 2 observations:

  • first, we have learned a great deal about digital markets; and in many respects, there is now less uncertainty than there was, say, 5 years ago. We now know enough, for instance, to make a compelling case that a new regulatory framework for the most powerful platforms is required to promote competition in a number of digital markets. In terms of our casework, we have carried out a number of reviews of our merger decisions to ensure the lessons from past cases are brought to bear on our current work. We need to continue to develop and build our baseline understanding of these markets: to reduce uncertainty, to support our work, and to inform wider policymaking. The CMA’s recent work on the implications of algorithms for competition and consumer policy, on which we published a report in January, is an example of that. [footnote 10]

  • second, we must recognise that a degree of uncertainty is inevitable, particularly when it comes to our merger assessment. We must acknowledge it. We must minimise it where we can. But its presence cannot preclude us from acting. Wherever we act in the face of uncertainty, there will be a risk of over-enforcement, and we recognise, and seek to guard against, the consequent risks of a ‘chilling effect.’ But it must be balanced against the risk of under-enforcement that comes from doing nothing.

In many digital mergers the risk of consumer harm from under-enforcement is high. We know that many markets in this space can ‘tip’ towards monopoly. We know that can happen quickly. We know that ‘self-correction’ through market forces is unlikely; and correction through antitrust enforcement is often too slow to avoid irreversible harm.

In the CMA’s advice to the UK Government on a new Digital Markets Unit, which will be responsible for applying a new pro-competition regulatory regime to digital markets, we have also made recommendations about changes to the merger regime for the most powerful digital platforms — that is, those we think carry the highest risk of consumer harm from under-enforcement. The additional concerns we might reasonably be expected to have about acquisitions by such companies (such as the risk of further cementing their market power for many years) could be reflected in a lower standard of proof than we would normally apply before intervening, and this is what we have recommended to Government.

Many of us are now familiar with the statistic that – between 2008 and 2018 of the 400 acquisitions made globally by the 5 largest digital firms – none has been blocked by competition authorities. But it remains a powerful one. It is very hard to look at those numbers, to look at the state of the relevant markets today, and conclude with hindsight that the balance has been struck correctly.

Competition and regulation

Let me now say something about that final principle: that competition is preferable to regulation.

It may be surprising to hear – given that we have recently advised the Government on new regulation for digital platforms – but I remain convinced that this principle, too, remains a good starting point. From pharmacies to legal services, we have seen through our work over the years that well-intentioned but poorly-designed rules can end up protecting incumbents at the expense of consumers and potential competitors.

Effective competition, on the other hand, can often get the same or better results for consumers. By providing the incentives to innovate, to keep prices low and to keep quality high, it is in itself (as Ron Bannerman said) a powerful regulator.

But equally, we need to get away from the simplistic idea that regulation is usually harmful to competition. Designed well, regulation can help competition work better, and improve outcomes for consumers and society at large. In banking, for instance, CMA intervention through interoperability to enable customers to access and share their banking data – essentially a regulatory intervention – has paved the way for new products and services.

Likewise, the CMA, and others, including the ACCC, have concluded that the traditional competition policy toolkit – antitrust enforcement and merger control – is insufficient to address the competition issues arising from the growth of the most powerful digital firms. Procompetitive regulatory intervention (such as interoperability or non-discrimination in rankings) is justified, and indeed necessary, if we are to sustain innovation and growth, and avoid consumer harm, in this increasingly important sector.

With this in mind, we welcome the UK Government’s commitment to establish a Digital Markets Unit within the CMA from April 2021 to oversee a new regulatory regime for the most powerful digital firms. We are working hard to support Government in its work to establish this new regulatory function and framework. And we have today published a refresh of our Digital Markets Strategy which sets out our priorities across our digital work for the coming year.

For competition authorities with responsibilities for consumer law enforcement, it is important also to examine what more can be done to sustain consumer trust, confidence and engagement with digital markets.

Digitalisation has provided huge benefits to consumers, including a great deal more choice. But it’s also provided new opportunities for our choices to be manipulated: drip pricing, scarcity messaging, fake reviews, unlabelled social media endorsements. If the decisions consumers face are too complex; if they cannot rely on the information they are presented with; or if they are manipulated into making choices against their interests, this harms the competitive process. So the 2 efforts, more robust consumer protection and more robust competition, are complementary and support each other.

Both the CMA and ACCC are doing that through their casework. For example, we’ve both acted on fake and misleading online reviews. And we have both taken cases against the secondary ticketing platform, Viagogo, to ensure it gives consumers accurate information. [footnote 11]

So, to sum up so far. Staying ahead of the curve requires us to get closer to consumers; to build and develop our knowledge, especially of digital markets; to give due weight to the risk of under-enforcement; to adapt and update our toolkit; and to contemplate procompetitive regulatory intervention where traditional competition policy is falling short.

As if that wasn’t enough, it also requires something else.

Strong international co-operation

Many of the challenges arising from the digital revolution are shared.

There is also a shared idea of what we need to do; that, for the largest digital platforms, robust competition policy needs to be bolstered by ex ante regulatory tools.

That has been the conclusion in the EU, in the UK, in Germany, in Australia, and elsewhere.

The task now is to achieve coherence.

That’s why, as part of our advice to Government, we made clear the importance of the new Digital Markets Unit being able to share information and forge networks with international competition and consumer agencies.

This is not about having identical rules.

It is about recognising and managing differences; and working in a spirit of constructive co-operation to meet a common objective.

The largest international banks are supervised according to different requirements in different jurisdictions. But with a strong shared objective (in this case, financial stability) and close regulatory co-operation, this can be made to work.

Likewise, so too can our approach to the largest digital firms.

That co-operation will also be important in the CMA’s traditional antitrust work as it takes on major cases, including many in digital markets.

I look forward to continuing to work closely with the ACCC as we seek, together, to stay ahead of the curve; and to make markets – new and old – work well for consumers.

[footnote 1] Bannerman, R.M. (1985) Development of Trade Practices Law and Administration, Australian Economic Review, 18: p.85.

[footnote 2] Thomas Philippon has shown persuasively how a decline in competitive pressure in the US has led to higher prices, lower investment and lower productivity growth (The Great Reversal, Belknap Press, 2019). John Van Reenen and others have demonstrated a robust empirical relationship between competition and productivity, with one of the principal mechanisms being improvements to management practices generated by greater rivalry between firms (see, for instance, Van Reenen, J. (2011) Does competition raise productivity through improving management quality? International Journal of Industrial Organization, 29(3), p.306-16).

[footnote 3] See, for instance, a 2003 note from the OECD Secretariat for the Global Forum on Competition, which found “Among OECD countries, there appears to be a shift away from use of competition laws to promote what might be characterised as broad public interest objectives, and use of public-interest based authorisation procedures, exemptions or political over-rides (collectively, “public interest objectives”) in competition laws, that contemplate a consideration of factors which extend well beyond what appear to be the generally accepted “core” competition policy objectives of promoting and protecting the competitive process, and attaining greater economic efficiency (the “ core competition objectives”). This shift is reflected either in an elimination of, or less frequent or more restricted use of, legal tests or political over-rides in domestic competition laws that would permit (i) an anticompetitive merger or restrictive trade practice to proceed on the basis of broader public interest considerations; or (ii) a procompetitive merger or trade practice to be blocked or remedied on the basis of such concerns.

[footnote 4] Some go further and argue that under-enforcement is intrinsically less costly because it can correct itself through the market. Again, that idea found analytical expression in the error-cost framework pioneered by Frank Easterbrook. But it reflects a more general and widely-held idea that has its origins in classical liberal economic thought: namely, that precipitous intervention by the state in markets can do more harm than good. Jonathan Baker (2019), “The Antitrust Paradigm: Restoring a Competitive Economy” convincingly summarises many erroneous arguments against enforcement in US antitrust (chapter 5).

[footnote 5] Bannerman, R.M. (1982), Competition as the regulator, University of New South Wales Law Journal, 61: p.79. This too has Chicago School pedigree, in George Stigler’s seminal work, which showed the tendency for regulators to become captured by vested interests, at the expense of consumers.

[footnote 6] See, for instance, Lina Khan (2018), “The New Brandeis Movement: America’s Antimonopoly Debate” or Michelle Meagher (2020), “Competition is Killing Us.”

[footnote 7] A recent example comes from the American Economic Liberties Project, in their paper The courage to learn” (January 2021), which makes a case for stronger regulation across a number of US sectors, including digital (“Google, Facebook, and Amazon are almost completely unregulated, and their consolidation of power undermines democracy”).

[footnote 8] See, for instance, letter from Lord Tyrie (then CMA Chair) to Greg Clark, 21 February 2019,

[footnote 9] Transcript of FTC Hearings on Competition and Consumer Protection in the 21st Century, 21 September 2018, p.44.

[footnote 10] CMA, Algorithms: how they can reduce competition and harm consumers, January 2021.

[footnote 11] In the UK, we think that there is a strong case to strengthen consumer protection law. For example, unlike in Australia, fines cannot be levied on firms that break consumer law. Strengthening our consumer protection enforcement powers so that they are brought into line with our competition law powers will help us to deal more swiftly and robustly with harmful practices, both online and offline. We’ve made proposals to the Government and are hopeful they will be taken forward. And we continue to examine what further legislative change may be necessary to protect consumers online.


This site uses Akismet to reduce spam. Learn how your comment data is processed.