Assessing dynamic competition in merger control
September 2023 | TALKINGPOINT | MERGERS & ACQUISITIONS
Financier Worldwide Magazine
September 2023 Issue
FW discusses dynamic competition in merger control with Dennis Beling, Andrew Swan, Guillaume Duquesne and Zita Vasas at Compass Lexecon.
FW: Could you explain the concept of dynamic competition? How does it differ from potential competition?
Vasas: Dynamic competition and potential competition are different forces – and they can be confusing when blended together. I find it helpful to distinguish actual competition from potential competition on the one hand, and static competition from dynamic competition on the other. Actual competitors are already present in the market, whereas potential competition refers to the expected or likely entry of new players, or expansion by existing competitors. The main distinction here is who competes; whether a firm or a product is already present in the market or not. On the contrary, the difference between static and dynamic competition is what firms compete on. In static competition, rivals offer lower prices, or superior quality products to attract customers. In dynamic competition, rivals primarily focus their efforts on research and development (R&D) activities so that they can create demand for new products and services. That is, they innovate, competing in creating something new.
FW: Why is it important to protect dynamic competition?
Beling: Economics suggests that the introduction of new products creates substantial value for society, and that disruptive firms generate substantial innovation. The recent focus on protecting dynamic competition in innovative industries is understandable, and there is much at stake.
Swan: The starting point is recognising that innovation is what really matters for advancing consumer welfare. That can take different forms – inventing new products and services, new methods of production and distribution, or new ways of doing and organising business. These benefits are larger than the benefits that come from improvements in static competition and the impact of mergers on the dynamic competition to innovate should rightly be a central concern of merger enforcement policy.
“Dynamic competition and potential competition are different forces – and they can be confusing when blended together.”
FW: Could you outline why a merger might harm dynamic competition?
Beling: A merger could harm dynamic competition if it reduces incentives to innovate. For instance, firms that mainly compete by investing to develop the next disruptive product may invest less in innovation after a merger. This is because the threat of losing sales to a rival’s innovative new product in the future is less than it would be if the merging parties still competed on a standalone basis. Similarly, the benefits of developing one’s own new product may decrease.
Swan: Most of the cases that involve a dynamic competition angle have been in the technology and pharmaceutical industries. And they often involve nascent competitors, with Meta/Giphy, Illumina/Pacific Biosciences, Roche/Spark Therapeutics, Illumina/Grail, and Facebook/Within recent examples. However, dynamic concerns have not been restricted to the technology and pharmaceutical industries, or to acquisitions of nascent competitors. Sika/MBCC is a recent example where concerns were about the potential for innovation to disrupt an established industry. In this case, sustainability was identified as an increasingly important parameter of competition. The European Commission (EC) approved the merger on condition of divestiture of MBCC’s chemical admixtures business – an input into producing green cement – across multiple regions, as well as innovation centres and R&D facilities globally.
FW: What factors are leading to growing concern about dynamic competition?
Duquesne: The growing importance of technology companies, and the increase in their M&A activity, has triggered interest in dynamic theories of harm. In part, that is because digital markets often feature characteristics which favour concentration or a tendency to tip to one single platform, which makes them harder to contest through static competition. The main mechanism left to discipline large digital platforms is dynamic competition ‘for’ the market – potential and actual entry of an alternative platform may mitigate the ability of the incumbent to exert market power. In fast-evolving markets, that threat can be powerful, as even small firms could rapidly become major competitors.
Vasas: Another reason why the technology sector is more prone to concern about dynamic theories of harm is that the innovation process is less structured, so disruptive innovation can happen relatively rapidly and more frequently than in other sectors. For instance, in the pharmaceutical sector, innovation is important, but the process of developing and launching new products is lengthy and heavily regulated. As digitalisation spreads and creates possibilities for transformational innovation beyond high-tech products, concern about dynamic competition may spread to other sectors.
Swan: Apart from the recognition that innovation is central to advancing welfare, the concern stems from a perceived lack of scrutiny of mergers by competition authorities, particularly on this issue. Only a small fraction of mergers is scrutinised, and an even smaller fraction of those are challenged and blocked or approved with conditions. Of course, on its own, lack of intervention is not evidence that there has been underenforcement, or that acquisitions have harmed innovation. In principle, economics tells us that a merger could reduce incentives to innovate, but it could also amplify those incentives at least when taking account of potential efficiencies. In practice, the available empirical evidence is at best mixed. Clear-cut examples of mergers that have been permitted and reduced innovation are lacking. This is not surprising given the challenges of disentangling counterfactual scenarios, particularly where the activities of the firms in question involve complementarities. But this does not mean that there should not be increased concern with dynamic competition theories of harm – the importance of maintaining and fostering innovation means that there should.
“The growing importance of technology companies, and the increase in their M&A activity, has triggered interest in dynamic theories of harm.”
FW: To what extent does concern about dynamic competition explain increased enforcement?
Swan: Growing concern about dynamic competition explains some of the perceived and actual increase in merger enforcement, particularly by the UK’s Competition and Markets Authority (CMA), but not all of it. The CMA signalled that dynamic concerns would feature more prominently in its merger enforcement. Its revised Merger Assessment Guidelines (MAGs) devote an entire section to dynamic competition and refer to digital platforms and pharmaceuticals as examples where such concerns are likely to arise. Its previous MAGs made no reference to loss of dynamic competition as a theory of harm. Similarly, in the US, authorities have also signalled that dynamic competition issues should receive more focus in updated merger guidelines. However, dynamic concerns do not explain tightened enforcement on their own. Various statistics show that merger enforcement activity and abandonments have been increasing across jurisdictions in the past few years, including based on ‘traditional’ horizontal unilateral effects and vertical foreclosure concerns.
FW: What challenges arise when analysing concerns about dynamic competition?
Beling: In an assessment of dynamic competition, the standard tools used in static competition assessments – shares of supply in the relevant market and concentration indices – are typically still the starting point of the analysis. However, they are much less informative for assessing dynamic competition. What matters most are the incentives to invest and innovate, rather than the pricing incentives that are usually the focus of assessments of static competition. Market definition can also have less relevance: where innovation is an important parameter of competition, a relevant product market may not yet exist or could be difficult to delineate. Revenues and market shares may not be accurate indicators because companies with small revenues could develop into important constraints.
Duquesne: When analysing dynamic competition, it is common that markets – particularly digital ones – have specific characteristics that need to be taken into account, such as multi-sidedness, network effects, data-driven economies of scope, and so on. The tools and evidence used for traditional markets will often need to be updated or adapted to assess these factors. An added complication is that analysing dynamic competition is inherently uncertain. Not only do authorities need to predict what impact the merger might have, they also need to predict what will happen to the market absent the merger. In mature markets, that counterfactual is typically the status quo – the past can be a reliable guide to the imminent future. This is not likely in dynamic and innovative digital markets. The number of credible counterfactuals can quickly multiply and attaching a credible probability to each of these scenarios becomes extremely difficult. This is an inherently speculative and difficult task, especially in fast-evolving markets and for nascent companies. Concern about innovation in general has led to a flourishing of new specific dynamic theories of harm. The risk is that any possible theory of harm becomes worth investigating – while these theories may apply only in very specific circumstances, if at all, many of them have received little empirical support. This can make the review process extremely burdensome. The inherent difficulty and uncertainty in assessing dynamic competition in digital mergers and ultimately the risk of getting it wrong have led to a perceived increasing asymmetry in the standard of proof in digital mergers. Competition authorities have been willing to express concerns in one among several possible scenarios, while parties have to demonstrate efficiencies in many possible scenarios. In practice, this means dismissing any efficiencies, which may be sizeable in such markets.
“Most of the cases that involve a dynamic competition angle have been in the technology and pharmaceutical industries.”
FW: What evidence can be used to analyse dynamic competition?
Beling: It is crucial in an assessment of dynamic competition to identify clearly what the merging parties compete on – a step the UK’s Competition Appeals Tribunal (CAT) referred to as ‘identification of the dynamic element’. This may be R&D efforts to develop a new medicine, developing a new mechanism to attract social media users, or a programme of entering local markets. To understand the extent to which the merging parties innovate in the same area, or can be expected to in future, it is also crucial to analyse the parties’ respective innovation capabilities. In some cases, it is possible to use quantitative indicators for this purpose. In cases involving R&D, useful indicators might involve the size and qualification of research teams, and it may also be possible to gauge to what extent the merging parties’ research overlaps by analysing patent data. However, in many cases such a quantitative approach will not be possible, and the assessment then focuses on records of past innovation and the firms’ own innovation plans. Internal documents, including those of the merging parties and their competitors, are therefore typically key evidence in dynamic competition assessments.
Duquesne: It is essential to form a view about the counterfactual and assessing the likely future competitive constraints that the target could potentially have imposed upon the acquirer’s products, and vice-versa, in different scenarios. This places more emphasis on case-specific dynamic economic modelling, making the assessment of such mergers much more complex and involved. Internal documents and survey evidence can play a critical role in calibrating those models and grounding them in the facts. Applying advanced data science techniques to market data can also provide extremely useful insights. Some commentators have suggested that the target’s valuation is a useful indicator. It can provide some insight into the rationale for the transaction, especially in the case of nascent competitors. Does a high valuation reflect potential synergies and efficiencies from the transaction and the bargaining power of each party? Alternatively, does it reflect the advantage of neutralising a potentially disruptive competitor? These valuation analyses need to be run carefully, especially in digital markets where efficiencies may be difficult to quantify at the level of detail required. Data on offers from other bidders may be useful in this assessment.
FW: To what extent do authorities assess dynamic competition in different ways?
Beling: The CMA and EC have come to different substantive conclusions in identical matters: Microsoft/Activision and Illumina/Grail. But it is difficult to relate that to specific differences in their underlying frameworks. In principle, more relevant guidance exists in the UK. In addition to the CMA’s revised MAGs, there is relevant guidance from the CAT, which set out relevant factors for assessing dynamic competition in its Meta/Giphy decision. It is fairly uncontroversial to say that significant uncertainty regarding the practical details of an assessment of dynamic competition in matters before the CMA remains. Less guidance is available from the EC. The most relevant cases focus on innovation competition in industries where R&D plays a key role, such as Dow/Dupont and Bayer/Monsanto. For such cases, there is now a relatively clear analytical framework in place, but the EC’s enforcement record shows that concerns about dynamic competition extend to other industries, where that framework is less helpful. Overall, there is significant uncertainty in cases before the EC. When reviewing matters involving dynamic competition, both the CMA and EC will be able to cast a wider net and raise concerns they would not have raised when applying traditional static competition analysis. For companies contemplating transactions, it is advisable to carry out a thorough review of potential future overlaps – in terms of overlapping products and innovation pipelines, and also in terms of overlapping capabilities – that pays close attention to areas in which complaints from customers or competitors might be expected.
Swan: The framework adopted by the CAT in Meta/Giphy does highlight some relevant factors to assess in these cases, in particular where they involve the acquisition of a nascent player. This includes the important ‘cross checks’ to weed out ‘duds’ and to consider the ‘disbenefits of intervention’. However, it is far from a complete framework for assessing the range and complexity of dynamic competition considerations, both potentially pro- and anti-competitive. It remains far too focused on the activities of the specific parties to the merger as opposed to the wider innovation environment they operate in, their positions and capabilities within that, and those of their competitors. Without that wider context, particularly in cases where there is uncertainty about the relevant product markets in question, it is difficult to do a meaningful assessment of the closeness of competition between the parties in the innovation activities and the extent to and by which firms will continue to be constrained post-merger.
Vasas: The much-awaited draft update to the EC’s 1997 market definition guidelines repeatedly refers to the ‘dynamics of competition’ – and it reflects important developments such as digitalisation, the rise of multi-sided platforms, and the importance of innovation in certain sectors. The EC sets out that it is the expected structural changes that it considers could alter the dynamics of market demand and supply. It considers that such structural changes could be driven by either technological or regulatory developments. In both cases, the EC would apply a forward-looking assessment, which for technological developments would likely involve evidence on firms’ R&D and innovation activities. The draft notice acknowledges that an assessment relying on hypothetical changes may be less reliable and reserves the right not to reach a definitive conclusion on what the relevant market is.
“A merger could harm dynamic competition if it reduces incentives to innovate.”
FW: What implications does sector regulation have for dynamic mergers?
Beling: The most relevant piece of sector regulation is the European Union’s Digital Markets Act (DMA) which recently entered into force, and similar pieces of regulation that have been discussed in other jurisdictions and which may be introduced going forward. The DMA imposes a range of rules on the conduct of large tech firms, seeking to lower entry barriers and facilitate the ability of entrants to challenge established firms with innovative products and services. There are two implications for mergers facing dynamic theories of harm. First, under article 14 of the DMA, large tech firms must notify the EC of all acquisitions, enabling greater scrutiny. Although the DMA does not include a provision that fundamentally changes the substantive review of acquisitions by large tech firms – such as reversing the burden of proof in such cases and prohibiting them unless the acquirer can demonstrate no harm to competition – it does mean no acquisitions will go unnoticed. The acquirer will have to start thinking at an early stage about the impact on competition of an acquisition as the DMA requires that information on the ‘nature and rationale’ of the deal be provided. Second, the provisions in the DMA aimed at facilitating contestability – interoperability requirements to reduce the strength of network effects and lower entry barriers – aim to protect dynamic competition and so may reduce concerns about the impact of dynamic competition of a merger. This could have implications for remedy design. For instance, where the DMA already imposes interoperability requirements, it may not be necessary to have dedicated merger remedies on interoperability.
FW: In your opinion, is the future of assessing dynamic competition looking clearer?
Duquesne: While developments in the UK and Europe have helped in setting the direction of travel, there is still a long way to go. There is still no clear agreement on how to assess dynamic competitive effects, particularly in fast-evolving digital markets. Developing robust and widely accepted tools for analysing these impacts, that can be applied in practice, is a key challenge for economic practitioners.
Dennis Beling is a senior vice president with Compass Lexecon and head of its Brussels office. Previously based in Compass Lexecon’s Singapore office, he continues to assist clients across Asia Pacific with regular travel to the region. Mr Beling has more than 18 years’ experience as a competition economist gained in both public and private sector roles. His work has covered various industries including high technology, financial services, construction, chemicals, energy and transport. He can be contacted on +852 3768 4610 or by email: dbeling@compasslexecon.com.
Andrew Swan is a senior vice president based in Compass Lexecon’s London office. He has more than 14 years’ experience as a professional economist specialising in competition policy, with particular expertise and interest in the economics of network industries, and vertical and conglomerate issues. Before joining Compass Lexecon, he was an economist in the policy and research division at the South African Competition Commission based in Pretoria, and at consultancies in Johannesburg and London. He can be contacted on +44 (0)78 6070 4469 or by email: aswan@compasslexecon.com.
Guillaume Duquesne is a senior vice president at Compass Lexecon. Since joining in 2013, he has advised clients on a wide variety of competition policy issues, covering sectors as diverse as energy, telecommunications, financial services and travel agencies. His recent work includes advising on the competitive effects of the mergers Alstom/Bombardier Transportation, E.On/Innogy, Thales/Gemalto, EDF/Areva and Amadeus/Navitaire before the European Commission, and Cofigeo/William Saurin before the French Autorité de la Concurrence. He can be contacted on +32 (0) 2 274 25 11 or by email: gduquesne@compasslexecon.com.
Zita Vasas is a vice president based in the Paris and London offices. She has over 10 years of experience as a competition economist, including at a national competition authority, in the competition team of a sectoral regulator and in consulting. She has been involved in cases in a wide range of sectors, policy areas and jurisdictions. Her background is in competition, and she gained experience in applying behavioural economics and consumer policy at the Financial Conduct Authority. She can be contacted on +33 6 44 13 97 57 or by email: zvasas@compasslexecon.com.
© Financier Worldwide