Expert Q&A on horizontal and vertical merger enforcement

Expert Economists Joe Perkins, Salvatore Piccolo and Guillaume Duquesne participated in a Q&A with Financier Worldwide for their July issue on the effects of mergers on businesses, consumers and innovation, and how regulatory authorities could improve horizontal and vertical merger enforcement.

To what extent are industries across the economy becoming more concentrated and less competitive as a result of M&A? Could you explain how mergers typically affect consumers, workers, entrepreneurs and businesses?

Perkins: Mergers have two types of effect. They can allow businesses to produce new products, or to produce existing products more efficiently, as a result of synergies between firms. But they can also allow businesses to raise prices or to reduce service quality due to less competition. The first type of effect can benefit everyone – businesses, workers and consumers. But the second effect might benefit a company’s shareholders, but not its customers or employees. From a public policy perspective, striking the right balance between these effects is therefore crucial. Some critics of recent merger policy, such as Lina Khan, head of the US Federal Trade Commission (FTC), have argued that authorities have not got the balance right – they have allowed too many mergers, leading to higher prices, and increasing inequality. There is some evidence to back this up. Important economic studies have found increasing concentration and less competition overall in the US. But this finding is not universal, and any changes in Europe do not seem to be as dramatic.

Piccolo: The advent of the internet and the tremendous decrease in the cost of storing, processing and transmitting information has enabled the emergence of new industries, while also reshaping entire traditional sectors. The new digital economy is dominated by a few large conglomerate companies, commonly referred to as ‘Big Tech’ or the GAFAM – Google, Apple, Facebook, Amazon and Microsoft. The adoption of hybrid business models by these platforms and their aggressive acquisition strategies have renewed the policy interest for mergers. Vertical mergers, for example, raise competitive concerns since they may result in the anticompetitive foreclosure of non-integrated rivals. That is, vertically integrated suppliers may exploit their consolidated market power to exclude – partially or in full – their independent rivals to soften downstream competition: the so-called ‘foreclosure doctrine’. However, vertical mergers may also bring valuable efficiencies, mostly when they eliminate double marginalisation or when they help upstream suppliers overcome the hold-up problem that emerges when contracts are incomplete and suppliers make non-contractible, relationship-specific investments before dealing with their downstream units. The trade-off resulting from these forces is usually a complex phenomenon, and hard to evaluate in practice. Mergers that bring upward pressure on wholesale prices are likely to be blocked absent efficiencies. Mergers that soften double marginalisation or solve hold-up problems are more likely to bring efficiencies and thus are less likely to raise competitive concerns. But does this tell the whole story about vertical mergers involving complex digital ecosystems? Is the absence of double marginalisation sufficient to infer consumer harm in circumstances where ecosystem gatekeepers continuously innovate to improve the quality of their networks? What is the role played by network externalities in these mergers? Recent academic work triggered by these questions shows that traditional and consolidated presumptions fail when evaluating mergers in digital markets, suggesting that the above challenges can only be properly tackled if policymakers are guided by a renewed economic theory, taking into account the salient features of those emerging markets.

What role does M&A play specifically in digital industries?

Perkins: The stakes around merger control may be higher in the digital sector than in many other industries. The network effects in digital industries mean that the gains from bringing together complementary firms can be particularly significant. For instance, a merger can enable a small firm’s innovation to reach a very large audience quickly. But network effects also mean that there is a greater risk of ‘tipping points’, with a market tipping toward monopoly once one firm becomes particularly large. Competition authorities have blocked few digital mergers historically. But there is some evidence of a change in approach recently. The UK Competition and Markets Authority (CMA) blocked Facebook’s proposed acquisition of video-sharing site Giphy in November 2021. It argued that the merger would reduce dynamic competition. Facebook is currently appealing this decision, and the outcome of the appeal could have a major impact on future mergers in digital industries.

Piccolo: While most platforms’ products – such as search, cloud computing, e-commerce, social networks, mobile OS and hardware, for example – are the result of considerable investments and bring large benefits to consumers and business users, there are growing worries that digital platforms take advantage of their control over their networks and ecosystems to consolidate market power, for example through aggressive acquisition strategies. In vertical mergers, for example, one traditional concern is that gatekeeper platforms may foreclose potential rivals to soften competition within their marketplaces, thereby extracting supra-competitive prices from final users, including consumers, third-party sellers, as well as developers. Yet, this logic neglects some salient aspects of digital markets. For example, it is widely recognised that network externalities are a fundamental pillar of the aggregative role played by digital platforms. One distinctive feature of these businesses is that they coordinate broad ecosystems – meaning groups of connected firms, often complementors – to lock-in their customers. Early platform adopters enjoy direct and indirect benefits as new users join their network. This is because trading and diversification opportunities expand with the network size, which may depend on the acquisition strategies of the platforms and, of course, on how strict or lenient the merger regime is. While on the one hand vertical and horizontal mergers may be driven by a traditional consolidation logic, platforms may also use this market power to internalise the benefits of coordination, enhance network externalities and exploit a large user base to design more efficient ecosystems.

Duquesne: Through acquisitions, digital platforms may strengthen their offering and boost the attractiveness of their ecosystem in order not to be displaced by actual or potential competitors. As such, digital mergers are in many instances another tool that digital platforms use to improve and enhance their value proposition, and ultimately compete in a fast-evolving market. It is frequently argued that digital platforms could develop all given innovations in-house – as opposed to acquiring a firm that already produces the innovation – which would imply that digital mergers can only be driven by the desire to dampen competition. However, this is unlikely to be true in all circumstances. First, large digital platforms may wish to acquire start-ups that produce products complementary to their own, if they believe that the complementary product or service has a particular characteristic that makes it stand out and that they would not be able to replicate this innovation in a timely fashion. Second, there is a limit to what large digital platforms can do. Digital platforms, like firms in many other industries, must manage scarce resources and are forced to prioritise, facing trade-offs between different projects. If an opportunity to improve its offering through an acquisition presents itself, the digital platform may well take it, but this does not mean that it would have organically developed the product without the acquisition.

How does the merger regime affect companies’ overall innovation and investment decisions?

Perkins: The merger regime can have important effects on the overall climate for innovation and investment. For instance, a start-up company that hopes to be bought out by a large tech firm has a strong incentive to try to develop a product that fits well with the tech firm’s existing products – but it may have less reason to invest in developing a competitor service. Mergers can also provide firms with an exit option when investment is risky. If a company is able to merge with another firm if customer demand proves to be lower than expected, it might be more prepared to make an investment in the first place. This could be particularly important in industries that require large, uncertain investments.

Piccolo: Ecosystem gatekeepers gain power by being nodal and hard to replace. But, to maintain a ‘bottleneck’ role and remain attractive to users, these platforms need to be constantly at the technological frontier to improve the quality of their final products and services, thereby benefitting all their participants, both final consumers and business users. The merger regime may have tremendous and irreversible effects on these innovation and investment decisions. The acquisition of potential competitors is an issue that has garnered considerable interest in recent years. The extent of the ‘killer acquisition’ phenomenon has been widely documented in economic literature, whereby an incumbent buys an entrant in the process of developing an overlapping treatment, only to shut down its efforts and preserve the initial monopoly. In the technology industry, the FTC reported in 2021 that the GAFAM companies have been acquiring start-ups at the pace of 60 per year over the period from 2010 to 2019, with some of these acquisitions raising serious concerns. Yet, most of these acquisitions were not killer acquisitions, since the acquirers have kept investing in the products. Another well-known manifestation of the link between innovation and M&A in digital markets is the entry for buyout phenomenon, where the prospect of being acquired by an incumbent provides stronger incentives to entrants. Besides providing stronger incentives to innovate, the prospect of being acquired may also enable entry, especially when firms face financial constraints. Allowing incumbents to merge with entrants is also likely to affect the direction of innovation – that is, the type of innovation brought about by entrants. In a nutshell, mergers may bias efforts toward incremental rather than radical innovation, toward development of close substitutes to the incumbent’s product, or toward technologies that are more complementary to those of a dominant incumbent. While some of these effects may be undesirable, the welfare impact of such mergers is ambiguous in general. As such, extreme proposals such as a blanket ban on acquisitions by the GAFAM have no basis in economic theory. Instead, recent economic literature identifies the key forces at play, thereby providing guidance for a more nuanced and thorough analysis of the various cases.

Duquesne: The merger control regime plays an important role to incentivise innovation in digital markets, in part because it has the potential to trigger chilling effects on dynamic competition. A possible benefit of digital mergers is to foster innovation from start-ups, which in turn helps digital platforms to improve their offering to the benefit of end customers. Many small firms launch their businesses and innovate with the objective of being acquired by large digital platforms. Payouts from acquisitions provide incentives for venture capitalists to invest, thereby playing an important role in promoting entrepreneurship and innovation. This is even more important if we consider that the acquired firm may also lack the necessary funds to fully develop the innovation. Through the acquisition, the digital platform may bring funding, alleviating these constraints and enabling full development of the technology. Killing this exit strategy through a perceived overly stringent merger regime could have a chilling effect on entrepreneurship and innovation. This is not to say that digital mergers may not in certain instances have the potential to negatively affect both the level and the direction of innovation in digital markets, as traditional mergers do, but that chilling effects on dynamic competition should be duly taken into account when reviewing digital mergers.

How have merger control regimes in developed countries changed in recent years? What major decisions are worth highlighting?

Perkins: Our assessment of the merger control regime in the European Union (EU) suggests that there have not been big shifts in the European Commission’s (EC’s) approach over time. Since 1990, the EC’s intervention rate has stayed fairly constant, at between 5 to 10 percent of mergers notified, with only a small proportion of those blocked outright. But there are signs of significant changes in coming years, linked to the UK’s exit from the EU. The UK CMA has made clear its willingness to take an activist approach to mergers, particularly where they might harm dynamic competition, and its intervention rate has risen to around 20 percent of mergers notified in recent years. The CMA decided to prohibit the recent planned mergers of Facebook with Giphy and Konecranes with Cargotec, despite approvals by some other authorities. It remains to be seen though whether this is the beginning of a broader trend.

Duquesne: The continuous growth of digital platforms, partly due to the acquisition of other digital companies, has fuelled claims of underenforcement in digital merger control. The concern has been framed in terms of the error-cost framework: competition agencies might have been too permissive, putting too much weight on the risk of an incorrect intervention, known as a ‘type I error’, over the risk of an incorrect clearance, known as a ‘type II error’, when assessing mergers in digital markets. The nature of competition in digital markets may indeed change the terms of the usual trade-off between these errors. Network effects and data-driven economies of scale often lead to highly concentrated markets, in which firms compete for the market. When this is the case, ensuring market contestability and protecting actual or potential competitors would be even more critical than in traditional markets, making type II errors particularly costly. According to this view, this would require considering significant changes in merger control and an increased scrutiny of digital mergers involving large digital platforms. Some commentators argue that the burden of proof should be shifted to the merging parties, with a presumption of illegality. The debate is still ongoing and further investigation is required. That said, there is not yet solid evidence that digital acquisitions are intrinsically anticompetitive and are systematically being underenforced through the current legal approach. Although claims of underenforcement may be justified in some instances, there is a risk that the one-sidedness of the current debate may swing the pendulum in the complete opposite direction, trading off clear efficiencies against far-fetched speculative anti-competitive effects, with unclear but potentially significant adverse consequences for dynamic competition, innovation and consumer welfare.

In your opinion, are the tools used by authorities to address digital mergers sufficient to keep up with the latest developments?

Perkins: The economic framework for analysing the immediate effects of standard horizontal mergers between competitors is well established. There is, of course, room for further development, but changes in this area are likely to be limited. However, there is much less agreement on how to assess dynamic competitive effects, such as the impact of a merger on potential competitors in a market. These effects are important across the economy but are likely to be especially crucial in the digital sector, both because of tipping point effects and because of the scope for small firms to quickly become major competitors. Developing robust and widely accepted tools for analysing such impacts, that can be applied in practice in the context of large mergers, is one of the key challenges for economic practitioners in the coming years. The lack of a rigorous framework could lead to significant uncertainty for firms, and allegations of ad hoc decision making by authorities.

Piccolo: While one of the main operational problems behind the assessment of a merger remains the proper definition of the counterfactual scenario, many of the existing economic techniques to evaluate mergers – including diversion ratio, merger simulation, and the gross upward pricing pressure index (GUPPI) – are unsuited for the digital sector in their traditional form. The competitive assessment of mergers in this sector must rely on techniques that account for their salient and innovative characteristics. For example, the tools for evaluating vertical mergers must account for ecosystem competition, especially when such rivalry involves innovation competition. Vertical mergers are likely to lead to increased investment and fiercer inter-ecosystem competition to the ultimate benefit of consumers. The positive welfare effects of such mergers are most significant when the merged entity faces competition from a superior ecosystem and consumers prefer ecosystem variety. These effects are hardly captured by the current analytical tools that do not explicitly account for features such as network externalities and ecosystem effects.

Duquesne: In many instances, the existing guiding principles for merger assessment are, at least in part, fit for purpose to assess digital mergers. However, tools used to assess traditional mergers may need to be updated to better account for the specific characteristics of digital markets – such as multi-sidedness, network effects, data-driven economies of scope, and so on – as well as digital platform and ecosystems competition, and the possible procompetitive and anticompetitive effects arising from digital mergers. For instance, in fast-evolving markets, static tools that mainly focus on price effects are likely to be of little relevance to appreciate the possible impact of digital mergers, where the question is more about possible impacts on dynamic competition and innovation. Also, the past is unlikely to be a reliable guide to the imminent future, which means that relying on historic pricing and sales data in assessing the counterfactual is unlikely to be appropriate. In digital mergers, there is likely to be a need to build more credible counterfactuals and to attach a credible probability to each. This will require placing more emphasis on case-specific economic modelling to evaluate the impact of digital mergers, making the assessment of such mergers much more complex and difficult. This calls for a possible revision of the current merger guidelines to identify potential new theories of harm, clear conditions under which they are more likely to materialise, how they could be tested empirically using the data available to the merging parties, and the set of evidence competition authorities would consider relevant.

Could you provide an overview of how regulatory authorities are looking to evaluate and potentially update their horizontal and vertical merger guidelines? What are the key areas of focus?

Perkins: A wide range of issues have emerged in authorities’ recent discussions on updating their merger guidelines, but three important ones are worth highlighting. First, there has been increased interest in issues of potential or dynamic competition, particularly in the digital sector, along with a greater willingness to intervene even where market developments are uncertain. Second, authorities have highlighted some of the wider impacts of mergers beyond their effects on consumers. For instance, the recent US Department of Justice (DOJ) and FTC consultation discusses the impacts of buyer power, particularly on labour markets, suggesting an interest in how workers’ bargaining ability might be reduced by a merger. Finally, there has been a reduction in emphasis on the process of defining the boundaries of a market in assessing mergers. The CMA’s new guidelines relegate market definition to the final chapter and give it much less prominence than in previous iterations.

In your opinion, what additional steps should enforcement agencies consider to improve horizontal and vertical merger enforcement?

Perkins: There are several ways in which the process of merger control could usefully develop in the coming years. First, there is lots of scope to develop and use exciting new tools for assessing competitive effects, based on data science. We have found that techniques such as natural language processing (NLP) can greatly enrich the evidence base for merger decisions, allowing rigorous analysis of thousands or even millions of documents. Second, increasing the clarity of guidance, particularly around relatively new issues such as dynamic competition or impacts on labour markets, could significantly increase confidence in authorities’ methods, and ensure that merger control does not inadvertently stifle innovation and investment. Third, it is important for authorities to develop a reflective approach to learning what works and what does not, based on evaluation evidence. Both the EC and the CMA have commissioned important evaluations of recent mergers, which should lead to dynamic improvements to merger enforcement over time.

Piccolo: The trade-off between the pros and cons of vertical mergers poses new intellectual challenges for policymakers wishing to update merger guidelines. Can they confidently make competition policy decisions based on traditional and consolidated presumptions when evaluating mergers in digital and related markets? If not, why? What are the risks associated with an overly conservative approach? What are the potential drawbacks of a too-lenient approach to mergers and acquisitions in digital and data-intensive sectors? Answering these questions requires stronger collaboration between policy and academic circles and, therefore, greater reliance of merger guidelines on solid economic theory. In addition to the need to understand how merger policy affects platforms’ conduct in digital markets, a necessary step is the development of techniques that incorporate the salient features of those emerging markets into economic models that deliver robust counterfactual scenarios. For example, given the large variety of business models that exist in the digital sector, understanding how to adapt standard techniques, such as GUPPI, to different hypotheses on business models is a fundamental step to obtaining counterfactuals that fit actual industry structures and characteristics.

In light of current developments, what considerations do companies need to make when evaluating potential mergers, assessing antitrust risk and preparing their defence strategies? What advice would you offer?

Perkins: The risks around merger control have risen in recent years, with some authorities showing increased willingness to intervene. In this context, it is more important than ever to develop and present robust evidence of how a merger will benefit consumers, both through lower prices and through dynamic impacts on innovation and investment. This can be supported by providing a clear description of a merger’s rationale and expected effects, in advance of engaging with competition authorities. Moreover, it is critical to understand the global aspects of merger control – recent experience has shown that major global mergers can be derailed by decision making in one jurisdiction, perhaps long after agreement has been reached elsewhere. Companies should ensure they understand well the priorities and timings of different authorities, and coordinate these as far as possible.

Piccolo: The common language is and will always be economics. Companies must understand that being rigorous has great merits in interacting with enforcement agencies. The only way to prevail is to prepare defence strategies grounded in formal and robust arguments based on good economic theory. Establishing defence strategies on these arguments will help win the battle on the ‘cultural’ ground and have the merit of guiding the collection of evidence corroborating the defence line itself.

Duquesne: There is a growing concern that digital mergers are only driven by the desire to dampen competition. While there is no reason for this to be systematically the case, it means that potential merging parties need to get ready to rebut this presumption. This implies working on the economic rationale for the transaction and evidencing likely efficiencies brought forward by the contemplated transaction. It also means conducting more advanced economics at the time of the pre-notification to rebut potential theories of harm that may be difficult to rebut straight away in phase I.

This Q&A was originally published by Financier Worldwide and is available on their website and in PDF format here. The views expressed are those of the authors only and do not necessarily represent the views of Compass Lexecon, its management, its subsidiaries, its affiliates, its employees, or clients.