Andrea Pezzoli: “Market Definition And Market Power Evaluation In Dynamic Industries”

The Network Law Review is pleased to present you with a Dynamic Competition Initiative (“DCI”) symposium. Co-sponsored by UC Berkeley, EUI, and Vrije Universiteit Amsterdam’s ALTI, the DCI seeks to develop and advance innovation-based dynamic competition theories, tools, and policy processes adapted to the nature and pace of innovation in the 21st century. The symposium features guest speakers and panelists from DCI’s first annual conference held in April 2023. This contribution is signed by Andrea Pezzoli, Vice Secretary General at Italian Antitrust Authority.


1. Preliminary Remarks

This brief contribution will focus on three aspects of dynamic competition. Firstly, I’ll try to provide a tentative definition of dynamic industries/markets, identifying their main characteristics. Secondly, I’ll emphasize the main challenges raised by dynamic industries for market definition and, hence, for the assessment of market power. Finally, I’ll provide some practical examples emerging from the experience of the Italian Competition Authority.

2. Tentative Definitions of Dynamic Industries

A number of different characteristics may be helpful to trace the boards of what we can look at as dynamic industries/markets:

  • Existing markets where competition is mainly based on innovation and where price competition plays a residual role. In this sense, it is possible to consider not only digital markets, where quite often we deal with zero-price sides, but also the pharmaceutical industry, where innovation, quality, and therapeutic performance play a major role;
  • “Not existing markets” where competition may take place at the R&D level and final products are not identified yet.1See, for instance, the assessment made by the European Commission of Bayer’s acquisition of Monsanto. Commission (2018) M.8084 Bayer/Monsanto, and G. Federico (2013),” Horizontal Mergers, Innovation and the Competitive Process”, Journal of European Competition Law and Practice, 8.
  • New markets whose development mainly depends on data and where it is not that easy to identify the boundaries of the relevant markets;
  • Industries where antitrust enforcement (especially in merger review) needs a longer time horizon and has to cohabit with a degree of uncertainty, which sometimes may be unbearable and inconsistent with the need for legal certainty;
  • Industries where enforcement has to be extremely careful: we all agree that there is a danger to discouraging the “new Google”, but when we realized that the “new Google” is not emerging yet (and what is more, seems out of sight), we start thinking that it might be too late to intervene with ex-post tools,2The idea that antitrust enforcement is just an ex-post tool is at least questionable. There are so many examples of antitrust interventions that require an ex-ante approach that make the distinction between regulation as an ex-ante tool and antitrust as a mere ex-post tool quite meaningless. privileging ex-ante and future-proof regulation (how ambitious!). We start looking at market power from a different perspective, and the temptation to be concerned by bigness as such begins to matter again.

Though these characteristics are by no means typical of dynamic industries, we should not forget that dynamic competition may also take place in “traditional” markets when disruptive innovation radically changes their business model. In these cases, should we talk of Dynamic Industries/Markets?

Not to say of the need for a dynamic evaluation when we deal with cases of Failing Firm Defence, and we desperately look for a sound and reliable counterfactual, indispensable for an appropriate review of the merger. And we all know that the appropriate counterfactual seldom coincides with the status quo. In other words, having to define the market and evaluate market power, the need for a wider time horizon is not confined to innovative industries but is often needed for traditional ones as well.

3. The Main Challenges Raised By Dynamic Competition And The Need For Legal Certainty

It is evident – I’d say obvious – that dynamic industries raise significant challenges for market definition (which, in my view, may still play an important role)3In this sense, the new Market Definition Notice is certainly welcome. and above all, for the assessment of market power.

The main challenge comes from the need to predict the scope and the benefits of future competition. In this perspective, the increasing role played by scope and scale economies (mainly, but not only, because of the increasing role played by data) suggests that in defining relevant markets, we have to look more deeply at what happens on the supply side, taking for granted the risks related to a too broad market definition.4See Philips G. (2022), Scope, Scale and Concentration, The 21st Century Firm, NBER WP. However, the risks are worthwhile if we aim at a proper assessment of non-horizontal mergers, especially of conglomerate mergers, which no more can be reduced to mere diversification strategies.

In dynamic industries, the difficult task for antitrust agencies is the assessment of potential competition, in particular, the identification of potential competitors in a medium-long term horizon. This task may be quite challenging because not all the concerns can be quantified (nonetheless, they do exist) and, above all, because their difficult predictability might conflict with the need for legal certainty.

In this perspective, it may be understandable the request coming from the business community for “safe harbours” even when it may risk being a too rigid solution. However, what cannot be acceptable is to cope with these challenges just by freezing actual competitive market conditions and assuming away all unpredictable benefits of future competition, especially those which may derive from today’s “less efficient competitors”!

If it is not possible to quantify the impact, it does not mean that the impact does not exist. In this sense, it is almost impossible to resist the temptation to recall the short story of the guy who was looking for his key under a street lamp (because there was light) even though he was pretty sure that he had lost the key somewhere else (where, unfortunately, it was dark). Therefore, if we really want to approach dynamic competition, when we deal with market definition and market power, we cannot avoid “dancing in the dark.”5Springsteen, B. (2013), Dancing in the Dark.

4. My Experience As An Enforcer (And As An Economist)

In recent years, the Italian Competition Authority had to cope with market definition and the assessment of market power in dynamic industries in several cases.

In my view, one of the most interesting cases is a merger review concerning the joint venture between EnelX and Volkswagen aimed at developing electric mobility charging infrastructures.6C12404 – EnelX-Volkswagen Finance Luxembourg/JVC, Bullettin n° 51/2021 The Italian Competition Authority had to assess the impact of the merger on new and developing markets. An appropriate assessment of potential competition was key, and the identification and the role which potential entrants could play was not that simple. For this purpose, hearings and internal documents were revealed to be much more helpful than quantitative evidence.

On the other hand, what was quite clear is that market shares (meaning current market shares) in cases like this are not particularly meaningful for the evaluation of market power. In developing industries, market shares may change continuously and may be quite easily eroded unless there are other factors that determine their persistence and their stickiness.

The challenging definition of the relevant markets, mostly the markets affected by the competitive restrictions, was also crucial in a few proceedings concerning abusive conducts in the digital sector (whatever it means). In two cases, while it was relatively easy to assess the dominant position, it was much more complex the definition of the dynamic/innovative markets affected by the abusive conducts.7See A529 – Google/Compatibilità App EnelX Italia con sistema Android Auto, Bullettin n° 20/2021 and A552 – Google- Ostacoli alla portabilità dei dati, Bullettin n°  27/2022.

The case A529 – Google/Compatibilità App EnelX Italia con sistema Android Auto concerned again the services related to electric mobility. Very briefly, it could be said that the anti-competitive conduct was a refusal to access Android Auto, the system which would allow the EnelX’s app to be visible on the screen of the car. Without the possibility to be visible on the screen of the car EnelX’s app would have been confined to the screens of the mobile phones and could not be used, for safety reasons, while driving the car.

Therefore, even though EnelX’s app could provide a higher quality service for electric cars’ drivers – since it allows not only to map the charging stations and to know if they are vacant but also to book them – because of Google’s refusal to access it could not exercise its competitive constraint for the collection of data currently provided by Google Maps.

In this case, Google’s dominant position has been assessed in a “traditional” way (taking into consideration the distinctive features of multisided platforms). However, in the evaluation of market power, even more than the market shares held on the different sides, it was crucial to the “quasi-essential facility role” played by Android Auto(the app which can be used directly on the screen of the car and which allows the interoperability between the Infoteinement of the car and the apps which are consistent with Android Auto).

A552 – Google- Ostacoli alla portabilità dei dati, is a case that is still in progress; therefore, I can’t provide much details. However, the interesting point is that we should deal with the impact of Google’s conduct on the development of a new service: the service provided by intermediaries that users can authorize to collect their individual data from different platforms and resell them to advertisers, shifting part of the profits to the owners of the individual data. Something that might remind what Eric Posner and Glen Weil would describe as “digital trade unions for platforms’ users.”8See Posner, E.A, and Weil G. (2019), Radical Markets. Uprooting Capitalism and Democracy for a Just Society, Princeton University Press. The new market (or, better, the new intermediaries) could represent a competitive constraint for Google. The alleged exclusionary conduct might be considered a violation of art. 102 of the Treaty to the extent to which it may obstacle data collection by the new intermediaries. In this sense, the new intermediaries could be seen both as potential competitors or even as the origin of a new market;

Finally, it is worth mentioning another case where the Italian Competition Authority had to cope with market definition and market power in dynamic industries. More precisely, with the interplay between quite dynamic industries, such as e-commerce, online distribution, online marketplace, and apparently more traditional industries, such as logistics.

In A528 – FBA Amazon,9A528 – FBA  Amazon, Bullettin N° 49/2021. the Italian Competition Authority dealt with Amazon’s dominant position and the dilemma (but is it a real dilemma?) of the assessment of market power in multisided markets. The Italian Competition Authority opted for the “European way” (i.e., assessing market power side by side), and it was a sound exercise, and as often happens (though not always), convergent with the outcome of the “American way” (i.e., multi-sides approach).

However, from our perspective, the interesting point was the assessment of potential competition, i.e., those players operating in the logistics industry potentially affected by the tying strategy adopted by Amazon. The intriguing question was: to which extent can Amazon’s current logistics suppliers be considered Amazon’s as efficient competitors in the future? If the remedies identified by the Italian Competition Authority in her final decision and the commitments accepted by the Commission in the twin case concerning the remaining European countries reveal to be effective, we might have an empirical answer in the near future.

5. Some Concluding Remarks

Let me conclude with a more general remark: in my view, the challenges raised by market definition and the assessment of market power in dynamic industries are strictly linked with (and could be better dealt with) if we privilege a concept of competition that restores the intrinsic interplay between liberty and efficiency. A concept of competition that acknowledges economic freedom as a key source of social efficiency and that considers the protection of the competitive process its natural goal.10See Fox, E. (1981), “The Modernization of Antitrust: A New Equilibrium”, Cornell Law Review, Vol. 66.

Therefore, antitrust should not be focused only on static efficiency and should not be confined to the protection of the outcome of past competition. Especially – but not only – when we cope with mergers. In other words, and more directly: we have to be aware that today’s efficiency is not necessarily the best predictor of tomorrow’s efficiency. It might seem a paradox, but in this perspective the neo-brandeisian challenge to look at competition as an ongoing process might hide more than one similarity with the Austrian school and with Hayek’s view.11M. Grillo (2018), Competition, Efficiency and Liberty, Working Paper.

Andrea Pezzoli

Citation: Andrea Pezzoli, “Market Definition And Market Power Evaluation In Dynamic Industries”, Network Law Review, Summer 2023.


Related Posts