The New Market Definition Notice: Embracing Change
On 8 February 2024, the Commission published the long-awaited final text of the revised Market Definition Notice. Over a quarter of a century after the adoption of the first Market Definition Notice in 1997, we have a new and improved version ready for the modern age. The revised Notice covers a lot of ground: it not only consolidates knowledge and Commission practice of the last two decades, but it also introduces new concepts, signalling the Commission’s openness to incorporating new ideas, even when they may still need further development.
Getting here was no mean feat. The Commission went through various stages of study and consultation to get to this final text. In November 2022, the Commission released a first version – a draft Revised Notice – upon which I commented in a blog last year. In that blog, I argued that the Draft Revision brought the Notice further into the 21st century, but that there remained a few confusions to clear up and some open questions to address. In this blog, I want to reflect on the substance of the final text of the Notice. The post will focus on a few selected topics: the role of market definition, the inclusion of sustainability, and the contexts of digital markets, innovation and forward-looking assessments.
Greater clarity on the role of market definition: the variability of the market
A thorough grasp of the purpose of market definition is crucial to understanding its utility and scope as well as to select the most appropriate methodologies. For too long, reflections on market definition brushed over the role it plays in competition law. In recent years, however, a few scholars have queried the purpose of market definition, often in response to calls to abandon market definition (e.g. Robertson, Glasner & Sullivan, Podszun). I argued, for example, that the focus on market power in debates about market definition was short-sighted, as market definition’s utility goes beyond the indirect establishment of market power, allowing for the identification of competitive constraints which are relevant to assessing the feasibility of the conduct and anti-competitive effects in light of the theory of harm (the ‘purposive’ approach).
The revision of the Notice was an ideal opportunity, therefore, for the Commission to shed more light on the role of market definition. It is in this area that a clear evolution can be seen between the draft and the final text of the revised Notice.
The revised Notice expands on the role of market definition, in comparison to the 1997 version. The revised Notice still, understandably, puts market power at the centre of the exercise, referring to it as “the most notabl[e]” use of market definition (para 7) and even dropping the ‘inter alia’ wording which was present in the 1997 Notice (para 2 of the 1997 Market Definition Notice) which suggested that market definition had other functions beyond the calculation of market shares. However, in both its draft and final version, the revised Notice portrays market definition as an exercise inherently linked with the goals of the law and the focus of the analysis, beyond merely market power. Both the Draft and Final version of the Notice recognised that market definition is “closely related to the objectives” of the different competition law instruments (see para 8 of the Draft Notice, now para 9 of the final version of the Notice). In the same paragraph, it sets out why and when market definition is used (or not used) under Article 102 TFEU, the EUMR, and Article 101 TFEU.
What is most interesting about the revised Notice is the extent to which it recognises the purposive nature of market definition and, consequently, explicitly acknowledges that market definition may vary depending on the focus of the analysis in a case. Here, there is a notable evolution between the draft and the final text of the revised Notice. Although the Draft took some first steps towards explicitly recognising the purposive nature of market definition, and. thus, the possibility that market definition may vary depending on the focus of the analysis in a case, the Draft’s approach was inconsistent and confused. This inconsistency has been remedied in the final version.
If market definition is purposive and influenced by the different scopes and objectives of the different provisions of competition law, then there is likely to be some variation in the market definition exercise (depending on the legal basis, alleged conduct, and theory of harm). The Draft was not clear on this, however. In its paragraph 8 on the objectives and uses under each provision, it did not specify whether the use of market definition under each instrument could lead to different processes or outcomes. The further paragraphs of the Draft did not help clarify the question. Paragraph 11 of the Draft noted that “the outcome of market definition in a given case is usually unaffected by whether it takes place in the context of merger control or antitrust enforcement” (own emphasis). At the same time, the supplementary comment in footnote 20 noted that “in some cases, market definition may lead to different results depending on the type of analysis carried out”, particularly depending on the distinction between a focus on changes in market power or level of current or past market power/competition (further explanation and context of the SSNIP test was given in footnote 46 of the Draft). Moreover, in paragraph 15 the Draft referred to different situations in which the outcome of market definition can differ, including different parameters of competition.
The final text of the Revised Notice removes some of these ambiguities. In doing so, the Notice could be read as a closer acknowledgement of the variability of market definition. First, it removes the statement that the outcome is ”usually unaffected” from paragraph 14 (the new numbering of draft paragraph 11) and deletes the confusing footnote 20. Second, it adds to the wording of paragraph 18 (the new numbering of the Draft in paragraph 15). The wording it added explains why the outcome of market definition can differ: “as market definition seeks to identify the effective and immediate competitive constraints that are relevant for the competitive assessment of a specific conduct or concentration, the outcome of market definition can differ” (own emphasis). Like the Draft, the new text lists “parameters of competition” as factors which may influence the outcome of market definition. Yet, it does not stop there, by adding “the competitive concerns under consideration” as a factor in the list of things which may influence the outcome. Adding this explicitly to the list provides a clearer signal of the purposive nature of market definition.
Welcome to the future: a green and digital world full of innovation
The revised Notice brings the guidance on market definition into the modern day – and provides some glimpses of what is yet to come. The Notice reflects the wide range of considerations relevant to competition law, including sustainability, digital markets, and innovation and the quest to predict future market developments.
The green transition
Unsurprisingly, given the emphasis on digital markets in the lead-up to the revision, the revised Notice puts greater emphasis on non-price parameters of competition, such as innovation and quality (including privacy, image of security, or ability to integrate the product with other products). The transition to a digital economy is not the only evolution which influences market definition, however: environmental sustainability also features in the revised Notice. In the context of the calls to expand the remit of competition law, climate change has been the most notable success – something which is reflected in the revised Notice.
The revised Notice is explicit that there is a relationship between the objectives of competition law and market definition. In addition to the references to the objectives of the competition law instruments in paragraph 9 (which I mention above), the revised Notice also puts market definition in the context of competition law’s role in the twin green and digital transition (para 3). This wording was already present in the Draft (para 2), but the final version is even more expansive, acknowledging that competition law can complement the Union’s regulatory framework on environmental sustainability by considering sustainability as a factor of competition in the competitive assessment and in market definition (para 3 of the final text, see also para 15).
Digital markets
The revised Notice’s biggest changes remain in the spheres of digital markets and innovative industries. The revised Notice incorporates some lessons learned in cases involving digital markets. First, it provides guidance on the approach the Commission may take to the definition of markets in the context of multi-sided platforms (paras 94-96). The Notice suggests that the Commission may define a market for the platform as a whole or may define separate markets for each side of the platform (particularly where there are significant differences in the substitution possibilities on the different sides). The Commission acknowledges the importance of considering the interrelatedness of the sides (the network effects) in any case where relevant. Although the explanation here remains brief, there are several decisions which shed more light on the Commission’s actual practice.
Second, the Notice highlights the possibility that a service is supplied at a zero monetary price or negative prices and puts forward the use of alternatives to the SSNIP test such as an SSNDQ test (‘Small but Significant Non-transitory Decrease in Quality’) (paras 97-98). The Notice does not solve all challenges in the application of SSNDQ tests, a conceptual framework which is, after all, much less established than the SSNIP test and whose methodology still requires some further development. The Notice does provide a list of various aspects of quality which may be relevant (para 15). It also refers to the Google Android decision in which it used SSNDQ as a framework (footnote 54). In the same footnote, the Commission acknowledges that challenges remain in establishing a quantitative test on the same par as the SSNIP test, including the quantification of quality and the assessment of profitability.
The revised Notice also refers to a concept which is far less established in the practice of market definition: the ecosystem. Although ecosystems (theories) are increasingly becoming part and parcel of the discourse on competition law in digital markets, there is not yet a consensus on how to identify them for competition law enforcement and there is only little decisional practice so far in which they have played a role in the assessment (e.g. Google Android, Booking.com/eTraveli). It is not surprising, therefore, that the Notice does not revolutionise market definition in the context of ecosystems. Instead, it acknowledges that ecosystems “can, in certain circumstances, be thought of as consisting of a primary core product and several secondary (digital) products whose consumption is connected to the core product” (para 104) and/or the secondary products may be offered as a bundle. Thus, the Notice puts forward, that it may be possible to apply the principles of after-markets or bundles for the definition of markets in an ecosystem, although this approach will not always fit. Although the Notice does not provide extensive guidance on how to define markets for ecosystems, the sole fact of its inclusion in the Notice is a big step: it signals that ecosystems can play a further role in the future and that we need to actively consider how to incorporate them in the analysis.
Forward-looking assessments
The revised Notice devotes considerable space to forward-looking market definition.
It refers in various paragraphs to markets which may undergo structural transitions, such as the emergence of new products or technology or regulatory changes which affect supply and demand. The Notice raises the possibility that the Commission will take into account “sufficiently foreseeable changes” in demand and competitive constraints (para 8) where the short- or medium-term structural transitions would lead to ”effective changes in the general dynamics of supply and demand” within the period relevant to the assessment (para 21). The introduction of new products or processes may lead to such structural market transitions, so that the Commission may consider expected changes in substitution possibilities (and not just existing substitutes) (para 55).
When it comes to the evidence for such changes, there is a notable difference in wording between the draft and final version of the text: whereas the draft required “sufficient probability” or sufficient “certainty” that the changes will take place, the final version merely requires “sufficient likelihood” (para 21 of the final text, compared to para 16 of the draft; paragraph 77).
In addition to the paragraphs on structural transitions, the revised Notice has a specific section dedicated to assessment in the context of significant R&D. The Notice distinguishes between two scenarios in which innovation may be relevant to market definition. First, there may be pipeline products which, although not yet available, can be sufficiently identifiable to consider their future possibility as substitutes in an existing product market or for a new product market (para 91). Second, undertakings may be engaging in innovation efforts – R&D processes – where it is not yet possible to specify the resulting future products. Although it would not be feasible, therefore, to identify ‘product’ markets, the Notice considers that in some cases it may still be beneficial to the assessment to identify the boundaries of innovation competition (para 92).
The inclusion of this second type of ‘market’ definition is not merely theoretical, but reflects the practice of the Commission in the Dow/Dupont (M.7932) and Bayer/Monsanto (M.8.84) mergers, where it identified ‘innovation spaces’. Although the Commission’s wording seemed cautious in these decisions, careful not to call these innovation spaces ‘markets’, I would argue that it adopted approaches very reminiscent of traditional market definition. The revised Notice acknowledges that it is challenging to identify a “product market in the strict sense”, but its inclusion in the Notice signals that is market definition of some sort – even if we forego the ‘traditional’ focus on products.
This does not seem to be the end of the discussion on market definition in the context of innovation. Much is still left to be fleshed out, for example on forward-looking assessments outside of merger control, or whether it is possible to identify ‘innovation spaces’ for digital markets where R&D efforts may not have the same features as in agrochemical or pharmaceutical industries.
Conclusion
The revised Notice covers a greater variety of issues and more clearly illustrates the complexity of market definition. It is no surprise, then, that the Notice has nearly doubled in length, from 58 paragraphs in 1997 to 115 paragraphs in 2024.
As competition law becomes more complex (integrating a wider range of considerations and parameters), market definition will, too. Although some may raise questions about this direction of travel, it is commendable that the Commission provides a realistic overview of various issues which may be considered when defining relevant markets. The revision of the Notice was meant to provide comprehensive and clear guidance. The revised Notice certainly covers a lot of ground. Whether it will also provide sufficient clarity on what can actually be expected in a specific case remains to be seen.
The Commission has made great strides in bringing market definition into the 21st century. The journey of market definition is by no means at its end. It is a road we will all travel together, with the Notice as a necessarily incomplete roadmap. As I have argued elsewhere, objectivity in market definition is unlikely to be achieved through predictable outcomes, making it particularly important that the Commission provides enhanced transparency about the process. As new problems are solved, the Commission may wish to update the Notice more regularly.
