Unveiling nuances in market definition: a closer look at the EU's updated notice

The public consultation on the 23-year-old Market Definition Notice was launched over the summer of 2020. The objective behind this consultation is to determine whether the Notice is still fit for purpose and if/how it can be improved to better reflect new market developments.

Market definition often plays a pivotal role in the application of the competition rules. There is a continued need for a market definition notice that provides methodological guidance on defining relevant markets for the purpose of competition law. For undertakings, legal practitioners, national competition authorities and the national courts, the Notice is an important tool to validate their approach.

However, since 1997, the case-law has brought new perspectives and methods on how to define a relevant market for the purposes of EU competition law. We explain in our contribution how this can be reflected in the revised Notice.

In terms of next steps, the Commission aims to publish the results of the evaluation phase in mid-2021. Adoption of the new Notice could take place in the course of 2022.

Our contribution

As a preliminary remark, please note that most of the public consultation questions were formatted in such a way that the respondent first selects a “yes / no / other etc.” answer (answer chosen underlined in the below) and then provides an explanation for this answer.

I. General questions on the notice

I.1. In the last five years, have you or your company / (business) organisation been required to assess the relevant product and geographic market for competition law purposes?

Yes – No – Do not know – Not applicable

I.2. If your reply to question I.1. was yes, please specify the type of competition law assessment

Assessment of a concentration between undertakings under Council Regulation Nº 139/2004 (the EU Merger Regulation); assessment of concerted practices and agreements between companies under Article 101 of the Treaty; assessment of abuse of dominance by an undertaking under Article 102 of the Treaty; assessment under the national competition law of one of the 30 states of the European Economic Area; assessment under the national competition law of a jurisdiction outside of the European Economic Area; other: civil law proceedings (to assess whether undertakings are competitors, eg, in relation to tort of unfair competition or breach of confidentiality); state aid analysis (to assess potential impact on competition).

I.3. How often do you consult the Notice?

Frequently (several times per year).

I.4. Do you consult the Notice for any purpose other than competition law assessment?

Yes – No – Do not know – Not applicable

In certain civil law proceedings (as specified above under I.2.1).

II. RELEVANCE

II.1. Is there still a need for a Notice to provide correct, comprehensive and clear guidance on market definition?

Yes – No – I do not know

II.1.1. Please explain your reply

We believe there is a continued need for a market definition notice that provides methodological guidance on defining relevant markets for the purpose of competition law. Market definition often plays a pivotal role in the application of the competition rules (eg, to assess the applicability of a block exemption). For undertakings, legal practitioners, national competition authorities and the national courts, the Notice is an important tool to validate their approach.

This being said, it is important to emphasize that the Notice is 23 years old and needs to be revised to ensure that it also provides sufficient guidance when delineating dynamic and innovative markets. The Notice focusses mainly on short-term demand substitutability, and it lacks proper guidance as to the qualification of innovation as an important parameter of competition in dynamic market environments.

In practice, partly due to the abstract nature of the Notice and partly due to practical difficulties in gathering the required data for making a price elasticity analysis, we consider that it is often difficult to apply the Notice. For this reason, the market definitions adopted in previous cases are often a better suited tool to predict how the relevant market should be defined. In particular, the Commission’s database of merger decisions is used extensively by practitioners.

We believe that it would be beneficial if the Notice were to be supplemented by additional tools to aid with market definition. In this regard, we think it would be very helpful if a more extensive database would be available, containing product and geographic market definitions from past cases and categorized by NACE code, with a brief explanation of the relevant factors. We are aware of the existence of certain commercial services, but we believe a database should be available that is accessible to all undertakings free of cost. This could, in our view, significantly reduce (legal) advisory and compliance costs.

III. Effectiveness

III.1. Have the following aspects within “Definition of relevant market” (paragraphs 7- 12) provided correct, comprehensive and clear guidance?

• Definition of relevant product market and relevant geographic market (7-9): Yes – Partially – No – I do not know.
• Concept of relevant market and objectives of Community competition policy (10-11): Yes – Partially – No – I do not know.
• Differences between market definition in assessing past behaviour (antitrust) and in assessing a change in the structure of supply (merger control) (12): Yes – Partially – No – I do not know.

III.1.1. Please explain your reply, including, if applicable, how the guidance may be incorrect, incomplete or unclear

In spite of the rise of emerging dynamic markets, the definition of the relevant product and geographic market and the paragraphs providing background to the objectives of market definition (paras. 7-11) in our view largely remain unchanged.

In relation to para. 7, we note that the current Notice defines a relevant product market as those products which a consumer regards as substitutable on the basis of the product’s (i) characteristic, (ii) price, or (iii) use. Yet, further clarification should be given in relation to whether these are alternatives or if a hierarchy exists between them (which appears not to be the case in light of para. 25 of the Notice). However, undertakings and legal practitioners are often left with the sense that the Commission’s own practice suggests the SSNIP test is the “gold standard” by which market definition is outlined. In our view, it is not clear when the Commission can simply revert from a SSNIP test to a characteristics test to define the market and / or how much weight a competition authority should place on a product’s characteristic or use in such a price analysis – and the wording of para. 36 is not overly informative as to how this weighting process should take place.

In relation to para. 12, we believe the current text is open to misinterpretation. We agree that there may be differences when making a retrospective analysis (aimed at defining the relevant market for a situation occurring in the past) as opposed to a prospective analysis (taking into account anticipated future developments). However, we consider it important to point out that a prospective analysis is not only required in relation to concentrations, but also often when advising on undertakings’ anticipated behaviour from the perspective of Art. 101 or 102 TFEU (eg, in relation to a new cooperation to be set up or a new policy to be adopted). As currently drafted, para. 12 could be misinterpreted as indicating a distinction between market analysis in merger cases as opposed to cartel or abuse of dominance cases, whereas the proper distinction should be between past, current and future situations, regardless of the area of competition law.

III.2. Have the following aspects within “Basic principles for market definition” (paragraphs 13-24) provided correct, comprehensive and clear guidance?

• Competitive constraints (13-14): Yes – Partially – No – I do not know.
• Demand-side substitutability (15-19): Yes – Partially – No – I do not know.
• Supply-side substitutability (20-23): Yes – Partially – No – I do not know.
• Potential competition (24): Yes – Partially – No – I do not know.

III.2.1. Please explain your reply, including, if applicable, how the guidance may be incorrect, incomplete or unclear

Para. 14 indicates that demand substitution is the most immediate and most effective disciplinary force on undertakings. Whilst in traditional markets this will often be correct, it is not always the case in dynamic, fast-evolving tech markets. In those markets, the threat of potential market entry of fully new (disruptive) technologies and products is often equally or more important than the competitive constraints exercised by existing substitutes. This dynamic aspect of market definition in our view is not sufficiently reflected in the Notice.

Paras. 15-19 on demand substitution are exclusively focussed on price competition and therefore do not capture other important factors of competition. Non-price competition may result in equally important competitive constraints, for instance in relation to sustainability, and should in our view also be reflected in the Notice.

In addition, the Notice should in our view provide guidance for markets where the SSNIP-test cannot easily be applied for other reasons. We believe it would be beneficial if the Notice were to address how demand substitution can be measured in tipped markets, multi-sided markets and zero price markets.
A tipped market is characterized by network effects, and user demand is locked in due to high adoption rate of the same product by other users. In such markets, price competition is often not a relevant factor. This is even more the case for multi-sided markets, where an undertaking’s competitive behaviour on one side of the market may not be driven by price elasticities at all, if it is driven by considerations relating to the other side of the market. The SSNIP-test also cannot be applied to the increasing number of zero-priced markets developing, in particular in relation to online platforms. For zero-priced markets, a potential alternative for assessing demand elasticity could be an analysis focussing on quality, namely based on a small but significant non transitory decrease in quality (SSNDQ).

Paras. 20-23 on supply substitution also focus on price changes as an incentive for market entry. These paragraphs fail to capture other relevant factors for market entry (such as economies of scale and scope in production, increasing network effects and levels of control of data) (or, indeed, for not entering a market, even if, from a pure price perspective, it might be profitable).

We believe the Notice should also address other competitive restraints than supply and demand substitution. In particular, in digital and dynamic markets, competitive pressure also arises from non-substitute products, services and business models. R&D and innovation competition can form important competitive restraints. Dynamic market contexts require the market analysis to incorporate behaviour that takes place before a relevant product market has properly emerged, due to the fact that this can have significant impact on innovation and competition.

Para. 24 of the Notice in our view is too dismissive of the role of potential competition – which is excluded from the market definition. This approach fails to capture that in dynamic markets, competitive constraints arising from potential competitors can be significant and should be incorporated in the market definition. It should further be borne in mind that especially in digital markets, undertakings often compete by supplying non-substitute products or highly imperfect substitutes. In particular, competitive pressure might be exercised by products relying on different technological infrastructures or supported by distinct business models. If competition on innovation and the uncertainties attached are addressed at the market definition stage, this prevents market shares from being relied upon too much in a dynamic market context.

One further point is that para. 24, in our view, inappropriately excludes potential competition from the market definition assessment – while at the same time, potential competition is properly included in other areas of competition policy, such as when defining the category of agreements which are capable of falling foul of Article 101 TFEU (ie, in the form of agreements between potential competitors) or subject of a block exemption regulation (eg,, Article 1(1)(c) of Regulation 330/2010). To the extent that there are justifiable policy reasons for sanctioning agreements between potential competitors, it is neither intuitive or consistent that the policy underpinning market definition would entirely exclude potential competition from its analysis.

III.3. Have the following aspects within “The Process of defining the relevant market in practice” (paragraphs 25-35) provided correct, comprehensive and clear guidance?

• Product dimension (25-27): Yes – Partially – No – I do not know.
• Geographic dimension (28-31): Yes – Partially – No – I do not know.
• Market integration in the Community (32): Yes – Partially – No – I do not know.
• The process of gathering evidence (33-35): Yes – Partially – No – I do not know.

III.3.1. Please explain your reply, including, if applicable, how the guidance may be incorrect, incomplete or unclear

As a general comment in relation to Section III of the Notice, we note that the Notice clearly reflects the Commission’s position in gathering information and evidence for the purpose of market definition. It should be kept in mind that the Commission has a highly privileged position in this regard in comparison to undertakings, legal practitioners and (national) courts. While the Commission (as most NCAs) has the option of launching market surveys to gather evidence, bolstered by an obligation to cooperate for the addressees of surveys, this option is not, or only to a significantly lesser extent, available to undertakings and legal practitioners. Undertakings in particular have no way of making other market participants cooperate with surveys or information requests and very often they will find market participants reluctant to do so.

In particular, in view of the self-assessment approach adopted by the Commission since 2004, the limited availability of market information often poses a significant impediment to undertakings and their legal advisers to be able to advise on the application of competition law with sufficient precision. For this reason, as mentioned under II.1.1. above, we believe the Notice should be supplemented with additional tools such as a well-organized database of previous market definition decisions.

In relation to para. 25, we believe that the open approach to empirical evidence expressed in the last sentence is not fully aligned with the Commission’s approach in practise. We believe that the approach taken (correctly, in our view) by the Commission and NCAs is to value evidence derived from actual behaviour higher than evidence derived from surveys. This approach reflects the fact that surveys are generally more susceptible to an incorrect outcome due to biases or alternate interests of participants. We believe this approach should be reflected in the Notice.

With respect to the last sentence of para. 29 we repeat the comment made under III.2.1. above that an exclusively price-focussed analysis may miss important other factors, on the basis of which customers may or may not decide to switch their demand to suppliers located elsewhere.
We note that it has become practice for the Commission and many NCAs to rely on undertakings’ internal documents as a source of evidence. This should be incorporated in the Notice. In particular, the Notice should specify which types of internal documents may potentially be used and reflect on their evidentiary value.

III.4. Have the following aspects within “Evidence to define markets – product dimension” (paragraphs 36-43) provided correct, comprehensive and clear guidance?

• Introductory paragraphs (36-37): Yes – Partially – No – I do not know.
• Evidence of substitution and quantitative tests (38-39): Yes – Partially – No – I do not know.
• Views of customers/competitors and consumer preferences (40-41): Yes – Partially – No – I do not know.
• Barriers and costs of switching (42): Yes – Partially – No – I do not know.
• Different categories of customers and price discrimination (43): Yes – Partially – No – I do not know.

III.4.1. Please explain your reply, including, if applicable, how the guidance may be incorrect, incomplete or unclear

In the introduction (paras. 36-37) we believe an extra paragraph should be included which stipulates that competitive pressure in digital markets can be exerted by complementary and non-substitute products, services, and business models.

With respect to paras. 40-41, we refer back to our comments made under III.3.1 above in relation to the limited availability of and practical difficulties in obtaining such evidence. We believe more guidance would also be welcome on the value to be attached to market and consumer surveys as sources of evidence. In particular in relation to consumer surveys, we consider that the often low response rate and the corresponding issues for the representativeness of such surveys (ie, how to exclude that the survey is mostly responded to by a small minority of biased consumers while the majority of neutral consumers do not respond?) should be addressed.

III.5. Have the following aspects within “Evidence for defining markets – geographic dimension” (paragraphs 44-52) provided correct, comprehensive and clear guidance?

• Evidence of diversion to other areas (45): Yes – Partially – No – I do not know.
• Demand characteristics and views of customers and competitors (46-47): Yes – Partially – No – I do not know.
• Geographic patterns of purchases and trade flows (48-49): Yes – Partially – No – I do not know.
• Barriers and costs of switching (50): Yes – Partially – No – I do not know.
• Examples from Commission practice and relevance of different factors (51-52): Yes – Partially – No – I do not know.

III.5.1. Please explain your reply, including, if applicable, how the guidance may be incorrect, incomplete or unclear

With respect to evidence of diversion to other areas (para. 45) and current purchase patterns (para. 48) we note that indications of such diversion can indeed, as suggested in the Notice, be an indication of a wider geographic market, but that the opposite is not necessarily true. In the absence of a reason in the past to divert orders, customers may have chosen local suppliers but in itself that is not indicative of customer behaviour in response to a reason to divert (such as a price rise, reduced availability or other relevant factors).

The paragraph on switching barriers (para. 50) is focussed mainly on trade in physical goods (ie, transport costs, custom tariffs). We believe also other types of switching barriers, in particular in relation to (digital) services should be addressed in the Notice.

In relation to (digital) services, customers’ switching costs and the barriers to purchase from foreign companies form less of a switching barrier. Other types of switching barriers that are more present within the digital services spectrum are consumer privacy, personal data protection and the lock-in effects of dominant platforms and gatekeepers.

III.6. Have paragraphs 53 to 55 on the “Calculation of market share” provided correct, comprehensive and clear guidance?

Yes – Partially – No – I do not know

III.6.1. Please explain your reply, including, if applicable, how the guidance may be incorrect, incomplete or unclear

In relation to para. 53, we refer back to our comments made under III.3.1 above in relation to the limited availability of market data. Unlike the Commission’s position, it is not a realistic option for undertakings to ask their competitors about sales data (which moreover may be competitively sensitive information that cannot be exchanged in view of the cartel prohibition). In many cases, firms typically have a good general sense of their market share. However, the difficulties for firms to self-analyse their market share cannot be overstated and raise wider policy issues beyond the scope of the Notice, such as fairness and due process, as to when a firm’s market share is used to identify or impose legal obligations. Such issues typically occur in the context of the application of merger control thresholds that adopt a market share test (eg, Spain and Portugal) or in the application of a block exemption regulation (eg, Regulation 330/2010). Often it is very difficult in practice to define the relevant market and to establish an undertaking’s market share with the degree of precision that is required to apply hard thresholds carrying important legal consequences.

While, in itself, it seems correct, as mentioned in para. 55, that sales value information often may give a better impression of the relative position of undertakings, it should be kept in mind that in practice most often there is no option to choose, as sales value information often is not available.

III.7. Have paragraphs 56 to 58 on the “Additional considerations” provided correct, comprehensive and clear guidance?

Yes – Partially – No – I do not know

III.7.1. Please explain your reply, including, if applicable, how the guidance may be incorrect, incomplete or unclear

We believe that the points addressed in this section on primary and secondary markets and chains of substitution are valid and important. We do believe, however, that other additional considerations ought to be included in the Notice, in particular in relation to trade patterns in digital markets. In this regard, we refer back to our comment made under III.2.1. above in relation to tipped markets, multi-sided markets and zero price markets.

III.8. Do you consider that there are any major points of continuity (for example legal, economic, political, methodological, or technological) that have not changed since 1997 and that you consider should continue guiding the principles of the Market Definition Notice going forward?

Yes – No – I do not know

III.9. If yes, please identify in the following table the major points of continuity that have not changed since 1997 and that you consider should continue guiding the principles of the Market Definition Notice going forward.

	Major points of continuity	Short explanation/concrete examples	Paragraphs of the Notice where those ideas are expressed
1	Product and geographic aspects of relevant market	The product and geographic dimension of the market are almost always a suitable starting point to enable parties to self-assess the relevant markets within which competition occurs.	paras. 25-27 paras. 28-31
2	Use of SSNIP test as default parameter for product market	The SSNIP test is often a good tool for identifying substitutability and indirectly is often also a relevant proxy for other competition-relevant features, eg, quality and pace of innovation.	para. 15
3	Focus on trade patterns and trade flows for geographic market	Significant (international) trade flows generally should support the hypothesis of a broader geographic market. Potential increases in the level of trade can discipline producers in the event of price increases. This is also applicable in more dynamic markets.	para. 29

III.10. Do you consider that there are major trends and developments (for example legal, economic, political, methodological, or technological) that have affected the application of the Notice but are currently not reflected in it?

Yes – No – I do not know

III.11. If yes, please identify in the following table the major trends and developments that you consider have affected the application of the Notice but are currently not reflected in it. Please describe the specific shortcomings of the Notice in this regard, including concrete examples.

	Major trends/changes	Short explanation/concrete examples	Paragraphs of the Notice that may require an update	Specific shortcoming of the Notice
1	Impact of potential competition	Many markets have become more dynamic in character since 1997, which increases the competitive pressure exercised by potential market entry. It would be appropriate to reflect this in the Notice.	para. 24	Potential competition is not taken into account when defining markets under the current notice. Potential competition is only considered in the subsequent phase of the analysis. In markets with a dynamic character where innovation is at a stage at which it will reach the market with a certain predictability, potential competition may be an actual disciplinary factor on the existing market and should be taken into account in the market delineation phase. The Notice should reflect this.
2	Developments in the way in which (mostly digital) products and services are offered and marketed	Developments not addressed by the Notice are tipped markets, multi-sided markets and zero price markets.	paras. 15 - 19	The Notice should address how demand substitution can be measured in tipped markets, multi-sided markets and zero price markets.

III.12. Is there any area for which the Notice currently does not provide any guidance, but which would be desirable?

Yes – No – I do not know

III.12.1. Please explain your reply

As mentioned above under III.2.1., we believe the Notice should address instances where classic market definition methodologies do not or not fully apply (eg, in relation to tipped markets, multi-sided markets and zero price markets) and set out the approach to be followed in relation to such markets.

IV. EFFICIENCY

IV.1. Are the net benefits – benefits net of costs - associated with following the guidance described in the Notice positive (compared to a situation without the Notice in place)?

Yes, the net benefits are positive (the benefits of having the Notice in place exceed the costs thereof).

IV.1.1. Please explain your reply and, if possible, quantify the magnitude of the (positive or negative) net benefits.

We believe the Notice provides important guidance on market definition which in itself is a significant benefit. As observed above, the methodologies described in the Notice may not in all cases be suitable or even possible to apply in practice, in particular in the context of a self-assessment. Nevertheless, even in such cases the Notice may have added value by identifying alternative methodologies and approaches. A new version of the Notice, addressing more alternative options, might increase its added value.

V. COHERENCE

V.1. How well do the different components set out in the Notice operate together?

The different components of the Notice work well together without apparent contradictions.

V.1.1. Please explain your reply, especially if you have identified any contradictions

We do not believe the Notice to be inherently contradictory. As observed above, we do believe the Notice to be incomplete in view of developments since 1997, in particular in digital markets.

V.2. Is the Notice coherent with other instruments that provide guidance on the interpretation of the EU antitrust rules (based on Articles 101 and 102 TFEU)?

Yes – No – I do not know

V.2.1. Please explain.

We believe that the Notice is, and should remain, value-neutral in relation to Articles 101 and 102 TFEU although the Notice raises certain inconsistencies in relation to potential competition – see III.2.1 above.

V.3. Is the Notice coherent with the Merger Regulation and with other instruments that provide