An extract from The Dominance and Monopolies Review - Edition 9

Market definition and market power

Since the major overhaul of German competition law by the 10th Amendment to the ARC in 2021, there are now three different concepts of market power in German competition law.

  1. The assessment of 'absolute' single or collective dominance requires a definition of the relevant market as well as the assessment of market power on the relevant market based on the prevailing market structure and market conditions. In the assessment of dominance, the FCO and the German courts continue to place considerable importance on market shares and have only slowly started to adopt the more sophisticated economic analyses used by the EU Commission.
  2. In contrast, the assessment of 'relative' market power focuses more on the distribution of bargaining power between potential addressees of the provisions on relative market power in vertical relationships or 'superior market power' in relation to small or medium-sized competitors of the potential addressee.
  3. The new concept of 'paramount cross-market significance' is designed to enable the FCO to pre-emptively intervene against activities of large platforms even in markets where they are not yet dominant, but could become dominant quickly due to their resources, strategic positioning and the 'specificities of markets of the digital economy', namely network effects, data advantages and self-reinforcing effects (tipping) in digital ecosystems. As opposed to traditional market power analysis, cross-market significance is not tied to a clear market definition and does not require dominance in at least one market. The definition of cross-market power does not differ significantly from the definition of dominance on multi-sided markets.
i Market definition

In defining relevant product markets, the German courts and the FCO primarily analyse the substitutability of goods and services from a demand-side perspective based on the intended use, characteristics and price of the relevant products. In some cases, the FCO also refers to the 'small but significant and non-transitory increase in price' test as an additional, but not the only or the principal, criterion for market definition.23 The concept of supply-side substitution (i.e., other manufacturers being able and willing to adjust their production within a short time and without significant cost) is also taken into account if the demand-side perspective leads to overly narrow markets.24

As under EU law, the relevant geographic market comprises the area in which the enterprises concerned compete, in which the conditions of competition are sufficiently homogeneous, and that can be distinguished from neighbouring areas because of appreciably different competitive conditions.25

In practice, ex post behavioural enforcement tends to take a somewhat narrower view on market definition than merger control, given that the perspective of specific customers or competitors potentially harmed by the conduct at issue that cannot switch to alternative suppliers, can sometimes influence the assessment.

The German legislator has also clarified that a relevant market (and consequently also market power on it) may be found even if the relevant services are rendered free of charge.26

ii Dominance

As previously noted, German competition rules on unilateral conduct apply to companies in a position of single or collective dominance, and to companies enjoying 'relative' market power in vertical relationships or over small and medium-sized competitors. Section 18 of the ARC defines single and collective dominance.

Single dominance

According to Section 18(1) of the ARC, single dominance exists if a company is either without competitors, or not exposed to significant competition or in a 'superior market position' as compared with its competitors (which can exist even if there is significant competition in the market). Single dominance exists where the market power of an enterprise enables it to act without sufficient constraints from the marketplace (i.e., a situation in which an enterprise is able to act to an appreciable extent independently of its competitors, customers, suppliers and, ultimately, consumers).

Section 18(3) of the ARC lists the following criteria that may in particular be taken into account for the assessment of whether a company is in a 'superior market position':

  1. the enterprise's market share;
  2. its financial resources;
  3. its access to competitively relevant data;
  4. its access to input supplies or downstream markets;
  5. its affiliations with or links to other enterprises;
  6. legal or factual barriers to market entry;
  7. actual or potential competition by domestic or foreign enterprises;
  8. its ability to shift its supply or demand to other products; and
  9. the ability of the enterprise's customers or suppliers to switch to other enterprises.

The 10th Amendment clarified that the extent of an undertaking's access to competitively relevant data is also a factor in assessing market power in traditional markets that are not multi-sided.

In practice, the FCO and the German courts tend to focus on whether an enterprise has sufficient market power to determine the most important business parameters. An appraisal of market shares (both in absolute and relative – compared with competitors – terms) is still the most important factor. The rebuttable market share-based presumption pursuant to Section 18(4) of the ARC provides an important first indication of possible dominance where the market share of a company exceeds 40 per cent.27 While not impossible, it is often difficult in practice to rebut the presumption with economic arguments, especially in the case of high market shares substantially above the presumption threshold. The underlying reason is that German law expressly stipulates that a dominant position can be based on a 'superior' market position, even if the company concerned faces significant competition from its rivals.

Additional criteria for the assessment of market power in multi-sided markets and networks28 are provided in Section 18(3a) of the ARC. In particular, the following criteria must be taken into account when assessing a company's market position on multi-sided and network markets:

  1. direct and indirect network effects;
  2. the parallel use of more than one service (multi-homing) and the difficulties faced by users in switching services;
  3. economies of scale in connection with network effects;
  4. the company's access to data relevant for competition; and
  5. competitive pressure driven by innovation.

Market shares typically play less of a role for the determination of dominance on digital markets with free (e.g., paid by advertisement) online services and multi-homing, whereas other criteria, such as user numbers, are typically taken into account on these markets.

In the new Section 18(3b) of the ARC, the German legislator recently introduced the concept of 'intermediation power' to the list of criteria for assessing market power in digital markets. Intermediation power describes the importance of a company that is active as an intermediary between different market sides (e.g., suppliers and customers without direct contact with each other) for the access to upstream or downstream markets. The explanatory notes to the 10th Amendment underline in particular the importance of rankings for companies in this context.

The FCO applied these additional criteria in its decisions against German ticketing system operator CTS Eventim and Facebook. In CTS Eventim, the FCO found that the ticket platform enjoyed a dominant position with regard to event organisers and ticket offices in the two-sided platform market for ticketing services in Germany. In Facebook, the FCO based its dominance analysis of the German market for social media networks, in particular, on three factors:

  1. direct network effects resulting from Facebook's large number of users (creating a 'lock-in effect' for its users, as they would lose all their existing contacts if they switched to another social network – creating high entry barriers);
  2. indirect network effects that Facebook enjoys in relation to its advertisement customers (given the large number of Facebook users, advertisers cannot easily switch to another social network to reach as many users); and
  3. Facebook's access to users' personal data.

While the DCA has suspended the FCO's Facebook decision, it did not challenge the FCO's finding of market power in the interim proceedings. Upon the FCO's appeal, the FCJ quashed the DCA's decision concluding that there was 'no doubt' that Facebook had a dominant position in the German market for social media networks and abused it through the relevant user conditions (see Section IV.ii).

In its recent proceedings regarding Amazon's terms and conditions and its behaviour towards retailers on its German marketplace platform http://www.amazon.de, the FCO focused on Amazon Marketplace's role as an intermediary between resellers and consumers and gatekeeper to consumers who purchase their products online, in its assessment of market power on the two-sided market for the provision of online marketplace services.

Collective dominance

According to Section 18(5) of the ARC, collective dominance exists where there is no significant competition between the two or more largest companies in a market and where they jointly are not constrained sufficiently by competition from third parties. Collective dominance is defined as a few companies in an oligopolistic setting engaging in tacit coordination or collusion with the result that they effectively do not compete with each other.

Section 18(6) of the ARC also provides for market share-based legal presumptions for collective dominance. Thus, three or fewer companies are presumed to be collectively dominant if they have a market share of at least 50 per cent; and five or fewer companies are presumed to be collectively dominant if they have a market share of at least two-thirds of the market. These presumptions are rebuttable, and the companies can show that substantial competition exists between them individually or that they are jointly sufficiently constrained by outsiders or customers. For a non liquet, the presumption stands.

German courts have so far rarely addressed collective dominance issues outside of merger cases. The FCO and the German courts generally employ the criteria established by the EU General Court in Airtours v. Commission29 in determining collective dominance (albeit in a somewhat modified form).

'Relative' dominance and 'superior market power'

As noted above, going beyond the scope of Article 102 of the TFEU, Section 20 of the ARC prohibits exclusionary (and discriminatory) conduct not only by companies that are dominant in 'absolute' terms, but also (1) the abuse of 'relative dominance' in vertical relations by companies on which 'other companies depend' as suppliers or purchasers of certain kinds of goods or commercial services (Section 20(1) ARC), and (2) the abuse of 'superior market power' at horizontal level by companies enjoying 'stronger market power in comparison to their small and medium-sized competitors' (Section 20(3) ARC). These provisions aim at protecting (small and medium-sized) companies against anticompetitive conduct by their trading partners or larger competitors.

Relative market power in vertical relationships pursuant to Section 20(1) of the ARC requires that an enterprise is dependent upon another company as supplier or customer without adequate alternatives and the dependency is not offset by corresponding countervailing market or negotiation power. Section 20(1) of the ARC considers the following forms of dependency: the trading partner (1) is dependent on the other company's 'must have' brands or 'must have' products; (2) is in a position of 'company-specific dependency' (e.g., because it has specialised and designed its product or service to the needs of a particular purchase ('lock-in') after transaction-specific investments); or (3) is dependent due to scarcity of the product. Section 20(1), third sentence of the ARC establishes a presumption of dependency if a purchaser of goods frequently receives rebates or similar bonuses from its suppliers that go beyond customary rebates granted to other purchasers.

The 10th Amendment to the ARC extended the application of the concept of relative market power significantly. Relative market power no longer only exists in relation to small and medium-sized companies, but is now also possible in relation to all companies that are dependent on their trading partner. Hence, large companies can also invoke this provision if there is a significant asymmetry in the relationship with the relatively dominant company. Such dependency may also be based on the fact that a company is dependent on platforms with intermediary power for access to upstream and downstream markets if they cannot switch to other platforms (Section 20(1), sentence two). Pursuant to Section 20(1a), such dependency also exists if a company is reliant on access to data controlled by another company, even if the data has not been shared with any third parties.

Section 20(3) of the ARC further enlarges the scope of the abuse of dominance legal framework to companies that do not even qualify for relative market power in vertical relations, but that have particular power (i.e., 'superior market power' at horizontal level over certain smaller, but not all, rivals), and prohibits exclusionary conduct such as tying/bundling, predatory pricing and margin squeeze. Its main application area is retail food trade. An example of prohibited exclusionary conduct is frequent pricing below cost.30 The 10th Amendment to the ARC also prohibits the abuse of superior market power by gatekeepers and digital intermediaries by hampering the realisation of network effects by competitors. This marks a significant departure from the general principle that internal growth of companies is not sanctioned.

Paramount cross-market significance

As its centrepiece, the 10th Amendment to the ARC introduced the entirely new regulatory approach of paramount cross-market significance, which is designed to enable the FCO – as a preventive measure – to intervene against activities of large platforms even in markets where they are not yet dominant, but could become dominant quickly due to their resources, strategic positioning and the 'specificities of markets of the digital economy', namely network effects, data advantages and self-reinforcing effects (tipping) in digital ecosystems. The new concept of paramount cross-market significance applies to companies with significant activities in multi-sided, though not necessarily digital, markets or networks. As opposed to traditional market power analysis, cross-market significance is not tied to a clear market definition or dominance in at least one market. Rather, paramount cross-market significance is based on the following non-exhaustive, non-cumulative list of criteria to determine this position on a case-by-case basis: (1) dominance in one or several markets (though there is no need for the FCO to establish dominance in any relevant market); (2) financial resources and access to inputs; (3) vertical integration and activities in otherwise interrelated markets; (4) access to competitively relevant data; and (5) relevance of the company for the market access of third parties. However, there are no quantitative criteria such as turnover thresholds, meaning that the FCO will have significant discretion in its finding of paramount cross-market significance. While the law is specifically aimed at large gatekeepers and intermediaries in digital markets (in particular, the GAFA companies), it is also possible for other, smaller companies to be caught by it.

Unlike the traditional concept of market dominance, which is assessed ex post based on market shares and other relevant factors on a clearly defined market (see above), the new concept of paramount cross-market significance requires a declaratory order (which is valid for five years) formally establishing a company's paramount cross-market significance (Section 19a(1) ARC), which helps to deal with the legal uncertainty raised by the very abstract definition of 'paramount cross-market power'. This first order provides the FCO with the legal basis for a subsequent second order, prohibiting certain types of specifically listed practices, even in markets where the company is not (yet) dominant (Section 19a(2) ARC). The FCO can also join the declaratory order and the prohibition order in one decision. There is no need for the FCO to prove competitive harm of the conduct. In contrast, the company concerned bears the burden of proof to show that its conduct is objectively justified. The application of Section 19a of the ARC does not exclude the parallel application and issuance of fines under the traditional abuse of dominance rules. Section 19a(2) provides for an exclusive list of prohibited conduct, which includes self-preferencing, hindering other companies, tying and bundling, certain strategies with respect to data processing, interoperability, non-transparency and requests for unfair advantages.

Concerning the special regime for large digital companies with paramount cross-market significance, the German legislator has clearly taken inspiration from the European Commission's proposal for the Digital Markets Act (DMA). The new Section 19a of the ARC and the European Commission's draft DMA overlap in their scope of application but are not congruent. It is as yet unclear whether the DMA, once enacted, will prevail over Section 19a of the ARC, or whether the FCO will still be free to prohibit behaviours that would be allowed under the DMA (see further Section VIII).