The application of EU competition law to online platforms has been subject to much debate in the course of the past two years. The EU Commission and several national competition law authorities have launched multiple studies in order to identify the possible challenges in this context and establish an action plan. The distinguishing characteristics of online platforms including: the reliance on the two-multisided market model, the use of personal data and zero pricing indicate that the application of competition law may require some adjustments. The extent of such adjustments is far from being settled as well as any agreement with regard to anything resembling an action plan. Reality however does not wait until such plan is developed as it can be seen by the recent Dutch case in the matter of Funda Real Estate (in Dutch only) where the Dutch courts had to assess an alleged abuse of dominance by an online platform due to the application of differential treatment among its trading partners. The judgment by the Dutch court constitutes an important contribution to the application of competition law to online platforms. The findings of the Dutch court indicate that the current literature as well as the guidance paper developed by the Bundeskartellamt, for the market definition in cases concerning online platforms, may not be suitable for application in practice. Furthermore, the judgment establishes an important precedent indicating that online platforms, despite their distinguishing characteristics, will nonetheless be met by the same legal scrutiny as any other undertaking when it comes to the qualification of business practices as abusive. Accordingly, differential treatment by dominant undertakings, including online platforms, is prohibited solely when capable of creating a competitive disadvantage among the trading parties of the concerned undertaking.
Facts of the case
The case of Funda Real Estate concerns an action claiming the alleged abuse of dominance by the Funda online platform for real estate offerings. The claim was launched by the VBO group of Dutch real estate agents that publish their offers on the Funda platform. The abuse of dominance concerned the differentiated treatment granted by Funda to the various real estate agent groups that place their offers on the platform. According to the claim VBO real estate agents were charged higher prices to put their offers on Funda than NVM real estate agents (agents that are part on the Dutch national association of real estate agents). Furthermore the VBO real estate agents claimed the ranking system of the platform systematically placed their offers in less favorable position in the search results compared to those of the NVM real estate agents, making VBO real estate agents less attractive to potential sellers.
Establishing dominance- how many relevant markets?
In cases where an alleged abuse of dominance under art. 102 TFEU is claimed the first matter that needs to be assessed is whether the concerned undertaking is a dominant. Establishing dominance requires defining the relevant market as a prerequisite to the finding of dominance (Case T-62/98 Volkswagen v Commission).
In the case of online platforms the matter of market definition is a rather complex one as online platforms are two-or multisided markets. Online platforms interact with two or more separate customer groups and meet the demands of these groups by facilitating interaction between them. Accordingly, demand-side substitutability for the platform can be assessed with regard to more than one customer group, meaning that the market definition process might result in multiple relevant markets. Consequently defining the relevant market for online platforms first requires establishing how many markets need to be defined with respect to the online platform. Generally speaking the choice comes down to defining a single relevant market for the platform of separate relevant markets with regard to the customer groups participating on the platform.
The choice between these possibilities has significant consequences to subsequent test of substitutability and consequently to the scope of the relevant market. Determining that a single relevant market needs to be defined for the online platform means that the all the alternatives that will be considered as potential substitutes only if these can meet the demand of all the customer groups participating on the platform. Therefore the potential alternatives for the online platform must also be (online) platforms. In such cases the core product or service for which substitution is tested, is in essence the interaction between those distinct customer groups in some form of matchmaking. By contrast if separate relevant markets are defined with regard to the various customer groups of the platforms substitutability is tested separately with regard to these groups. Therefore in this second scenario the potential alternatives for the online platform with regard to each of the customer groups of the platform may also include non-platform undertakings.
In the case of Funda the Dutch court found, based on an unpublished expert report, that the relevant market in the case entailed the market for real estate websites in the Netherlands. With other words, a single relevant market was defined for the platform, which includes solely other platforms that bring together potential buyers and real estate agents. With regard to this relevant market the Dutch courts concluded the Funda had a dominant position. In this regard the findings of the Dutch courts provide proof that defining a single relevant market for a two-multisided (online) platform is not merely a theoretical possibility as it often portrayed to be. Furthermore the findings of the Dutch courts indicate that the choice between a single and multiple markets in the case of online platforms is not depended on whether the market is a transaction or a non-transaction one nor whether the platform is a matching or advertisement platform. In this case the Funda platform is in fact a non-transaction advertisement platform that, according to current theories by scholars and the Bundeskartellamt would require the definition of multiple relevant markets. Luckily however the Dutch courts chose a different approach, which fits in better with the commercial reality in this case. Furthermore, the market definition process in this case did not include all the various customer groups served by the Funda platform, which is a multi-sided online platform. Accordingly, the market definition was performed solely with regard to real estate agents and residential real estate buyers. Funda, however, facilitates additional interactions between renters and tenants of residential real estate as well as sellers and buyers of commercial real estate that were not considered relevant for this case. Consequently, this judgment indicates that in the case of multi-sided online platforms the definition of the relevant market may be restricted to only some of the ‘sides’ of the platform; something which was also not considered a possibility until now.
Differentiated treatment – can a dominant undertaking discriminate?
Pricing
Online platforms are two-multisided markets thus will almost inherently make use of a skewed pricing scheme where some customer groups pay more than other in order to participate on the platform. The question in this case however was whether the platform may charge different prices from trading parties within the same customer group, namely the real estate agents. In this regard a degree of price discrimination was found (according to the claimant – in a period of 8 year the price difference has amounted to approx. 640%). However the Dutch court, relying on the opinion of AG Wahl (Case C-525/16 MEO), indicated that a price difference does not inherently entail a competitive disadvantage as is required to prove for the purpose of establishing an abuse of dominance. This approach to differentiated prices was later also confirmed by the CJEU which followed the findings of the AG in this matter [par. 25-30]. According to the Dutch court the claimant did not provide sufficient evidence to prove such a competitive disadvantage and therefore concluded that the pricing scheme was not abusive [par. 2.16].
Platform ranking
In the matter of the differential placement of the offers placed by real estate agents, the Dutch court similarly required proof that the ranking practice resulted in a competitive disadvantage. Despite the fact that the differentiated ranking on the platform resulted in an average of 16% less inquiries by potential buyers, the Dutch court found that such practices did not distort competition among real estate agents. According to the court sellers do not chose real estate agents based on their raking on the Funda platform but rather based on other aspects such as price and client relations. Furthermore based on the study presented to the court the non-NVM real estate agents were still prominent in certain parts of the Netherlands and there market share has not declined in the past eight years [par. 2.17]. Consequently in this matter as well the Dutch court found no proof of an abuse of dominance.
Conclusion
The Dutch case of Funda is an interesting one when it comes to the market definition of online platforms. Although the economic analysis which resulted in this specific market definition in this case was not published, the outcome is sufficient to indicate that the approaches suggested by the Bundeskartellamt as well as the ACM (Dutch competition authority) may not be suitable in practice.
On the matter of abuse of dominance the Funda case does not introduce much novelty but rather confirms that an abuse of dominance requires proof of a distortion of competition and cannot be established in a formalistic manner. This effects based approach follows the logic of 102 (c) TFEU that specifically refers to a competitive disadvantage thus indicating that differentiated prices and/or trading conditions are not per se abusive even when exercised by dominant undertakings. Furthermore, this approach is equally compatible with the economic theory on (price) discrimination indicating that such practices are not always detrimental to competition or consumer welfare. However for a more detailed interpretation and application of art. 102(c) TFEU we must wait until the CJEU answers the questions referred in the preliminary procedure in the MEO case (Case C-525/16). With some luck the CJEU will address the relationship between reduced profitability and competitive disadvantage in such cases. With respect to online platforms, the Dutch judgement confirms that (dominant) online platforms are free to determine their own governance provided that such choice does not distort competition. With other words, dominant online platforms will be subject to same scrutiny under art. 102 TFEU as dominant non-platform undertakings.
Finally, although not put forward by the claimant, one may wonder why no (alternative) claim of excessive pricing was made in this case. Despite the complications involved in the legal test of excessiveness, if the price difference was indeed so substantial as portrayed by the claimant such a claim could have been an interesting option to consider.