A few days ago the European Commission opened a preliminary investigation against Amazon to assess whether the American company is engaging in anticompetitive conducts. The Commission is concerned by the dual role of Amazon, which is at the same time owner of the biggest online market place and a seller of its private label products therein. Since Amazon obtains a significant amount of data from the transactions occurring on its market place, DG Comp suspects that Amazon might be using this information to better tailor its private label offer and eventually exclude the other retailers active on the market place.
The preliminary investigation launched against Amazon is only the latest episode of DG Comp´s enforcement agenda in the digital environment. The Commission already investigated and fined, most notably, Google (Google Shopping and Android cases), Facebook (submission of false information during the investigation of its acquisition of Whatsapp), Apple and Amazon (state aid). Why so much focus on this industry? In a recent interview Commissioner Vestager noted that the digital economy is still a fluid unregulated environment where players with huge economies of scale may endanger competition on the market. In this context, antitrust intervention may be used as a tool that allows to identify in these formative years pro and anti-competitive conducts, so that eventually citizens and customers will be able to enjoy the full benefits of a completely new economy and new ways of doing business.
In light of the above, the Amazon probe has not come totally unexpected. After all, Amazon started its private-label business (“AmazonBasics”) in 2009 with batteries – shortly outselling its major competitors in the online market – and today it commercialises more than 70 private label brands at very competitive or even below-cost (according to this award-winning paper) prices. Amazon´s business strategy has certainly been successful, but it triggered the curiosity of competition watchdogs, especially in Europe.
Is Amazon dominant?
The first questions in the Amazon case arise on the relevant market: is Amazon really dominant under the meaning of article 102 TFEU? In which market should Amazon´s power be measured? Will DG Comp (i) identify separate relevant markets for the retail distribution of each specific product (as done in the E-Book case), thereby limiting its intervention only to certain conducts of Amazon related to specific products, or it will (ii) envisage a broader market for e-retail platforms, in order to target Amazon´s entire business model? These are strategic issues that will need to be addressed by the Commission.
Are data a relevant factor under EU competition law?
Another issue to be considered here is the relevance of data under EU competition law. Should ownership of massive data-sets be considered as an indicator of market power? Or, alternatively, is data relevant under EU competition law only when it is misused? This is a topic relevant not only to the digital sector, but also to other industries, most notably automotive.
In her interview Commissioner Vestager noted that when big players enjoy network effects and very low marginal costs, data may be a resource, an input, the starting point for all innovation or also a barrier to entry. The circumstance that data is perceived as an entry barrier implies that it may be considered as a source of market power (always keeping in mind that there is no harm in owning market power provided that it is not abused). After all, the circumstance that some companies have access to a bigger amount of data than smaller companies enable them to better tailor their offer, increase their revenue, invest more and increase the gap with the competitors (this was well noted in the report on competition law and data published two years ago by the German and the French Competition Authority).
Someone in the European Commission recently suggested that access to certain data might be even mandated through competition enforcement, suggesting that data may amount to an essential facility to do business in certain markets. Although the burden of proof to demonstrate that data amounts to an essential facility appears at least burdensome, the fact that someone at DG Comp has come out with such a concept and the thoroughness of the recent clearance of Apple´s acquisition of Shazam suggests that we might expect some new theories of harm based on the relevance of data.
Amazon´s conduct under investigation
Generally speaking, there is no much public information at the moment, aside from the fact that the Commission sent questionnaires to online retailers to narrow the scope of its investigation. My take is that the Commission is verifying whether AmazonBasics and the small businesses using Amazon´s market place were granted access to Amazon´s dataset at discriminatory conditions.
In general, if the Commission establishes that Amazon´s data amounts to an essential facility for small businesses active in the retail market, Amazon will be obliged to grant access to its data and a refusal to deal will amount to an abuse per se. According to the case law of the CJEU on this issue, access to a facility can be mandated if that facility/input is indispensable for carrying on a certain business. In Bronner, the CJEU ruled that a product or service is indispensable only if there are no alternative products or services and there are technical, legal or economic obstacles that make it impossible or unreasonably difficult for any undertaking seeking to operate on the downstream market to develop, possibly in cooperation with other companies, products or services. The essentiality test is usually met by facilities such as telecommunication networks, rails and similar infrastructure, thus it would be a prime for datasets.
If the Commission does not manage to meet the burdensome test described above, it will have to evaluate the effects of Amazon´s conduct on competition. In this regard, the CJEU recently ruled in MEO that only discrimination that “tends to distort competition on the downstream market” is prohibited, thereby shifting on the Commission the burden to prove that a certain differential treatment is so substantial as to be capable of distorting competition in the market.
Is there a need to adapt competition law to the digital market?
Over the past years we have assisted to an increasingly interesting debate on whether competition law should reinvent itself to deal with challenges coming from the digital market and, especially, from multi-sided platforms (here and here). Indeed, even if most of the platforms offer great services to the consumers, often for free, on the other hand this “for free” or “low price” strategy serves the purpose to cross-subsidise the other side of the platform until gaining market power. If this strategy succeeds, it will be almost impossible for new players to challenge the incumbent as a consequence of the significant network effects enjoyed by the platform. Let´s try to imagine for one second how a new comer could challenge today Google, Facebook or Amazon. This is a process that leads to efficiencies but also to market concentration and potential dominance. Competition law allows to tackle abusive conducts by digital platforms only when they become dominant and the barriers to entry have already become almost insurmountable.
While most jurisdictions still appear to stick to the old good competition law rules, some countries are trying to make one step forward. For example, the German government commissioned a study on the need to reform the norms on abusive practices. Among the suggestions proposed by the advisors, one is that dominance should arise not only from a strong position in supply or demand, but also from power in intermediation. This is an interesting suggestion that might allow competition authorities to step in at an earlier stage if multi-sided platforms engage in abusive conducts.