On line (r)evolution? The prospects for selective distribution in an internet future


The application of competition law to online distribution models has been something of a theme for several competition bodies recently. The French, German and UK authorities have all either taken decisions or made statements about various aspects of online retailing. Reflecting the considerable engagement of national competition authorities (NCAs) with this issue, the CJEU has dealt with various Article 267 references: recently, it has criticised some aspects of selective distribution in the PierreFabre case and given guidance on selective distribution issues in the motor vehicle sector in the Auto 24 judgment.

Just before Christmas (20 December), in a pre-Christmas message of the type so beloved of competition regulators, the OFT announced a consultation on proposed commitments to be given in respect of its investigation into online booking of hotel rooms (see here). The Competition Commission’s Provisional Findings Report in its private motor insurance investigation also touches on the issue of the impact on competition of Platform MFNs in price comparison websites (see paras 68 onwards) and the OFT’s Mobility Scooters case also reflects current NCA focus on internet retailing (here).

The most recent attempt on an EU wide level to grapple with the competition analysis of behaviour in online retail, and with the related thorny topic of selective distribution, was in the Vertical Restraints Guidelines, published in 2010.

Against this background, and the growth of internet retailing generally, the German competition authority, the Bundeskartellamt (which has been very active in assessing various online business practices including those of companies such as Sennheiser and, more recently, Gardena, published a discussion paper for its working group on Competition Law which discusses vertical restraints in the online context.

The BKA reviews a variety of provisions encountered in vertical supply/distribution contracts and how they may particularly affect competition in the internet environment. It is an interesting overview of the classic analysis of vertical restraints (including the problems of double-marginalisation and free-riding). It also considers ways in which the growing use by consumers of the internet may require a rethink of previous approaches to aspects of vertical contracts such as MFNs and selective distribution. The paper as a whole is worth reading for those with an interest in the competition approach to the distribution of consumer goods. Notably, however, the BKA appears to be particularly sceptical of the benefits of selective distribution, commenting that case law suggests that, unless objectively justified, “selective distribution systems are to be considered as restraints of competition by object, which necessarily affect competition in the Common Market” – although it does conclude the this needs to be assessed on a case by case basis.

The BKA is not persuaded of the consumer benefits of provisions in selective distribution agreements preventing the use of online market places. It notes (with some apparent disappointment) that such prohibitions are not regarded as hardcore by the EU Commission, but takes comfort from the Pierre Fabre case, remarking that, as the CJEU was sceptical about the arguments that competition restrictions are justified to protect brand image, this “challenges the explanation in the guidelines on this matter.”

Of real concern for brand owners as a possible straw in the wind, the BKA specifically comments that: “Irrespective of the scope of the hardcore restriction, the benefits of the Block Exemption Regulation can be withdrawn in cases where an ‘excessive clause’ has effects in a specific case that are incompatible with Article 101(3) TFEU.” This is of interest because NCAs (as well as the EU Commission) have the power to withdraw the protection of the Vertical Restraints Block Exemption in certain circumstances. One of the specific examples of grounds on which a block exemption may be withdrawn by a Member State (mentioned in the recitals to the block exemption regulation itself) is where there are parallel networks of selective distribution agreements which adversely affect competition.

Of course, it is only possible for an NCA to exercise this power where it considers the conditions of competition are peculiarly national so that there is a “distinct geographic market”. However, the facts that:

  • the BKA has explicitly mentioned withdrawal as a possibility, and
  • that it is critical of the effect of any restrictions on internet trade on consumer welfare, and that
  • it (and other NCAs) have taken the lead on challenging the approach taken by branded goods manufacturers to on line sales in a number of cases

are relevant and interesting – and should be of some concern. Branded goods suppliers took enormous trouble during the consultation period before the adoption of the Vertical Block Exemption and Guidelines to explain the long run benefits of selective distribution and of certain restraints on internet sales. Some of those arguments appear now to be likely to be powerfully challenged. Those who believe that selective distribution of branded goods has a positive role to play in the on-line era and that it can deliver real benefits for consumers should regard the BKA paper as an invitation to engage powerfully in the debate, and to do so at an early stage. The BKA’s summary (on page 26 of its Background paper) speaks for itself:

The manufacturers of these products are facing increased price competition and are justifying their controversial practices by claiming that they are trying to maintain well-established specialist trade structures, specific services or a certain brand image. The interpretation of the VRBER’s hardcore restrictions are of particular significance in this respect. Another issue of growing importance is the question as to what extent selective distribution is the right form of distribution for the respective products and if the selection criteria are the most suitable.