16 October 2018

Lessons From PostNL: SSNIP Test Prevails

A Dutch administrative court of appeal recently annulled an ACM market analysis decision imposing obligations on PostNL. The ACM had established that PostNL, which originated from the formerly state-owned Dutch mail company, has significant power in the market for 24-hour business mail and consequently imposed several obligations on PostNL, including entry and tariff requirements. However, in defining the relevant product market, the ACM excluded digital mail; had this been included, PostNL would not be considered dominant. The appeal centred on the question of whether the ACM had rightfully concluded that digital mail was not part of the same relevant product market as 24-hour business mail. The court ultimately found that the ACM had not sufficiently proven that digital mail was not part of the relevant market, and annulled the ACM’s decision. This ruling highlights the crucial nature of the methods used to define a market when analysing dominance.

A noteworthy aspect of this case was the court’s examination of what role a quantitative analysis should play in the ACM’s determination of dominance. The ACM’s analysis was confined to a qualitative analysis, looking at product characteristics, rather than using a “small but significant and non-transitory increase in price” test – known as an SSNIP test – to determine the relevant market. The court held that the ACM does not have a general obligation to carry out a quantitative analysis each time it defines a relevant market; however, where a definition of the relevant market is debatable, an SSNIP test may be required. When challenging the ACM’s market definition, PostNL submitted quantitative data supporting its view, including an SSNIP test carried out at its request. According to the court, it was for the ACM to refute the conclusions drawn from that quantitative analysis.

The ACM argued that an autonomous trend favouring digital communications over physical mail is taking place independently of the pricing of both these services, meaning an SSNIP test would not have produced meaningful data. In that respect, the court held that the fact that large groups of users are switching from old to new technology (regardless of the price level) does not mean that research cannot be conducted into the effects of prices on any marginal user, for whom a price change can indeed make a difference. The court ruled that the ACM had not been able to prove that no meaningful SSNIP test could be carried out, and therefore annulled the ACM’s market analysis decision and repealed the obligations placed on PostNL.

The annulment of the ACM’s decision means that empirical evidence plays a significant role when assessing whether conduct may be qualified as anti-competitive, at least in the context of investigations started by the ACM. It remains to be seen whether such detailed analyses become the norm in compliance efforts.