Chillin'Competition

Last week’s post discussed (here) the ‘as-efficient competitor principle’, which is key to make sense of the case law of the past decade. It also briefly pointed out that it tends to be conflated with the ‘as-efficient competitor test’, which has a much narrower scope of application.

The ‘as-efficient competitor test’ is relevant only in a very specific scenario, which is that of conditional rebates. In that context, it might be useful as a more or less accurate filter allowing authorities to spot the schemes that are more likely to cause anticompetitive effects.

The logic of the test is not difficult to grasp. It aims to identify those rebate schemes that work, in practice, as an exclusivity obligation (in the sense that rivals would only be in a position to supply customers benefitting from the rebates at a loss and can therefore be deemed to put the said customers out of their reach).

One of the issues that was abundantly commented at the GCLC conference was that of whether the test is necessary to establish an abuse. There should be little doubt that, as a matter of positive law, the Court has repeteadly and consistely rejected the idea that it is a pre-requisite for a finding of abuse.

What is more, Unilever clarified that a competition authority is not required to conduct the test where a dominant firm seeks to prove that a scheme is incapable of restricting competition on the basis of the test. On the other hand, the authority is under a duty to assess its ‘probative value‘.

In the end, it looked like there was wide agreement at the conference room about the above (given the clarity of the case law, consensus on this point is not so surprising). If anything, there was some scope for discussion around whether, generally speaking (or as a rule), an authority must rely on the test (on this point, there appear to be conflicting readings of the relevant judgments).

There is another point that was not mentioned during the conference and that could have usefully clarified the place and value of the ‘as-efficient competitor test’ in the case law. It is important to emphasise not only that the test is not necessary to establish an abuse, but also that it is insufficient to do so.

In other words, it is not because the test shows that an equally efficient rival would be forced to supply its products at a loss that the rebate scheme is necessarily abusive.

It is necessary to complete the assessment by evaluating other criteria. Some of the criteria such as the extent of the dominant position and, in particular, the coverage of the practice, shed light on whether the scheme is capable of causing anticompetitive effects.

Other criteria would shed light on the object of the conduct (in this sense, evidence of an anticompetitive strategy, mentioned back in AKZO and reiterated in Intel, would be a key consideration).

Absent these other factors, which would confirm the anticompetitive object and/or the effect of the practice, an abuse would not be established to the requisite legal standard.

This point, while seemingly obvious, tends to be forgotten, even though Intel makes it clear that the test is just a filter (the Guidance treated it in the same way: the Commission never suggested in that document that the administration of the test makes is unnecessary to consider other factors).

The question about the ‘as-efficient competitor test’ is why it acquired its mythical status, when it was never meant to be anything other than an imperfect filter with a narrow scope of application.