If AG Kokott in Post Danmark II was a 102 hawk – ordoliberal-redux, fossilizing form over function, economics on the “too difficult pile” for authorities and courts – then AG Wahl firmly sets out his stall as the 102 economics dove (§117, fn 87).  Less of an opinion, more of a manifesto.  It calls for a common sense approach to 102, and its re-centering within mainstream economic thought:

  • No per se Rebates can be pro or anticompetitive.  After all, they result in a reduction in price, generally to the consumer’s benefit.  So there is no sense in having per se rules to condemn rebate schemes.
  • Examine effects, not form. The Intel General Court (GC) was thus wrong to find a category of per se illegal “quasi-exclusivity” rebate schemes.  That is to penalize form, not effect.  Rather, rebates linked to exclusivity or high market share obligations should be judged in the round, just like any target-linked rebate.
  • Abuse requires evidence of likelihood of harm, and not just that it is a mere possibility or appears more likely than not.
  • A customer wanting a rebate is not evidence of illegal foreclosure. All customers want a price reduction.  That’s not the question.  The test must be whether competition is illegally foreclosed. That requires a far fuller analysis.
  • Market coverage and duration are factors, but not necessarily decisive ones. 14% market coverage does not necessarily indicate foreclosure.  Nor do long term schemes.  The question is whether customer can easily switch, and competitors can viably match, the rebate cheme.
  • A price cost test – though also not decisive – is an important factor, especially if other criteria produce ambiguous results. The as-efficient-competitor (AEC) price cost test (asking whether a rival can viably match the rebate’s value over a smaller base of “contestable” sales if it had the dominant company’s cost base) is useful.  If the Commission conducts an AEC test, and other criteria produce an ambiguous result, then the GC must review the AEC test results.  It cannot simply ignore the AEC test, as the Intel GC did.

The Advocate General’s opinion is here.


When 3 become 2: The “Supercategory” rebate

AG Wahl takes issue with the Intel GC’s three buckets (1) per se illegal “quasi exclusivity”, (2) presumptively lawful “volume linked” rebates and (3) “all the circumstances” third category rebates (§80-82). This is a misreading of Hoffmann-La Roche he opines (§83-84).

There are two categories only.  One, where the cost benefits are volume linked, are presumptively lawful.  The other, loyalty rebates, must be proven to be exclusionary based on a consideration of all the circumstances.  There is no “per se” quasi exclusivity bucket, a “supercategory” rebate (§84), as the Intel GC found (and indeed the EUCJ in Post Danmark II), just one comprising both quasi-exclusivity and “all the circumstances” Intel GC third category rebates.

Rebates must be assessed on all the circumstances as to whether or not it is exclusionary.  There is no legal short cut to a per se presumption.  This is appropriate because (i) effects must be part of the assessment or objective justification would be impossible (§86-88) (ii) rebates can be pro-competitive so per se condemnation would be inappropriate (“Experience and economic analysis do not unequivocally suggest that loyalty rebates are, as a rule, harmful or anticompetitive”) (§89-93); (iii) the harmfulness of rebates is context dependent so an in-the-round assessment is essential (“contemporary economic literature commonly emphasises that the effects of exclusivity are context-dependent”) (§94-100); and (iv) it would otherwise result in a different standard being applied to price based abuses as between predation, margin squeeze and rebates (see, eg, Post Danmark I) (§101-105)


Post Danmark II a diversion from Post Danmark I’s economics-driven reasoning

The Grand Chamber decision in Post Danmark I (applying an as-efficient-competitor test to determine whether selective price cutting was exclusionary) is held up as the poster child for a coherent economic approach.  By contrast, AG Wahl encourages the EU Court of Justice (EUCJ) to depart from Post Danmark II’s per se condemnation of exclusivity rebates.  This is necessary, he opines, “to ensure a coherent jurisprudential approach to the assessment of conduct falling under the purview of Article 102 TFEU.” (§105)


An Effects Based Assessment

A frontal assault on mealy-mouthed standards of harm

AG Wahl attacks head on the least satisfactory strand of 102 jurisprudence.  The failure of the EU Courts to articulate a justiciable yardstick to measure likelihood of harm. The rebates cases have never demanded proof of actual effects.  Just that conduct “tended to” or “had the capability“ of excluding.  Once a rebate of a particular form was identified, it was almost always branded “capable” of exclusionary effects, with no actual proof being required.

The AG calls out these formulations as an unacceptable judicial shorthand that penalised form over substance:

“[T]he special responsibility of dominant undertakings … cannot be taken to mean that the threshold for the application of the prohibition of abuse laid down in Article 102 TFEU can be lowered to such an extent as to become virtually non-existent. That would be the case if the degree of likelihood required for ascertaining that the impugned conduct amounts to an abuse of a dominant position was nothing more than the mere theoretical possibility of an exclusionary effect, as seems to be suggested by the Commission. If such a low level of likelihood were accepted, one would have to accept that EU competition law sanctions form, not anticompetitive effects.” (§118)

To show abuse requires evidence of likelihood of harm, and not just that it is a mere possibility or appears more likely than not (§117)

A rigorous analysis of likely exclusionary effect

The AG demands a more rigorous evidence lead proof of exclusion.

First, he dispenses with false flags.  Evidence that customers found rebates attractive is not decisive. (§124-126)  Offering customers lower prices is the essence of competition. It cannot of itself demonstrate exclusion of rivals.  So too, evidence that a rival enjoyed commercial success does not preclude abusive conduct being present (§158-160).  It is also not appropriate to conjoin separate alleged abuses as a single complex infringement, unless the abusive nature of each such conduct is proven. Two unproven abuses cannot make a proven one. (§187-193)

Second, he considers relevant criteria:

  1. Market coverage is an important criterion. If market coverage is low, other evidence may be required to show exclusionary impact.  In Intel 14% market coverage did not show exclusionary effect in itself.  It would need to be determined whether this represented, for example, flagship customers:

“Where loyalty rebates target customers that are of particular importance for competitors to enter or expand their share of the market, even modest market coverage can certainly result in anticompetitive foreclosure. Seen in that light, a market coverage of 14% may or may not have an anticompetitive foreclosure effect.  What is certain, however, is that such market coverage cannot rule out that the rebates in question do not have an anticompetitive foreclosure effect. This is so even assuming that the rebates and payments in question target key customers.  Quite simply, 14% is inconclusive.” (§142-143) (emphasis in original)

2. Duration should also be considered. But it may not be decisive as evidence of foreclosure. There can be many reasons for a long term agreement.  The key is whether the customer can switch without penalty to other suppliers.  If the rival cannot sell its products and match the rebate except by making a loss, then that is an indication of a tie (§153). The length of the relationship does not – of itself – answer that question (§154-157).

3. The AEC test “cannot be ignored” (§164-172). AG Wahl resuscitates the AEC test laid down in the Article 102 Enforcement Priorities paper after it’s marginalisation in Tomra and Post Danmark II.  He  concludes this is an important consideration where other factors in the Commission’s assessment are not determinative of whether there was anticompetitive foreclosure.  Where “the other circumstances assessed by the General Court do not unequivocally support a finding of an effect on competition … the AEC test cannot simply be ignored as an irrelevant circumstance.” (§169)

The AG then sets out a two step approach to assessing illegal foreclosure:

  • Market coverage, duration and other factors indicative of foreclosure should be considered. If the AEC test has been conducted by the Commission that should be verified;
  • If the scheme fails to meet any one of these criteria (low market coverage, short duration, positive AEC test) then a more thorough economic assessment is required. (§172)

The AG then goes on to conclude that the GC failed to apply this framework for analysis and the case should be remitted to the GC for review.


Other issues: extra-territorial jurisdiction and failure to disclose relevant evidence

 The AG also faults the Commission for taking long arm jurisdiction in relation to Intel’s rebate scheme with Lenovo, finding that there was too remote a nexus with China (where the scheme was implemented). The AG finds considerations of international comity should lead to restraint in extraterritorial overreach (§308-327)

AG Wahl further faults the Commission for not keeping records of potentially exculpatory evidence – in this case a meeting with a key customer, Dell. He concludes this was such an essential defect of procedure that the case should be remitted to the GC for further consideration as to whether the entire decision should be annulled or just the findings in relation to Dell. (§251-277)



The AG Opinion offers a welcome reappraisal of the spirit and intent of Article 102 analysis of rebate schemes.  The straightjacket of the three categories in Intel and Post Danmark II are rightly criticised as overly formalistic and likely to deter pro-competitive price cutting via rebate schemes.  The cursory factual examination of a scheme’s form is no substitute of a solid economic analysis of its likely foreclosure effects, in particular due to its market coverage and the viability of competitors matching the rebate (generally over a smaller volume of sales).  The AEC test is meant to assess the latter and is an important advance in the legal analytical framework.  AG Wahl rightly recognises this.  The question therefore is whether the EUCJ accepts this recommendation to revise and revisit old case law, and the new formulations in Post Danmark II, to engage in a more meaningful economic assessment.  Or whether the form based categories of the Intel GC will prevail.




Leave a Reply

Your email address will not be published. Required fields are marked *