On 6 September the European Court of Justice (ECJ) upheld Intel's appeal against the General Court's judgment that had confirmed that Intel's exclusivity/loyalty rebates were an abuse of its dominant position on the market for x86 CPUs. The ECJ rejected the General Court's acceptance of the rebates as an abuse of dominance without first requiring an examination of all of the circumstances. In effect, the ECJ treats such rebates in much the same way as "by object" restrictions are treated under Article 101 – they are presumed to have anti-competitive effects, but it is still possible for the dominant firm to rebut this presumption. If the dominant firm raises arguments that its conduct was not capable of having the alleged effects, then the enforcer must assess potential effects.
The importance of an effects-based approach
The ECJ has made clear in its Intel judgment that exclusivity and loyalty rebates do not automatically breach Article 102. They are presumed harmful but, where the dominant firm argues that they are not capable of affecting competition, a further effects-based analysis is required. The ECJ indicates that this will necessitate an analysis of the extent of the dominant firm's position on the market, the share of the market affected by the practice, the conditions and arrangements for granting the rebates as well as their duration and amount. The analysis also requires an assessment of the "possible existence of a strategy aiming to exclude competitors that are at least as efficient as the dominant undertaking". This reinforces the ECJ's statement (in Intel and in earlier case-law) that Article 102 does not seek to ensure that competitors less efficient than the dominant undertaking remain on the market. The ECJ elaborates that not all exclusionary effects are necessarily detrimental to competition - competition on the merits may lead to the departure or marginalisation of competitors that are less efficient and so less attractive to consumers. In other words, the exclusion of less efficient competitors should not automatically be ascribed to the conduct of the dominant firm.
The judgment also confirms that an effects analysis has a place in determining whether conduct, which in principle falls within the Article 102 prohibition, is justified. This assessment involves balancing the favourable and unfavourable effects of the conduct to see whether the detriments to competition are counterbalanced or outweighed by efficiencies which benefit consumers.
These findings bring the case-law on exclusivity and loyalty rebates into line with previous case-law concerning aggressive pricing strategies, such as margin squeeze and predatory pricing. What is not as clear from the judgment is whether the ECJ has endorsed the "as efficient competitor" test as the necessary or decisive test of whether conduct is exclusionary, or whether other benchmarks are possible. Future judgments may clarify this.
This is not the end of the case for Intel, as the ECJ has referred it back to the General Court to re-examine whether the rebates are capable of restricting competition, in light of the facts and economic evidence put before it.
How do these findings play out in the UK?
Shortly before the Intel judgment, the UK's Competition and Markets Authority (CMA) published its "no grounds for action" decision in Impulse Ice Cream, in which it conducted an effects-based analysis of Unilever's promotional deals giving retailers who purchased a minimum quantity of single-wrapped impulse ice-creams further quantities either free of charge or at reduced prices (e.g. buy 2 cases get 1 case free). The CMA records that there is no general presumption that offers of this type are abusive, but that it was required to assess whether Unilever's offers could produce an exclusionary effect. It continues by saying that potentially anti-competitive rebates will be abusive if they tend to restrict competition or are capable of having that effect, but it is not necessary to establish a concrete anti-competitive effect. In this analysis anti-competitive intent is one factor that may be taken into account, but not a prerequisite for establishing abuse. It concluded that Unilever's offers were unlikely to have exclusionary effects as they were made in February and March when ice cream consumption is low, and were unlikely to have a significant effect on purchasing decisions outside those months.
The CMA's decision is based on much of the ECJ case-law cited in the Intel judgment, which it foreshadows in its effects-based approach. We can therefore expect to see more of this type of analysis in both Article 102 and Chapter 2 cases. Overall it, reasonably, raises the bar for demonstrating that a loyalty/exclusivity rebate infringes Article 102 by comparison with the General Court, bringing the legal test more in line with the potential economic effect of the conduct. In practical terms, potentially dominant firms are closer to where they were before the General Court's judgment – with the question of context for their rebates more firmly in the forefront of their design, than the focus being on their form.