What’s the scoop? A summary of the most important findings and implications of the preliminary ruling in the Unilever case

Written by Ruben Elkerbout and Philine Wassenaar

Summary

In its latest preliminary ruling of the Unilever case, the EU Court of Justice (the Court) has clarified three essential elements in its assessment of Article 102 TFEU. It has firstly clarified the relevance (or lack thereof) of the single economic entity doctrine in the case of distribution networks. Here, the Court emphasises the importance of unilaterally imposed policies and contracts to attribute the actions of distributors to a dominant undertaking. Secondly, the Court expanded the applicability of the effects based test as articulated in the seminal Intel case to exclusivity clauses. We think that this is another potential nail in the coffin of the enforcement of more traditional Article 102 TFEU cases, as this provides dominant undertakings with yet another possibility to rebut the anti-competitive nature of exclusivity clauses with a variety of economic evidence. Setting aside said economic evidence must be motivated by the competition authority. Finally, this judgement is the fourth within the duration of a little more than a year to address the relevance of the so-called As Efficient Competitor test (AEC-test).

Background:

The Court was asked two questions in Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati by the Italian Council of State. The case involves a € 60 million fine imposed by the Italian Competition Authority, the Autorita Garante della Concorrezo e dei Marcati (AGCM) upon Unilever for abusing their dominant position on the market of individual packaged ice cream distribution to specific sales outlets, such as beach resorts and bars. According to the AGCM, Unilever included exclusivity clauses in contracts with outlet operators requiring them to obtain all individual packaged ice cream supplies from Unilever and offered rebates/commissions for specific Unilever product sales.

The AGCM attributed the alleged abusive behaviour solely to Unilever, even though it was executed by the distributors, considering Unilever and its distributors as a “single economic unit” due to their contractual links. Moreover, the AGCM considered that it was under no obligation to analyse the economic studies produced by Unilever in order to demonstrate that the exclusivity clauses did not have exclusionary effects vis-à-vis it’s at least as efficient competitors. The AGCM argued that the nature of such an exclusivity clause alone was sufficient to establish the infringement of Article 102 TFEU. After appealing this decision and subsequently losing that appeal, Unilever took the case to the Italian Council of State, which referred two questions to the Court: 1) whether the “single economic unit” concept is applicable in this case, and 2) if the competition authority has a duty to consider dominant companies’ economic analyses on the ability of conduct to exclude efficient competitors from the market, and specifically if an “as efficient competitor” test must be carried out.

The Court’s main conclusions

1. Conduct of Unilever’s distributors attributed to Unilever

Firstly, the Italian Council of State asked whether the actions of Unilever’s distributors as part of its distribution network could be attributed to Unilever. The Court explains that there are scenarios like the one at hand where an undertaking in a dominant position may be held responsible for the conduct of its distributors with which it has contractual relations[1]. The Court emphasises the importance of Unilever unilaterally drawing up standard contracts for all its distributors in determining whether the conduct of the distributor should be attributed to Unilever[2]. The fact that Unilever has ‘legal and economic’ links to its distributors and gives said distributors instructions to implement its policy was enough for the Court to determine that Unilever should bare the risks of (illegal) conduct carried out by the distributors[3]. The attribution of the distributor’s conduct to Unilever is not surprising in this case. The Court made a logical decision to clarify that this was an issue surrounding the attributability of Unilever’s conduct, not one concerning the single economic entity doctrine.

2. Intel’s effects based test is also applicable to exclusivity clauses

The second question regards examining economic evidence to rebut possible anti-competitive effects of exclusivity clauses[4]. This is not the first time that Unilever is punished for its exclusivity clauses in the ice cream market. In the Van den Bergh Foods case, Unilever had provided retailers with free freezer cabinets and prohibited them from storing non-Unilever ice creams in those cabinets. Here, the General Court found that the aforementioned exclusivity clause, having considered the economic context, has a distortive (dissuasive) effect on the market[5].

It follows from previous case law such as Hoffman La-Roche that exclusivity clauses by their very nature constitute an exploitation of a dominant position[6]. However, in Intel[7], the Court seemed to take a different approach by clarifying that where the dominant undertaking argues, supported by evidence, that the conduct is not capable of producing the alleged foreclosure effect, the competition authority must assess this evidence. In that case, loyalty rebates were not considered naked restrictions.

In the Unilever judgement, the Court expands this effects-based approach to exclusivity clauses[8]. As a result, in case the dominant undertaking in question submits supporting evidence during the administrative procedure maintaining that the exclusivity clauses are not capable of excluding equally efficient competitors, the respective competition authority must ensure that the exclusivity clause was genuinely able to produce these effects[9]. The use of the word ‘ensure’ does raise an important question however, namely what does this entail in practice? How does a competition authority ‘ensure’ that the clause was capable of producing these effects? How high is this threshold? In practice, this will be a question for (national) competition authorities to answer.

Moreover, evidence which justifies said conduct, which include objective justifications and potential efficiencies must also be considered in the assessment of whether the exclusivity clause restricts competition[10]. Additionally, despite its anti-competitive nature as determined in Hoffman La Roche, this qualification apparently does not provide a free pass for competition authorities to skip a capability test in a situation where the dominant undertaking rebuts the anti-competitive effects with evidence. More importantly, the competition authority is under a further obligation to motivate why the study does not contribute to proving that the practice was incapable of generating anti-competitive effects[11]. According to the Court, the obligation to examine this evidence derives from the right to be heard[12]. This clarification to motivate its reasons may seem to be a win for dominant undertakings in theory. That being said, in practice, this obligation to motivate is unclear and ambiguous, since the Court does not state concretely what this obligation entails. The Court provides no clarity on when the competition authority has sufficiently motivated its claim and at what point the right to be heard is actually respected. This is a question for (national) competition authorities to answer.

As a result, despite the position of dominant undertakings being strengthened based on the conclusions of the Court, the way in which competition authorities must prove or motivate its findings is left unaddressed and therefore exposed to the inconsistent application and interpretation of (national) competition authorities.

3. The role of the ‘As Efficient Competitor’ test is clarified

The Court has also clarified the role of the AEC test in the context of exclusivity clauses[13]. The AEC test refers not to one particular test but ‘various’ tests which have the aim of assessing ‘the ability of a practice to produce anti-competitive exclusionary effects’[14]. Here, it concerns the ability of a hypothetical competitor of the dominant undertaking, which is as efficient as the dominant undertaking in terms of cost structure, to offer customers a rate which is ‘sufficiently advantageous to encourage them to switch supplier (…) without causing the competitor to incur losses’[15]. The Court clarified that the AEC test is not always appropriate in the case of non-pricing practices, especially in the case of a refusal to deal or in markets with significant entry barriers[16]. The Court also explains that there are many methods to assess the anti-competitive exclusionary effects of certain conduct[17]. Therefore, the competition authorities are not under a ‘legal obligation’ to apply the AEC test[18], as previously stated in Post Danmark[19], but may consider the test to be useful in quantifying the effects of the exclusivity clause in question[20]. Due to this, the Court concludes that where an AEC analysis is provided by the dominant undertaking, the competition authority may not disregard said analysis without assessing its probative value[21].

This ruling is the fourth judgement (!) within the span of a year which criticises and clarifies (the execution of) the AEC test. The Intel[22] judgement concluded that the European Commission’s (the Commission) analysis was incomplete, contained errors and had not established that the rebates at issue were likely to have anti-competitive effects. The Commission’s decision in Qualcomm[23] was also annulled by the General Court due to procedural irregularities and errors in the analysis of the AEC test. Finally, Google Android[24] also shed light on numerous errors of reasoning in applying the AEC test. All the aforementioned judgements demonstrate that the European courts can maintain a rather stringent approach when it comes to conducting an AEC assessment. This ruling by the Court in Unilever is a valuable addition to these three judgements.

Key implications of the ruling

The ruling has a number of implications. First of all, it clarifies that dominant companies are responsible for conduct that is strictly speaking executed by their distributors on the downstream markets when the dominant undertaking is basically responsible for this conduct. Dominant companies cannot therefore escape liability by pointing toward their independent buyers.

Secondly, the fact that exclusivity clauses are also more or less subject to an effects-based test, thereby expanding the applicability of the Intel ruling to other clear-cut exclusionary abuses, will have significant consequences for the assessment of exclusionary practices in the future. The ruling welcomes the ability to counter the anti-competitive nature of such practices, which will without a doubt compel dominant undertakings to rebut the anti-competitive effects of exclusivity clauses with a myriad of economic evidence. This can be seen as a victory for dominant companies.

In addition, the ruling sends another clear message to competition authorities that they do not have unchecked freedom in the assessment of the relevance of certain evidence. As a natural consequence, the competition authority’s burden to examine evidence will also increase as it must also motivate why evidence of the dominant undertaking is not relevant in proving pro-competitive effects, as well as assess the probative value before disregarding the AEC analysis submitted by the dominant undertaking. This will increase the quality of the assessments carried out by the authorities and the subsequent decisions as a result, but will also lead to significant additional work and will cost much more time and money. Moreover, as mentioned above, the extent to which a competition authority must examine evidence and motivate its (lack of relevance) is unclear. All these considerations may cause competition authorities to drop Article 102 TFEU cases and complainants to cease their potential pursuits for justice in courts, ultimately leading to less enforcement of an Article which (according to some) was already under-enforced.

Therefore, the ruling also reduces a different significant burden, i.e. the obligation to apply the AEC test in the first place. By confirming once again that the AEC test is not an obligation, the Court shows that there is more than one way to skin a cat. That being said, where the competition authority choses to apply the AEC test, it must consider all relevant circumstances and economic analyses which are produced and conduct the test in a very strict and rigorous manner.

[1]C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 27.

[2] C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 30-31.

[3] C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 31.

[4] C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 34.

[5] Case T-65/98, Van den Bergh Foods Ltd v Commission of the European Communities, 23 October 2003, para 98.

[6] Case 85/76, Hoffmann-La Roche & Co. AG v Commission of the European Communities, 13 February 1979, para 89.

[7] C-413/14 P Intel Corp v Commission, 6 September 2017, para138.

[8] C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 49-51, see to that extent C-413/14 P Intel Corp v Commission, 6 September 2017, para139-140.

[9] C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 52.

[10] C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 49.

[11] C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 55.

[12]C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 54

[13] C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 59.

[14] C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 56.

[15]C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 56.

[16] C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 57.

[17] C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 57.

[18] C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 58

[19] C-23/14, Post Danmark A/S v Konkurrenceradet, 6 October 2015.

[20] C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023, Para 59.

[21] C-680/20 Unilever Italia v. Autorità Garante della Concorrenza e dei Mercati, 19 January 2023,Para 60.

[22] See in that regard: T-286/09, Intel Corp v Commission, 26 January 2022.

[23] T-235/18, Qualcomm v Commission, 15 June 2022.

[24] T-604/18, Google and Alphabet v Commission, 14 September 2022