CJEU rejected AG Wathelet's proposal for vicarious liability for agent's behaviour in competition law: a more stringent test, but how stringent? (C-542/14)

In its Judgment of 21 July 2016 in VM Remonts and Others, C-542/14, EU:C:2016:578, the Court of Justice of the European Union (CJEU) issued an important clarification of the rules applicable to the attribution of (vicarious) liability for infringements of EU competition law, thus expanding its case law on the subjective elements (ie mens rea-like requirements) of the prohibition of anticompetitive behaviour in Art 101(1) TFEU.

In doing so, the CJEU rejected the proposal for stringent vicarious liability formulated by AG Wathelet (see my criticism here) and formulated a more stringent test for the attribution of anticompetitive behaviour of an independent agent. The test formulated by the CJEU raises some interpretative issues, though, and it deserves some comment.

It is worth reminding that the case addressed issues concerning the imputability of anticompetitive practices in which a third party services provider is engaged to the 'client' undertaking that hired those services (ie how to make the 'client' undertaking liable for the anticompetitive behaviour of one of its services providers). 

The case was quite convoluted because it concerned the imputability of a bid rigging offence to a supplying company that engaged a consultant to help it formulate a bid in a tender for a public contract. After the fact, it became apparent that the consultant engaged in collusion with other tenderers in the same bid. The question was, thus, to what extent the bidder should be liable for the collusion that resulted from the allegedly independent activity of the consultant (third party services supplier) and, in any case, what level of proof of anticompetitive intent would be necessary to impose liability on the 'client' undertaking.

In addressing this issue, the CJEU rejected a parallelism between the rules applicable to an undertaking's employees to its agents, and determined that 'where a service provider offers, in return for payment, services on a given market on an independent basis, that provider must be regarded, for the purpose of applying rules aimed at penalising anti-competitive conduct, as a separate undertaking from those to which it provides services and the acts of such a provider cannot automatically be attributed to one of those undertakings' (C-542/14, para 25, emphasis added).

However, the CJEU stressed that this different treatment is based on the independence of market activity of the service provider and, consequently, it would not be justified where the client undertaking exerted significant control over the apparently independent service provider. To that effect, the CJEU determined that

Article 101(1) TFEU must be interpreted as meaning that an undertaking may, in principle, be held liable for a concerted practice on account of the acts of an independent service provider supplying it with services only if one of the following conditions is met:

–  the service provider was in fact acting under the direction or control of the undertaking concerned, or
– that undertaking was aware of the anti-competitive objectives pursued by its competitors and the service provider and intended to contribute to them by its own conduct, or
–  that undertaking could reasonably have foreseen the anti-competitive acts of its competitors and the service provider and was prepared to accept the risk which they entailed (C-542/14, para 33, emphasis added).

Of particular relevance in the field of public procurement, the CJEU also provided some clarification regarding the unauthorised disclosure of commercially sensitive information by the agent, by stressing that

Whilst it is true that [an undertaking is liable for a competition infringement] when that undertaking intended, through the intermediary of its service provider, to disclose commercially sensitive information to its competitors, or when it expressly or tacitly consented to the provider sharing that commercially sensitive information with them ... the condition is not met when that service provider has, without informing the undertaking using its services, used the undertaking’s commercially sensitive information to complete those competitors’ tenders (C-542/14, para 32, emphasis added).

In my view, the VM Remonts Judgment should be welcome for what it does not do. That is, for its rejection of AG Wathelet's proposal for a reversal of the burden of proof, to the effect that the 'client' undertaking would have been considered liable unless it could adduce sufficiently convincing evidence (i) relating to the fact that the agent (services provider) had acted outside the scope of the functions that had been entrusted to it, (ii) regarding the precautionary measures taken by the ‘client’ undertaking at the time of designation of the agent and during the monitoring of the implementation of the functions in question, and (iii) regarding the ‘client’ undertaking's conduct upon becoming aware of prohibited behaviour--so as to demand a public distancing and positive reporting, under the analogous rules of Dansk Rørindustri and Others v Commission, C-189/02 P, C-202/02 P, C-205/02 P to C-208/02 P and C-213/02 P, EU:C:2005:408.

However, regarding the positive test that it sets for the assessment of whether anti-competitive activity by an agent can be imputed to the client undertaking, the VM Remonts Judgment seems less satisfactory, in particular due to the last condition of the test in its paragraph [33], whereby 'an undertaking may, in principle, be held liable for a concerted practice on account of the acts of an independent service provider supplying it with services ... if  ... that undertaking could reasonably have foreseen the anti-competitive acts of its competitors and the service provider and was prepared to accept the risk which they entailed' (emphasis added).

This seems to be an adaptation of the test developed in Commission v Anic Partecipazioni, C-49/92 P, EU:C:1999:356, paragraph [87], to which the CJEU refers in VM Remonts to stress that 'an undertaking may be held liable for agreements or concerted practices having an anti-competitive object when it intended to contribute by its own conduct to the common objectives pursued by all the participants and was aware of the actual conduct planned or put into effect by other undertakings in pursuit of the same objectives or that it could reasonably have foreseen it and was prepared to accept the risk' (C-542/14, para 29, emphasis added).

The adaptation of this test to cases of anticompetitive behaviour by an agent seems problematic because it stretches its last part concerning the acceptance of a risk of occurrence of anticompetitive behaviour by third parties (in that case, co-conspirators). In Anic, the undertaking concerned had been attending meetings with other undertakings that formed part of a cartel. Therefore, the assessment of whether the undertaking could reasonably foresee specific types of anti-competitive conduct by its co-conspirators (formally, third parties) derives from its own participation in meetings--that is, derives from its own observation of the behaviour of other entities that participate in the anti-competitive practice.

This cannot be the case in a scenario such as that presented by VM Remonts, where the client undertaking does not participate in any meetings and where it has no (proven) knowledge of the activity of the agent. In these cases, it would seem that the first two prongs of the VM Remonts test would suffice: ie the client undertaking is liable for the anticompetitive behaviour of the agent if (a) it controls the agent or (b) is aware of the anti-competitive behaviour between the agent and third parties, and aims to contribute to it. Introducing the third condition, according to which the client undertaking can also be liable if (c) it could have reasonably foreseen anticompetitive behaviour between its agent and third parties and was prepared to accept the risk which they entailed, seems to far fetched. 

Whereas in an Anic-like scenario the reasonable prediction of anticompetitive behaviour by co-conspirators derives from information directly acquired in the meetings in which the undertaking participates--that is, can be presumed under logical rules--in a VM Remonts-like scenario, any claim as to the undertaking's duty to foresee anticompetitive behaviour would be pure speculation.

If the client undertaking has no positive knowledge of the anticompetitive behaviour in which the agent [otherwise, the prong (b) of the test would apply], how is it ever going to be possible to determine that it ought to have foreseen it? If this is on the basis of its relationship with the agent, this dangerously reopens the door to a test like the one developed by AG Wathelet or, worse, creates a sort of culpa in eligendo of its agent that is equally troublesome.

If (factual) speculation is to be avoided and the imposition of vicarious liability is rejected by the CJEU in VM Remonts (para 26, although see para 27, which makes it less clear-cut), the only reasonable interpretation of the prong (c) of the test developed in paragraph [33] of VM Remonts is that it can simply never be applied. In which case, one can be forgiven for wondering if the CJEU did not pay sufficient consideration to the adaptation of the Anic test to a situation involving an independent service provider.

AG Wathelet proposes creation of excessive presumption of liability for third party infringement of Art 101 TFEU (C-542/14)

In his Opinion of 3 December 2015 in case VM Remonts and Others, C-542/14, EU:C:2015:797 (not yet available in English), Advocate General Wathelet advised the Court of Justice of the European Union (CJEU) on issues concerning the subjective elements (ie mens rea-like requirements) of the prohibition of anticompetitive behaviour in Art 101(1) TFEU. 

In particular, the case addresses issues concerning the imputability of anticompetitive practices in which a third party services provider is engaged to the 'client' undertaking that hired those services (ie how to make the 'client' undertaking liable for the anticompetitive behaviour of one of its services providers). In my view, AG Wathelet's proposal is clearly excessive (see critical assessment below) and deserves closer inspection. 

The case is quite convoluted because it concerns the imputability of a bid rigging offence to a supplying company that engaged a consultant to help it formulate a bid in a tender for a public contract. After the fact, it became apparent that the consultancy engaged in collusion with other tenderers in the same bid. The question was, thus, to what extent the bidder should be liable for the collusion that resulted from the allegedly independent activity of the consultancy (third party services supplier) and, in any case, what level of proof of anticompetitive intent would be necessary to impose liability on the 'client' undertaking.

In AG Wathelet's Opinion, it is not necessary to prove a personal behavior of any corporate officer of the 'client' undertaking, or his knowledge or consent to the behavior of the external services provider that also acted on behalf of other participants in a possibly prohibited agreement. AG Wathelet proposes to create a presumption of (vicarious?) liability, so that it is incumbent upon the 'client' undertaking to adduce sufficiently convincing evidence to rebut that presumption and escape liability. 

In particular, AG Wathelet considers that the necessary proof concerns (i) evidence relating to the fact that the third party (services provider) has acted outside the scope of the functions that had been entrusted to it, (ii) evidence regarding the precautionary measures taken by the 'client' undertaking at the time of designation of the third party and during the monitoring of the implementation of the functions in question, and (iii) evidence regarding the 'client' undertaking's conduct upon becoming aware of prohibited behavior.

AG Wathelet's VM Remonts Opinion follows the expansive/strict interpretation of the subjective elements in the prohibition of Art 101(1) TFEU in recent cases such as AC-Treuhand v Commission (C-194/14 P, EU:C:2015:717, re liability of a cartel facilitator, see an interesting comment here); Schenker and Others (C-681/11, EU:C:2013:404, re reliance on third party advice, see comments here); or Kone (C-557/12, EU:C:2014:917, re extension of 'umbrella' liability for damages to third parties to a cartel, see comments here). 

This is a very relevant opinion, with potentially very significant effects commensurate to those of the presumption of liability of the parent company, which has shaken competition law enforcement in the EU for the last 5 years or so.  

Therefore, it is interesting to look at AG Wathelet's reasoning in some more detail:

60. In my view, two extreme positions must be rejected. On the one hand, the automatic imputation of responsibility to the undertaking for the actions of a third party, regardless of the degree of involvement of the undertaking, which would go against fundamental principles governing the imposition of competition law sanctions (in particular the principles of personal responsibility and legal certainty), and, on the other hand, the obligation of the competent competition authority to demonstrate convincingly that the undertaking receiving the services from the third party was aware of the criminal acts committed by the latter or had consented to them, which would create a risk of seriously undermining the effectiveness of competition law.

61. Indeed, "... the prohibition on participating in anti-competitive practices and agreements and the penalties which infringers may incur are well known, it is normal that the activities which those practices and agreements involve take place in a clandestine fashion, for meetings to be held in secret, frequently in a non-member country, and for the associated documentation to be reduced to a minimum. " Therefore, it would be too easy to "hide" behind a third party in order to go unpunished under competition law.

62. Moreover, the importance of keeping free competition allows for companies that entrust third parties with functions such as those at issue in the present case [ie public procurement consultancy services] to be required to take all possible precautions to prevent such third parties from infringing competition law, avoiding, in particular, any negligence or recklessness in the definition or in the monitoring of these functions.

63. In line with this, the solution I propose for cases such as that in the main proceedings is to establish a rebuttable (iuris tantum) presumption of liability of the undertakings for acts contrary to competition law committed by a third party whose services it has engaged and which cannot be considered its subsidiary or ancillary body. Such a presumption can maintain the balance between, on the one hand, the objective of effectively suppressing behavior contrary to the competition rules, in particular to Article 101 TFEU, and to prevent their recurrence bearing in mind that respect for these rules requires an active corporate behavior at all times and, on the other hand, the requirements arising from the fundamental rights regarding the imposition of sanctions. Such a presumption would apply even if the acts performed by the third party were different from the functions entrusted to it, and even when it was not possible to demonstrate that the undertaking that used the services was aware of the acts of the provider or consented to them.

64. This presumption should apply to an undertaking from the moment the authority responsible for the enforcement of competition rules proves the existence of an act contrary to competition law committed by a person working for (or providing services to) the undertaking but which does not, directly or indirectly, form part of its organisational chart.

 65. In order to respect the balance to which I referred in point 63 of this Opinion, the undertaking may rebut the presumption of liability by submitting all elements supporting its claim that it was unaware of the illegal behavior  in which the third party service provider engaged, and by demonstrating that it took all necessary measures to prevent such a breach of competition law precautions, and this in three stages.

66. The first is when his appointment or hiring occurs. It refers in particular to the choice of supplier, the definition of the functions and the monitoring of its implementation, the conditions (or exclusion) of recourse to subcontracting, obligations to ensure respect for the law, in particular, competition, and the sanctions for breach of contract, as well as whether authorization was required for  any act not provided for in the contract.

67. The second stage includes the period of execution of the functions entrusted to the third party, ensuring that the latter strictly sticks to its functions as defined in the contract.

68. The third stage is when the third party commits a breach of competition law, even if committed at the back of the undertaking. The undertaking cannot simply ignore that behaviour, it should distance itself publicly from the forbidden act, prevent its repetition or report it to the administrative authorities. Indeed, as stated by the Court: "... passive modes of participation in the infringement, such as the presence of an undertaking in meetings at which anti-competitive agreements were concluded, without that undertaking clearly opposing them, are indicative of collusion capable of rendering the undertaking liable under Article [101(1) TFEU], since a party which tacitly approves of an unlawful initiative, without publicly distancing itself from its content or reporting it to the administrative authorities, encourages the continuation of the infringement and compromises its discovery" (C-542/14, paras 60-69, references omitted, own translation from Spanish, emphasis added).

In my own opinion, the creation of the presumption proposed by AG Wathelet goes way too far. In simple conceptual terms, it excessively erodes the principle of personal responsibility and falls short of meeting the desirable balance that the AG presents himself. The 'client' undertaking and the third party service provider are, in these cases, completely independent undertakings and the creation of the presumption would go beyond the acceptable limits of expansion of the concept of (functional) single economic entity. 

Plainly, it is excessive to impose this type of burden of proof (probatio diabolica) on undertakings that simply lack the knowledge and manpower required to monitor the execution of the activities contracted out to the third party to the standard created by AG Wathelet. This applies at least in stages one (design of the contract) and two (monitoring of execution), where the 'client' undertaking will in many cases be affected by significant asymmetries of information and gaps in human capital. Otherwise, what would be the economic rationale for contracting out something the undertaking could carry out on its own?

I would thus prefer the CJEU to deviate from the proposal of AG Wathelet in this case and to reject the creation of such rebuttable presumption of liability for the anticompetitive behaviour of third parties to which a 'client' undertaking has outsourced certain types of functions. The competent competition authority should always be obliged to demonstrate, at least at the level of sufficient indicia (balance of probabilities, but for?), that the recourse to the third party aimed to circumvent the prohibition of Art 101(1) TFEU--ie, that there was an anticompetitive agreement (by object) between the 'client' undertaking and the third party services provider because the outsourcing had the object of creating further restrictions of competition (on the issue of prohibitions by object, see here). Thus, if the contracting out arrangement was not genuine or if there are indications that the outsourcing aimed at a restriction of competition, then the burden of proof could be reversed. But to create a presumption of liability in the way that AG Wathelet proposes is excessive.

CJEU's new formulation of legal test to distinguish between anticompetitive agreements by object and by effect (C-345/14)

In its Judgment of 26 November 2015 in Maxima Latvija, C-345/14, EU:C:2015:784, the Court of Justice of the European Union (CJEU) has returned to the debate on the distinction between anticompetitive agreements by object and by effect for the purposes of the application of Art 101(1) TFEU [for background commentary of the case, see here].

This is a field of EU competition law of notable and growing complexity, fundamentally due to the lack of clear demarcating lines between both types of anticompetitive conduct [see S King, Agreements that restrict competition by object under Article 101(1) TFEU: past, present and future (2015) PhD Thesis, London School of Economics]. 

Some of the consequences of the distinction remain obscure or highly theoretical, not least because Art 101(1) TFEU covers both types of anticompetitive practices, which should therefore be subjected to similar sanctions (moderated only by the magnitude of the effects, where they exist). Two main areas of debate have emerged, though, around the (practical) boundaries where the categorisation carries legal weight: on the one hand, the possibility to exempt de minimis "by object" infringements; and, on the other hand, the standard of proof applicable to "by object" anticompetitive conduct.

Regarding the possibility to exempt de minimis "by object" infringements, the debate focuses on the implications of the CJEU Judgment of 13 December 2012 in Expedia (C-226/11, EU:C:2012:795), as well as the guidelines issued by the Commission in trying to clarify the state of the law [see the Guidance on restrictions of competition "by object" for the purpose of defining which agreements may benefit from the De Minimis Notice]. For the purposes of this discussion, it is worth stressing para 36, where the CJEU clarified that "the distinction between ‘infringements by object’ and ‘infringements by effect’ arises from the fact that certain forms of collusion between undertakings can be regarded, by their very nature, as being injurious to the proper functioning of normal competition" (emphasis added).

Regarding the standard of proof applicable to cases concerning "by object" anticompetitive conduct, the discussion revolves around the Judgment of 11 September 2014 in CB v Commission ('Cartes bancaires'; C-67/13 P, EU:C:2014:2204) [for discussion from a legal perspective, see here; and for an economic perspective, see here]. The significance of the Cartes bancaires Judgment is commonly seen as resting in the recasting of the case law in para 51, where the CJEU ruled that "it is established that certain collusive behaviour, such as that leading to horizontal price-fixing by cartels, may be considered so likely to have negative effects, in particular on the price, quantity or quality of the goods and services, that it may be considered redundant, for the purposes of applying Article [101(1) TFEU], to prove that they have actual effects on the market" (emphasis added).

In its Judgment in Maxima Latvija, the CJEU has followed the same general approach to the delineation between anticompetitive practices by object and by effect, and has consolidated the Cartes bancaires test. More specifically, the CJEU has ruled that

18 ... the concept of restriction of competition ‘by object’ ... must be interpreted restrictively and can be applied only to certain types of coordination between undertakings which reveal a sufficient degree of harm to competition that it may be found that there is no need to examine their effects ... That case-law arises from the fact that certain types of coordination between undertakings can be regarded, by their very nature, as being harmful to the proper functioning of normal competition ...

19 ... certain collusive behaviour, such as that leading to horizontal price-fixing by cartels, may be considered by their nature as likely to have negative effects, in particular on the price, quantity or quality of the goods and services, so that it may be considered redundant, for the purposes of applying Article 101(1) TFEU, to prove that they have actual effects on the market ... Experience shows that such behaviour leads to falls in production and price increases, resulting in poor allocation of resources to the detriment, in particular, of consumers.

20 ... the essential legal criterion for ascertaining whether an agreement involves a restriction of competition ‘by object’ is ... the finding that such an agreement reveals in itself a sufficient degree of harm to competition for it to be considered that it is not appropriate to assess its effects (C-345/14, paras 18-20, references omitted and emphasis added).

In my view, the formulation of this test is conceptually appealing. However, it is equally problematic because it relies on a test of 'balance of economic probabilities' that actually requires, in most cases, at least a quick look at the effects the anticompetitive practice has (or could have) generated. In terms of alleviating the burden of proof, this is less than clear cut. These complications are self-evident in the Maxima Latvija Judgment itself, where the CJEU applies this newly (re)formulated test as follows:

21 ... Maxima Latvija [an operator of large shops and hypermarkets] is not in a competitive situation with the shopping centres with which it has concluded the [lease] agreements at issue in the main proceedings. Although ... a fact of that nature in no way precludes an agreement from containing a restriction of competition ‘by object’ ... it must, however, be stated that the agreements at issue in the main proceedings are not among the agreements which it is accepted may be considered, by their very nature, to be harmful to the proper functioning of competition ...

22 Even if the clause at issue in the main proceedings could potentially have the effect of restricting the access of Maxima Latvija’s competitors to some shopping centres in which that company operates a large shop or hypermarket, such a fact, if established, does not imply clearly that the agreements containing that clause prevent, restrict or distort, by the very nature of the latter, competition on the relevant market, namely the local market for the retail food trade.

23 Taking account of the economic context in which agreements ... are to be applied, the analysis of the content of those agreements would not ... show, clearly, a degree of harm with regard to competition sufficient for those agreements to be considered to constitute a restriction of competition ‘by object’ within the meaning of Article 101(1) TFEU (C-345/14, paras 21-23, references omitted and emphasis added).

Looking at the application of the test in the Maxima Latvija case, two difficulties emerge. Firstly, there is no existing list of anticompetitive agreements by object (neither an open, closed or indicative list). This creates a significant complication and legal uncertainty, not least because of the potentially moving target of what constitutes "agreements which it is accepted may be considered, by their very nature, to be harmful to the proper functioning of competition".

Secondly, there is no clear indication as to what "degree of harm" (ie presumed anticompetitive effects) suffices to allow for an agreement to be categorised as "by object" rather than by effect. The need to consider the anticompetitive agreements in the "the economic context in which [those] agreements ... are to be applied" is a puzzling requirement from a legal perspective. If (one of) the main point(s) concerning the distinction between anticompetitive agreements "by object" and "by effect" is to clarify the burden of proof required for each of them (implicitly, in order to reduce the requirements for a finding of an anticompetitive agreement "by object"), then this requirement does not really help and ends up resulting in a slippery slope of relaxation of requirements depending on the general consensus (?) that some agreements are by their nature harmful.

Given the ongoing uncertainty, the test for the distinction between anticompetitive agreements by object and by effect is not likely to reduce litigation at all. Overall, then, this seems to be a never-ending discussion. In terms of legal simplicity, it could be preferable to return to a state of the law where it was clear that Art 101(1) TFEU covers both types of anticompetitive practices and triggers an enforcement procedure that needs to comply with the same requirements, regardless of whether ex post and in a specific case, it may look like some of the probatory efforts could have been spared. 

The classical justification for per se prohibitions and not rule of reason approaches is fundamentally based on enforcement costs--and that is, at least in part, why they are progressively abandoned in the US (see here). It seems clear to me that trying to create that distinction in this area is, unfortunately, not leading to any savings and, in that case, it may be worth avoiding a shaky per se rule...