Provisional text
OPINION OF ADVOCATE GENERAL
RANTOS
delivered on 5 October 2023 (1)
Case C‑298/22
Banco BPN/BIC Português SA,
Banco Bilbao Vizcaya Argentaria SA, Portuguese branch,
Banco Português de Investimento SA (BPI),
Banco Espírito Santo SA (in liquidation)
Banco Santander Totta SA,
Barclays Bank Plc,
Caixa Económica Montepio Geral – Caixa Económica Bancária, SA,
…..
Unión de Créditos Inmobiliários, SA, Establecimiento Financiero de Crédito, Portuguese branch,
Caixa Geral de Depósitos, SA,
Caixa Central de Crédito Agrícola Mútuo CRL,
Banco Comercial Português SA
v
Autoridade da Concorrência,
intervener:
Ministério Público
(Request for a preliminary ruling from the Tribunal da Concorrência, Regulação e Supervisão (Competition, Regulation and Supervision Court, Portugal))
(Reference for a preliminary ruling – Competition – Agreements, decisions and concerted practices – Article 101 TFEU – Agreements between undertakings – Restriction of competition by object – Exchanges of information between credit institutions – Information concerning commercial conditions and production values)
I. Introduction
1. This request for a preliminary ruling concerns the interpretation of Article 101(1) TFEU and the conditions under which an exchange of information between competing undertakings may be classified as a ‘restriction of competition by object’.
2. The request was made in proceedings between several banking institutions and the Autoridade da Concorrência (the Competition Authority, Portugal; ‘the Competition Authority’), the defendant in the main proceedings, concerning the Competition Authority’s decision to impose a fine on those institutions for an infringement of national provisions of competition law and of Article 101 TFEU, which consists of participation in a concerted practice, in the form of informal coordination between competitors via the exchange of sensitive and strategic information.
3. A specific feature of the present case lies in the fact that the Competition Authority characterised a ‘standalone’ exchange of information as a restriction of competition by object without having established the existence of a cartel, a legal characterisation disputed by the banking institutions, which maintain that that exchange of information did not reveal the degree of harm required to be so characterised and that account should therefore be taken of not only the object of that exchange but also its effects. Since it took the view that the case-law of the Court contains no precedents capable of providing useful guidance for handling the present case, the referring court made a reference to the Court of Justice in that regard.
4. The present case provides the Court with the opportunity to develop its case-law on the analysis of exchanges of information between competitors under Article 101(1) TFEU. The Court will therefore have the opportunity to examine, once again, the concept of restriction of competition by object, which, while it has long been debated, continues to present significant conceptual ambiguities and raise questions of interpretation.
II. Legal framework
5. The referring court refers to Lei n.º 19/2012 que aprova o novo regime da concorrência) (Law No 19/2012 establishing the new competition rules) of 8 May 2012 (‘the Competition Law’) (2) which replaced Lei n.º 18/2003 (establishing the legal rules on competition) of 11 June 2003 (3). Article 9 of the Law on competition, under the heading ‘Agreements, concerted practices and decisions by associations of undertakings’ (as well as the former Article 4 of Law No 18/2003, under the heading ‘Prohibited practices’), in essence, reproduces the content of Article 101 TFEU.
III. The dispute in the main proceedings, the questions referred for a preliminary ruling and the procedure before the Court
6. On 9 September 2019, the Competition Authority adopted a decision by which it imposed a fine on the applicants in the main proceedings for having participated in a standalone exchange of information (4) contrary to Article 101 TFEU and the equivalent national provisions.
7. In reaching that conclusion, the Competition Authority considered that the exchange of information in question constituted a restriction of competition by object, which relieved the authority of the obligation to investigate its possible effects on the market. Moreover, that authority did not criticise the undertakings concerned for having participated in any other form of practice restrictive of competition to which the exchange of information could be linked, such as a pricing or market-sharing agreement.
8. The applicants in the main proceedings brought an action against that decision before the Tribunal da Concorrência, Regulação e Supervisão (Competition, Regulation and Supervision Court, Portugal), the referring court, on the ground that the exchange of information at issue could not be regarded as being, in itself, sufficiently harmful for an examination of its effects not to be necessary. In that regard, the applicants claimed that the Competition Authority failed, in particular, to take into account the economic, legal and regulatory context in which that exchange was implemented, whereas that context ought to have been taken into account before being able to conclude that there was a restriction of competition by object.
9. On 28 April 2022, the referring court delivered an interim judgment of almost 2 000 pages in which it indicated which of the facts contained in the Competition Authority’s decision were to be regarded as established. In its request for a preliminary ruling, that court summarised that judgment, dividing its description into five subheadings concerning, respectively, the nature of the information exchanged, the form of the coordination, the objective pursued, the legal and economic context, and the alleged existence of procompetitive effects.
10. First, as regards the nature of the information exchanged, that information related to the home loans market, the consumer credit market and the corporate lending market. Two types of information were exchanged concerning those markets:
– current and future commercial ‘conditions’, namely charts of credit spreads, the borrowing capacities of customers and risk parameters, which, given the exhaustive nature and the systematic organisation of the information exchanged, were not in the public domain at the time of the exchange;
– ‘production volumes’, that is to say, individualised figures of the entities concerned, showing the amount of loans granted in the preceding month. Those data were communicated in a disaggregated format and were not available in that form or from any other source at the time of the exchange or subsequently.
11. Second, as regards the duration and form of the exchange of information, the referring court states that the exchange took place between May 2002 and March 2013. It took the form of institutionalised bilateral or multilateral contacts, carried out by means of telephone communications or emails, with the full knowledge of management.
12. Third, as regards the objective of that exchange, given that the exchange enabled the banks concerned to obtain detailed, organised, up-to-date and accurate information about their competitors’ offers, that court inferred therefrom that the objective of that exchange was to reduce uncertainty with regard to their strategic conduct with a view to reducing the risk of commercial pressure.
13. Fourth, as regards the legal and economic context of that exchange of information, during 2013 the six largest credit institutions in Portugal, all of which participated in the exchange of information, managed 83% of total national bank assets. Moreover, from mid 2008, contrary to the trend in Euribor, the index reflecting interbank interest rates within the eurozone, which had fallen sharply at that time, the credit spreads applied by the financial institutions to new home loans increased significantly, which mitigated the decrease in interest rates to end customers. (5) The summary of the interim judgment states, also under the heading ‘legal and economic context’, that the exchanges of information at issue were regular and organised within a closed loop. Furthermore, the exchanges concerned strategic information which was not publicly available or was difficult to obtain or to organise. The information exchanged was distinct from that which credit institutions provided in accordance with their obligations to provide information to consumers.
14. Fifth, with regard to whether there were potentially positive or, at the very least, ambivalent effects on competition, that summary states that the banks in question had been unable to demonstrate or establish that (i) the exchanges of information had generated efficiencies; (ii) those efficiencies had been passed on to consumers; or (iii) the restrictions of competition were indispensable.
15. Lastly, even though the referring court itself notes that the exchange in question is likely to contribute to a reduction in commercial pressure and uncertainty regarding the strategic conduct of competitors on the market, which may lead to informal coordination restricting competition, it takes the view that the present reference for a preliminary ruling is justified by the absence of precedents in the case-law of the Court relating to the present situation.
16. In those circumstances, the Tribunal da Concorrência, Regulação e Supervisão (Competition, Regulation and Supervision Court) decided to stay the proceedings and to refer the following questions to the Court of Justice for a preliminary ruling:
‘(1) Does Article 101 TFEU … preclude the classification as a restriction of competition by object of a comprehensive, monthly exchange between competitors of information concerning commercial conditions (in particular, current and future credit spreads and risk variables) along with (monthly, individualised and disaggregated) production figures on home loan offers, corporate lending offers and consumer credit offers, exchanged regularly and in reciprocal fashion, in the retail banking sector, in the context of a concentrated market with barriers to entry, which has artificially increased transparency and reduced uncertainty with regard to the strategic conduct of competitors?
(2) If the answer is in the affirmative, does Article 101 TFEU preclude such classification where it has been impossible to identify or establish any gain in efficiency or any uncertain or positive effect on competition resulting from that exchange of information?’
17. Written observations were submitted to the Court by the applicants, the Competition Authority and the Ministério Público. Written observations were also submitted by the Portuguese, Greek, Italian and Hungarian Governments, the European Commission and the EFTA Surveillance Authority. Oral argument was presented at the hearing on 22 June 2023 by the applicants, the Competition Authority, the Portuguese and Greek Governments, the EFTA Surveillance Authority and the Commission.
IV. Analysis
A. Preliminary observations
18. As a preliminary point, it should be noted that almost all the applicants devoted a substantial part of their written observations to challenging the referring court’s account of the facts in the main proceedings, going so far as to argue that the Court of Justice is required to amend the factual circumstances described by the referring court in order to provide a useful response to that court. (6)
19. In that regard, it should be recalled that it is clear from the settled case-law of the Court that, in the context of the procedure referred to in Article 267 TFEU, which is based on a clear separation of functions between the national courts and the Court of Justice, it is not for the Court, but for the national court, to ascertain the facts which have given rise to the dispute. (7) It follows that, since it is empowered solely to rule on the interpretation or validity of an EU text, the Court cannot verify the accuracy of the facts set out by the national court or rule on the substance of the allegations by certain parties seeking to contest the relevance of the factual circumstances described by the referring court in its request for a preliminary ruling.
20. That said, the interpretation which the Court is called upon to give to a provision of EU law in the factual context described by the referring court does not give rise to any presumption that that situation is indeed the one at issue in the main proceedings. Accordingly, it is always ultimately for the referring court to verify that the facts which it has placed before the Court actually correspond to the situation at issue in the main proceedings.
21. That conclusion cannot be called into question by the obligation on the national courts, to which the applicants refer, accurately to describe the factual context in which the questions referred for a preliminary ruling have arisen. While such an obligation is intended to enable the Court to ensure that the reference for a preliminary ruling is not inadmissible, the fact remains that, according to settled case-law, in order for a request to be inadmissible, the interpretation of EU law sought must bear no relation to the actual facts of the main action or to its purpose, the problem must be hypothetical or the Court must not have before it the factual or legal material necessary to give a useful answer to the questions submitted to it, (8) which is not the situation in the present case.
22. Consequently, there is no need to rule on the criticisms made by the applicants regarding the relevance of the factual circumstances proposed by the referring court. The same applies to the applicants’ requests to reformulate the questions referred for a preliminary ruling, by which they request the Court to (re)examine the description of the facts given by the referring court and to reclassify them, which is exclusively the role of the referring court.
23. Lastly, it should be noted that the wording of the questions by the referring court seems to suggest that the second question referred should be answered only if the first question is answered in the affirmative. In my view, in a case such as that in the main proceedings, in which the main issue is whether an exchange of information with the characteristics observed in the present case constitutes a restriction of competition by object, it is appropriate to address together the matters dealt with in the two questions referred for a preliminary ruling. Accordingly, the efficiencies or procompetitive effects alleged and which are referred to in the second question will be relevant in the analysis of the legal and economic context in which the exchange of information must be assessed, in order to determine whether it constitutes a restriction of competition by object.
B. The first and second questions referred
24. The two questions referred by the referring court concern the legal classification as a restriction of competition by object of an exchange of information having the characteristics described in points 10 to 14 of the present Opinion.
25. It should be borne in mind in that regard that, in the context of the procedure referred to in Article 267 TFEU, the role of the Court is limited to interpreting the provisions of EU law referred to it, in the present case Article 101(1) TFEU. Therefore, it is not for the Court of Justice, but for the referring court, to determine in the end whether, taking account of all of the information relevant to the situation in the main proceedings and the economic and legal context of which it forms part, the agreement at issue has as its object the restriction of competition. (9) However, the Court, when giving a preliminary ruling, may, on the basis of the information available to it, provide clarification designed to give the national court guidance in its interpretation in order to enable it to decide the case before it. (10)
26. Accordingly, before examining those questions, it will be helpful, in my view, to recall the scope of the concept of ‘restriction by object’ and to provide some clarification concerning the framework for applying that concept to exchanges of information.
1. The concept of restriction of competition by object
(a) The general principles set out in the case-law of the Court
27. In order to be caught by the prohibition laid down in Article 101(1) TFEU, an agreement, a decision by an association of undertakings or a concerted practice must have the ‘object or effect’ of appreciably preventing, restricting or distorting competition within the internal market. (11) It must be recalled, in that regard, that the anticompetitive object and effect of an agreement are not cumulative but alternative conditions for assessing whether such an agreement comes within the scope of the prohibition laid down in Article 101(1) TFEU. Accordingly, the alternative nature of that requirement, as shown by the conjunction ‘or’, leads to the need to consider the precise object of the agreement. (12)
28. In accordance with the settled case-law of the Court, certain types of coordination between undertakings reveal a sufficient degree of harm to competition so that there is no need to examine their effects. That case-law arises from the fact that certain forms of coordination between undertakings can be regarded, by their very nature, as being harmful to the proper functioning of competition. (13)
29. In order to determine whether an agreement between undertakings or a decision by an association of undertakings reveals a sufficient degree of harm to be considered a restriction of competition by object as provided for in Article 101(1) TFEU, regard must be had to the content of its provisions, its objectives and the economic and legal context of which it forms part. (14)
30. It should be pointed out, however, that the concept of restriction of competition by object must be interpreted restrictively. It follows that, where the analysis of coordination between undertakings does not reveal a sufficient degree of harm to competition, the effects of the coordination should be considered and, for it to be caught by the prohibition, it is necessary to find that factors are present which demonstrate the restriction of competition. (15)
31. Lastly, it should be recalled that, from a substantive point of view, there is no difference between the conduct of undertakings which is considered to restrict competition following an examination of its effects or its object, since both types of conduct are prohibited under Article 101(1) TFEU. The dichotomy between restriction of competition by object and restriction of competition by effect is first and foremost a procedural device meant to guide the Competition Authority on the analysis to be carried out under Article 101(1) TFEU and the resources which it must apply depending on the circumstances of the case. (16)
(b) The need for there to be ‘reliable and robust experience’ for a practice to be classified as a restriction of competition by object
32. Among the issues raised by the referring court in its first question is that of the need for there to be reliable and robust experience, within the meaning of the case-law of the Court, in order to be able to classify a ‘standalone’ exchange of information, such as that at issue in the main proceedings, as a restriction of competition by object. In other words, the referring court is seeking to ascertain whether there must be a precedent in order to find that a certain type of conduct constitutes a restriction of competition by object.
33. That question must be answered in the negative.
34. It must be recalled, in the first place, that it is clear from the very wording of Article 101(1) TFEU and, in particular, from the term ‘in particular’ that that provision does not contain an exhaustive list of agreements which have as their ‘object’ or ‘effect’ the restriction of competition. Other types of agreements may also be classified as restrictions ‘by object’ where such a classification may be made in accordance with the requirements stemming from the case-law of the Court. (17)
35. It must be noted, in the second place, that, although in several of those judgments – including in particular the judgments of 11 September 2014, CB v Commission (C‑67/13 P, EU:C:2014:2204; ‘the judgment in CB v Commission’), and the judgment in Budapest Bank – the Court has, in fact, emphasised the need for there to be sufficiently reliable and robust experience in order for an agreement to be regarded, by its very nature, as harmful to the proper functioning of competition, and therefore to be able to be classified as a restriction of competition by object, (18) the Court has nonetheless clearly recognised that the fact that the Commission did not take the view that a certain kind of agreement was, by virtue of its very object, restrictive of competition cannot, in itself, prevent it from doing so in the future following an individual, detailed examination of the practices at issue. (19) Any other interpretation would be tantamount to obstructing the application of a Treaty provision which is drafted in such a way as to cover new categories of restriction of competition which could arise in the future.
36. It is, therefore, appropriate to reject the argument of some of the applicants that the existence of reliable and robust experience is a precondition for a practice to be regarded as a restriction by object and that, consequently, national courts or competition authorities must establish that there is a precedent in order to classify conduct on the market as a restriction of competition by object. (20)
37. It is necessary to recall, in the third place, that the essential legal criterion for ascertaining whether an agreement or concerted practice involves a restriction of competition by object as provided for in Article 101(1) TFEU is the finding that such an agreement or practice reveals in itself a sufficient degree of harm to competition for it to be considered that it is not necessary to assess its effects. (21)
38. Agreements or practices which are comparable to conduct or categories of conduct whose harm to competition is beyond doubt in the light of experience will, as a rule, more easily fulfil that criterion. The existence of reliable and robust experience regarding the harmful nature of an anticompetitive practice ‘increases’ the likelihood that a practice which has the same characteristics as another which has previously been classified as a restriction of competition by object will also be so classified. (22) However, as I stated in point 34 of the present Opinion, the absence of precedents does not prevent competition authorities from classifying as a restriction of competition by object agreements which, following an individual and detailed examination, are found to be detrimental to competition.
39. Moreover, the fact that not all the characteristics of an agreement which is the subject of proceedings by a competition authority are identical to the characteristics of a practice which has previously been classified as a restriction of competition by object does not mean that there is not sufficiently reliable and robust experience in that regard. Indeed, to require that all the characteristics of such agreements match absolutely (including in relation to the relevant markets), as some of the applicants appear to maintain, would unjustifiably limit the scope of the concept of restriction by object and make it particularly difficult for competition authorities to apply that concept.
40. Nevertheless, in view of the requirement to interpret the concept of restriction of competition by object restrictively, although practices for which there are no precedents may be regarded as restrictions by object, that classification should be restricted to cases where the anticompetitive nature of an agreement or practice is obvious or where the practices in question have no credible explanation other than to restrict competition on the market. (23)
(c) Taking into account the legal and economic context when assessing a restriction of competition by object and how it differs from the analysis of restrictive effects on competition
41. A second major issue in the present case is the taking into account of the legal and economic context when assessing a restriction of competition by object and the distinction between taking into account the context and examining the effects, when assessing a restriction of competition.
42. It should be recalled, first, that the main purpose of taking into account the legal and economic context when analysing a restriction of competition by object is to confirm or invalidate an initial finding that a given practice has an anticompetitive object, which was made on the basis of other elements specific to that practice.
43. It should be noted, in that regard, that, in its judgment in Budapest Bank, the Court referred to the Opinion of Advocate General Bobek, in accordance with which the finding of a restriction of competition by object requires a two-step analysis. (24) Accordingly, the competition authorities must determine, as a first step, whether, given the content and objectives of the agreement, that agreement falls within a category of agreements which are detrimental to competition, in the light of a reliable and robust wealth of experience (or, failing that, which are clearly detrimental to competition). (25) As a second step, those authorities must carry out a ‘basic reality check’ in order to verify whether specific circumstances of the legal and economic context of the agreement concerned may cast doubt on the presumed harmful nature of that agreement. (26)
44. The examination of the legal and economic context seeks to avoid the risk of ‘false positives’ which may stem from a form-based analysis of an agreement, which is detached from the ‘economic reality’ and the legal and legislative landscape in which it occurs. The object of an agreement must be assessed not in the abstract but in the circumstances of the individual case and in the light of the actual conditions in which the market functions, having regard to all relevant factors. (27) That approach reflects, more specifically, the development of the Court’s case-law and the move from a broad and formalistic interpretation of the concept of restriction of competition by object to a more restrictive interpretation of that concept based on logic and experience. (28)
45. Second, the analysis of the economic context in which the practice occurs must not be confused with an analysis of its effects, since the latter entails an additional burden of proof and a more detailed examination of the effects of the agreement on the market in order to establish the existence of a restriction of competition. Were it otherwise, the dichotomy between restriction ‘by object’ and restriction ‘by effect’ would no longer have any meaning.
46. Although that distinction may seem relatively simple in theory, its practical implementation nonetheless proves more complex. Taking account of the economic and legal context in which an agreement occurs may, in certain cases, obscure in particular the dividing line between where the examination of the agreement in the light of its object ends and where the examination of the agreement in the light of its effects begins. Furthermore, the fact that in its case-law the Court considers that the assessment of the effects of agreements or practices under Article 101 TFEU implies, as for a finding of a restriction by object, the need to take into consideration the actual context in which they occur, in particular the economic and legal context in which the undertakings concerned operate, the nature of the goods or services affected, as well as the real conditions of the functioning and the structure of the market,(29) may also lead to confusion.
47. In essence, the difference between the two categories of restriction of competition lies in the intensity with which they are examined. Accordingly, in cases where the anticompetitive object is easy to perceive, the analysis of the economic and legal context in which the practice occurs ought to be limited to what is strictly necessary to confirm or cast doubt on the harmful nature and anticompetitive object revealed by the analysis of the content and objectives of the practice in question. (30) It follows that such an analysis certainly cannot, as a rule, remedy a failure actually to identify an anticompetitive object by demonstrating the potential effects of the measures in question. (31)
48. As recalled in point 28 of the present Opinion, in order to conclude that an agreement has an anticompetitive object, it must be possible to establish that it is in fact capable of restricting competition without having to examine its effects. Therefore, an analysis of the anticompetitive object of an agreement must ‘switch’ to an analysis of the anticompetitive effects of that agreement only where it becomes apparent that it is impossible to establish, despite an analysis of all the relevant inherent contextual factors, that that agreement is capable of restricting competition. (32) That would be the case, for example, if the analysis of the legal and economic context gave rise to doubts as to whether the agreement concerned was particularly harmful in nature (during the first stage of the test) or the analysis revealed effects which are at least ambiguous.
49. Third, as the Court recently recalled in the judgment in HSBC, (33) where the parties to an agreement rely on its procompetitive effects, those effects must, as elements of the context of that agreement, be duly taken into account for the purpose of its characterisation as a restriction by object, in so far as they are capable of calling into question the overall assessment of whether the concerted practice concerned revealed a sufficient degree of harm to competition and, consequently, of whether it should be characterised as a restriction by object. (34) However, a mere unsubstantiated assertion concerning the procompetitive effects of an agreement at issue is insufficient to rebut its characterisation as a restriction by object. (35) Therefore, even assuming that they are demonstrated, relevant and specifically related to the agreement concerned, those procompetitive effects must be sufficiently significant, such that they justify a reasonable doubt as to whether the agreement concerned caused a sufficient degree of harm to competition, and, therefore, as to its anticompetitive object. (36)
50. Fourth, although taking account of claimed efficiency gains or procompetitive effects forms part of the legal and economic context in which the exchange of information must be assessed, it should be clarified that that stage of the analysis is different from that carried out under Article 101(3) TFEU, which aims to examine, once a restriction of competition has been established, whether the criteria for exemption have been satisfied. (37) Taking account of procompetitive effects is intended not to undermine characterisation as a restriction of competition within the meaning of Article 101(1) TFEU, but rather to appreciate the objective seriousness of the practice concerned and, consequently, to determine the means of proving it. (38)
2. Application of the concept of restriction of competition by object to exchanges of information
51. As a preliminary point, it should be noted that it is clear from the settled case-law of the Court that the definitions of ‘agreement’, ‘decisions by associations of undertakings’ and ‘concerted practice’ referred to in Article 101(1) TFEU are intended, from a subjective point of view, to catch forms of collusion having the same nature which are distinguishable from each other only by their intensity and the forms in which they manifest themselves. (39) It follows that the criteria laid down in the Court’s case‑law for the purpose of determining whether conduct has as its object or effect the prevention, restriction or distortion of competition apply irrespective of whether the case involves an agreement, a decision or a concerted practice. (40)
52. Moreover, as regards the definition of a concerted practice, the Court has held that such a practice is a form of coordination between undertakings by which, without it having been taken to the stage where an agreement properly so-called has been concluded, practical cooperation between them is knowingly substituted for the risks of competition. (41) Furthermore, it has been established since the judgment of 16 December 1975, Suiker Unie and Others v Commission (40/73 to 48/73, 50/73, 54/73 to 56/73, 111/73, 113/73 and 114/73, EU:C:1975:174, paragraph 288), that exchanges of information may constitute an autonomous infringement falling within the scope of Article 101(1) TFEU.
53. It follows from Article 101(1) TFEU that the concept of a concerted practice implies, in addition to the participating undertakings concerting with each other, subsequent conduct on the market and a relationship of cause and effect between the two. However, the Court has held that, subject to proof to the contrary, which the economic operators concerned must adduce, it had to be presumed that the undertakings taking part in the concerted action and remaining active on the market take account of the information exchanged with their competitors in determining their conduct on that market. (42)
54. Exchanges of information which reduce or remove the degree of uncertainty as to the operation of the market in question, with the result that competition between undertakings is restricted, are therefore considered to be contrary to Article 101(1) TFEU. (43) The TFEU provisions relating to competition entail a requirement that economic operators should be free to act independently. Although that requirement does not deprive economic operators of the right to adapt themselves intelligently to the existing and anticipated conduct of their competitors, it does, however, strictly preclude any direct or indirect contact between such operators the object or effect of which is either to influence the conduct on the market of an actual or potential competitor or to disclose to such a competitor the course of conduct which they themselves have decided to adopt or contemplate adopting on the market. (44)
55. Nevertheless, while the Court has, on a number of occasions, been called upon to examine the compatibility of exchanges of information with Article 101(1) TFEU, it has not always stated clearly whether the criteria to which it refers, in particular the criterion of the reduction or removal of uncertainty as to the operation of the market, relate to the concept of restriction in general, whether the aforementioned criteria cover only restrictions by effect or whether they are also capable of establishing an anticompetitive object. (45) However, the fact that that criterion is applied for identifying both a restriction of competition by object and by effect is not surprising. As I mentioned in point 31 of the present Opinion, the significance of the distinction between restriction of competition by object and by effect is primarily evidential.
56. It should be pointed out, however, that in certain judgments in which reference has been made to the concept of restriction by object in the context of an exchange of information, the Court has sought to clarify further the circumstances in which a classification as a restriction of competition by object should be made. In particular, the Court has held that an exchange of information which is capable of removing uncertainty between participants as regards the timing, extent and details of the modifications to be adopted by the undertakings concerned in their conduct on the market must be regarded as pursuing an anticompetitive object, (46) and, consequently, as capable of influencing directly the commercial strategy of competitors or of affecting normal competition on the market. (47) That is the case, in particular, where the exchange of information relates to particularly sensitive elements from the point of view of competition, such as future pricing or one of the components of those prices, as was the case in the judgments in T-Mobile and Dole or, more recently, in the judgment in HSBC.
57. However, where the exchange of information concerns elements which are less sensitive from the point of view of competition law or where the anticompetitive object is not clear from an analysis of the content, objectives and legal and economic context in which those exchanges occur, the Court has held that it is necessary to carry out an analysis of the effects. That was the solution adopted, for example, in the ASNEF case, in which the Court held that, given its characteristics, the exchange of certain data between banking institutions intended to create a banking register, did not have the object of restricting competition and that it was therefore necessary to analyse its effects (48). In the same case, the Court held, furthermore, that, in view of certain measures taken by the banks in question not to disclose sensitive data, the files exchanged between them were not capable of revealing either the respective market positions of the competing undertakings or their commercial strategies. (49)
58. In the light of the foregoing, the following observations must be made.
59. In the first place, it should be pointed out that not every exchange of information between competitors can be regarded as restrictive of competition. Exchanges of information are a common feature of several competitive markets. Moreover, economic theory shows that transparency between traders may contribute to competition intensifying, and make it possible to solve problems of information asymmetries and generate various types of efficiencies, thereby making markets more efficient. (50)
60. In the second place, it is clear from the case-law of the Court cited in point 54 of the present Opinion that, as regards exchanges of information between competitors, the reduction or removal of uncertainty as to the strategic conduct of a competitor on the market is the decisive criterion for assessing whether there has been an infringement of Article 101(1) TFEU.
61. Assessment under that criterion mentioned above is highly dependent on the nature of the information exchanged between competitors. Only the exchange of strategic (or commercially sensitive) information can reduce uncertainty on the market and have an impact on the decision-making independence of the parties, thereby reducing competition. While there is no precise definition of the concept of strategic (or commercially sensitive) information, it is accepted, in principle, that information related to prices and quantities is the most strategic, followed by information about costs and demand. (51) Moreover, the strategic usefulness of the data exchanged may also depend on a series of other factors such as the degree of concentration in the market concerned, whether the information is aggregated or disaggregated, its age and the frequency of the exchanges in question. (52)
62. In the third place, although an exchange of information may fall within the scope of Article 101(1) TFEU, the fact that such an exchange concerns strategic data which are capable of reducing uncertainty on the market does not automatically mean that it is classified as a restriction of competition by object.
63. In view of the need to interpret the concept of restriction of competition by object restrictively, that classification can be applied only to exchanges of information in respect of which it is clear and unambiguous that, in view of their characteristics and without it being necessary to examine their effects, the criterion of the reduction or removal of uncertainty on the market is satisfied, with the result that those exchanges can influence directly the commercial strategy of competitors by enabling them to adapt their conduct on the market. As I have mentioned in point 56 of the present Opinion, that criterion will be deemed to be satisfied where the exchange of information concerns key elements of competition, such as future capacities and pricing.
64. In the light of the foregoing, it must be concluded that an exchange of information can constitute a practice which restricts competition by object where it is apparent from an analysis of its content, its objectives and the legal and economic context in which it occurs that that exchange presents a sufficient degree of harm to competition. Moreover, the fact that that exchange is a ‘standalone’ exchange, in the sense that it is not associated with a finding of a cartel, is not such as to call into question the finding that there is a restriction of competition by object, provided that the exchange presents a sufficient degree of harm. (53)
3. Assessment of the classification as a restriction of competition by object in the present case
(a) Preliminary observations
65. As a preliminary point, it should be noted that it is apparent from the order for reference that, in view of their characteristics as described in points 10 to 14 of the present Opinion, the exchanges in question related to current and future data which were strategic from the point of view of competition law and provided the applicants with accurate information about their competitors’ offers, thereby reducing their uncertainty with regard to strategic conduct and facilitating alignment by means of informal coordination.
66. That description is, however, contested by the applicants who argue that, contrary to the findings of the Competition Authority and the referring court, the characteristics of the information exchanged did not permit such coordination on the market. (54)
67. It should, however, be recalled, first, that it is not for the Court to verify the accuracy of the facts set out by the referring court (55) and, second, that it is for the referring court to make the final assessment, having regard to all the relevant factors characterising the situation in the main proceedings and the economic and legal context of which it forms part, as to whether the exchange of information in question had the object of restricting competition. (56)
68. Having made those clarifications, I propose to examine, as a first stage, the part of the exchange of information relating to the commercial conditions of loans applied for (in particular those relating to credit spreads) before going on to examine, as a second stage, the exchanges relating to production volumes and, as a third and final stage, the conditions under which a single exchange concerning those two types of information, being examined jointly, could be regarded as pursuing an anticompetitive object.
(b) The information on ‘commercial conditions’
(1) The content of the information exchanged
69. As a preliminary point, it should be noted that it is apparent from the order for reference that the ‘credit spreads’ on which the banks exchanged information are an essential element of pricing. (57) It also follows from that order that, by communicating between competitors one of the components of the price which they were going to apply, the applicants contributed to increased transparency on the market by reducing uncertainty with regard to their current or future strategy, which allowed each of the participating banks to use that information in determining its commercial strategy and to align itself, at any time, by means of informal coordination.
70. It should be noted at the outset that it is clear from the case-law of the Court referred to in point 56 of the present Opinion that an exchange with such characteristics is capable of being classified as a restriction of competition by object. Accordingly, contrary to the position taken by some of the applicants, it must be held that there is sufficiently reliable and robust experience to support the view that such exchanges concerning future pricing (or certain parameters of pricing) are inherently anticompetitive, in view in particular of the especially high risk of collusion which they entail, with the result that they can be classified as a restriction of competition by object.
71. It should also be borne in mind that, in accordance with the case-law of the Court, the assessment of an exchange of information between competitors must be made in the light of the concept inherent in the Treaty provisions relating to competition, that each economic operator must determine independently the policy which it intends to adopt on the market (58) However, the exchange of information between competitors on factors determining price blatantly contravenes that independence requirement, in particular where that information relates to future pricing intentions which allows undertakings to anticipate a competitor’s strategic commercial moves and to adapt to them, thereby reducing competitive pressure on the market.
72. It should also be noted that, in addition to being confidential at the time of the exchange, the information relating to credit spreads is particularly relevant for determining the loan offers which banks make to their customers. Accordingly, notwithstanding the fact that the banking market is highly regulated, banking institutions have a degree of decision-making freedom as regards determining credit spreads, which ensures strategic differentiation of each bank and therefore constitutes a key parameter of competition between those institutions. (59)
73. The content of that exchange therefore presents, in itself, a sufficient degree of harm to competition and may be regarded, by its very nature, as being detrimental to the proper functioning of normal competition, which is sufficient for it to be concluded that conduct has occurred which distorted the competitive process on the relevant markets. (60)
74. Moreover, contrary to the view taken by some of the applicants, it is not necessary for a concerted practice to cover every parameter of competition. Such a practice can have an anticompetitive object even where it concerns only individual parameters of competition, such as the credit spread. (61) The fact that the final price includes other components which may not have (all) been the subject of an exchange of information is not such as to call into question the finding that there is a restriction of competition by object.
75. The strategic and commercially sensitive nature of the data exchanged is not, moreover, called into question, even if it were to be established, as a number of the applicants maintain, that some of the exchanges in question related neither to the final prices charged by the banks nor to the credit spreads actually offered to customers, but rather concerned a range of indicative rates which were used as starting points for individual negotiations with each customer according to his or her particular risk profile. The disclosure of such data can be sufficient to reveal strategic intentions regarding future pricing behaviour and thereby to facilitate collusive behaviour between competing undertakings. (62)
(2) The purpose of the exchange of information
76. It is settled case-law of the Court that, to determine whether an agreement or exchange of information comes within the prohibition laid down in Article 101(1) TFEU, close regard must be paid in particular to its objective aims. (63) Moreover, those objective aims, which must be clear from the practices at issue, should not be confused with the subjective intentions of whether or not to restrict competition or with any legitimate objectives pursued by the undertakings in question. It is, furthermore, established that an agreement may be regarded as having an object restrictive of competition even if it pursues other legitimate objectives. (64)
77. In that regard, it should be recalled that the Competition Authority found that, given the nature of the information covered by the exchange, that exchange could have no objective other than to restrict competition. That finding is disputed by the applicants, which maintain that the exchange of information was an informal means of facilitating benchmarking by the banks, enabling them to compare their respective offers, (65) while reducing the costs associated with such a comparison exercise, which was likely to have procompetitive effects, (66) with the result that the purpose of the exchange was not inherently anticompetitive.
78. That argument does not seem credible and should therefore be rejected.
79. While an exchange of information can generate efficiencies and make undertakings more efficient, in particular by making it possible for them to compare their respective practices and thus improve both their internal efficiency and their position on the market, it goes without saying that initiatives such as benchmarking do not justify practices which are anticompetitive in themselves, such as exchanging information which is confidential and strategic from the point of view of competition law, such as information relating to actions planned by undertakings concerning pricing.
80. I also find it difficult to follow the banks’ reasoning regarding the objectives pursued by the exchange of information in question. The usefulness of such an exchange is questionable since, according to the applicants, the information exchanged, on the one hand, would be made public by the banks at the same time as (or immediately after) the exchange in question and, on the other hand, could not be taken into account in order to modify credit spreads, in view of the internal procedures applicable within the banks. Apart from the fact that that finding is disputed by the Competition Authority and is not apparent from the order for reference, such an exchange of information would lack all commercial rationale. It is therefore necessary to question the motivation which could have led the applicants to expose themselves to a not inconsiderable risk from the point of view of competition law in order to obtain information which was, according to them, of no real commercial interest.
(3) The legal and economic context
81. As a preliminary point, it should be stated that the applicants criticise the Competition Authority for failing to take into account the economic, legal and regulatory context of the banking sector during the infringement period. Assessed in the relevant legal and economic context, the exchanges in question were, according to the applicants, actually procompetitive or, at the very least, such an assessment should have raised doubts regarding whether those exchanges were harmful, which would have called into question the finding of a restriction of competition by object. (67)
82. It should also be recalled, as I stated in point 47 of the present Opinion, that where the anticompetitive object is easy to perceive, as appears to be the case for exchanges relating to credit spreads, the analysis of the economic and legal context in which the practice occurs should be limited to what is strictly necessary to ascertain whether there are any particular circumstances capable of giving rise to doubt as to the harmful nature of that practice. It follows that only those elements which are truly relevant to the analysis of the legal and economic context must be examined by a competition authority and that the Competition Authority is not required to examine arguments which are purely hypothetical or distinct from the legal and economic context in which the exchange of information or the conduct in question takes place. (68)
83. The applicants dispute, in the first place, the assessments made by the Competition Authority and the referring court as to the degree of concentration (and market shares) of the various participants during the period covered by the exchanges and the fact that those exchanges took place within a ‘closed loop’.
84. While it is not for the Court to take the place of the referring court in assessing the method used by the Competition Authority and the validity of the analysis carried out by it, it should be stated, first of all, that the degree of concentration is one of the factors which may prove to be relevant in the analysis of a restriction of competition (69). Accordingly, the Court has already held that in a highly concentrated market the exchange of certain information can, depending in particular on the type of information exchanged, be such as to enable undertakings to understand the position and commercial strategy of their competitors on the market, thereby distorting rivalry on that market and increasing the probability of collusion, or even facilitating it. (70)
85. In addition to the collusive potential of that exchange, the fact that it took place within a closed loop also poses a risk of excluding banks which do not participate in the exchange and which therefore do not have access to the same data in order to assess existing and future conditions on the relevant market. Such an exchange within a limited group of participants is also liable to make it more difficult for new operators to enter the market, (71) in particular where that market has the characteristics identified by the referring court. (72)
86. Moreover, the fact that, unlike the exchanges of information on production volumes which were regular, the exchanges concerning the credit spread tables were sporadic, as the applicants argue, does not in itself rule out the possibility that an exchange of information has an anticompetitive object. The Court has held that a meeting on a single occasion may, in view of the structure of a market, constitute a sufficient basis for undertakings to concert their conduct. (73)
87. In the second place, the applicants submit that, given the characteristics of the Portuguese banking market, the conduct at issue could not have any effect on competition, nor therefore could it lead to collusive conduct. (74) More specifically, in view of the internal procedures which the credit institutions have to follow in order to modify their offer, those institutions could not react immediately to such information.
88. In that regard, it should be noted that it is apparent from the case-law of the Court that subject to proof to the contrary, which the parties must adduce, (75) it must be presumed that the undertakings take account of the information exchanged with their competitors in determining their conduct on the market when they remain active on it (76) and that that is all the more the case where the undertakings concert together on a regular basis over a long period. (77)
89. It seems to me that the applicants are seeking to rebut that presumption by relying on it being ‘factually impossible’ to take that information into account in order to adapt and modify their conduct on the market. However, even if that were the case (which is not apparent from the order for reference), such an argument would not be sufficient in itself to rule out a finding of a restriction of competition by object.
90. It should be recalled, first, that Article 101 TFEU, like the other competition rules of the Treaty, is designed to protect not only the immediate interests of competitors or consumers but also to protect the structure of the market and thus competition as such. Therefore, in order to find that a concerted practice has an anticompetitive object, there does not need to be a direct link between that practice and the prices paid by end users. (78) Accordingly, even conduct which may give rise to a certain reduction in the price of the relevant products or services (or have a neutral impact on competition) may, under certain circumstances, be regarded as being inherently anticompetitive. (79)
91. Second, the fact that the credit spreads would be made public shortly after the exchanges in question in no way alters the fact that that information was confidential and not accessible to the public at the time it was actually exchanged. Moreover, it is apparent from the order for reference that the information related to the future pricing intentions of the banks concerned and was therefore strategic and particularly sensitive from the point of view of competition law.
92. Even supposing that the argument put forward by the applicants had been proven, I am of the view that it would not be sufficient to call into question the finding that there was an infringement consisting in a restriction by object but that that could be taken into account in calculating the fine and, depending on the circumstances, could lead to a reduction in the fine imposed.
93. In the third place, as regards the arguments based on the specific nature of the banking sector, the applicants maintain that, by reducing uncertainty on the market, the exchange of information at issue could give rise to efficiencies and have procompetitive effects which are beneficial to consumers.
94. It should be noted, in that regard, as I recalled in point 49 of the present Opinion, that the possibility of there being procompetitive effects highlighted by the applicants does not call into question the conclusion that a specific agreement constitutes a restriction of competition by object, unless the effects in question are proven, relevant, specific and so significant as to be capable of calling into question the conclusion that the exchange of information is inherently harmful to competition.
95. Although the applicants identify in a general and rather theoretical fashion certain allegedly procompetitive aspects which follow from the exchanges at issue, they do not appear to be in a position to provide proof thereof. There is no evidence in the file that those exchanges made it possible to improve the functioning of the market or correct its failures. (80) However, even supposing that the banks in question had passed on certain advantages to their customers, which is not apparent from the order for reference, that would not rule out the nature of the conduct in question being anticompetitive. (81)
96. Moreover, in view of the particularly sensitive nature of the information relating to credit spreads described in points 71 to 74 of the present Opinion, it is unlikely that the procompetitive aspects relied on by the applicants give rise to reasonable doubt as to whether the exchanges in question are harmful to competition. (82)
97. In the fourth place, the applicants maintain that, contrary to the findings of the referring court, the increase in interest rates was not due to the exchange of information at issue, but was explained by other factors linked to the global financial crisis which occurred during 2008 and the fiscal consolidation measures implemented by Portugal in the wake of that crisis.
98. For reasons comparable to those already set out in point 90 of the present Opinion, I am of the view that such an argument must also be rejected. (83) Moreover, it must be noted that, although it is clear from the order for reference that the exchanges in question intensified during the period of the economic crisis, the fact remains that those exchanges began in 2002, well before the onset of the financial crisis and the subsequent interventions by the associated regulatory bodies in response to that crisis.
99. In the fifth place, as regards the regulatory environment in which the exchange of information took place, the applicants criticise the referring court for failing to take account of the fact that the legislative framework applicable to the banking sector in Portugal includes a series of rules intended to guarantee a certain level of transparency on the market in order to avoid systemic crises. Those rules were, moreover, according to the applicants, introduced by EU consumer protection law.
100. That argument likewise cannot succeed, in my view, since it is clear from the order for reference that the information exchanged by the applicants was distinct and went beyond the information communicated by them in the context of their regulatory obligations. It should also be noted that, contrary to the assertions of some of the applicants, no rules of EU law could require banks to exchange information such as credit spreads. (84)
(c) Information on ‘production volumes’
101. There being no need to set out once more the ‘traditional analytical approach’ to a restriction of competition by object, which includes an examination of the content, objectives and legal and economic context in which an agreement occurs, I consider it important to provide some clarification regarding the exchanges relating to production volumes, since some of the parties to the main proceedings appear to argue that those exchanges could in themselves (and irrespective of the analysis of the exchanges on credit spreads) have an anticompetitive object.
102. It should be pointed out, first, that data relating to production volumes can, in principle, constitute strategic and sensitive information from the point of view of competition law, provided that the nature of the information exchanged and the context of that exchange make it possible to reduce uncertainty regarding the strategic conduct of a competitor on the market. (85)
103. Second, it must be noted that, unlike exchanges of information relating to credit spreads, the exchange in question does not relate to future practices but to data from the preceding month. Although it is for the referring court to make the final assessment concerning the temporal attributes of that information, in the light of the specific circumstances of the banking market in question, it should be borne in mind that, in principle, the exchange of past (or historic) data is unlikely to lead to a collusive outcome and is less harmful from the point of view of competition law, as it is unlikely to be indicative of the competitors’ future conduct or to provide a common understanding on the market. (86)
104. In that regard, it should be stated that there is no predetermined threshold for when data becomes historic, that is to say, old enough not to pose risks to competition. Whether data is genuinely historic depends on the specific characteristics of the relevant market and in particular the frequency of price renegotiations in the industry. (87) It follows that, although improbable, it cannot be ruled out that exchanges relating to past events could also constitute restrictions by object as provided for in Article 101 TFEU. That would be the case if the exchange of recent individualised information on strategic variables revealed trends, awareness of which would be capable of reducing or eliminating the parties’ uncertainty as to their future intentions on the market, in which case such an exchange could be equivalent to the exchange of information on future data.
105. Third, it is accepted that exchanges of genuinely aggregated data, that is to say, where it is sufficiently difficult to distinguish information relating to a particular company, are much less likely to lead to restrictive effects on competition than exchanges of data regarding specified companies. Accordingly, the risk that the exchange of strategic information may reduce uncertainty on the market and therefore restrict competition is higher when that information is disaggregated. (88)
106. Fourth, the fact that the information exchanged went beyond the regulatory obligations of the banks concerned and related to data which were not available to the public is not in itself sufficient to characterise those exchanges as anticompetitive. It is still necessary to establish that the information exchanged made it possible to reduce or remove uncertainty on the market (and did so in a clear and unambiguous manner in order for it to be concluded that there was a restriction of competition by object).
107. It follows that, although it cannot be ruled out that the exchange of recent and disaggregated data on production volumes may be strategic in nature and may be sensitive from the point of view of competition law, in particular where that exchange takes place in a highly concentrated market and where the frequency of those exchanges is high, the order for reference contains nothing allowing it to be established clearly, as is required by the need to interpret the concept of restriction by object restrictively, that that exchange is of a particularly harmful nature from the point of view of competition and would (in itself) have made it possible to reduce strategic uncertainty as to participants’ future conduct on the market. (89)
(d) Joint analysis of the information exchanged
108. It follows from the foregoing analysis that, while the part of the exchange of information described by the referring court which relates to credit spreads is likely to fall within one of the categories of agreements or concerted practices covered by the concept of restriction by object, such a finding is not so obvious as regards the production data, if those exchanges are analysed in isolation and separately.
109. It is clear, however, both from the Competition Authority’s initial decision and from the order for reference, that, while the referring court drew a distinction between the two types of information exchanged, it did not consider that each of those exchanges was in itself restrictive by object, but rather that they formed part of a single exchange which it classified as a restriction of competition by object. Moreover, it should be noted that, on the basis of the findings of the referring court itself as to the inherent harm to competition of the exchange, (90) such an exchange is likely to constitute a restriction of competition by object. Nonetheless, in order for such a legal classification to be made, two other conditions, which are not apparent from the case file in the present case, must also be satisfied.
110. First, it is important, from the point of view of legal certainty, to ensure that the theory of harm on the basis of which an anticompetitive practice is declared unlawful by a competition authority is clear, in particular in so far as concerns establishing that such a practice has an anticompetitive object. (91)
111. Second, the interaction between exchanges relating to those two types of information which make it possible to support the theory of harm adopted by a competition authority must emerge unambiguously from the analysis conducted by that authority. The Competition Authority must therefore establish a sufficiently clear link between exchanges relating to those two types of information and explain how an exchange with such characteristics is sufficiently harmful to competition to justify classification as a restriction of competition by object. In other words, the Competition Authority must show how those exchanges, taken together, form part of a manifestly anticompetitive ‘plan’ and are such as to allow the conduct of the banks concerned to converge. (92)
V. Conclusion
112. In the light of the foregoing considerations, I propose that the Court answer the questions referred for a preliminary ruling by the Tribunal da Concorrência, Regulação e Supervisão (Competition, Regulation and Supervision Court, Portugal) as follows:
(1) Article 101 TFEU must be interpreted as meaning that it does not preclude the classification as a restriction of competition by object of an exchange between competitors of information concerning the commercial conditions applicable to transactions (in particular, current and future credit spreads and risk variables) and production figures for home loan offers, corporate lending offers and consumer credit offers in the banking sector, where such a practice has artificially increased transparency and reduced uncertainty as to the functioning of the market.
(2) Article 101 TFEU does not preclude such classification where it has been impossible to identify or establish any gain in efficiency or any uncertain or positive effect on competition resulting from that exchange of information.