Contents
Judgment of the General Court (Seventh Chamber)23 July 2025note(Competition — Agreements, decisions and concerted practices — Sector of Foreign Exchange (Forex) spot trading of G10 currencies — Decision finding an infringement of Article 101 TFEU and Article 53 of the EEA Agreement — Exchanges of information — Agreements or concerted practices relating to G10 foreign exchange activities — Restriction of competition by object — Single and continuous infringement — Principle of sound administration — Rights of the defence — Fines — Basic amount — Proxy for value of sales — Article 23(2) and (3) of Regulation (EC) No 1/2003 — Unlimited jurisdiction)In Case T‑84/22,UBS Group AG, the successor in law to Credit Suisse Group AG, established in Zurich (Switzerland),UBS AG, the successor in law to Credit Suisse AG, established in Zurich,Credit Suisse Securities (Europe) Ltd, established in London (United Kingdom),represented by R. Wesseling and F. Brouwer, lawyers,vEuropean Commission, represented by A. Boitos, C. Zois and T. Franchoo, acting as Agents,THE GENERAL COURT (Seventh Chamber),composed of K. Kowalik-Bańczyk, President, E. Buttigieg (Rapporteur) and B. Ricziová, Judges,Registrar: I. Kurme, Administrator,having regard to the written part of the procedure,further to the hearing on 7 March 2024,gives the following
Judgment
1By their action under Article 263 TFEU, the applicants, UBS Group AG, the successor in law to Credit Suisse Group AG, UBS AG (‘UBS’), the successor in law to Credit Suisse AG, and Credit Suisse Securities (Europe) Ltd, seek, first, annulment of Commission Decision C(2021) 8612 final of 2 December 2021 relating to a proceeding under Article 101 TFEU and Article 53 of the EEA Agreement (Case AT.40135 – FOREX (Sterling Lads)) (‘the contested decision’) and, secondly, the reduction of the amount of the fine which, in that decision, was imposed jointly and severally on Credit Suisse Group, Credit Suisse and Credit Suisse Securities (Europe) (together, ‘Credit Suisse’).
I.Background to the dispute
A.The procedure giving rise to the contested decision
2The procedure was initiated following an application for a marker under the Commission Notice on Immunity from fines and reduction of fines in cartel cases (OJ 2006 C 298, p. 17; ‘the Leniency Notice’), submitted on 27 September 2013 by UBS. In that application, UBS informed the European Commission of the existence of an alleged infringement in the sector of foreign exchange spot trading of G10 currencies.
3Following that application, UBS made oral statements and produced documentary evidence. On 2 July 2014, the Commission granted UBS conditional immunity from fines pursuant to point 8(a) of the Leniency Notice.
4On 11 October 2013, 14 October 2013 and 17 July 2015, Barclays plc, Barclays Execution Services Limited and Barclays Bank PLC (together, ‘Barclays’), HSBC Holdings plc and HSBC Bank plc (together, ‘HSBC’) and NatWest Group plc and NatWest Markets Plc (together, ‘RBS’) respectively submitted applications for the reduction of fines under the Leniency Notice. Following those applications, they made oral statements and produced documentary evidence.
5On 25 July 2014, 1 March 2016 and 29 April 2016, the Commission sent requests for information to Credit Suisse and to all the undertakings referred to in paragraphs 3 and 4 above.
6On 27 October 2016, pursuant to Article 11(6) of Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles [101 and 102 TFEU] (OJ 2003 L 1, p. 1), the Commission initiated infringement proceedings against Credit Suisse and all the undertakings referred to in paragraphs 3 and 4 above, requesting that they state their interest in participating in a settlement procedure pursuant to Article 10a of Commission Regulation (EC) No 773/2004 of 7 April 2004 relating to the conduct of proceedings by the Commission pursuant to Articles [101 and 102 TFEU] (OJ 2004 L 123, p. 18), as amended.
7Credit Suisse and all the undertakings referred to in paragraphs 3 and 4 above decided to participate in a settlement procedure and in meetings with the Commission, on 9 November 2016 and 7 February 2018. On 19 February 2018, Credit Suisse informed the Commission that it was ending its participation in that procedure. The Commission therefore reverted to the standard procedure with regard to Credit Suisse under point 19 of its Notice on the conduct of settlement procedures in view of the adoption of Decisions pursuant to Article 7 and Article 23 of Regulation No 1/2003 in cartel cases (OJ 2008 C 167, p. 1).
8On 24 July 2018, the Commission sent Credit Suisse a Statement of Objections as part of the standard procedure. After it was granted access to the file, Credit Suisse submitted its written observations in response to the Statement of Objections and expressed its views at the hearing which took place on 7 December 2018.
9On 18 March 2021, the Commission sent Credit Suisse a supplementary Statement of Objections.
10Having had access to the file, Credit Suisse submitted, within the time limit prescribed, its written observations in response to the supplementary Statement of Objections. The second hearing took place on 8 June 2021.
11On 25 June 2021, Credit Suisse replied to the request for information made during the second hearing.
12On 2 December 2021, the Commission adopted the contested decision, which was addressed solely to Credit Suisse.
13On the same day, the Commission adopted Decision C(2021) 8613 final relating to a proceeding under Article 101 TFEU and Article 53 of the EEA Agreement (Case AT.40135 – FOREX (Sterling Lads)) (‘the settlement decision’), the addressees of which were the parties which had agreed to participate in the settlement procedure and had made a proposal for a settlement, namely Barclays, HSBC, RBS and UBS.
B.The contested decision
1.Relevant products and sector concerned
14The conduct referred to in the contested decision concerns the sector of foreign exchange spot trading of G10 currencies, namely: the euro (EUR); the Australian dollar (AUD); the Canadian dollar (CAD); the Swiss franc (CHF); the Danish krone (DKK); the British pound (GBP); the Japanese yen (JPY); the Norwegian krone (NOK); the New Zealand dollar (NZD); the krona (SEK); and the United States dollar (USD).
15The type of trading referred to in paragraph 14 above is defined in the contested decision as an agreement between two parties to exchange two currencies, namely to buy a certain amount (the ‘notional amount’) of one currency in exchange for its equivalent in another currency at the value at the time of the agreement (‘exchange rate’). The exchange rate of one currency is quoted against the amount of another currency for which it can be exchanged. Currencies are thus always traded in pairs and each therefore has a relative value. Exchange rates vary according to information about their fundamental value. In the short term, they are mainly determined by traders’ order flows, while market fundamentals determine exchange rates in the longer term.
16Since electronic trading activity (in the sense of trades which are automatically booked or executed either by the relevant bank’s proprietary electronic trading platforms or by computer algorithms) is excluded from the scope of analysis in the present case, the contested decision concerns only ‘voice trading’ activity.
17Foreign exchange spot trading transactions are concluded on the over-the-counter market, so that they remain decentralised with a high level of liquidity and constant trading. Those transactions are generally initiated when a financial institution is contacted by an end-customer, namely, in particular, financial institutions, which are considered to be informed customers since, given the scale of their transactions, they have strong incentives to acquire information liable to influence the evolution of exchange rates. Those financial institutions set up specific sales desks which serve as an interface between end-customers and the traders employed by brokers to trade currencies. End-customers may also contact traders directly.
18The banks, the traders of which were involved in the exchanges concerned in this case, act as market makers on the foreign exchange spot trading market and stand ready to enter into trades with their customers at any moment to ensure market liquidity.
19The types of transactions initiated by customers include:– immediate orders (allowing immediate entry into trades for a certain amount of currency based on the prevailing market rate, the settlement of which generally takes place within two days of entering into the transaction);– conditional orders (the settlement of which occurs only if certain market conditions are met, for example if the value of a currency reaches a certain level);– orders to be executed at a ‘fixing’, which in this case concerned only the WM/Reuters Closing Spot Rates and the European Central Bank (ECB) reference rates (the settlement of which occurs at a specific reference rate).
20Traders generally offer two-way prices (a bid price or buyer price, and an ask price or seller price), depending on the size of the transaction and the currencies being traded. The revenue generated by foreign exchange transactions depends on the volume of currency exchanged and the difference between the bid price and the ask price of a single currency (‘the bid-ask spread’). The bid-ask spread is the compensation received by the trader for the immediacy of the service provided and for the risk he or she subsequently bears by having a notional amount of a given currency in his or her trading book.
21After buying one currency in exchange for another, a trader finds him or herself in a ‘long position’, in the sense that he or she holds a positive notional amount of a currency in his or her trading book, until he or she ‘closes’ that position, that is to say sells that currency. After selling one currency in exchange for another, the trader finds him or herself in a ‘short position’, in the sense that he or she holds a negative notional amount of a currency in his or her trading book, until he or she ‘closes’ that position, that is to say buys that currency. Accordingly, a ‘risk position’ remains open until an opposite trade takes place. However, on account of the rapid fluctuations in exchange rates on the market, those positions present risks for traders, with the result that, on that market, traders compete with one other in the expert management of the risks associated with those positions.
22Traders determine their strategy and, in particular, the setting of bid-ask spreads, or even prices, in the light of their trading book and the information available to them. On the one hand, it is clear from the evidence in the file that market makers must actively adjust prices in relation to inventory by altering prices and not simply bid-ask spreads, with the objective of achieving rapid inventory turnover and not accumulating significant positions on one side of the market. On the other hand, access to market information is important for traders in their decision-making. Exchange rate movements are in fact influenced, in the long term, by macroeconomic variables and, in the short term, by the flow of customer orders or, in other words, by microeconomic variables. In so far as concerns the latter variables, the parties also agree that orders from informed customers – that is to say, other financial institutions – indicate to traders the direction in which the market is likely to move, thereby encouraging traders in possession of that information to trade in the same direction as the informed customer. Information relating to the flows of customer orders is, in principle, confidential.
23Moreover, traders may also be willing to create, keep or increase their (open risk) positions in their proprietary trading book. Proprietary trading occurs when traders primarily engage in trading activities with the banks’ own money rather than on behalf of their clients, and look to exploit a competitive advantage in the market by building open risk positions which would enable them to earn excess returns.
24Lastly, when traders are not servicing incoming orders from end-customers in their market making capacity, they typically trade with each other on the ‘interdealer’ market, that is to say a structure of the foreign exchange spot trading market reserved exclusively for financial institutions (‘dealers’), with the aim of adjusting their own inventory positions or taking a position for the purposes of their proprietary trading activities. In that case, traders are each other’s counterparties. They can communicate directly (bilaterally), usually via an electronic system. Prices are quoted on a take-it-or-leave-it basis. They may also communicate via a system of brokers who announce the best bid-ask spreads on screens and make the information available on their platforms.
2.The conduct alleged against Credit Suisse
25In the contested decision, the Commission referred to three types of conduct that took place on the private and multilateral online chatroom, namely Sterling Lads (‘the chatroom at issue’), between five undertakings active in the banking and financial sector, namely Barclays, Credit Suisse, HSBC, RBS and UBS. That conduct concerned, in the words of the contested decision, (i) an underlying understanding; (ii) exchanges of information, and (iii) occasional instances of coordination. According to the contested decision, the discussions in that chatroom relating to that conduct took place from 25 May 2011 to 12 July 2012 and Credit Suisse participated in the discussions in that chatroom from 7 February to 12 July 2012 (‘the relevant period’).
26In the first place, as regards the underlying understanding, the Commission found that participation in the chatroom at issue entailed forming part of a circle of trust, mutual expectations and mutual benefits and, therefore, compliance with a series of tacit rules, which could be summarised, in essence, as follows: (i) traders would meet in the chatroom at issue to exchange information throughout the day; (ii) the information exchanged in the chatroom would not be disclosed to other competing traders, (iii) the information exchanged could be used to the participating traders’ benefit, and (iv) that information should not be used against the persons who shared it. The Commission took the view that that conduct constituted an agreement, which it characterised as a ‘restriction by object’ within the meaning of Article 101(1) TFEU on the ground, in essence, that it entailed rational ex ante expectations designed to remove the uncertainties inherent in the market in question for the mutual benefit of the traders and revealed, therefore, a sufficient degree of harm to competition.
27In the second place, in so far as concerns the exchanges of information, the Commission referred to numerous contacts between the traders of the banks concerned. According to the Commission, those contacts involved, on an extensive and recurrent basis, exchanges of current or forward-looking commercially sensitive information about their trading activities, which enabled the banks concerned to be informed, when taking their market decisions, of their competitors’ positions, intentions and constraints, which helped them in their subsequent decisions.
28In that context, the Commission found that the general objective of the exchanges of information was to influence two essential parameters of competition in spot trading on the relevant market – namely price and expert risk management – thereby removing the inherent uncertainties of that market. In so doing, the participating traders knowingly substituted practical cooperation between them for the risks of autonomous competition on those parameters, to the detriment of the other market participants, thereby revealing a sufficient degree of harm to competition. Therefore, it characterised those exchanges of information as agreements and/or concerted practices which are restrictive of competition ‘by object’ within the meaning of Article 101(1) TFEU.
29In the third place, as regards the occasional instances of coordination, the Commission found that the exchanges of commercially sensitive information enabled the traders participating in the chatroom at issue to identify situations in which they risked interfering with the interests of others, thereby facilitating the possibility for them of occasionally coordinating their actions through a form of ‘standing down’. That conduct reduced the risk that traders would not obtain the desired outcome and avoid trading simultaneously in opposite directions. The Commission found that such conduct constituted an agreement and/or concerted practice, which it characterised as a ‘restriction by object’ within the meaning of Article 101(1) TFEU, on the ground, in essence, that such conduct reduced the normal uncertainties inherent in the relevant market, in that the undertakings implicated in the matters could identify situations where one of them could derive a profit when the others refrained from trading in order not to interfere with its trading strategy, and, accordingly, revealed a sufficient degree of harm to competition.
30In the fourth place, the Commission also found that the aforementioned three restrictions of competition together formed part of a single and continuous infringement on the ground that they pursued a common objective of mitigating the normal uncertainties inherent in the foreign exchange spot trading market in order to reduce the risk and to comfort the undertakings participating in the chatroom at issue (‘the participating undertakings’) in their pricing and risk management decisions, in order not to compete independently. The Commission asserted that the various instances of conduct referred to above formed part of an ‘overall plan’ since they were implemented following the same modus operandi via the chatroom at issue, implying the continuity of participation.
31First, the Commission found that Credit Suisse had shown an intention to contribute to the common objective pursued, which was to restrict competition on the market for foreign exchange spot trading of G10 currencies. Secondly, it found that Credit Suisse was aware of the underlying understanding and the exchanges of information, with the exception of occasional instances of coordination that took place before the relevant period. The Commission next found that Credit Suisse was liable for the exchanges of information forming part of the single and continuous infringement but did not find it liable for (i) the underlying understanding, and (ii) the occasional instances of coordination.
32Thus the Commission found, in Article 1 of the contested decision, that Credit Suisse had infringed Article 101(1) TFEU by participating, during the relevant period, in exchanges of ‘current or forward-looking commercially sensitive information which constitute agreements and/or concerted practices (within a wider single and continuous infringement) having the object of restricting and/or distorting competition regarding foreign exchange spot trading of G10 currencies covering the entire EEA’.
3.The fine
33The Commission stated that it had determined the amount of the fine imposed on Credit Suisse in accordance with the Guidelines on the method of setting fines imposed pursuant to Article 23(2)(a) of Regulation (EC) No 1/2003 (OJ 2006 C 210, p. 2; ‘the Guidelines on the method of setting fines’).
(a)The basic amount of the fine
34In so far as concerns the determination of the basic amount of the fine, given the particularities of the foreign exchange spot trading sector and, in particular, the fact that foreign exchange spot trading in G10 currencies does not generate sales in the usual sense of the term, the Commission calculated the value of the sales of the undertakings that participated in the infringement at issue by means of a ‘proxy for the value of sales’ (‘the proxy’).
35Next, the Commission took into account a gravity factor of 16%, since the infringement was, by its very nature, among the most harmful restrictions, and the duration of that infringement, using a multiplier of 0.42 years.
36Lastly, after adding an additional amount of 16% of the value of sales, for the purposes of deterring participation in such practices, irrespective of the duration of the infringement, the Commission set the basic amount of the fine at EUR 86765000.
(b)The final amount of the fine
37In so far as concerns setting the final amount of the fine, the Commission found that Credit Suisse was liable solely for the exchanges of information forming part of the single and continuous infringement, and deducted 4% from the total basic amount of the fine on account of mitigating circumstances due to the absence of liability, on the part of Credit Suisse, for (i) the underlying understanding, and, (ii) the occasional instances of coordination, in the amount of 2% in respect of each of those circumstances.
38The Commission therefore imposed, in Article 2(a) of the contested decision, a fine of EUR 83294000 on Credit Suisse.
II.Forms of order sought
39The applicants claim that the Court should:– annul the contested decision;– in the alternative, annul Article 1 of the contested decision in part and reduce the fine imposed in Article 2;– in any event, reduce the amount of the fine imposed in Article 2 of the contested decision;– order the Commission to pay the costs or, in the alternative, an appropriate proportion of their costs.
40The Commission contends that the Court should:– dismiss the action in its entirety;– order the applicants to pay the costs.
III.Law
41In support of their action, the applicants put forward five pleas in law. The first three pleas – alleging infringement of Article 101 TFEU and infringement of the obligation to state reasons – and the fifth plea – alleging infringement of the principle of sound administration and the rights of the defence – seek, in essence, to challenge the Commission’s finding in Article 1 of the contested decision that Credit Suisse infringed Article 101 TFEU. The fourth plea – alleging infringement of Article 23 of Regulation No 1/2003, the Guidelines on the method of setting fines, the principles of proportionality and equal treatment and the obligation to state reasons – seeks to contest the fine imposed on Credit Suisse by the Commission in Article 2 of the contested decision.
A.Article 1 of the contested decision
1.The first plea, alleging infringement of Article 101 TFEU and a failure to state sufficient reasons as regards the classification of the online exchanges of information as anticompetitive agreements and/or concerted practices
42By their first plea, the applicants argue, in essence, that the exchanges of information which took place between the traders in the chatroom at issue do not constitute anticompetitive agreements and/or concerted practices, whether they are considered in isolation or together with the underlying understanding. This first plea consists of two parts.
43In the first part of their first plea, the applicants argue that the Commission failed to prove that there was an underlying understanding and therefore that the online exchanges of information constitute an anticompetitive agreement and/or concerted practice.
44In the second part of their first plea, they argue that the Commission failed to adduce the requisite evidence to establish that the online exchanges of information constitute standalone agreements and/or concerted practices.
(a)Preliminary observations
45It should be recalled that the criteria of coordination and cooperation necessary for determining the existence of a concerted practice are to be understood in the light of the notion inherent in the TFEU provisions on competition, according to which each economic operator must determine independently the policy which he or she intends to adopt on the common market (see judgment of 19 March 2015, Dole Food and Dole Fresh Fruit Europe v Commission, C‑286/13 P, EU:C:2015:184, paragraph 119 and the case-law cited).
46While it is correct to say that this requirement of independence does not deprive economic operators of the right to adapt themselves intelligently to the existing or anticipated conduct of their competitors, it does, nonetheless, strictly preclude any direct or indirect contact between such operators by which an undertaking may influence the conduct on the market of its actual or potential competitors or disclose to them its decisions or intentions concerning its own conduct on the market where the object or effect of such contact is to create conditions of competition which do not correspond to the normal conditions of the market in question, regard being had to the nature of the products or services offered, the size and number of the undertakings involved and the volume of that market (see judgment of 19 March 2015, Dole Food and Dole Fresh Fruit Europe v Commission, C‑286/13 P, EU:C:2015:184, paragraph 120 and the case-law cited).
47The Court of Justice has therefore held that the exchange of information between competitors was liable to be incompatible with the competition rules if it reduced or removed the degree of uncertainty as to the operation of the market in question, with the result that competition between undertakings is restricted (see judgment of 19 March 2015, Dole Food and Dole Fresh Fruit Europe v Commission, C‑286/13 P, EU:C:2015:184, paragraph 121 and the case-law cited).
(b)The first part of the first plea, alleging that the Commission failed to prove that there was an underlying understanding and therefore that the online exchanges of information constitute an anticompetitive agreement and/or concerted practice
48The applicants put forward three complaints. By their first complaint, they argue that the contested decision is based on the premiss that the underlying understanding is a prerequisite for the classification of the exchanges of information as anticompetitive agreements and/or concerted practices for the purposes of Article 101(1) TFEU. By a second complaint, they claim that the Commission has failed to prove that information was exchanged in the chatroom at issue pursuant to that understanding. In support of that complaint, they challenge the Commission’s interpretation of the contemporaneous evidence, namely online discussions between traders, and argue that neither those discussions, nor the settlement submissions, nor the concurrent indicia prove the existence of that understanding. By a third complaint, they argue that the Commission cannot exclude the underlying understanding from judicial review on the ground that Credit Suisse was not held liable for that understanding, since this is an essential element of the Commission’s argument against Credit Suisse.
49As regards the first complaint put forward by the applicants, it should be noted that it is apparent from an overall reading of the contested decision and, in particular, recitals 337 to 341, 344, 345 and 347 to 349, read in conjunction with recitals 394 to 405 and 411 to 417, that the Commission carried out separate legal classifications, on the one hand, of the underlying understanding as an agreement restricting competition and, on the other hand, of the exchanges of information as agreements and/or concerted practices which are restrictive of competition. In recitals 487, 489 and 491 to 499 of that decision, the Commission also characterised those two forms of conduct as separate constituent elements of the single and continuous infringement (see paragraphs 26 to 31 above). It should next be recalled that Credit Suisse was held liable only for the infringement comprising the conduct relating to the exchanges of information.
50It should also be noted that, although the exchanges of information involving Credit Suisse took place in an online chatroom, that link with that aspect of the underlying understanding cannot be treated in the same way as the existence of a prerequisite, on which the applicants rely, capable of influencing the separate legal classifications of the underlying understanding and the exchanges of information.
51More specifically, as is apparent from recitals 337 to 341 of the contested decision, the Commission’s legal classification, in that decision, of the exchanges of information as anticompetitive agreements and/or concerted practices is based neither on the rules of the underlying understanding nor on the tacit acceptance of such rules by the traders participating in those exchanges of information (‘the traders in question’). Contrary to what the applicants argue, the contested decision therefore cannot be interpreted as establishing the underlying understanding as a prerequisite for the classification of the exchanges of information.
52The content of recitals 346 and 369 to 371 of the contested decision, cited by the applicants in support of their arguments, cannot invalidate that finding.
53In that connection, first, as regards recital 346 of the contested decision, which relates to the classification of the underlying understanding as an agreement, it should be noted that that understanding admittedly establishes, in essence, that the exchanges of information allowed the traders participating in the chatroom at issue to monitor any deviations from rules. However, it makes no reference to the legal classification relied upon in that decision in so far as concerns the exchanges of information.
54Secondly, it follows from recitals 369 to 371 of the contested decision, the sole purpose of which recitals being to respond to the arguments put forward by Credit Suisse during the administrative procedure, that the Commission based the existence of that understanding on indirect evidence, namely the exchanges of information, in so far as the underlying understanding constitutes a ‘logical premiss’ for those exchanges. Thus, the reference to the ‘logical premiss’ must be understood simply as being used for the same purpose of demonstrating its existence in the present case. Consequently, the fact that the existence of the underlying understanding can be inferred from the exchanges of information must also be understood as seeking to prove its existence, without prejudice to the separate legal classifications of the exchanges of information, on the one hand, and the underlying understanding, on the other hand.
55Moreover, contrary to the applicants’ arguments, the Commission was not required to explain in the contested decision the reasons why Credit Suisse exchanged information in a manner different from that of the other banks involved in the conduct complained of, as compared with the considerations in the settlement decision.
56In that regard, it should be recalled that the settlement procedure and the standard administrative procedure are distinct. Thus, when the decision in respect of Credit Suisse was adopted following the standard administrative procedure, first, the Commission was bound solely by the Statement of Objections and, secondly, it was required, in observance of the principle of audi alteram partem, to take account of all the relevant circumstances, including all the information and arguments that had been put forward by Credit Suisse during the exercise of its right to be heard, with the result that the Commission was required to review the file in the light of those elements (see, to that effect, judgment of 20 May 2015, Timab Industries and CFPR v Commission, T‑456/10, EU:T:2015:296, paragraphs 90, 96 and 107).
57In so far as concerns the adoption of the contested decision in relation to Credit Suisse following the standard administrative procedure, it should be noted that it is apparent from the Statement of Objections, as supplemented by the supplementary Statement of Objections, that the Commission carried out a separate classification of the underlying understanding and the exchanges of information as anticompetitive agreements and/or concerted practices and that Credit Suisse did not raise any objections in that connection during the administrative procedure. Consequently, the Commission was not required to review the file in so far as concerns those legal classifications.
58It follows from the foregoing that, in the light of the separate legal classifications of, on the one hand, the exchanges of information and, on the other hand, the underlying understanding for which Credit Suisse was not held liable (see paragraph 32 above), there is no need to examine the second and third complaints in the first part of the first plea, which are based on the incorrect premiss that the underlying understanding is, in the contested decision, a prerequisite for the classification of the exchanges of information as anticompetitive agreements and/or concerted practices for the purposes of Article 101(1) TFEU.
59Accordingly, the first part of the first plea must be rejected.
60That said, in so far as, in the second part of the first plea, the applicants rely on the claim that the exchanges of information are not anticompetitive and that they are independent of the underlying understanding, the analysis of which requires an examination of the content and interpretation of those exchanges, the fact that the applicants’ objections concerning the scope and interpretation of the contemporaneous evidence are put forward in the context of their arguments based on the underlying understanding, which are advanced in support of the first part of the first plea, cannot prejudice the Court’s analysis of them, in the context of the second part of that plea, in order to determine the merits of the applicants’ objections to the classification of the exchanges of information as standalone agreements or concerted practices.
(c)The second part of the first plea, alleging a lack of evidence establishing that the online exchanges of information constituted standalone agreements and/or concerted practices
61This second part is divided, in essence, into three complaints. By their first complaint, the applicants criticise the Commission’s interpretation of certain items of contemporaneous evidence, namely certain online discussions, on which the Commission relied in finding that the online exchanges of information constituted agreements and/or concerted practices, and argue, in particular, that those discussions were not commercially sensitive. By a second complaint, the applicants maintain that the online exchanges of information cannot be classified as anticompetitive agreements and/or concerted practices, by arguing, in essence, that the alleged commercially sensitive nature of that information and the reduction of uncertainty among competitors which could result from such information are not sufficient for the purposes of such a classification. By a third complaint, they submit that the Commission erred in law in finding that Credit Suisse’s legitimate explanation for the exchanges of information at issue was irrelevant.
62In the contested decision, the Commission presented and interpreted the content of more than 100 discussions. Accordingly, the exchanges of information analysed in that decision were divided into four categories on the basis of the subjects discussed, namely (i) bid-ask spreads (recitals 245 to 251), (ii) customer orders (recitals 183 to 228), (iii) open risk positions (recitals 234 to 238) and (iv) current or planned trading activity (recitals 258 to 285). The Commission found, in essence, that all those exchanges concerned current or forward-looking commercially sensitive information about the trading activities of the traders involved, which enabled the banks concerned to be informed of their competitors’ positions, intentions and constraints, which helped those banks in their subsequent decisions. On the basis of that analysis, the Commission concluded, in recitals 337 to 341 of the contested decision, that those exchanges of information constituted an anticompetitive concerted practice for the purposes of Article 101(1) TFEU.
63The first two complaints should be examined together.
(1)The first complaint, alleging that the Commission misinterpreted certain items of contemporaneous evidence, and the second complaint, alleging that the online exchanges of information do not constitute anticompetitive agreements and/or concerted practices
(i)The admissibility of certain arguments and items of evidence put forward by the applicants
64The Commission disputes the admissibility, first, of the applicants’ line of argument relating to the misinterpretation of certain exchanges of information and of Annex A.8 to the application, on which that line of argument is based, and, second, of the line of argument claiming that the exchanges of information do not constitute an infringement, and of Annex A.9 to the application and Annexes C.2 and C.3 to the reply, on which that line of argument is based.
65In that regard, it must be recalled that, under Article 21 of the Statute of the Court of Justice of the European Union and Article 76(d) of the Rules of Procedure of the General Court, an application must state the subject matter of the proceedings and a summary of the pleas in law on which the application is based. According to settled case-law, that statement must be sufficiently clear and precise as to enable the defendant to prepare its defence and the General Court to rule on the application (judgment of 7 March 2017, United Parcel Service v Commission, T‑194/13, EU:T:2017:144, paragraph 191).
66It should also be recalled that, in particular, it is necessary, for an action before the General Court to be admissible, that the basic matters of law and fact relied on by the applicant be indicated, at least in summary form, coherently and intelligibly in the application itself. Whilst the body of the application may be supported and supplemented on specific points by references to extracts from documents annexed thereto, a general reference to other documents, even those annexed to the application, cannot make up for the absence of the essential arguments in law which, in accordance with the provisions recalled in paragraph 65 above, must appear in the application (see judgment of 7 March 2017, United Parcel Service v Commission, T‑194/13, EU:T:2017:144, paragraph 192 and the case-law cited).
67Thus, the annexes may be taken into consideration only in so far as they support or supplement pleas or arguments expressly set out by the applicant in the body of his or her pleadings and in so far as it is possible to determine precisely what are the matters they contain that support or supplement those pleas or arguments (see judgment of 9 September 2015, Samsung SDI and Others v Commission, T‑84/13, not published, EU:T:2015:611, paragraph 33 and the case-law cited). Furthermore, it is not for the Court to seek and identify in the annexes the pleas and arguments on which it may consider the action to be based, since the annexes have a purely evidential and instrumental function (see judgment of 17 September 2007, Microsoft v Commission, T‑201/04, EU:T:2007:289, paragraph 94 and the case-law cited).
68In the first place, as regards the admissibility, essentially disputed by the Commission, of the applicants’ line of argument relating to the misinterpretation of certain exchanges of information analysed in the contested decision, it should be noted that, as is apparent from the application, in support of their argument that ‘the Commission continues to misunderstand trading jargon and FX spot market functioning’, the applicants refer to Annex A.8 to the application, which is an expert report by a former chair of the Interbank Division of the Global FX Committee, and cite three examples from that report to illustrate their argument.
69At the reply stage, in response to the Commission’s arguments concerning inadmissibility, the applicants reiterate the aforementioned argument and subsequently add that Annex A.8 also demonstrates, first, that ‘the existence of an anticompetitive overall plan or an object to restrict competition cannot be easily or directly inferred from the “documentary evidence”’ and, secondly, that the Commission had inferred, solely on the basis of its interpretations, ‘the existence of the underlying understanding’. Again at the reply stage, they argue that the fact that the Commission was not able to distinguish between past and future orders ‘is a … factor to be taken into account as it casts doubt on the diligence and robustness of the Commission’s investigation’. Lastly, the applicants rely on the relevance and specific nature of their arguments by reference to paragraphs 18, 69 and 78 of the application. Those paragraphs concern the alleged pro-competitive nature of the exchanges of information – as was recognised, it is argued, in the leniency applications (paragraph 18) – the Commission’s misinterpretation or misunderstanding of trading jargon and the functioning of the foreign exchange spot trading market (paragraph 69) and, in essence, the four rules of the underlying understanding (paragraph 78).
70At the hearing, in response to the question put by the Court, the applicants maintained their position that their arguments are set out clearly and specifically in their pleadings.
71In that regard, it should be observed that the applicants put forward a line of argument, as set out in paragraphs 68 and 69 above, which is confined to general and unsubstantiated statements that are often couched in one or two sentences and are even only the subject of a general reference to Annex A.8, without indicating coherently and intelligibly the effects which that line of argument – and still less the expert report on which it is based – might have on the validity of the Commission’s findings set out in the contested decision. In those circumstances, it must be held that those arguments do not satisfy the requirements referred to in paragraphs 65 to 67 above.
72Consequently, the applicants’ arguments referring to Annex A.8 and relating, in essence, to the misinterpretation of certain exchanges of information analysed in the contested decision must be rejected as inadmissible and, in the light of the general reference to that annex, that annex must not be taken into account.
73In the second place, as regards the admissibility of the line of argument that the exchanges of information were not anticompetitive, which is based on a reference to Annex A.9 to the application and Annexes C.2 and C.3 to the reply, it should be held that, contrary to what the Commission maintains, that line of argument must be regarded as sufficiently clear and precise. It is in fact clear from the applicants’ written pleadings before the Court, albeit only summarily, that, by means of the categorisation of the exchanges of information, which they divided into (i) general discussions which cannot be regarded as ‘commercially sensitive’ on the foreign exchange spot trading market, (ii) (more) specific or forward-looking discussions which deal with relatively low trading volumes and which could not have provided relevant information on the expected alleged changes in exchange rates and (iii) discussions which concern larger trading volumes but which could have had a single hypothetical anticompetitive purpose, namely a recipient trader using the discussions to try to make a profit before the trader who shared that information, the applicants seek to dispute the Commission’s finding that the online exchanges of information were anticompetitive in the light of their content, their object or their capacity to ‘distort’ competition. Accordingly, that line of argument put forward by the applicants must therefore be regarded as consistent with the requirements of the case-law referred to in paragraphs 65 and 66 above, and is therefore admissible.
74Moreover, in the light of the case-law cited in paragraph 67 above, it is only in so far as Annexes A.9, C.2 and C.3 support or supplement the arguments expressly set out by the applicants in their pleadings that they will be taken into account.
(ii)The anticompetitive nature of the discussions specifically disputed by the applicants
75The applicants specifically dispute the Commission’s interpretation of the discussions which they classified in the first category (see paragraph 73 above), namely the discussions of 14 June, 5 August and 4 October 2011; 7 February and 1 June 2012; 14 May 2012; 11 April 2012; 13 April 2012; 9 May 2012; 5 June 2012; 25 April 2012; 10 February 2012; and 25 April, 19 June and 3 and 4 July 2012, on the ground that the information exchanged during those discussions is ‘market intelligence’ and not commercially sensitive information, and that it is not anticompetitive in nature.
76In order to determine whether the Commission was right to find that the traders exchanged commercially sensitive information of an anticompetitive nature, it is necessary to assess the nature of those exchanges. More specifically, it is necessary to analyse whether those exchanges related to non-public information which the traders could not already have known, where appropriate aggregated and anonymised, and to examine, in particular, the accuracy and purpose of the exchange of such information.
–The discussions of 14 June, 5 August and 4 October 2011 and 7 February and 1 June 2012
77The discussions of 14 June, 5 August and 4 October 2011 and 7 February and 1 June 2012 set out in recitals 107, 111, 122, 139, 150 and 504 of the contested decision, the Commission’s interpretation of which is disputed by the applicants, were not considered by the Commission to be of a commercially sensitive nature.
78As regards, in the first place, the discussions of 14 June, 5 August and 4 October 2011 and 7 February and 1 June 2012 set out in recitals 107, 111, 122, 139 and 150 of the contested decision, it should be noted that the Commission found that those discussions demonstrated, in essence, that access to the chatroom at issue was based on mutual trust between the participating traders and that those traders were acting in a closed circle intended to favour their own interests. According to the Commission, the exchanges set out in recitals 139 and 150 of that decision also demonstrated that the Credit Suisse trader who participated in the chatroom at issue before being employed by Credit Suisse had rejoined that chatroom as an employee with full knowledge of how it worked and of what his participation entailed.
79As regards, in the second place, the discussion of 5 August 2011 set out in recital 504 of the contested decision, the Commission considered that this demonstrated that the Credit Suisse trader, without actively participating in the exchanges, was aware of them.
80Accordingly, the applicants’ argument alleging incorrect classification of the discussions referred to in paragraph 77 above, regarded as commercially sensitive, has no basis in fact and must therefore be rejected.
–The discussion of 14 May 2012
81As regards the discussion of 14 May 2012 reproduced in recital 209 of the contested decision, read in conjunction with recitals 196 to 202 and 337 to 341, the Commission found that, during that discussion, the traders shared information concerning their positions, outside the framework of potential transactions between them, and concerning speculations on the price of a GBP/USD currency pair. It found that such information, which concerned customer orders, was commercially sensitive and capable of contributing to the anticompetitive nature of the entirety of the exchanges of information which took place in the chatroom at issue.
82The applicants dispute the Commission’s interpretation of that discussion, arguing that it concerns general information, contains no information relating to the prospective orders or positions of the traders, and thus amounts to the disclosure of ‘market intelligence’. They add that the references to ‘rhs light’ and ‘LHS samlls cable’ do not relate to amounts, pricing or customers and are therefore anonymised. They conclude that that discussion fails to establish that the information exchanged in the chatroom at issue is anticompetitive.
83The Commission disputes the applicants’ interpretation of the discussion in question.
84It should be noted that, as is apparent from the excerpt from the discussion in question, the Credit Suisse trader was present in the chatroom at issue on that day and that, at 14:50:48, he told the other participants that he had orders to sell a small quantity of an unspecified currency pair to be executed at the fixing (‘rhs lights’). At 14:52:57, a trader asked him whether it was a GBP/USD currency pair (‘cable??’) and, at 14:55:51, shared information relating to customer orders he held, namely orders to sell a small quantity of a GBP/USD currency pair to be executed at the ‘fixing’ (‘LHS samlls cable’). During the same conversation, another trader expressed his view on the increase in the value of the GBP/USD currency pair (‘feels like its just being put higher wsith soft eur/gbp and eur/usd running into some bids …’), stating that he had no orders for that currency pair.
85As is apparent from the transcript of the discussion in question, the traders exchanged information that was not public and related to specific orders that they had, and specified the order type (sell order to be executed at the fixing), the relevant currency pair (GBP/USD) and their quantity. That discussion therefore reflects running commentary on the traders’ activities which gave rise to an exchange of commercially sensitive information.
86Even if the information exchanged is, as the applicants maintain, anonymised, that circumstance does not prevent it from being classified as commercially sensitive information in so far as that information, by virtue of its subject matter and its level of detail and the fact that it is not accessible to competitors not present in the chatroom at issue, confers a commercial advantage on the recipients of that information.
87Such information is capable of informing specialist traders about market movements relating to the GBP/USD currency pair prior to the fixing, in view of the fact that the smaller sell orders to be executed at the fixing are less able to have a significant influence on exchange rate movements. The communication of such information thus enables traders to adapt their trading strategies accordingly and to mitigate the uncertainties inherent in the foreign exchange spot trading market.
88Accordingly, the Commission was right to find that the discussion of 14 May 2012 concerned commercially sensitive information capable of contributing to the anticompetitive nature of the exchanges of information which took place in the chatroom at issue.
–The discussion of 11 April 2012
89As regards the discussion of 11 April 2012 reproduced in recital 223 of the contested decision, read in conjunction with recitals 214 to 221 and 337 to 341, the Commission found that it related to a specific customer’s immediate order and that the disclosure of that commercially sensitive information removed some of the uncertainties inherent in the foreign exchange spot trading market, thereby allowing traders participating in the chatroom at issue better to anticipate market movements and to adapt their strategies in order to take advantage of the disclosed information and of the resulting increase in market transparency and, accordingly, was capable of contributing to the anticompetitive nature of the exchanges of information at issue.
90The applicants argue that the discussion of 11 April 2012 concerns general information which does not include prospective orders or traders’ positions. They add that the comment ‘lost some X’ contained in that discussion indicates no amounts or prices and that several customers were doing the rounds on the market, which did not allow any specific customer information to be inferred from it. They conclude that that discussion concerns market intelligence which was not anticompetitive in scope.
91The Commission disputes the applicants’ proposed interpretation of the discussion in question.
92In that connection, it should be noted that the excerpt from the discussion of 11 April 2012, set out in recital 223 of the contested decision, confirms that, at 06:50:23, a UBS trader informed the traders participating in the chatroom at issue that he had just concluded a transaction by selling euros against British pounds (‘lost some X’, that is to say a EUR/GBP currency pair), specifying at 06:50:23 that the transaction involved a counterparty who usually went from one trader to the other (‘usual geezer who does the rounds’). A number of participants in the chatroom, including the Credit Suisse trader, said thank you for that information (‘ta’).
93That discussion must be regarded as revealing an exchange of precise, current and confidential information, in that it disclosed to the other traders participating in the chatroom at issue the type of order concerned (immediate), the time it was carried out (the present), the relevant currency pair (EUR/GBP) and a reference to the type of customer for whom the transaction had just been concluded.
94In so far as concerns the latter reference to a type of customer, the applicants’ argument that this did not make it possible to infer specific customer information cannot be accepted. The UBS trader’s reference to a counterparty who ‘does the rounds’ must be understood, in the context of that exchange, as sufficiently specific for the purposes of his identification by those traders, who are professionals in the field, thus enabling them to take into account the nature of the customer concerned, in particular whether or not he was an informed customer. Such a conclusion is appropriate having regard, first, to the reference to the fact that the counterparty was a regular one (‘usual geezer’) and, secondly, to the fact that none of the traders who thanked the USB trader for that information asked for details concerning the reference to that customer. However, as the Commission found in recital 40(b) of the contested decision, read in conjunction with footnote 381, without that being contested by the applicants, making information available to other traders on whether or not the customer concerned is an informed customer must be regarded as capable of allowing the recipients of that information to anticipate possible market movements.
95In those circumstances, the information shared during the discussion in question cannot be regarded as general solely on the ground that it did not specify amounts traded or exchange rate levels, or relate to prospective orders. In that regard, it is sufficient to note that ‘market intelligence’ consists, in essence, of general and anonymised observations on the state of the market, from which information such as that communicated during the discussion in question manifestly deviated.
96It must therefore be held that the information exchanged in question was commercially sensitive, current and specific, and capable of removing some of the uncertainties inherent in the foreign exchange spot trading market. It therefore allowed traders participating in the chatroom at issue better to anticipate market movements for the EUR/GBP currency pair and to adapt their strategies in order to take advantage of the information disclosed and was therefore capable of contributing to the anticompetitive nature of the exchanges of information at issue.
97Accordingly, the Commission was right to find that the discussion of 11 April 2012 concerned commercially sensitive information capable of contributing to the anticompetitive nature of the exchanges of information which took place in the chatroom at issue.
–The discussion of 13 April 2012
98In so far as concerns the discussion of 13 April 2012 referred to in recital 224 of the contested decision, read in conjunction with recitals 214 to 221 and 337 to 341, the Commission found that it concerned current or forward-looking commercially sensitive information related to customers’ immediate orders, in communicating in the chatroom at issue the volume or direction of a specific, non-aggregated order or the type of customer, thereby removing some of the uncertainties inherent in trading on the foreign exchange spot trading market and increasing the level of transparency of changes in the exchange rate for the relevant currency pair. The Commission concluded that that discussion was such as to contribute to the anticompetitive nature of the exchanges of information at issue.
99The applicants argue that the exchange of information which took place during the discussion on 13 April 2012 concerns ‘market intelligence’, since it does not relate to the prospective orders or positions of the traders. Thus, the expression ‘lose more cable’ used by the Credit Suisse trader is anonymised and concerns approximate amounts, without any price information or customer information. According to the applicants, the term ‘corp’ is a generic term used to designate an undertaking and therefore does not make it possible to identify a customer.
100The Commission disputes the applicants’ interpretation of the discussion in question.
101In that regard, it should be noted that the excerpt from the discussion of 13 April 2012, reproduced in recital 224 of the contested decision, confirms that, at 09:56:42, the Credit Suisse trader communicated information to another trader present in the chatroom at issue relating to his current transaction to sell 40 million in a GBP/USD currency pair to a large undertaking (‘lose more cable 40 odd corp’), for which that other trader thanked him (‘tks ratty’).
102Thus, the discussion in question clearly reveals an exchange of precise and current information that was not in the public domain, which was known only to the parties to the transaction, namely the trader and his customer. The information on the transaction communicated by the Credit Suisse trader included the specific amount (40 million), the relevant currency pair (GBP/USD), the direction of the transaction (sale) and a reference to a type of customer (large undertaking).
103In that regard, the applicants’ argument that the term ‘corp’, used by the Credit Suisse trader during the discussion in question, does not make it possible to infer the specific identity of the customer concerned cannot succeed. The trader nevertheless indicated the type of customer, that is to say, in particular, that it was a large undertaking. It should be noted that order flows of some large undertakings and international undertakings may be regarded as informative, since they enable traders in possession of that information to anticipate exchange rate movements for a relevant currency pair, thereby removing some of the uncertainties inherent in the foreign exchange spot trading market and increasing transparency on that market for the benefit only of recipients of that information.
104Moreover, the usefulness of information relating to a type of customer, in this instance ‘corp’, is supported by other exchanges referred to in the contested decision (see, in particular, recitals 278 to 280 of that decision), from which it is apparent that such information was appreciated by the traders participating in the chatroom at issue.
105In those circumstances, it must be held that the information shared by the traders during the discussion on 13 April 2012 cannot be regarded as constituting ‘market intelligence’ solely because it does not relate to prospective orders.
106Accordingly, the Commission was right to find that the discussion of 13 April 2012 concerned commercially sensitive information such as to contribute to the anticompetitive nature of the exchanges of information which took place in the chatroom at issue.
–The discussion of 9 May 2012
107As regards the discussion of 9 May 2012 referred to in recital 276 of the contested decision, read in conjunction with recitals 252 to 257 and 337 to 341, the Commission found that, during that discussion, the Barclays trader asked the Credit Suisse trader about the completion of his trading strategy, to which the Credit Suisse trader replied in the affirmative, providing additional details regarding the levels traded. In the Commission’s view, under normal market conditions, the Barclays trader would have had no reason or means to be in possession of the information that the Credit Suisse trader had recently ‘cleared his bids’. The exchange of that commercially sensitive information conferred a competitive advantage on the Barclays trader, inasmuch as it could have had an influence on his subsequent pricing behaviour, enabling him to adapt his bids to a less competitive level, in the knowledge that the Credit Suisse trader had been able to fill all his bids.
108The applicants argue that the discussion of 9 May 2012 consists of general statements, in view of the fact that the references to orders do not include amounts, prices or information about the customers concerned and are therefore anonymised. Nor does that discussion relate to traders’ prospective orders or their positions.
109Moreover, in response to a question put by the Court by way of a measure of organisation of procedure, the applicants submitted that the purpose of the discussion in question may have been to explore the possibility of matching trades between the traders participating in that discussion but stated that they were not in a position definitively to confirm that interpretation. In the alternative, they reiterated their position that the purpose of the discussion in question was to share ‘market intelligence’ in order to facilitate those traders’ understanding of the market. In any event, the information exchanged, even assuming that it was commercially sensitive, was general, anonymous, historic and non-specific and would not allow the traders to engage in anticompetitive conduct.
110The Commission takes the view that the information shared during the discussion in question is commercially sensitive information which goes beyond the scope of ‘market intelligence’, in so far as it relates, in essence, to specific and current levels of trading.
111In response to a question put by the Court by way of a measure of organisation of procedure, the Commission reiterated its position that the discussion in question concerned the exchange of non-public and specific information, enabling the recipient of that information to adapt the price he was offering to his own clients, even trying to increase it, in the knowledge that it was unlikely that his competitor, having completed his trading strategy, would continue trading the currency pair in question. The Commission added that the Credit Suisse trader’s indication of a fall in the exchange rate would not have made sense if the Barclays trader did not know in which currency pair the Credit Suisse trader’s bids had just been executed.
112In that regard, it should be noted that, during the discussion in question, the Credit Suisse trader, in stating the trading levels, indeed communicated information relating to his trading which is, by its very nature, of a commercially sensitive nature (‘yeh all filled’; ‘feels like its dropped 200 pips its only 40’). However, contrary to the Commission’s argument, the foregoing excerpt from the discussion does not suggest that the exchange of that commercially sensitive information conferred a competitive advantage on the Barclays trader, in so far as it could have had an influence on his subsequent pricing behaviour by enabling him to adapt his bids.
113First, the Commission’s assertion that the information relating to exchange rate levels ‘would … not make sense if the two [traders] had been discussing whether their bids were filled in general, without referring to a specific currency pair’ is, in the light of the content of that discussion, speculative in nature. Indeed, there is nothing in the discussion, or in any other discussion cited in the contested decision, to suggest that the traders were discussing the filling of bids at the levels referred to by the Credit Suisse trader for a specific currency pair.
114Secondly, there is nothing in the discussion in question to support the Commission’s assertion that the trader who received the information shared by the Credit Suisse trader could have known that it was unlikely that his competitor would keep on trading the currency pair in question. That assertion is therefore also speculative in nature.
115Such speculations do not satisfy the evidential burden on the Commission to demonstrate to the requisite legal standard the existence of the circumstances constituting an infringement (see judgment of 22 November 2012, E.ON Energie v Commission, C‑89/11 P, EU:C:2012:738, paragraph 71 and the case-law cited).
116Accordingly, in the absence of more specific information, it cannot be held that that discussion reduced or removed uncertainty on the market in such a way as to confer a competitive advantage on the recipient of the information.
117In those circumstances, it must be held that the Commission has failed to demonstrate to the requisite legal standard that the discussion of 9 May 2012 was capable of contributing to the anticompetitive nature of Credit Suisse’s conduct. The applicants therefore rightly maintain that the Commission could not rely on that discussion against Credit Suisse.
–The discussion of 5 June 2012
118In so far as concerns the discussion of 5 June 2012 set out in recital 281 of the contested decision, read in conjunction with recitals 252 to 257 and 337 to 341, the Commission found that that discussion concerned commercially sensitive information, given that a UBS trader exchanged information relating to a recently executed order, providing specific details on the currency pair, the exchange rate and the type of customer which, without that discussion, the Credit Suisse trader would have had no reason to know or means of knowing. However, knowledge of such information placed the Credit Suisse trader at a competitive advantage, if the same type of customer requested a trade in the same currency pair, knowing that the customer had agreed to trade at the exchange rate previously proposed by the UBS trader. Thus the exchange of information during that discussion was such as to contribute to the anticompetitive nature of the exchanges of information at issue.
119The applicants submit, in essence, that the information exchanged during that discussion concerns ‘market intelligence’ in view of the fact that the references to customer orders were anonymised and indicated no amounts or customer information, but only approximate exchange rate levels. They add that it is not clear that the reference to ‘rooskies’ concerns a specific customer.
120The Commission disputes the applicants’ interpretation of the discussion in question.
121In that regard, it should be noted that the excerpt from the discussion in question confirms that, at 08:38:40, a UBS trader asked the Credit Suisse trader whether he had received any sell orders (‘are you seeing much eurusd selling Ratty’), stating, at 08:44:28, that a type of customer had just bought British pounds against US dollars at a specific exchange rate (‘rooskies buying cable at 2530’). In response, at 09:06:14, the Credit Suisse trader stated that he had not received any such orders (‘sry mate no’), adding, at 09:06:24, that he had observed a decrease in the level of the exchange rate for that currency pair (‘nothing just seen it melt 100 pips’). At 09:06:44, the UBS trader thanked him for that information (‘tks bro’).
122As regards the interpretation of the content of the discussion in question, it must be stated that, as is apparent from paragraph 121 above, the information exchanged related to precise, current and confidential information, known only to the parties to the transaction, namely the trader and his customer, inasmuch as it concerned the type of orders, their quantity, the relevant currency pair, the exchange rate level for the relevant currency pair and a type of customer involved in the transaction.
123None of the applicants’ arguments can invalidate the foregoing finding.
124First, as regards the reference to ‘rooskies’, which did not clearly relate to a specific client, it must be held that that reference, made during the discussion in question, is to be understood as identifying a specific type of customer, having regard, first, to the thanks expressed by the UBS trader for that information and, second, to the fact that it did not give rise to any further questions on the part of the other traders. Moreover, that conclusion is supported by the overall context of the exchanges referred to in the contested decision, from which it is apparent that a reference to such customers (‘rooskies’ or ‘roosky’) (see recitals 263 and 272 of the contested decision) was always appreciated by other traders, thus giving rise to the presumption that, in their minds, that reference was easily identifiable as indicating a specific type of relevant customer, namely whether informed or not, such as to allow traders, who are professionals in the field, to anticipate possible market movements and be comforted as to their strategic decisions.
125Secondly, with regard to the exchange rate levels which, according to the applicants, were referred to in an approximate way in the discussion in question, it should be observed that the reading of that discussion, and in particular the part relating to the GBP/USD currency pair, set out in paragraph 121 above, clearly contradicts the interpretation suggested by the applicants, since the figure mentioned is very precise.
126In those circumstances, the information exchanged during the discussion in question cannot be regarded as general or as relating to ‘market intelligence’, solely because it did not concern prospective orders. Accordingly, the Commission was right to find that the discussion of 5 June 2012 concerned commercially sensitive information such as to contribute to the anticompetitive nature of the exchanges of information which took place in the chatroom at issue.
–The discussion of 25 April 2012
127As regards the discussion of 25 April 2012 referred to in recital 195 of the contested decision, read in conjunction with recitals 176 to 182 and 337 to 341, the Commission found that, during that discussion, the Credit Suisse trader did not provide any information. However, he was present in the chatroom at issue when that discussion on commercially sensitive information took place or had access to that discussion during a subsequent connection. The commercially sensitive information on conditional orders exchanged that day in the chatroom at issue was likely to increase the Credit Suisse trader’s level of confidence in a potential change in the market price for a GBP/AUD currency pair, thereby enabling him to make better-informed decisions concerning his commercial strategy and thus take advantage of that level of privileged – and, therefore, anticompetitive – information.
128The applicants argue that the discussion in question concerns ‘market intelligence’, in view of the fact that it contains no information relating to the traders’ prospective orders or their positions. Most of the information shared contains speculation about future movements. They add, on the one hand, that ‘hypo’ refers not to a customer, but to a counterparty encountered on electronic platforms who deals with several counterparties on the market and, on the other hand, that ‘russia selling eur’ may refer to a publicly known activity of the Central Bank of the Russian Federation.
129The applicants also claim in their application and then in their reply, in essence, that the Commission failed to provide sufficient reasons for its finding that that discussion and others that are analysed in paragraphs 137 to 162 below enable it to prove its allegations that those discussions are commercially sensitive and anticompetitive in nature, and that it is neither for the applicants nor for the Court to identify the section of an excerpt from a discussion on which the Commission relies in the absence of precise time stamping.
130The Commission disputes the applicants’ interpretation of the discussion in question.
131In that connection, in so far as the claim of the applicants summarised in paragraph 129 above may be understood as criticising the Commission for failing to state sufficient reasons in the contested decision, it should be noted that recital 195 of that decision – which is, admittedly, set out succinctly – shows clearly and unequivocally the Commission’s reasoning in so far as concerns the commercially sensitive and anticompetitive nature of the discussion of 25 April 2012, from which an excerpt is reproduced in recital 191 of the contested decision. It is apparent from recital 195 of that decision that that discussion was relied upon against Credit Suisse on the ground that that discussion enabled its trader to increase his level of confidence as to the potential change in the market price for the GBP/AUD and GBP/NZD currency pairs, thereby enabling him to make better-informed decisions concerning his trading strategy.
132Furthermore, it should be observed that a reading of the transcript of another excerpt from the discussion of 25 April 2012 submitted by the applicants before the Court confirms that the traders exchanged information relating to their current ‘pending buy orders’, specifying the currency pairs concerned with the trigger points for their execution, as is apparent from a UBS trader’s question (at 08:42:19: ‘what’s the trigger point for all those gbp/aud and gbp/nzd longs to puke’) and the responses of the HSBC and RBS traders (respectively, at 08:42:36: ‘looking at 1.6040 as a bit of a lvi’ and at 08:42:56: ‘1.5500 gpaud …’).
133The information shared during the discussion in question cannot, contrary to what the applicants claim, be regarded as ‘market intelligence’, in so far as such information manifestly goes beyond observations on the general state of the market.
134The discussion in question clearly reveals an exchange of very precise, current and confidential information, inasmuch as the information shared concerns the types of orders of the customers, their direction, their trigger levels and the currency pairs concerned, which are known only to the parties to the transaction, namely the trader and his customer. Thus, that discussion reflects running commentary on the traders’ activities which gave rise to an exchange of commercially sensitive information, capable of increasing transparency between the traders and of reducing uncertainty on the market to their sole benefit.
135That finding is not invalidated by the applicants’ mere claim that some of the information provided during the discussion in question concerns speculation on the trend in future movements. Aside from the general nature of that argument, the fact that some of the information shared is not regarded as commercially sensitive does not mean that that discussion as a whole is of a general nature. It should also be noted that not knowing to what the references ‘hypo’ or ‘russia selling eur’ specifically relate does not mean that that exchange is of a general nature either, since it is sufficient that such a reference makes it possible to specify whether or not a particular customer is informed (see paragraph 94 above), which the applicants do not dispute.
136Accordingly, the Commission was right to find that the discussion of 25 April 2012 concerned commercially sensitive information such as to contribute to the anticompetitive nature of the exchanges of information which took place in the chatroom at issue.
–The discussion of 10 February 2012
137As regards the discussion of 10 February 2012 referred to in recital 228 of the contested decision, read in conjunction with recitals 214 to 221 and 337 to 341, the Commission found that, during that discussion, the Credit Suisse trader did not provide any information. However, he was present in the chatroom at issue when that discussion took place or had access to that discussion subsequently. In the Commission’s view, the commercially sensitive information on immediate orders exchanged that day in the chatroom at issue was likely to increase the Credit Suisse trader’s level of confidence in a potential change in the market price, thereby enabling him to make better-informed decisions regarding his commercial strategy.
138The applicants argue that the exchange of information in question concerns ‘market intelligence’, in view of the fact that it contains no information relating to the prospective orders or positions of the traders. According to the applicants, during that exchange, traders speculated about the future trading of a specific counterparty (‘JP’), without having any certainty. They add, in particular, that ‘corp’ is a generic and anonymised term for a corporation, but that it is not possible to conclude from this that it refers to a specific client.
139The Commission disputes the interpretation of the content of the exchange in question as put forward by the applicants.
140In that connection, as a preliminary point, it is noted that recital 228 of the contested decision does not contain a transcript of the discussion in question. In that recital, the Commission stated that it concerned an exchange of information relating to a counterparty who sought to obtain a specific trade and tried to enter into a sell order (‘exchange of information regarding JPMorgan (“JP”) getting a specific trade and still trying to sell’). That discussion was held against Credit Suisse on the ground that that discussion enabled its trader to increase his level of confidence as to the potential change in the market price, thereby enabling him to make better-informed decisions on his trading strategy.
141That said, it should be noted that a reading of the excerpt from the exchange of 10 February 2012, produced by the applicants, confirms that, at 10:12:14, a UBS trader stated that a specific counterparty had just obtained a trade (‘JP had the trade’). At 10:12:46, an HSBC trader thanked him for that information and requested further information in order to ascertain whether it was worth selling a GBP/USD currency pair (‘thx’, ‘cable worth a sell?’), to which, between 10:13:35 and 10:14:08, a UBS trader replied in the affirmative, specifying also that that specific counterparty had an immediate order for a EUR/GBP currency pair and its exchange rate level (‘yep’, ‘eurgbp held here at 57-60 pops draghi’, ‘but i thik that plmub in JP is still rying to sell X’). Finally, at 10:15:32, a Barclays trader informed the other traders in the chatroom that he had just sold EUR 100 million to a large undertaking (‘sold ton eur for a corp’).
142Accordingly, the discussion which took place on 10 February 2012 reveals an exchange of precise, current and confidential information, in that it disclosed to the competitors present in the chatroom at issue the customer orders which the other participants had or had just filled, stating the currency pairs concerned, their exchange rate levels and a specific counterparty to those orders. That discussion, which therefore reflects running commentary of the traders’ activities and positions, gave rise to an exchange of commercially sensitive information.
143Moreover, as is clear from the excerpt from the discussion quoted above, the traders exchanged views on the advisability of selling British pounds against US dollars. Such an exchange of information, which is intrinsically linked to the determination of trading strategy, is by its very nature contrary to the independence which an economic operator must, according to the case-law cited in paragraph 45 above, demonstrate on the market when determining the policy which it intends to follow on that market.
144In that regard, the applicants’ argument that the reference to a ‘corp’, which is allegedly anonymised and generic, is not specific to a customer cannot be accepted. Irrespective of whether that reference designates a specific client, it is sufficient to recall, as has been stated in paragraph 94 above, that a type of customer, namely whether informed or not, such as in this instance a large undertaking, is sufficient to allow traders, who are professionals in the field, to anticipate possible market movements and be comforted as to their strategic decisions.
145Accordingly, the Commission was right to find that the discussion of 10 February 2012 concerned commercially sensitive information capable of contributing to the anticompetitive nature of the exchanges of information which took place in the chatroom at issue.
–The discussions of 25 April, 19 June and 3 and 4 July 2012
146In so far as concerns the exchanges of 25 April, 19 June and 3 and 4 July 2012 referred to in recital 285 of the contested decision, read in conjunction with recitals 337 to 341, the Commission found that the Credit Suisse trader, who did not provide any information during those exchanges, was present in the chatroom at issue and therefore had access to those exchanges. It found, in essence, that the cumulative exchanges of commercially sensitive information increased the trader’s level of confidence as to the potential direction of the market price of the currency pairs discussed, which thereby enabled him to make better-informed decisions as to his trading strategy. In particular, the traders shared information on trading activities relating to currencies, in particular EUR/JPY and USD/JPY (exchange of 25 April 2012), GBP/USD (exchange of 19 June 2012), EUR, USD/JPY and GBP/USD (exchange of 3 July 2012) and EUR and GBP/USD (exchange of 4 July 2012).
147The applicants argue that the aforementioned exchanges concern general market information. That information is anonymised and does not relate to prospective orders or to the positions of traders and clients. In so far as concerns the exchanges of 19 June and 3 and 4 July 2012, they do not relate to volumes traded or prices. The applicants add that the terms used in those exchanges, such as ‘lev’ (exchanges of 25 April and 19 June 2012) or ‘models’ (exchange of 4 July 2012), are also anonymised and generic. The applicants submit that it is not clear whether the terms ‘m/e’, ‘rm’, ‘rusky’, ‘Russia’, ‘roosides’ (exchange of 19 June 2012), ‘rusky’ (exchange of 3 July 2012), ‘models’, ‘range guy’ and ‘desert’ (exchange of 4 July 2012) refer to specific customers. Lastly, the expression ‘51.3 is the nod’ (exchange of 4 July 2012) is a reference to economic data rather than a market price.
148The Commission disputes the interpretation of the exchanges in question as put forward by the applicants.
149First, as regards the discussion of 25 April 2012, it should be noted that it is apparent from the excerpt from that discussion, produced by the applicants, that the traders exchanged precise and current information concerning their recent trading activities and, in particular, the amounts traded and the relevant exchange rate levels (an RBS trader at 09:56:44: ‘i lost a coupel of ton around 40 earlier’; and a Barclays trader at 10:05:44: ‘lost ton’), customers and the relevant currency pair (a UBS trader at 10:03:25: ‘lev buys usdjpy – gt track record’; and an HSBC trader at 10:04:54: ‘anyone get paid there in euro’). Moreover, a number of traders discussed strategies to adopt on the market, suggesting in particular not to buy one currency against another (the RBS trader at 09:57:11: ‘but def not right time to be long yet’; the Barclays trader at 09:57:24 to 09:57:26: ‘nope’ and ‘agree’) or to buy a particular currency (a Barclays trader at 10:13:50: ‘get eur now’).
150Such information is, contrary to what the applicants argue, commercially sensitive and likely to increase transparency on the market to the sole benefit of the traders participating in the chatroom at issue and to comfort those traders as to their strategic decisions. The exchange of that information is thus, by its very nature, contrary to the requirement of independence which an economic operator must, according to the case-law cited in paragraph 45 above, demonstrate on the market when determining the policy which it intends to follow on that market.
151In that regard, the applicants’ argument that the term ‘lev’ does not suggest the specific identity of a customer cannot succeed. Even if that were the case, that term indicates the customer type, whether informed or not, that is to say, for example, that it is an alternative investment fund (a fund using leverage), thus enabling traders with that information to anticipate exchange rate movements for a relevant currency pair (see paragraph 94 above). Moreover, the usefulness of information relating to a type of customer, in this instance ‘lev’, is supported by other exchanges referred to in recitals 264 and 268 of the contested decision, from which it is apparent that such information was appreciated by the traders participating in the chatroom at issue.
152Accordingly, the Commission’s conclusion concerning the anticompetitive scope of the information exchanged between the traders on 25 April 2012 must be upheld.
153Secondly, in so far as concerns the discussion of 19 June 2012, it must be noted that, as is apparent from the excerpt from that discussion produced by the applicants, the traders exchanged information concerning their trading activities, including, in particular, information on the type of customers and the exchange rates for specific currency pairs (a UBS trader at 06:42:12: ‘m/e sniffing in eurusd’, and at 06:42:20: ‘passes here on teh 15th price’; another UBS trader at 07:14:24: ‘got cable fromrm high 80s, lively last time he dealt’; an HSBC trader at 07:15:02: ‘Russia gave me a fex clips off highs too’; and a Barclays trader at 08:09:53: ‘lev guys selling eur’).
154Thus, the information communicated during the discussion in question relates to current and prospective transactions and specifically states the currency pairs and types of customer concerned by those transactions. Such information, known in principle only to the trader and his customer, is of a commercially sensitive nature.
155Contrary to what the applicants claim, the fact that the references to customers which emerge from the discussion in question do not make it possible to designate specific customers is not such as to call into question the commercially sensitive nature of that discussion. A type of customer is sufficiently relevant for the purpose of determining whether the transaction entered into with that customer is informative as to market movements, thereby conferring an advantage on traders who have that information compared to their competitors who do not participate in the chatroom at issue. Moreover, the relevance of communicating the type of customer in question is supported by the thanks from the Credit Suisse trader after he received that information (at 06:42:25: ‘ta’).
156Consequently, it must be held that, in view of its commercially sensitive nature, accuracy and timeliness, the information disclosed during the discussion on 19 June 2012 cannot be regarded as mere general market information.
157Thirdly, it is clear from the excerpt from the discussion of 3 July 2012 that traders shared information with each other on customer order flows in so far as concerns currency pairs and the type of customers concerned (‘rusky sells euros’, ‘seeing few usdjpy sellers’, ‘rusky sold cable’). It is true that the information exchanged does not indicate the amounts traded or specific exchange rates. However, that fact alone does not make the rest of the information exchanged general in nature. The disclosure of information on orders – in this instance buy and sell orders, a relevant currency pair and the type of customer – must be regarded as commercially sensitive. Moreover, the applicants put forward no explanation capable of casting doubt on the usefulness of sharing such confidential information in the chatroom, as an alternative to the explanation, on which the Commission relied in the contested decision, relating to the advantage which traders receiving that information could have gained in terms of their trading strategies.
158In those circumstances, the Commission’s finding in relation to the anticompetitive nature of the discussion which took place between the traders on 3 July 2012 must therefore be upheld.
159Fourthly, it is apparent from the excerpt from the exchange of 4 July 2012 that, at 08:46:30, a Barclays trader communicated to the other participants in the chatroom at issue information concerning a type of customer buying euros (‘range guy buys eur’) and that, at 08:46:44, a UBS trader, for his part, provided information relating to a purchase of GBP/USD by another type of customer (‘desert buys cable’). In the following minutes, the Credit Suisse trader thanked them for that information (‘ta’). At 10:12:00, the Barclays trader asked the other traders in the chatroom at issue if they were aware of the euro sales (‘anyone seeing this eur selling’). At 10:25:39, the same trader stated that he was aware of a type of customer who was again selling euros. At 13:08:19 to 13:08:21, a UBS trader provided information about a type of customer who was selling British pounds against US dollars (‘models in selling on the toy here’, ‘betty’), to which the Barclays trader replied by confirming that he had made the same observation (‘same’).
160In that regard, it should be observed that, admittedly, the information shared does not concern the volumes or exchange rates applicable to those trades. However, the fact remains that information of that confidential nature cannot, contrary to what the applicants claim, be regarded as general and anonymised. It informs traders participating in the chatroom at issue of the trading of specific currencies by those traders according to customer types which are similarly specific in terms of whether they are informed or not and which are, in principle, known only to the traders involved in such transactions.
161Moreover, the question put by the Barclays trader demonstrates the usefulness of sharing such information between traders participating in the chatroom at issue in order to give them an advantage over other market participants who did not have access to that information. In those circumstances, as the Commission found, making that information available to traders could therefore allow them, as professionals in the field, to increase their level of confidence as to the potential direction of the market price of the currency pairs discussed, thereby enabling them to make better-informed decisions on their trading strategies.
162Accordingly, the Commission was right to find that the discussions of 25 April, 19 June and 3 and 4 July 2012 concerned commercially sensitive information such as to contribute to the anticompetitive nature of the exchanges of information which took place in the chatroom at issue.
–Conclusions on the discussions specifically disputed by the applicants
163It follows from all of the foregoing that the Court must reject the applicants’ arguments that the exchanges of information were not anticompetitive and that the contemporaneous evidence is insufficient to prove that they were indeed anticompetitive, in so far as those arguments are directed against the specific discussions falling within the first category referred to in paragraph 73 above, with the exception of the arguments concerning the exchange of information of 9 May 2012, in so far as that exchange, contrary to the Commission’s finding in the contested decision, cannot substantiate the anticompetitive nature of the exchanges of information alleged against Credit Suisse. As regards the latter exchange, it will be necessary to determine below whether that error on the part of the Commission is such as to have an impact on the merits of the findings that Credit Suisse committed an infringement and the validity of the fine imposed as a result.
(iii)The anticompetitive nature of the discussions not specifically disputed
164The applicants dispute generally the anticompetitive nature of the discussions which they classified in the second and third categories referred to in paragraph 73 above. First, they argue that the discussions analysed in recitals 126 to 129, 142, 147, 183, 185 to 195, 208, 210, 225 to 228, 234 to 238, 245 to 250, 258 to 275, 277 to 280, 282 to 285, 291 to 295 and 473 of the contested decision did not provide information on expected changes in exchange rates in the light of the relatively small volumes involved. Second, they claim that the discussions analysed in recitals 113 to 121, 184, 195, 204 to 207, 211 to 213 and 228 of the contested decision could have had only one anticompetitive purpose, namely ‘front-running’ the trader who disclosed the information.
165First of all, as regards the discussions set out in recitals 126 to 129, 142, 147 and 291 to 295 of the contested decision, which fall within the second category referred to in paragraph 73 above, and the discussions set out in recitals 113 to 121 of the contested decision, which fall within the third category referred to in paragraph 73 above, they do not fall within the scope of the anticompetitive conduct identified by the Commission in recitals 337 to 341 of that decision.
166First, the Commission took the view, in essence, that the discussions set out in recitals 126 to 129 of the contested decision demonstrated that the traders participating in the chatroom at issue understood and expected that the information that they exchanged would be used in their common interest and not against their respective interests. Secondly, as regards the discussion set out in recital 142 of the contested decision, the Commission stated in recital 144 that that discussion showed that traders apologised when they did not act in the expected manner. Thirdly, according to the Commission, the discussion set out in recital 147 of the contested decision showed that information was exchanged by the traders in the chatroom at issue in order to guide their competitors. Fourthly, the discussions set out in recitals 291 to 295 of the contested decision are indicative of occasional instances of coordination and not of the anticompetitive conduct found by the Commission against Credit Suisse. Fifthly, according to the Commission, the discussions set out in recitals 113 to 121 of the contested decision demonstrated, in essence, that the members of the chatroom at issue formed part of a closed circle of trust.
167Next, as regards the discussions falling within the second category referred to in paragraph 73 above and analysed in recitals 183, 185 to 195, 208, 210, 225 to 228, 234 to 238, 245 to 250, 258 to 275, 277 to 280, 283 to 285 and 473 of the contested decision, the content of which is not disputed by the applicants, it is apparent from those discussions that the traders participating in the chatroom at issue exchanged, on an extensive and recurrent basis, information relating to current, including recently executed, or prospective transactions, such as specific currency pairs, volumes, directions, trading conditions and the types of customers concerned by those transactions. On occasions, they sought advice from other traders on how to trade and on a particular currency or currency pair, indicated that a particular transaction by one of the other participating traders had moved the value of a specific currency triggering their stop orders, shared their views on how to trade and communicated the level at which the transaction had been executed.
168Moreover, traders exchanged information relating to open ‘short’ or ‘long’ risk positions, the specific currencies involved in those positions and, on some occasions, their intentions or even their strategies, that is to say that they preferred not to sell a specific currency held in a ‘long’ position and to hold it in their trading book, that they were not going to close their ‘short’ positions because they expected the value of a specific currency to depreciate further, and that they were not going to buy a specific currency against another until an exchange rate fell to a specific level.
169As regards the discussions falling within the third category referred to in paragraph 73 above and set out in recitals 184, 195, 204 to 207, 211 to 213 and 228 of the contested decision, the content of which is not disputed by the applicants, it is apparent from those discussions that the traders exchanged, on an extensive and recurrent basis, current or forward-looking commercially sensitive information relating to the various types of orders received from their customers for specific currency pairs and exchange rates, to volumes, and to the type of customers concerned.
170The content of the discussions which took place between the traders participating in the chatroom at issue, including those referred to in paragraphs 167 and 169 above, shows that those traders disclosed, spontaneously or on request, commercially sensitive information which could be used by other traders to carry out their trading activities or define their trading strategy, by anticipating changes in exchange rates. Thus the exchanges of such information increased transparency solely between the participants in the infringement at issue.
171It follows that, contrary to what the applicants claim, the exchanges of information, in particular those that come under the third category referred to in paragraph 73 above, do not have the hypothetical single anticompetitive purpose of front-running the trader who shared the information by making the market move against the latter’s interest, but instead reduce the uncertainty that existed on the market to the benefit of the participating traders and to the detriment of their customers and other competitors.
172Moreover, the applicants’ argument that the allegedly low volume of transactions discussed in the context of certain exchanges could not have provided information on expected changes in exchange rates, must be rejected. That argument put forward by the applicants runs counter to the finding made in the contested decision, which they do not dispute, that the exchanges of information held against Credit Suisse related to large transactions in terms of volumes.
173In any event, if it were found that only the criterion relating to the volume of the transactions discussed in the exchanges of information examined in the contested decision is relevant for the purpose of assessing the anticompetitive nature of those exchanges, this would result in the exchanges as a whole not being examined and the exclusion, without justification, of a large number of those exchanges from the scope of the infringement of Article 101(1) TFEU, which, in the light of their essentially confidential and strategic nature, should not occur.
174Accordingly, the anticompetitive nature of the discussions, the interpretation of which is definitive and established, will be taken into account for the purposes of examining the second and third pleas, put forward by the applicants, relating to the existence of a ‘restriction by object’ and a single and continuous infringement, which will be examined below.
175It must be concluded that, contrary to what the applicants claim, the participating traders, through their direct, in-depth and recurrent contacts by means of frequent exchanges in the chatroom at issue, which related to current or forward-looking commercially sensitive information, such as that examined in paragraphs 81 to 106, 118 to 162 and 167 to 169 above, expressed their common intention to cooperate on the foreign exchange spot trading market, thereby knowingly substituting practical cooperation between themselves for the risks of competition, such as to influence their conduct on the market and disclose the conduct which their competitors in the chatroom at issue planned to adopt. In the light of the case-law cited in paragraphs 45 to 47 above, the Commission therefore rightly found that those exchanges could be classified as anticompetitive agreements and/or concerted practices for the purposes of Article 101 TFEU.
176It should also be noted that the Commission’s finding that the exchanges of information could be classified as anticompetitive agreements and/or concerted practices for the purposes of Article 101 TFEU remains justified even without taking into account the discussion of 9 May 2012, analysed in recital 276 of the contested decision and examined in paragraphs 107 to 117 above. The error of assessment made by the Commission as regards the nature of the information exchanged during that discussion therefore has no bearing on the contested decision.
177None of the applicants’ arguments is such as to invalidate the conclusion set out in paragraph 175 above.
178In that connection, the applicants’ argument that the absence of any instances of coordination during the period of Credit Suisse’s participation in the exchanges of information in the chatroom at issue excludes the classification of those exchanges as agreements and/or concerted practices must be rejected in the light of the case-law referred to in paragraphs 45 to 47 above.
179In so far as concerns the argument that the traders did not act in a ‘closed circle’, the applicants rely on ‘evidence in the Commission’s file’. However, in support of that argument, the applicants refer to a non-existent section of their application and confine themselves to a general reference to that evidence in the Commission’s file. Even if that evidence were included in the file submitted to the Court, such a general reference cannot remedy the absence of evidence in support of the applicants’ arguments, in so far as it is not the task of the Court to search through all the elements in the file for those which are capable of supporting those arguments (see, to that effect, judgment of 27 September 2006, Roquette Frères v Commission, T‑322/01, EU:T:2006:267, paragraph 209).
180Moreover, in so far as the applicants confine themselves, at the reply stage, to stating that examples of information which was shared outside the chatroom at issue are set out in Annex A.8, it should be recalled that their general reference to that annex does not make it possible, for the purposes of the case-law cited in paragraph 67 above, to determine precisely what the matters are in that annex which support or supplement their arguments. Accordingly, it cannot be taken into account in support of the applicants’ argument.
181It follows from all of the foregoing that the Commission was right to find that the exchanges of information which it had analysed in the contested decision justified the finding of the existence of agreements and/or concerted practices of an anticompetitive nature for the purposes of Article 101 TFEU.
182The first and second complaints in the second part of the first plea must therefore be rejected.
(2)The third complaint, alleging that the Commission erred in law in finding that Credit Suisse’s legitimate explanation for the exchanges of information at issue was irrelevant
183In their third complaint, the applicants argue, in essence, that they gave legitimate explanations for each of the discussions cited in the contested decision. They refer in that regard to the paragraphs in the application concerning the categorisation of the discussions at issue set out in paragraph 73 above and submit that the information exchanged in the chatroom at issue was ‘market intelligence’. They also state that the vast majority of the Credit Suisse trader’s order information was not shared in the chatroom at issue, which casts doubts on the probative value of the evidence on which the Commission relies. The Commission failed to engage meaningfully with the legitimate explanations provided by the applicants. As a result of that error, the Commission wrongly considered, in recital 141 of the contested decision, that that trader ought to have distanced himself from that chatroom.
184The Commission disputes the applicants’ arguments.
185First, the applicants’ argument that the Commission failed to engage meaningfully with their legitimate explanations for each of the exchanges and therefore wrongly considered that the Credit Suisse trader ought to have distanced himself cannot succeed. It must be noted that, in the contested decision, the Commission examined, in particular in recitals 455 to 480, the question whether the information exchanged during the discussions at issue was ‘market intelligence’, that is to say, information which could legitimately be disclosed, as the applicants claimed, and found that this was not the case.
186Furthermore, it should be recalled that the applicants’ arguments based on their categorisation of the discussions at issue, and on the fact that the information exchanged during certain discussions at issue is merely ‘market intelligence’, have been examined in the context of the first and second complaints in paragraphs 75 to 182 above and rejected.
187Secondly, as regards the applicants’ argument that the vast majority of the Credit Suisse trader’s order information was not shared in the chatroom at issue, it should be observed, as the Commission has, that as long as commercially sensitive information was regularly shared in that chatroom, it is irrelevant that most of the information relating to that trader’s orders was not shared there.
188It follows from all of the foregoing that the third complaint in the second part of the first plea must be rejected.
189In addition, in so far as the applicants refer in the heading of the present plea to a failure to provide sufficient reasons as to the classification of the online exchanges of information as anticompetitive agreements and/or concerted practices, and even if such a reference, which is unsubstantiated, could constitute an admissible argument within the meaning of Article 76(d) of the Rules of Procedure, it must be held that the Commission did not infringe the obligation to state reasons under the second paragraph of Article 296 TFEU. It is clear from the examination of the present plea that the explanations provided in the contested decision enabled the applicants to understand the Commission’s reasoning concerning that classification and enabled the Court to exercise its power of review in that regard, thus satisfying the requirements of that provision (see, in that connection, judgment of 23 November 2023, Ryanair v Commission, C‑210/21 P EU:C:2023:908, paragraph 105 and the case-law cited).
190Consequently, the first plea must be rejected in its entirety.
2.The second plea, alleging infringement of Article 101 TFEU and failure to state sufficient reasons as to the characterisation of the online exchanges of information as a restriction by object
191By their second plea, the applicants dispute the characterisation of the exchanges of information as a ‘restriction by object’. This plea is divided into two parts. The first part alleges that the Commission failed to discharge its burden under Article 101 TFEU of proving that the online exchanges of information restricted competition by object. The second part alleges that the Commission erred in law in concluding that the legitimate explanation for the conduct and the pro-competitive effects was irrelevant under Article 101(1) TFEU.
(a)Preliminary observations
192An 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 (see judgment of 19 March 2015, Dole Food and Dole Fresh Fruit Europe v Commission, C‑286/13 P, EU:C:2015:184, paragraph 122 and the case-law cited).
193It is thus apparent from the case-law that the provision of sensitive business information, such as information regarding pricing or information regarding supply and demand, including information in relation to future pricing or future supply and demand, makes it possible to reduce uncertainty as to the conduct of competitors on the market and to create conditions of competition which do not correspond to the normal conditions of the market and, consequently, gives rise to a concerted practice having as its object the restriction of competition within the meaning of Article 101(1) TFEU (judgment of 29 September 2021, Nippon Chemi-Con Corporation v Commission, T‑363/18, EU:T:2021:638, paragraph 195 (not published)).
194It must be recalled that, if a concerted practice is to be subject to the prohibition in principle laid down in Article 101(1) TFEU, a concerted practice must have as its ‘object or effect’ the prevention, restriction or distortion to an appreciable extent of competition within the internal market (judgment of 30 January 2020, Generics (UK) and Others, C‑307/18, EU:C:2020:52, paragraph 62).
195It follows that that provision, as interpreted by the Court of Justice, makes a clear distinction between the concept of ‘restriction by object’ and the concept of ‘restriction by effect’, evidence with regard to each of those concepts being subject to different rules (judgment of 30 January 2020, Generics (UK) and Others, C‑307/18, EU:C:2020:52, paragraph 63).
196Accordingly, as regards practices characterised as ‘restrictions by object’, there is no need to investigate their effects nor a fortiori to demonstrate their effects on competition in order to classify them as ‘restrictions of competition’, within the meaning of Article 101(1) TFEU, in so far as 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. Concerning such practices, all that is required is the demonstration that they can in fact be classified as ‘restrictions by object’, though mere unsubstantiated allegations are not however sufficient (judgment of 30 January 2020, Generics (UK) and Others, C‑307/18, EU:C:2020:52, paragraphs 64 and 65).
197On the other hand, where the anticompetitive object of an agreement, a decision by an association of undertakings or a concerted practice is not established, it is necessary to examine its effects in order to prove that competition has in fact been prevented or restricted or distorted to an appreciable extent (judgment of 30 January 2020, Generics (UK) and Others, C‑307/18, EU:C:2020:52, paragraph 66).
198It is also clear from the case-law of the Court of Justice that the concept of restriction of competition ‘by object’ must be interpreted strictly and can be applied only to some concerted practices between undertakings which reveal, in themselves and having regard to their objectives, and the economic and legal context of which they form part, a sufficient degree of harm to competition for the view to be taken that it is not necessary to assess their effects, since some forms of coordination between undertakings can be regarded, by their very nature, as being harmful to the proper functioning of normal competition. When determining that context, it is necessary to take into consideration the nature of the goods or services affected, as well as the real conditions of the functioning and structure of the market or markets in question (judgment of 30 January 2020, Generics (UK) and Others, C‑307/18, EU:C:2020:52, paragraphs 67 and 68).
(b)The first part of the second plea, alleging that there is no evidence of the existence of a restriction by object
199In recitals 394 to 454 of the contested decision, the Commission examined whether Credit Suisse’s conduct revealed a sufficient degree of harm to competition to be regarded as a restriction of competition ‘by object’. In recital 394 of the contested decision, the Commission recalled that, for that purpose, according to the case-law, it was necessary to take into account, inter alia, its content, its objectives and the economic and legal context of which it forms a part.
200First of all, as regards the content of that conduct, the Commission noted in the contested decision that the participating traders engaged in recurrent and extensive exchanges of information by way of which they disclosed to each other current or forward-looking sensitive information on confidential aspects of their conduct on the market, such as information on outstanding customer orders, open risk positions, bid-ask spreads and current or planned trading activities (recital 394 of the contested decision). The Commission also took the view that the flow of information exchanged had allowed the participating undertakings to have a full picture of what their competitors were or were not doing while actually trading in the market and had helped to reduce uncertainty between the participating undertakings as to the timing, extent and details of the intended conduct to be adopted on the market and the potential direction of the exchange rates involved. The Commission concluded that those exchanges went beyond exchanges of information in the public domain and were not justified by the traders’ objective of trading with each other as counterparties (recitals 396 and 397 of the contested decision).
201Next, the Commission considered that, in view of the nature, the frequency and the level of disaggregation of the information exchanged in the chatroom at issue, the purpose of the exchanges was to reduce normal market uncertainties for the participating undertakings on the foreign exchange spot trading market, in order to reduce risk and comfort them in their pricing decisions. Thus, in the Commission’s view, the continuous exchanges of commercially sensitive information provided the participating undertakings with the opportunity to subtract themselves from competition on the merits with regard to key parameters of competition (price and risk management). Moreover, that constant flow of exchanges of information within the chatroom also entailed an asymmetry of information between the participating undertakings and their non-participating competitors to the advantage of the former (recitals 398 to 401 of the contested decision).
202Finally, as regards the economic and legal context, the Commission noted, in the contested decision, the dynamic and liquid nature of the foreign exchange spot trading market. It considered that, in view of that context, the very frequent and detailed exchanges of commercially sensitive information which took place while the participating undertakings were operating in the market directly contributed to increasing transparency and reducing the uncertainty inherent in a competitive scenario (recitals 402 and 403 of the contested decision).
203The Commission concluded that the overall assessment of the exchanges of information at issue, in the light of the aims objectively pursued and the economic and legal context in which that conduct took place, revealed a sufficient degree of harm to competition to consider that the exchanges of current or forward-looking commercially sensitive information could be regarded as restrictive of competition by object within the meaning of Article 101 TFEU with respect to the foreign exchange spot trading of G10 currencies (recitals 404 and 405 of the contested decision).
204In the first part of the second plea, the applicants essentially criticise the Commission for having wrongly considered that the exchanges of information at issue were sufficiently harmful to competition to fall within that classification. In support of that part of the plea, they put forward, in essence, four complaints. By the first complaint, they submit that the Commission’s allegations that the exchanges of information at issue were intended to restrict competition are based on incorrect factual assumptions which the Commission failed to verify. By the second complaint, they argue that those allegations are not plausible in the light of the context of the market concerned. By the third complaint, they submit that the Commission committed an error of assessment in relying uncritically on the legal characterisation of the conduct presented in the evidence provided in order to obtain leniency and the settlement submissions. By the fourth complaint, they argue that the Commission cannot rely on its experience concerning exchanges of information in other market contexts or in other financial cases in order to reduce its burden of proof for establishing that the exchanges of information in the chatroom at issue restricted competition by object.
205It is appropriate first to examine the second complaint.
(1)The second complaint, alleging errors on the part of the Commission in its assessment of the context of the market concerned
206The applicants argue, in essence, that the Commission incorrectly assessed the factual and legal context of the foreign exchange spot trading market, or even that it failed to take that context into account when assessing the object of Credit Suisse’s conduct. They maintain that the Commission failed to assess correctly or take into account aspects of that market relating to (i) its transparency, (ii) its structure and (iii) the role of the traders operating on it.
207The Commission disputes the applicants’ arguments.
(i)The transparency of the foreign exchange spot trading market
208The first objection put forward by the applicants concerns the transparency of the market in question. In their view, during the period of the alleged infringement, the foreign exchange spot trading market was characterised by a limited degree of transparency. The applicants do not dispute that electronic platforms significantly increased transparency in the market. They nevertheless characterise the foreign exchange market as relatively opaque. According to the applicants, transactions take place on a number of platforms and there is no centralised system for real-time reporting of transactions on the foreign exchange spot trading market.
209The Commission disputes the applicants’ arguments. It claims, by referring in particular to recitals 12 to 16 of the contested decision, that, in the light of the development of screen-based electronic trading systems, which have made real-time prices (including the best available bid and ask prices) and corresponding trading volumes available on electronic platforms to virtually all market participants and which are constantly updated, the foreign exchange spot trading market had to be regarded as transparent at the time of the facts alleged.
210As the Commission stated in its defence, it is apparent from recitals 7, 14, 53 and 402 of the contested decision that, in its examination of the economic context in which the exchanges of information in the chatroom at issue took place, the Commission considered that the foreign exchange spot trading market was a large market, which was transparent, dynamic and very liquid. It is in the light of those factors, which are specific to the foreign exchange spot trading market, that the Commission concluded, in recitals 404 and 405 of the contested decision, that the exchanges of information constituted restrictions of competition by object.
211In that regard, it should be noted that, according to the material in the file, transparency of the market in question is defined as the ability of participants in that market to obtain information about the trading process, such as price, order volume, transaction volume, risk and the identity of the trader. It concerns the information available at the pre-trade stage as well as that available at the post-trade stage, and depends essentially on the willingness of participants and the ability of the exchange to display buy and sell orders publicly.
212Thus, according to the material in the file, it is necessary, on the foreign exchange spot trading market at the interdealer level (see paragraph 24 above), which is characterised by its hybrid structure, to distinguish between two trading channels, namely a direct (bilateral) channel and the broker channel, bearing in mind that transparency differs depending on the trading channel used. In particular, in the context of a direct (bilateral) trading channel, as in the present case, trading is not anonymised and the price and volume of transactions are kept secret by the parties to the transaction.
213It is true that the development of screen-based electronic trading systems considerably increased transparency on the market in question at the material time. However, the finding as to its transparency set out in recital 14 of the contested decision cannot, contrary to what the Commission asserts, be based, to the requisite legal standard, solely on the fact that prices and volumes are available on electronic platforms.
214In its written pleadings before the Court and at the hearing, the Commission itself acknowledged that conditions other than those relating to a real-time price and the corresponding trading volumes were not available to all market participants. In particular, it stated that, given the lack of access to information held by other traders, a certain degree of uncertainty remained in the market regarding the activities, positions, trading interests and intentions of end-customers and competing traders.
215However, the information to which the Commission refers in its written pleadings and at the hearing in order to conclude that there is a certain degree of uncertainty on the market is precisely the information which, where it is not held by traders, makes it possible to determine whether or not the foreign exchange spot trading market is transparent in nature for the purposes of the criteria referred to in paragraphs 211 and 212 above.
216More generally, in so far as the parties agree on the fact that the foreign exchange spot trading market is fundamentally asymmetrical and on the importance of the role of information on that market, the Commission cannot validly argue that the foreign exchange spot trading market, at the material time, was to be regarded as transparent. In a transparent market, information is immediately available (free of charge) to all market participants, which, as the Commission accepts, is not the case in the foreign exchange spot trading market having regard to the existence of a number of types of information.
217Accordingly, it must be held that the Commission made an error of assessment in finding, in recital 14 of the contested decision, that the foreign exchange spot trading market was transparent at the material time. It will therefore be necessary to examine whether that assessment could have a bearing, as the applicants claim, on the characterisation of the exchanges of information at issue as a ‘restriction by object’, since the transparency of that market is one of the factors which the Commission took into account in arriving at such a characterisation in recitals 404 and 405 of the contested decision (see paragraph 210 above).
(ii)The structure of the market in question
218The applicants’ second objection is that the Commission did not correctly assess the structure of the market in question, in terms of concentration. They submit, in essence, that, in such a fragmented market, in which the traders in question had minimal market shares, it was doubtful whether those traders could have created a significant asymmetry of information to the detriment of the other market participants. Those minimal market shares therefore cast doubt on the allegedly anticompetitive object of the exchanges of information at issue. They ultimately criticise the Commission for failing to explain why market shares were not relevant in the present case.
219The Commission challenges that objection made by the applicants.
220In that regard, although it is true that the structure of the market, as characterised by the level of market concentration, is a relevant factor in assessing whether the exchange of information restricts competition ‘by object’ (see, to that effect, judgment of 4 June 2009, T-Mobile Netherlands and Others, C‑8/08, EU:C:2009:343, paragraph 34), the only general principle applied in relation to the market structure is that supply must not be atomised (judgment of 2 October 2003, Thyssen Stahl v Commission, C‑194/99 P, EU:C:2003:527, paragraph 86).
221However, although the Court of Justice was initially able to take into consideration the oligopolistic or fragmented nature of the market on which the exchanges of commercially sensitive information took place, it is clear, in particular, from the judgment of 19 March 2015, Dole Food and Dole Fresh Fruit Europe v Commission (C‑286/13 P, EU:C:2015:184), that the EU judicature was able to recognise the anticompetitive object of such exchanges, even in the absence of oligopolistic market concentration (see judgment of 14 March 2013, Dole Food and Dole Germany v Commission, T‑588/08, EU:T:2013:130, paragraph 341 and the case-law cited).
222That said, it must be pointed out that, in view of the fact that competition law is concerned with the activities of undertakings (judgment of 7 January 2004, Aalborg Portland and Others v Commission, C‑204/00 P, C‑205/00 P, C‑211/00 P, C‑213/00 P, C‑217/00 P and C‑219/00 P, EU:C:2004:6, paragraph 59), the applicants’ argument that, as they confirmed at the hearing, the foreign exchange spot trading market is fragmented on account of the market shares of the traders who were employed by the banks involved in the infringement at issue, cannot be accepted. In any event, as is apparent from the material in the file, they have failed to establish the market shares of those banks for the ‘voice trading’ activities to which the analysis of the exchanges of information at issue is limited (see paragraph 16 above).
223It should also be noted that, in recital 28 et seq. and in recital 402 of the contested decision, the Commission, without expressing an opinion on the level of concentration in the light of the market shares of the participants in the infringement at issue, took into consideration the structure and specific features of the market in question as characterised by the role of the market makers who participated in the online exchanges of information. The role of market makers on that market, which was not disputed by the applicants, involved significant financial commitments which could not have been borne by a large number of participants, thereby ruling out the view taken by the applicants that the foreign exchange spot trading market was fragmented.
224Moreover, the General Court has already held, and the Court of Justice confirmed on this point, that exchanges of information are all the more sensitive in terms of competition where they take place between traders acting as market makers, in the light of their importance on the market, and in particular because they are generally and continuously active and therefore enter into a larger number of transactions than other market participants (judgment of 24 September 2019, HSBC Holdings and Others v Commission, T‑105/17, EU:T:2019:675, paragraph 145, confirmed on this point by the judgment of 12 January 2023, HSBC Holdings and Others v Commission, C‑883/19 P, EU:C:2023:11, paragraph 195). Therefore, from the point of view of preserving competition on the market, it is particularly fundamental that they determine their strategies and, in particular, their prices independently.
225The applicants’ objection that the Commission failed to take into account the structure of the market as characterised by its fragmentation must therefore be rejected.
(iii)The dual role of traders
226The applicants’ third objection relates to the Commission’s failure to take into account, when assessing the economic and legal context in question, the characteristic of the foreign exchange spot trading market that traders interact at all times in the dual role of competitor and counterparty. According to the applicants, while it is true that the reduction of uncertainty resulting from the exchange of information may indeed serve as an absolute criterion for an infringement ‘by object’ in markets in which any disclosure of commercially sensitive information among competitors creates conditions which do not correspond to the normal conditions of the market, it cannot serve as an absolute criterion where the normal conditions of the market dictate the regular exchange of commercially sensitive information among participants who act vertically and horizontally at the same time.
227The Commission challenges that objection made by the applicants.
228In that connection, it is true that, on the foreign exchange spot trading market, traders generally act both as counterparties when concluding trades between themselves, and as competitors in relation to potential customers. In response to a measure of organisation of procedure decided upon by the Court, the applicants stated, in essence, that while it is difficult to draw a formal distinction between the two roles, a clear instance of two traders being in competition is when they seek to conclude a trade with the same customer at the same time.
229Nevertheless, aside from the fact that the dual role of traders claimed by the applicants arises not in respect of the same transaction but in respect of two separate transactions, the delimitation of the scope of the assessment of the legality of the exchanges of information as set out in recitals 158 and 467 of the contested decision, which excludes exchanges between the traders in the normal course of business when they were assessing the possibilities of trading with each other, in principle runs counter to the applicants’ criticism recalled in paragraph 226 above.
230The applicants’ objection that the Commission failed to take into account the characteristic of the foreign exchange spot trading market concerning the dual role of traders must therefore be rejected as unfounded.
231It follows from the foregoing that the second complaint is well founded in so far as it alleges that the Commission erred in finding, in recital 14 of the contested decision, that the foreign exchange spot trading market was transparent at the material time and must be rejected as to the remainder. It will be necessary to examine, in the context of the first complaint, whether that assessment may have a bearing, as the applicants claim, on the characterisation of the exchanges of information at issue as a ‘restriction by object’.
(2)The first complaint, alleging that the Commission’s finding that the exchanges of information at issue had the object of restricting competition is based on inaccurate and unverified factual assumptions
232The applicants submit that the Commission failed to fulfil its obligation to demonstrate that it was possible to infer from the type of information exchanged and the characteristics of the market in question that the object of the exchanges in the chatroom at issue was anticompetitive. They argue that the Commission has not demonstrated that the exchanges of information comforted the traders in their daily pricing and expert risk management decisions. Nor, they argue, has the Commission established its premiss that the conduct of the participating traders was intended to create a significant asymmetry of information vis-à-vis the rest of the market. On the foreign exchange spot trading market, traders are encouraged to share information in order to inform their trading activity and to pass it on to their clients. In any event, the vast majority of the information relating to the Credit Suisse trader’s orders was not shared in the chatroom at issue.
233The applicants also claim that the discussions referred to in the contested decision, according to which the information exchanged enabled the traders participating in the chatroom at issue to wait before absorbing a short position, are inconsistent with the fact that market makers had, in the words of the contested decision, to ‘stand ready to trade with anyone needing foreign exchange at a moment’s notice’. They submit that most, if not all, of the discussions regarding bid-ask spreads referred to in the contested decision do not include any indication of reaching possible agreements on spreads and took place in the abstract, or even after a spread quote had already been provided.
234In order to respond to the applicants’ arguments, it is necessary to examine whether the various types of exchanges of information referred to in the contested decision revealed a sufficient degree of harm to justify their characterisation as a ‘restriction by object’.
(i)The exchanges of information relating to bid-ask spreads
235As regards the exchanges of information relating to bid-ask spreads, the interpretation of which was carried out in in recitals 245 to 251 of the contested decision, the Commission justified, in recitals 394 to 404 of that decision, the classification as a ‘restriction by object’ on the ground that they helped to remove the uncertainties inherent in the market in question as regards pricing, thereby increasing transparency concerning the higher bid-ask spreads which those traders could offer. In that way, such exchanges could influence the prices offered by those traders to customers, by reducing the risk associated with currency trading to their own advantage.
236In that regard, it must be noted – without, however, this being disputed by the applicants – that bid-ask spreads are a component of the final price paid by customers, in so far as they are applied to the mid-point between the bid and ask prices for a specific trade in a currency pair and determine the revenue received by the trader for the immediacy of the service provided and for the risk he or she subsequently bears by having a given currency in his or her trading book (see paragraph 20 above and recitals 48 and 239 of the contested decision). On the foreign exchange spot trading market, the ability of banks to assess their pricing risks is one of the essential parameters of competition, which the applicants likewise do not dispute.
237Accordingly, exchanges of non-public, precise information, in this instance concerning a pricing component, must be regarded as making it possible to reduce the normal uncertainties inherent in the market in question as regards pricing on that market, to create an asymmetry of information to the benefit only of the traders participating in the chatroom at issue and, consequently, those exchanges reveal a sufficient degree of harm to competition. Such a finding is sufficient for it to be accepted that those exchanges entailed a restriction of competition by object within the meaning of Article 101(1) TFEU (judgment of 12 January 2023, HSBC Holdings and Others v Commission, C‑883/19 P, EU:C:2023:11, paragraph 204).
238None of the applicants’ arguments can invalidate the foregoing conclusion.
239At the outset, the applicants’ arguments that, in essence, the exchanges of information on bid-ask spreads do not constitute a ‘restriction by object’ in view of the absence of any indication as to the possibility of entering into agreements concerning those spreads, must be rejected. Admittedly, it is not apparent from the exchanges of information relating to bid-ask spreads which took place in the chatroom at issue that traders agreed to apply, in a concerted manner, a particular level of spreads to a specific currency pair. This does not mean, however, that in disclosing to their competitors confidential and strategic information relating to the appropriate level of the bid-ask spreads for a specific currency pair or the spreads which they had just applied, the traders participating in the chatroom at issue did not reveal the conduct which they were likely to adopt on the market, thereby reducing uncertainty in the minds of those who had participated in the exchanges of information, within the meaning of the case-law cited in paragraph 47 above.
240Next, as regards the applicants’ argument that, in essence, since the final price quoted to customers is determined by a bank’s sales desk employees, it would not have been logical, from the point of view of the traders, who are rational actors, to engage in conduct consisting in quoting the widest possible spread, it should be pointed out that elements concerning the traders’ willingness to restrict competition or concerning their economic interest in doing so are normally of limited significance for the purpose of the characterisation of conduct as a ‘restriction by object’ (see, to that effect and by analogy, judgment of 7 January 2004, Aalborg Portland and Others v Commission, C‑204/00 P, C‑205/00 P, C‑211/00 P, C‑213/00 P, C‑217/00 P and C‑219/00 P, EU:C:2004:6, paragraph 335).
241It follows from the foregoing that the Commission was right to consider that the exchanges of information relating to bid-ask spreads which took place in the chatroom at issue contributed to the characterisation of the conduct in question, as a whole, as a ‘restriction by object’.
(ii)The exchanges of information on customer orders
242As regards the exchanges of information on customer orders, namely ‘conditional’ orders (recitals 183 to 195 of the contested decision), orders to be executed at the ‘fixing’ (recitals 206 to 213 of the contested decision) and ‘immediate’ orders (recitals 223 to 228 of the contested decision), the Commission justified, in recitals 394 to 404, read in conjunction with recitals 171 to 173 of the contested decision, their characterisation as a ‘restriction by object’ on the ground that, in view of their nature, frequency and level of disaggregation, they had removed uncertainties relating to the foreign exchange spot trading market and increased transparency concerning the trading strategies of traders participating in the chatroom at issue, in so far as the sharing of information on the direction of the market or the levels at which the market would show resistance could enable a trader to better position himself in order to profit therefrom or avoid a loss. Accordingly, those exchanges informed the traders, in particular, of price trends by reducing uncertainties as to the risks inherent in pricing (for example, by leading to a reduction in expected losses), whereas price is one of the parameters of competition on the foreign exchange spot trading market which the participating banks should have determined independently.
243In that regard, it should be recalled, as is clear from the material in the file, that information relating to customer orders is non-public information, known only to the parties to the transaction, namely the trader and the customer, and is important information which enables the trader who concluded the transaction to form his or her own opinion on exchange rate movements or determine his or her commercial strategy. The exchange of such information has the capacity to reduce the uncertainties inherent in the market in question, by enabling market makers not to determine independently (i) their strategy with regard to the specific currency pairs on which they have confidential and detailed information, and (ii) their conduct on the market, thereby enabling them to profit or to avoid the losses which might result from developments on the market.
244Consequently, the aforementioned exchanges of information must be considered, within the meaning of the case-law cited in paragraph 237 above, as revealing a sufficient degree of harm to competition to be characterised as a ‘restriction by object’.
245Such a finding cannot be invalidated by the applicants’ claim that there is a contradiction between the conclusion drawn from the exchange set out in recital 193 of the contested decision – namely that the information exchanged enabled participating traders to wait before absorbing a short position in their trading book (in order to maximise their chances of making a profit) – and the finding set out in recital 21 of that decision, namely that market makers had to ‘stand ready to trade with anyone needing foreign exchange at a moment’s notice’. In so far as the traders’ objective is to make a profit and limit their losses, it would make little sense for traders to act against their own commercial interests on the sole ground that, in principle, they must stand ready to trade at a moment’s notice. It cannot therefore be inferred that traders took positions solely on the basis of incoming customer orders, so as to exclude the characterisation of the exchanges of information relating to customer orders as a ‘restriction by object’.
246It follows from the foregoing that the Commission was right to consider that the exchanges of information on customer orders referred to in paragraph 242 above contributed to the characterisation of the conduct in question, taken as a whole, as a ‘restriction by object’.
(iii)The exchanges of information on open risk positions
247As regards the exchanges of information on open risk positions – the interpretation of which, carried out by the Commission, is set out in recitals 234 to 238 of the contested decision – the Commission justified, in recitals 394 to 404 of that decision, their characterisation as a ‘restriction by object’ on the ground that sharing that information had enabled traders to have greater certainty as to their respective commercial intentions concerning their potential hedging conduct, and had therefore removed uncertainty as to the potential changes in specific exchange rates, thereby conferring an advantage on participating traders over other market participants and helping them in the expert management of their risk, in respect of which they are in competition.
248In that connection, in the absence of any specific arguments put forward by the applicants against the characterisation of the exchanges of information on open risk positions as a ‘restriction by object’, it is sufficient to note, as does the contested decision, that the exchange of that information, in indicating a specific currency and the strategy of not closing the position or of waiting before closing it, must be regarded as capable of removing uncertainties inherent in the operation of the foreign exchange spot trading market.
249It follows that such exchanges of information, which are as a matter of principle contrary to the requirement of independence inherent in Article 101(1) TFEU, reveal a sufficient degree of harm to competition to be characterised as a ‘restriction by object’ within the meaning of the case-law cited in paragraph 237 above.
250It follows from the foregoing that the Commission was right to consider that the exchanges of information on open risk positions contributed to the characterisation of the conduct in question, taken as a whole, as a ‘restriction by object’.
(iv)The exchanges of information on current or prospective trading activities
251As regards the exchanges of information relating to current or prospective trading activities – the interpretation of which, carried out by the Commission, is set out in recitals 258 to 285 of the contested decision – the Commission justified, in recitals 394 to 404 of that decision, the characterisation as a ‘restriction by object’ on the ground that those exchanges had removed some of the uncertainties inherent in the foreign exchange spot trading market and increased the level of transparency of that market by providing the participants in the chatroom at issue with an insight into their competitors’ current conduct, thereby creating an asymmetry of information between the traders participating in the chatroom at issue and the other market participants. In that way, the former were comforted in their risk assessment when they were developing their own trading strategies, and in particular their pricing strategies.
252In that regard, it should be noted that the sharing of information on the details of transactions carried out by traders and the sharing of advice or views on the manner in which to trade, such as those in question, enable traders participating in the chatroom at issue to (i) acquire specific knowledge to better anticipate market movements, and (ii) adapt their trading strategies accordingly.
253Such exchanges of information are blatantly contrary to the requirement of independence, which must characterise the conduct of market makers on the market in question, operating within a system of effective competition. Those exchanges remove the uncertainties in their minds as to the essential parameters of competition on that market, namely price and expert risk management. Accordingly, those exchanges were able to increase transparency on the foreign exchange spot trading market to the benefit of the traders participating in the chatroom at issue and to the detriment of the other market participants. Consequently, they reveal a sufficient degree of harm to competition to be characterised as a ‘restriction by object’ within the meaning of the case-law cited in paragraph 237 above.
254It follows from the foregoing that the exchanges of information on bid-ask spreads, customer orders, open risk positions and current or prospective trading activities, which were analysed in the contested decision, reveal a sufficient degree of harm to competition. The Commission therefore rightly characterised them as a ‘restriction by object’.
255Contrary to what the applicants argue, the fact that the Credit Suisse trader did not share the vast majority of the information on his bids is irrelevant to the harmfulness of that bank’s conduct. In that regard, it is sufficient to note that the assessment of whether conduct reveals a sufficient degree of harm to competition is not based on any legal quantitative threshold of information shared with competitors.
256In those circumstances, notwithstanding the Commission’s error of assessment concerning the transparency of the foreign exchange spot trading market, the applicants’ arguments seeking to refute the classification of the exchanges of information at issue as a ‘restriction by object’ must be rejected.
257It follows from all of the foregoing that the first complaint must be rejected.
(3)The third complaint, alleging that the Commission erred in relying on the legal classification of the exchanges of information at issue which was presented in the evidence provided in order to obtain leniency and the settlement submissions
258In recitals 394 to 404 of the contested decision, read in conjunction with recitals 170 to 285, in support of the finding that there was a restriction of competition ‘by object’, the Commission relied on contemporaneous evidence made available to it by other banks involved in the infringement at issue, namely a number of transcripts of exchanges of information which took place in the chatroom at issue, on certain occasions relying, as regards their meaning, on statements made by those banks, in particular in the context of the leniency applications.
259The applicants complain that the Commission relied ‘uncritically’ on the leniency statements and settlement submissions. The applicants claim that those statements have little probative value and are provisional in nature and unfounded, pointing out that all the settling parties expressed their doubts as to the Commission’s understanding of the relevant market and the objective of the conduct at issue. The applicants maintain that the Commission relied on the leniency statements when these supported its interpretation of the facts.
260The Commission disputes the applicants’ arguments.
261As a preliminary point, it should be noted that, although, in the context of the present line of argument, the applicants cite passages from the leniency applications and the settlement submissions, as they confirmed at the hearing, they do not put forward any specific objection to the fact that the Commission relied on the settlement submissions.
262That said, in so far as the parties disagree as to the probative value of the leniency statements and whether the Commission should take them into consideration as evidence capable of confirming the existence of the infringement which Credit Suisse is alleged to have committed, it should be recalled that it is settled case-law that no provision or any general principle of EU law prohibits the Commission from relying, as against an undertaking, on statements made by other incriminated undertakings. If that were not the case, the burden of proving conduct contrary to Articles 101 and 102 TFEU, which is borne by the Commission, would be unsustainable and incompatible with the task of supervising the proper application of those provisions which is entrusted to it by the TFEU (see judgment of 8 September 2016, Goldfish and Others v Commission, T‑54/14, EU:T:2016:455, paragraph 96 and the case-law cited).
263In particular, where a person admits that he or she committed an infringement and thus admits the existence of facts going beyond those the existence of which could be directly inferred from the documentary evidence, that implies, a priori, in the absence of special circumstances indicating otherwise, that that person has resolved to tell the truth. Thus, statements which run counter to the interests of the declarant must in principle be regarded as particularly reliable evidence, unless they are not corroborated by other evidence (see, to that effect, judgment of 8 September 2016, Goldfish and Others v Commission, T‑54/14, EU:T:2016:455, paragraphs 98 to 100 and the case-law cited).
264Furthermore, where the Commission relies, in establishing an infringement of competition law, on documentary evidence, the burden is on the undertakings concerned not only to put forward a plausible alternative to the Commission’s view but also to allege that the evidence relied on in the contested decision to establish the existence of the infringement is insufficient (see judgment of 16 June 2015, FSL and Others v Commission, T‑655/11, EU:T:2015:383, paragraph 181 and the case-law cited).
265In that connection, it should be noted that, in their arguments, first, the applicants do not dispute the authenticity of the statements made in the context of the leniency applications and, secondly, they fail to demonstrate that there were special circumstances indicating that the leniency applicants had not resolved to tell the truth within the meaning of the case-law referred to in paragraph 263 above.
266As regards, more specifically, the passages in the leniency statements which, it is claimed, the Commission failed to take into account where they ran counter to its interpretation of the facts, the applicants refer to the documents produced by two other banks involved in the infringement at issue. It is clear from the statements made by those banks that they were seeking, in essence, either to warn the Commission of the potential repercussions on the functioning of the foreign exchange spot trading market of characterising the exchanges of information which took place in the chatroom at issue as a ‘restriction by object’, or to argue that those exchanges were ‘legitimate’, or even ‘pro-competitive’.
267However, it is clear that the aforementioned claims, on which the applicants rely, are intended, in essence, to ‘justify’ the exchanges of information at issue on the basis of their allegedly ‘legitimate’ or even ‘pro-competitive’ nature, which the Commission allegedly failed to take into account, yet no evidence has been adduced to cast doubt on the intrinsic capacity of those exchanges to affect competition on the market in question.
268On the contrary, it follows from the passages of the leniency statements referred to in paragraph 261 above and from others quoted in the contested decision, the factual accuracy of which is not disputed by the applicants, that the leniency applicants acknowledged that the exchanges of information which took place in the chatroom at issue were capable of increasing transparency on the market in question and, as a result, of conferring an advantage on the participants in that chatroom to the detriment of other market participants.
269In those circumstances, it must be held that the leniency statements which, as is clear from the contested decision, supported the contemporaneous evidence made available to the Commission by other banks constitute particularly reliable evidence within the meaning of the case-law cited in paragraph 263 above, which the Commission could validly take into account as factual context enabling it to assess the object of the exchanges at issue.
270Moreover, as is clear from the contested decision, the Commission used the leniency statements in question in order to confirm its own assessments based on the documentary evidence available to it. Consequently, the applicants’ argument that the Commission did not carry out its own assessment of the facts cannot be accepted.
271That finding cannot be called into question by any of the arguments put forward by the applicants.
272First, it is the contradictory statements contained in the applicants’ pleadings which highlight the unsubstantiated nature of their arguments. The applicants claim, first of all, that the statements made in the context of the leniency statements have a high probative value and, next, that the evidence provided in order to obtain leniency has a low probative value. When questioned about those contradictory statements at the hearing, they stated that that difference in approach was due to the fact that the first concerns the principle that statements made in the context of leniency applications have a high probative value, whereas the second concerns the low probative value of the relevant statements in the present case. However, that understanding of the applicants’ argument is contradicted by a reading of their written pleadings before the Court, which, in both cases, refer to case-law which emphasises the high probative value of such documents.
273As regards the allegedly low probative value of leniency statements which follows from the judgment of 13 July 2011, Kone and Others v Commission (T‑151/07, EU:T:2011:365), it should be noted that that argument put forward by the applicants is based on a misreading of that judgment. Indeed, contrary to what they claim, in that judgment the Court ruled on the ‘significant added value’ of leniency statements with respect to the evidence already in the Commission’s possession for the purposes of reducing the fine pursuant to the Leniency Notice, and not on the low probative value of leniency statements in general.
274Secondly and lastly, although the applicants claim that the Commission’s reliance on the statements of the banks seeking leniency is a source of concern on account of the complexity of the assessment required to determine the extent to which information may be exchanged on the foreign exchange spot trading market, it must be pointed out that they do not put forward any specific legal challenge in that connection.
275The third complaint must therefore be rejected.
(4)The fourth complaint, alleging that the Commission could not rely on its experience concerning exchanges of information in other market contexts or in other financial cases in order to lighten its burden of proof
276First, the applicants complain that the Commission lightened the burden of proof for establishing that the exchanges of information at issue restricted competition ‘by object’ by relying on its alleged experience concerning exchanges of information in other market contexts or in other financial cases.
277Secondly, the applicants submit that the Commission shifted to them the burden of proving that the alleged unlawful exchanges of information were necessary, instead of examining whether, on the foreign exchange spot trading market, those exchanges were in fact unlawful and were among the most harmful restrictions of competition.
278Thirdly, the applicants submit that the characterisation as a ‘restriction by object’ is possible only if experience and economic theory clearly indicate that a form of conduct is, by its nature, one of the most harmful types of anticompetitive conduct. However, the general economic principle that the exchange of non-public and detailed information can be expected to be inherently and significantly harmful to competition in most markets does not apply to the foreign exchange spot trading market.
279The Commission disputes the applicants’ arguments.
280First, as regards the applicants’ claim that the burden of proof had been lightened inasmuch as the Commission relied on its alleged experience concerning exchanges of information in other contexts in order to classify the exchanges of information at issue as a ‘restriction by object’, it is sufficient to point out that, as is clear from the considerations set out in paragraphs 235 to 254 above, the Commission rightly found that those exchanges of information were likely to create an asymmetry of information between the participating traders to the detriment of the other traders on the market and that they thus revealed a sufficient degree of harm to competition, without that conclusion being based on its experience in other markets.
281Secondly, in those circumstances, nor can the applicants allege that the Commission shifted the burden of proof for establishing that the exchanges of information were necessary or assert that the Commission based its finding on the object of those exchanges by reference to their ‘more granular or informative’ nature. It is clear from the contested decision that the Commission characterised the exchanges of information as a ‘restriction by object’ on account of their commercially sensitive nature, which reduced or even removed uncertainty on the market and, consequently, had an impact on the decision-making autonomy of traders on the market in question, and that the exchanges therefore revealed a sufficient degree of harm to competition.
282Thirdly, it is clear that, on the one hand, the economic literature to which the applicants refer in support of their arguments, summarised in paragraph 278 above, makes no reference to the circumstances of a foreign exchange spot trading market for the purposes of substantiating their line of argument.
283On the other hand, the Court recently held that the fact that the Commission had not, in the past, considered that an agreement similar to the agreement at issue was, by its very object, restrictive of competition is not such as to prevent it from doing so in the future (judgment of 25 March 2021, Lundbeck v Commission, C‑591/16 P, EU:C:2021:243, paragraph 130).
284Accordingly, the applicants cannot validly rely on the absence of an economic consensus to dispute the classification of the exchanges of information at issue as a ‘restriction by object’. Their arguments to that effect must therefore be rejected.
285It follows that the fourth complaint must be rejected.
(5)Conclusions on the first part of the second plea in law
286It follows from an examination of the complaints put forward in the first part of the second plea that, notwithstanding the Commission’s error as to the transparency of the foreign exchange spot trading market (see paragraph 217 above), the applicants’ arguments seeking to refute the classification of the exchanges of information at issue as a ‘restriction by object’ must be rejected and, consequently, without prejudice to the examination below of the arguments alleging that the effects of those exchanges were pro-competitive, the first part of the second plea must be rejected.
(c)The second part of the second plea, alleging that the Commission made errors in the context of the analysis of the legitimate or pro-competitive nature of the exchanges of information at issue
287In recitals 464 to 479 of the contested decision, the Commission observed, first of all, that any pro-competitive effects of conduct could be taken into account in the context of the assessment of Article 101(3) TFEU, but that Credit Suisse had not relied on the applicability of that article in the present case. In any event, as regards the characterisation of the exchanges of information at issue as ancillary restrictions, Credit Suisse had neither established that the foreign exchange spot trading market could not function without them, nor demonstrated any pro-competitive effects of those exchanges. Next, for the sake of completeness, the Commission examined Credit Suisse’s claims concerning the pro-competitive effects of the exchanges of information at issue.
288In that regard, the Commission recalled that the fact that the conduct pursues a legitimate objective does not preclude that conduct from being regarded as having an object restrictive of competition. Although the exchange of information was capable of narrowing bid-ask spreads and improving prices, there is a fundamental difference between such an exchange in a restricted circle of traders in a chatroom, as in the present case, and information accessible to all market participants. The pro-competitive effects of the exchanges of information alleged by the applicants would not apply to the exchange of information in a closed circle, since the information sharing in question created an asymmetry of information by providing a competitive advantage only to the participants in the chatroom at issue. In those circumstances, traders could decide not to lower prices but to increase their profits, since the bonuses which they receive are linked to the profit that they generate.
289First, the applicants dispute that assessment, arguing that the Commission applied the wrong legal framework, since it examined their arguments relating to the legitimacy and pro-competitive effects of the exchanges of information only for the sake of completeness. The applicants claim that, in so doing, the Commission took the view that the pro-competitive nature of those exchanges was not essential to the legal analysis of the exchanges of information at issue. According to the applicants, if a concerted practice pursues a legitimate or pro-competitive object, it cannot be characterised as an infringement by object.
290Secondly, they claim that the Commission erred in its assessment of their arguments and evidence in the light of the legitimate and pro-competitive nature of the exchanges of information in the chatroom at issue.
291The applicants submit, first, that the details concerned were exchanged for information purposes in order to be passed on to customers and to enable traders quickly to assess the inventory risk and thereby reduce the bid-ask spreads offered to customers.
292Second, it is argued that the Commission infringed the rules governing the burden of proof, in so far as, where conduct has atypical elements, it is easier for an applicant to disprove the Commission’s allegations by adducing evidence of another possible explanation. More generally, the applicants argue not that the conduct in question is remedied by its pro-competitive effects or by its ‘legitimate’ nature, but rather that its sole object is legitimate and pro-competitive.
293The Commission disputes the applicants’ arguments.
294As a preliminary point, it should be noted that, at the hearing before the Court, the parties were questioned as to the consequences to be drawn from the judgment of 21 December 2023, European Superleague Company (C‑333/21, EU:C:2023:1011) in so far as concerns the taking into account, in the present case, of the pro-competitive effects alleged by the applicants. In that regard, the parties submitted, in essence, that the pro-competitive effects must be taken into account in the context of the characterisation of the exchanges of information for the purposes of Article 101(1) TFEU. In particular, the applicants took the view that that judgment confirms their position that the pro-competitive effects of the exchanges of information at issue must be taken into account as elements of the economic and legal context in which those exchanges arise. The Commission contended that that judgment confirmed the analysis carried out in the contested decision.
295Again as a preliminary point, when questioned at the hearing with regard to certain arguments relating to the alleged ‘legitimate’ effects put forward in their written pleadings before the Court, the applicants confirmed that those effects must be understood as forming part of the same economic and legal context and as being, as such, relevant for the purpose of calling into question the characterisation of the exchanges of information as a ‘restriction by object’.
296That said, it should be recalled that, in order to determine, in a given case, whether an agreement, decision by an association of undertakings or a concerted practice reveals, by its very nature, a sufficient degree of harm to competition that it may be considered as having as its object the prevention, restriction or distortion thereof, it is necessary to examine, first, the content of the agreement, decision or practice in question; second, the economic and legal context of which it forms a part; and, third, its objectives (see judgment of 21 December 2023, European Superleague Company, C‑333/21, EU:C:2023:1011, paragraph 165 and the case-law cited).
297In that regard, it should also be borne in mind, first of all, in the economic and legal context of which the relevant conduct forms a part, that it is necessary to take into consideration the nature of the products or services concerned, as well as the real conditions of the structure and functioning of the sectors or markets in question. It is not, however, necessary to examine nor, a fortiori, to prove the effects of that conduct on competition, be they actual or potential, or negative or positive (see judgment of 21 December 2023, European Superleague Company, C‑333/21, EU:C:2023:1011, paragraph 166 and the case-law cited).
298Next, as regards the objectives pursued by the relevant conduct, a determination must be made of the objective aims which that conduct seeks to achieve from a competition standpoint. Nevertheless, the fact that the undertakings involved acted without having a subjective intention to prevent, restrict or distort competition and the fact that they pursued certain legitimate objectives are not decisive for the purposes of the application of Article 101(1) TFEU (see judgment of 21 December 2023, European Superleague Company, C‑333/21, EU:C:2023:1011, paragraph 167 and the case-law cited).
299Lastly, the taking into consideration of all of the aspects referred to in paragraphs 296 to 298 above must, at any rate, show the precise reasons why the relevant conduct reveals a sufficient degree of harm to competition such as to justify a finding that it has as its object the prevention, restriction or distortion of competition (see judgment of 21 December 2023, European Superleague Company, C‑333/21, EU:C:2023:1011, paragraph 168 and the case-law cited).
300It should be noted that it follows from the examination of the first plea and the first part of the second plea that the Commission analysed the content of the exchanges of information at issue, the economic and legal context of which they form a part and their objectives, and that it explained why taking all those aspects into account had led it to conclude that those exchanges had an anticompetitive object, in accordance with the case-law referred to in paragraphs 296 to 299 above.
301It should also be noted that, as is apparent from the case-law cited in paragraphs 297 and 298 above, contrary to what the parties claim, it is not necessary, first, to take into consideration the pro-competitive effects as such, for the purposes of assessing whether the exchanges of information at issue should be characterised as a restriction of competition ‘by object’ under Article 101(1) TFEU, including in the context of any examination of the existence of a degree of harm required for the purposes of such a classification. Secondly, consideration of the alleged legitimate objectives is not decisive in that assessment.
302Accordingly, the applicants cannot validly complain that the Commission made an error of law based on the ‘incorrect application of a legal framework’.
303In any event, it should be noted that, in recitals 465 to 479 of the contested decision, the Commission examined Credit Suisse’s arguments regarding the legitimate or pro-competitive nature of the exchanges of information. It is necessary to examine whether the applicants’ arguments are capable of calling the merits of that assessment into question.
304In the first place, as regards the economic theory referred to by the applicants in their written pleadings, which merely states that an increase in one undertaking’s efficiency ‘can’ have essentially pro-competitive knock-on effects, it must be held that the link between the conclusion resulting from that economic theory and the exchanges of information at issue cannot be regarded as established to the requisite legal standard. Moreover, the applicants’ claim is counter-intuitive from the point of view of the commercial logic and economic rationale of market makers, in so far as, when a restricted circle of competitors acting as market makers has additional information as compared with other market participants, the logic of maximising their profits in the light of that information leads to an alignment of the prices offered to their customers rather than to the offering of smaller spreads.
305Contrary to what the applicants claim, none of the material in the file shows that the traders in question intended to offer lower spreads to their clients in the light of the information exchanged. In that regard, on the one hand, the applicants refer to an exchange of information of 24 June 2011 reproduced in Annex A.12. Since that exchange of information predates Credit Suisse’s participation in the infringement at issue, it must be disregarded as irrelevant.
306On the other hand, as regards another example in Annex A.12, it should be noted that this dates from 19 June 2012, a date which falls within the relevant period. First, that example shows that an exchange took place between 10:40:33 and 10:41:38 during which the Credit Suisse trader asked an RBS trader participating in the chat room at issue about the bid-ask spread level applicable to an amount of 50 million for a EUR/GBP currency pair (‘whats spread 50 x these days’). In response to that question, the RBS trader indicated the applicable spread (‘3’). The Credit Suisse trader thanked him for that information (‘yeh thats what I thought’; ‘ta mate’).
307Secondly, it is apparent from another exchange reproduced in Annex A.12, which took place between 07:30:20 and 07:30:47 on 19 June 2012 on another chatroom, that two persons exchanged information relating to a bid-ask spread applicable to a EUR/GBP currency pair (‘50 eurgbp’; ‘sure’, ‘42 46’, ‘off’, ‘ref’). In that regard, the applicants submit that the exchange which took place between the Credit Suisse trader and the RBS trader, referred to in paragraph 306 above, probably led the Credit Suisse trader to propose narrower bid-ask spreads to a Credit Suisse client. However, as is apparent from the material in the file, the exchange between the Credit Suisse trader and the RBS trader took place after the exchange on the other chatroom. Moreover, there is nothing in the file to support the applicants’ claim that that exchange is linked to the exchange between the Credit Suisse trader and the RBS trader.
308However, even if those two examples of exchanges of information set out in Annex A.12 do concern a possible reduction in the spreads contemplated by the traders in question, it cannot reasonably be inferred from those examples that the exchanges of information relating to bid-ask spreads between the traders in question consistently benefitted customers, which would have made it possible to establish legitimate effects that are relevant and specifically related to the practice concerned and sufficiently significant.
309On the contrary, the applicants themselves acknowledge in the application that the exchange of information in chatrooms (outside the context of potential transactions) enables traders to manage trading risk and to set more accurate prices within a short space of time and, consequently, to reduce the risk associated with their role as market makers.
310While an exchange of information can indeed generate gains in efficiency and make banks more efficient, in particular by enabling them to compare their respective practices and improve their market positions, it goes without saying that such initiatives do not justify recourse to anticompetitive practices, such as the exchange of information which is confidential and strategic from the perspective of competition law, such as, in the present case, information relating to bid-ask spreads, customer orders, risk positions and the current or planned trading activities and positions of competing traders, which is relevant information for the purposes of pricing and expert risk management, namely the parameters in the light of which competition on the foreign exchange spot trading market is established.
311More generally, even if the exchanges of information at issue were to benefit customers as regards the prices which were offered to them, such an advantage would not in itself be such as to call into question whether those exchanges are sufficiently harmful to competition. It should be borne in mind that Article 101 TFEU, like the other competition rules of the Treaty, is designed to protect not only the immediate interests of individual competitors or consumers but also to protect the structure of the market and thus competition as such (judgment of 19 March 2015, Dole Food and Dole Fresh Fruit Europe v Commission, C‑286/13 P, EU:C:2015:184, paragraph 125). However, the applicants have in no way substantiated their claim that the exchanges of information benefitted the market more generally.
312Contrary to what the applicants claim, their burden of proving that it is permissible to have a reasonable doubt as to whether the exchanges of information at issue caused a sufficient degree of harm cannot be alleviated either in the light of the allegedly ‘atypical’ elements of those exchanges or in the light of the probative value of the evidence on which the Commission relied in the contested decision, since the probative value and sufficiency of that evidence have not, moreover, been called in question by the applicants.
313In the second place, it should be observed that, in support of their arguments set out in paragraphs 290 to 292 above, the applicants refer to the expert reports produced in Annexes A.5 and A.11 to the application (‘expert report 1’ and ‘expert report 2’ respectively).
314In that connection, it must be recalled that, according to the case-law, the probative value, inter alia, of the reports submitted – at a party’s request to support its claims – by a third party on the basis of the latter’s status as expert, is assessed in several respects. The author must ensure that he or she sets out his or her qualifications and experience and explains how they are relevant to his or her providing an opinion on the question under consideration. Moreover, in terms of its content the opinion must set out the reasons why it should be taken into consideration, whether as regards the reliability of the methodology used or the relevance of the answer given to that question for the purposes of the present case (judgment of 14 September 2022, Google and Alphabet v Commission (Google Android), T‑604/18, under appeal, EU:T:2022:541, paragraph 96).
315However, for the reasons set out below, the content of the expert reports cannot satisfy the second requirement set out in paragraph 314 above and, accordingly, support the applicants’ arguments.
316First, the findings set out, in particular, in expert report 1, according to which the exchange of information between traders directly benefits customers, are based on scientific papers which are not such as to corroborate the veracity of those conclusions. The passages from the papers, produced by the applicants during the administrative procedure before the Commission in response to its request for clarification, and also before the Court, in no way concern a scenario in which exchanges of commercially sensitive information, such as those in the present case, have purportedly favourable effects on narrower spreads, and still less on the reduction of the prices offered to customers. The same conclusion applies with regard to a publication cited by the applicants, in support of expert report 1, which, while establishing the link between narrower spreads and the possibility for a trader of closing his position at low cost, in no way finds there to be a link between the exchange of commercially sensitive information and lower prices offered to customers which are, logically, the only prices likely to benefit customers.
317Secondly, the analysis presented in expert reports 1 and 2 does not concern the situation in the present case – that is to say, exchanges of information within a closed circle of traders – but rather focuses more on general statements relating to exchanges of information between traders on the foreign exchange spot trading market or, furthermore, does not concern the confidential and strategic information that was exchanged in the present case, but refers to general information (‘market colour’, ‘spreads’, ‘aggregated and anonymous positions with respect to client orders’). More specifically, the reports refer to traders exchanging information in the chatroom at issue, in particular in order to rule out the possibility of an anticompetitive situation on account of the small market shares held by the traders and not in order to confirm the legitimacy of the effects, nature or aim of exchanges of commercially sensitive information within a restricted circle of traders.
318Contrary to what the applicants claim, expert report 2, which is relatively vague and unsubstantiated, is not based on any economic model establishing the legitimacy of the conduct which took place in the chatroom at issue where the information was exchanged within a closed group. This is a fortiori the case as regards expert report 1, which is not based on any other ‘scientific tool’ capable of supporting the claims relating to the purported legitimacy of the exchanges of information on customers or on the market more generally.
319In those circumstances, it must be held that the applicants have failed to demonstrate that the exchanges of information that took place in the chatroom at issue were of a legitimate or pro-competitive nature which may call into question whether they reveal a sufficient degree of harm to competition and, accordingly, their characterisation as a ‘restriction by object’. The Commission therefore did not make an error of assessment in rejecting the line of argument put to it in that regard during the administrative procedure.
320It follows from the foregoing that the second part of the second plea must be rejected.
321In addition, in so far as the applicants refer in the heading of the present plea to a failure to provide sufficient reasons as regards the characterisation of the online exchanges of information as a ‘restriction by object’, and even if such a reference, which is unsubstantiated, were to constitute an admissible argument within the meaning of Article 76(d) of the Rules of Procedure, it must be held that the Commission did not infringe the obligation to state reasons under the second paragraph of Article 296 TFEU. It is apparent from the examination of the present plea that the explanations provided in the contested decision enabled the applicants to understand the Commission’s reasoning concerning that classification and the Court to exercise its power of review in that regard, thus satisfying the requirements of that provision (see, in that regard, judgment of 23 November 2023, Ryanair v Commission, C‑210/21 P EU:C:2023:908, paragraph 105 and the case-law cited).
322Consequently, the second plea must be rejected in its entirety.
3.The third plea, alleging infringement of Article 101 TFEU and failure to state sufficient reasons as to the finding of a single and continuous infringement made in the contested decision
323In their third plea, the applicants challenge the finding of a single and continuous infringement made in the contested decision. This third plea consists of two parts. The first part alleges a lack of evidence and insufficient reasoning as to the existence of an overall plan pursuing a common objective to which Credit Suisse intended to contribute, of which it was aware or which it should have foreseen. The second part alleges an error of law concerning the classification of the underlying understanding as an element of the single and continuous infringement, which may be subject to different degrees of participation.
324In that connection, it is clear from well-established case-law that an infringement of Article 101(1) TFEU may be the result not only of an isolated act, but also of a series of acts or of continuous conduct, even if one or more aspects of that series of acts or continuous conduct could also, in themselves and taken in isolation, constitute an infringement of that provision. Accordingly, if the different actions form part of an ‘overall plan’ because their identical object distorts competition within the internal market, the Commission is entitled to impute responsibility for those actions on the basis of participation in the infringement considered as a whole (see judgment of 16 June 2022, Sony Corporation and Sony Electronics v Commission, C‑697/19 P, EU:C:2022:478, paragraph 62 and the case-law cited).
325An undertaking which has participated in such a single and continuous infringement, by its own conduct, which meets the definition of an ‘agreement’ or ‘concerted practice’ having an anticompetitive object within the meaning of Article 101(1) TFEU and was intended to help bring about the infringement as a whole, may also be responsible for the conduct of other undertakings in the context of that infringement throughout the period of its participation in the infringement. That is the position where it is shown that the undertaking intended, through its own conduct, to contribute to the common objectives pursued by all the participants and that it was aware of the offending 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 take the risk (see, to that effect, judgment of 24 June 2015, Fresh Del Monte Produce v Commission and Commission v Fresh Del Monte Produce, C‑293/13 P and C‑294/13 P, EU:C:2015:416, paragraph 157 and the case-law cited).
326On the other hand, if an undertaking has directly taken part in one or more of the forms of anticompetitive conduct comprising a single and continuous infringement, but it has not been shown that that undertaking intended, through its own conduct, to contribute to all the common objectives pursued by the other participants in the cartel or that it was aware of all the other offending conduct planned or put into effect by those other participants in pursuit of the same objectives, or that it could reasonably have foreseen all that conduct and was prepared to take the risk, the Commission is entitled to attribute to that undertaking liability only for the conduct in which it had participated directly and for the conduct planned or put into effect by the other participants, in pursuit of the same objectives as those pursued by the undertaking itself, where it has been shown that the undertaking was aware of that conduct or was able reasonably to foresee it and prepared to take the risk (see judgment of 24 June 2015, Fresh Del Monte Produce v Commission and Commission v Fresh Del Monte Produce, C‑293/13 P and C‑294/13 P, EU:C:2015:416, paragraph 159 and the case-law cited).
327That cannot, however, relieve the undertaking of liability for conduct in which it has taken part or for conduct for which it can undeniably be held responsible. Nor is the fact that an undertaking did not take part in all aspects of an anticompetitive arrangement or that it played only a minor role in the aspects in which it did participate material for the purposes of establishing the existence of an infringement on its part, given that those factors need to be taken into consideration only when the gravity of the infringement is assessed and only if and when it comes to determining the fine (judgment of 6 December 2012, Commission v Verhuizingen Coppens, C‑441/11 P, EU:C:2012:778, paragraph 45).
328In addition, to the extent that a finding of a single and continuous infringement leads to an undertaking being held liable for an infringement of competition law, it should be noted that, in the field of competition law, where there is a dispute as to the existence of an infringement, it is for the Commission to prove the infringements found by it and to adduce evidence capable of demonstrating to the requisite legal standard the existence of the circumstances constituting an infringement (see judgments of 22 November 2012, E.ON Energie v Commission, C‑89/11 P, EU:C:2012:738, paragraph 71 and the case-law cited, and of 28 November 2019, ABB v Commission, C‑593/18 P, not published, EU:C:2019:1027, paragraph 38 and the case-law cited).
329It follows from the case-law cited in paragraphs 324 to 328 above that three factors are decisive for the purposes of finding that an undertaking participated in a single and continuous infringement. The first concerns the very existence of the single and continuous infringement. The various forms of conduct in question must form part of an ‘overall plan’ with a single objective. The second and third factors concern whether or not the single and continuous infringement can be attributed to an undertaking. In order for that infringement to be attributed to it, that undertaking must, first, have been aware of the offending conduct planned or put into effect by other undertakings in pursuit of the same objectives or have reasonably been able to foresee that conduct and have been prepared to take the risks, and, second, have intended, through its own conduct, to contribute to the common objectives pursued by all the participants.
(a)The first part of the third plea, alleging lack of evidence and failure to state sufficient reasons as to the existence of an overall plan pursuing a common objective to which Credit Suisse intended to contribute, of which it was aware or which it should have foreseen
330In the first part of the third plea, the applicants claim that the Commission failed to prove, and failed to state sufficient reasons for, the existence of an overall plan pursuing a single anticompetitive objective, to which Credit Suisse intended to contribute, of which it was aware, or which it should have foreseen.
(1)The existence of an overall plan pursuing a common objective
331First, in recitals 481 to 484 and 488 to 491 of the contested decision, the Commission addressed the classification of a single and continuous infringement by examining the existence of an overall plan pursuing a common objective. It considered, in essence, that (i) the extensive and recurrent exchanges of information, (ii) the occasional instances of coordination and (iii) the underlying understanding constituted that infringement, which pursued a common objective of mitigating the normal uncertainty inherent in the foreign exchange spot trading market in order to reduce the risk and comfort the participating undertakings in their pricing and risk management decisions, in order that they would not compete independently. Moreover, the existence of an ‘overall plan’ was demonstrated by objective elements, namely the modus operandi of the chatroom at issue and the continuity in respect of the participating traders and the banks involved.
332The applicants complain that, as the basis for finding that there existed an overall anticompetitive plan, the Commission used the modus operandi of the chatroom at issue, namely the fact that the participating traders were in regular contact in the ‘private’ chatroom and that participation in that chatroom was ‘by invitation only’. They claim that the use of a multilateral chatroom was common and well known on the foreign exchange spot trading market. They claim that such chatrooms are ‘private’ in nature only in the sense that persons could participate in them after being invited by an administrator, who can be any other trader with access to a Bloomberg chatroom. They add that the composition of those chatrooms could frequently change and dispute the Commission’s claim in recital 497 of the contested decision there was a ‘high level of continuity in the participation of the persons concerned’ on the ground that there is nothing in the Commission’s file to indicate that the traders constituted a ‘closed group’. Moreover, they submit that, even if the traders exchanged information using an ‘almost daily’ pattern of communications, such a model corresponds to standard market-making conduct and, therefore, does not serve to prove the existence of an ‘overall plan’.
333The Commission disputes the applicants’ arguments.
334It should be recalled that several criteria have been identified by the case-law as relevant for assessing whether there is a single infringement, namely the identical nature of the objectives of the practices at issue, the identical nature of the goods or services concerned, the identical nature of the undertakings which participated in the infringement and the identical nature of the detailed rules for its implementation (see, to that effect, judgment of 19 December 2013, Siemens and Others v Commission, C‑239/11 P, C‑489/11 P and C‑498/11 P, not published, EU:C:2013:866, paragraph 243; see also judgment of 17 May 2013, Trelleborg Industrie and Trelleborg v Commission, T‑147/09 and T‑148/09, EU:T:2013:259, paragraph 60 and the case-law cited). Furthermore, whether the natural persons involved on behalf of the undertakings are identical and whether the geographical scope of the practices at issue is identical are factors which may be taken into consideration for the purposes of that examination (judgment of 17 May 2013, Trelleborg Industrie and Trelleborg v Commission, T‑147/09 and T‑148/09, EU:T:2013:259, paragraph 60).
335In that connection, it is apparent from the Commission’s findings in the contested decision, based, in particular, on discussions prior to and during the period of Credit Suisse’s participation in the infringement at issue, that, during that period, the objective pursued by the other conduct comprising, according to the Commission, the single and continuous infringement, namely, in particular, the occasional instances of coordination and the underlying understanding, consisted, in essence, in mitigating the normal uncertainties inherent in the foreign exchange spot trading market, and thus comforting traders in their pricing and risk management decisions, which is not disputed by the applicants. Moreover, as is apparent from paragraphs 235 to 254 above, the exchanges of commercially sensitive information, in which Credit Suisse participated, shared that same objective.
336Furthermore, the discussions that took place, in particular, during the period of Credit Suisse’s participation support the assessments made in the contested decision that, in essence, objective elements confirmed that the anticompetitive conduct adopted by the participants was linked and that that conduct contributed to the overall plan pursuing the anticompetitive objective described by the Commission.
337In that regard, it must be stated that, first, the conduct in question followed the same modus operandi, namely daily and frequent discussions on commercially sensitive information in the chatroom at issue. Moreover, that conduct took place within a ‘private’ chatroom. The applicants do not dispute, as they confirmed at the hearing, that access to the chatroom at issue was by personal invitation only (see, for example, recitals 107 and 111 of the contested decision).
338Secondly, the conduct in question involved a stable group of undertakings – namely Barclays, HSBC, RBS and UBS – and took place between the same natural persons involved on behalf of those undertakings during parallel or adjacent periods. That group was enlarged once the trader from one of those banks changed employer and took up a position with Credit Suisse, leading to Credit Suisse’s participation in the exchanges of commercially sensitive information in the chatroom at issue throughout the relevant period.
339The applicants’ vague and unsubstantiated arguments, which simply seek to ‘justify’ the conduct which took place in the chatroom at issue by reference to the allegedly common and well-known nature of the use of multilateral chatrooms or the allegedly ordinary nature of the daily pattern of communications, do not detract from the finding that there was indeed a private and stable group of persons and undertakings participating in the exchanges in question.
340Moreover, even if the participating traders were members of other chatrooms, chatrooms being widely used on the foreign exchange spot trading market, or the composition of those chatrooms changes frequently, the fact remains that the entirety of the exchanges of information analysed in the contested decision makes it possible to establish the identity of the traders who participated in the chatroom at issue throughout the duration of the infringement.
341Thirdly, the existence of an overall plan pursuing a single anticompetitive objective is also confirmed by other material in the file. As is apparent from the entirety of the contemporaneous evidence gathered by the Commission, all the conduct in question related to the same products, namely G10 currencies. That assessment is illustrated by a number of discussions analysed by the Commission in the contested decision, according to which, in the course of the same day, the same traders exchanged information about those currencies (see, for example, the discussions referred to in paragraphs 92, 101 and 132 above).
342Since the elements listed in paragraphs 337 to 341 above are relevant in assessing whether conduct forms part of an overall plan and is part of a single infringement under the case-law cited in paragraph 334 above, the Commission was right to conclude that there was an overall plan pursuing a common objective.
343Moreover, as regards the alleged legitimate and pro-competitive reasons for the exchanges of information, in so far as these reiterate the arguments put forward in the second part of the second plea analysed above, reference is made to the reasoning set out in paragraphs 287 to 319 above for the purposes of rejecting those reasons on the same grounds.
344Accordingly, the applicants’ line of argument seeking to invalidate the Commission’s finding as to the existence of an overall plan pursuing a common objective must be rejected.
(2)Knowledge of and intention to contribute to the common objective pursued
345In recitals 500 to 505 of the contested decision, the Commission examined Credit Suisse’s intention to contribute to the common objective and its knowledge of the offending conduct of the other participants. The Commission considered, in essence, first, that the Credit Suisse trader intended to contribute to that objective in so far as he received and provided commercially sensitive information in the chatroom at issue and, secondly, that he was aware of the exchanges of information and the underlying understanding and was aware or, in any event, should have reasonably foreseen that the exchanges of information allowed the traders to identify opportunities for coordination. Moreover, in recitals 507 to 538 of the contested decision, the Commission refuted several arguments of Credit Suisse.
346Lastly, in so far as concerns Credit Suisse’s liability for the single and continuous infringement, in recitals 539 to 547 of the contested decision the Commission excluded that bank’s liability for the underlying understanding and for the occasional instances of coordination, but held it liable for the extensive and recurrent exchanges of current or forward-looking commercially sensitive information, which formed part of that infringement.
347The applicants criticise the Commission for concluding that their trader was aware of an anticompetitive objective, or even of an overall anticompetitive plan, on the basis, first, of the findings that some discussions suggest that (i) that trader was aware of the identity of the administrator and of the members of the chatroom at issue (recital 527 of the contested decision); (ii) he was called by his nickname when he re-registered on that chatroom (recital 525 of the contested decision); (iii) he neither joined that chatroom nor was invited by chance (recital 155 of the contested decision); (iv) he had engaged in the same practices since rejoining that chatroom or he must have been aware that the object and rules of that chatroom remained unchanged (recital 521 of the contested decision); and, second, of the settlement submission of another bank, which was not an addressee of the contested decision but for which the Credit Suisse trader worked previously. In that regard, they argue that the Commission erred in law in relying on evidence of discussions and the settlement submission of another bank which predate the period of Credit Suisse’s participation in the infringement.
348The Commission disputes the applicants’ arguments.
349In the first place, on the one hand, it should be noted that most of the Commission’s findings in the recitals of the contested decision referred to in paragraph 347 above are based on the discussions of 7 February 2012, which therefore took place during the relevant period. On the other hand, it should be noted that the Commission rightly considered, in recital 519 of the contested decision, that the previous employment of the Credit Suisse trader formed part of the context and that it could take that fact into account in order to establish that Credit Suisse was aware of the infringement. As the Commission observed in its defence, it is clear from the case-law that events which occurred outside of the infringement period alleged against an undertaking constitute elements that form part of the body of evidence on which the Commission may rely in order to prove the anticompetitive nature of that undertaking’s conduct (see, to that effect, judgment of 2 February 2012, Denki Kagaku Kogyo and Denka Chemicals v Commission, T‑83/08, not published, EU:T:2012:48, paragraph 188).
350In the second place, it should be borne in mind that the finding of the existence of a single infringement is separate from the question whether liability for that infringement as a whole is imputable to an undertaking (see, to that effect, judgment of 26 September 2018, Infineon Technologies v Commission, C‑99/17 P, EU:C:2018:773, paragraph 174). Moreover, whether the single and continuous infringement as a whole is attributable to an undertaking must be assessed in the light of two factors, namely, first, the intentional contribution of that undertaking to the common objectives pursued by all the participants and, secondly, its knowledge of the infringing conduct planned or implemented by other undertakings in pursuit of the same objectives or the fact that it could reasonably have foreseen it and had been prepared to take the risk (see paragraphs 324 to 329 above).
351In the present case, in the light of the express terms of recitals 543, 545 and 547 of the contested decision, read in conjunction with the operative part of that decision, it must be held that, as the Commission confirmed at the hearing in response to a question put by the Court, the Commission does not attribute liability for the single and continuous infringement as a whole to Credit Suisse, but attributes liability for that infringement to Credit Suisse in so far as that undertaking participated in one of its constituent elements, namely the extensive and recurrent exchanges of current or forward-looking commercially sensitive information.
352Given that Credit Suisse was held liable for the infringement at issue not as a whole, but in so far as it participated in one of its constituent elements, it is entirely unnecessary, in the present case, to assess whether it was aware of the collusive conduct of the other members of that infringement and intended, through its own conduct, to contribute to the common objectives pursued by all the participants (see, to that effect, judgment of 26 September 2018, Infineon Technologies v Commission, C‑99/17 P, EU:C:2018:773, paragraph 177).
353Accordingly, it follows that the analysis in the contested decision of the ‘intention to contribute to the common objective and awareness’ has no bearing on whether the alleged infringement is attributable to Credit Suisse. Consequently, the applicants cannot effectively challenge the validity of the reasons given in that analysis in order to call into question whether the infringement is attributable to that undertaking.
(3)Distancing from the exchanges of information
354In recitals 529 to 533 of the contested decision, in response to the arguments put forward by Credit Suisse during the administrative procedure, the Commission found that that bank’s trader not only failed to distance himself from the exchanges of information at issue, but also resumed his participation in them after being employed by Credit Suisse.
355In the context of their arguments, the applicants maintain that the Commission erred in criticising the Credit Suisse trader for not having distanced himself from the exchanges of information in the chatroom at issue in order to avoid liability. They submit that the traders were automatically connected to all the chatrooms of which they were members when they connected to the Bloomberg system; however, awareness of the discussions in which the trader did not participate cannot be inferred solely from that connection.
356The Commission disputes the applicants’ arguments.
357In that respect, as regards the question whether the Commission may accept as evidence of anticompetitive conduct discussions which took place in the context of a chat room to which an undertaking was connected, but in which it did not actively participate, it should be borne in mind that, according to settled case-law, passive modes of participation in an infringement, such as the presence of an undertaking in meetings at which, without that undertaking clearly opposing them, anticompetitive agreements were concluded 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 (see, to that effect, judgment of 22 October 2015, AC-Treuhand v Commission, C‑194/14 P, EU:C:2015:717, paragraph 31 and the case-law cited).
358Admittedly, the nature of the infringement at issue, which took the form of multilateral contacts between the various players in the chatroom at issue, cannot be regarded as ‘participation’ in a ‘meeting’ for the purposes of the case-law cited in paragraph 357 above. However, the applicants have failed to put forward any reasons which prevent the application of that case-law to discussions held in the context of online chatrooms. The fact that the traders of the banks involved in the actions are not physically present but are connected only remotely to the chatroom at issue in which the exchanges at issue took place is irrelevant, since, as is clear from paragraphs 77 to 173 above, the content of those exchanges is anticompetitive.
359Moreover, in so far as it must be understood that, by their reference to the judgments of 30 November 2011, Quinn Barlo and Others v Commission (T‑208/06, EU:T:2011:701, paragraph 51), and of 10 October 2014, Soliver v Commission (T‑68/09, EU:T:2014:867, paragraph 105), the applicants seek to challenge the application of the case-law relating to the requirement of public distancing although the anticompetitive nature of those exchanges has not been established, that line of argument must be rejected as unfounded. For the reasons set out in paragraphs 77 to 173 above, the anticompetitive nature of the exchanges in question has been established by the Commission.
360Accordingly, in the present case, the Commission was entitled to consider that Credit Suisse was aware of the exchanges which took place in the chatroom at issue, to which its trader was connected, even though that trader had not actively participated in some of the exchanges analysed in the contested decision. Consequently, in the absence of any public distancing from the practices concerned or any reporting thereof to the administrative authorities, Credit Suisse could be held liable.
361The position would have been different only if the applicants had been able to demonstrate, by means of evidence that was certain and precisely time-stamped, that Credit Suisse had not in fact been aware of the offending exchange or exchanges, or had become aware of them only after a period of time such that the information contained in those exchanges was no longer sensitive.
362In that connection, no evidence to that effect has been submitted by the applicants, which merely argue that, of the more than 100 exchanges analysed in the contested decision, 36 correspond to situations in which the Credit Suisse trader (i) was not connected, (ii) was not active and (iii) was not employed by Credit Suisse, without, however, referring to any specific exchange of information. Such claims, which are unsubstantiated and directed in a general and non-specific manner against the 36 exchanges analysed in that decision, do not call into question the Commission’s finding that that trader participated in all the exchanges of information analysed in the contested decision over the relevant period and, consequently, the Commission’s finding as to the liability of Credit Suisse incurred as a result of that participation.
363Accordingly, the Commission’s finding must be held to be free from any error, and the applicants’ arguments in that regard must be rejected.
364In addition, in so far as the applicants refer, in the heading of the present plea and in the heading and wording of the present part, to a failure to provide sufficient reasons, in essence, as regards the finding of a single and continuous infringement made in the contested decision, and even if such a reference, which is unsubstantiated, were to constitute an admissible argument within the meaning of Article 76(d) of the Rules of Procedure, it must be held that the Commission did not infringe the obligation to state reasons under the second paragraph of Article 296 TFEU. It is clear from the examination of the present plea that the explanations provided in the contested decision enabled the applicants to understand the Commission’s reasoning concerning that finding and the Court to exercise its power of review in that regard, thus satisfying the requirements of that provision (see, in that regard, judgment of 23 November 2023, Ryanair v Commission, C‑210/21 P EU:C:2023:908, paragraph 105 and the case-law cited).
365The first part of the third plea must therefore be rejected.
(b)The second part of the third plea, alleging an error of law as regards the classification of the underlying understanding as an element of the single and continuous infringement capable of being the subject of different degrees of participation
366In the second part of the third plea, the applicants argue, in essence, that the underlying understanding does not constitute a type of conduct separate from the exchange of information in which Credit Suisse may in part have participated, but constitutes the modus operandi of the alleged infringement. They conclude that the Commission erred, or even contradicted itself, in considering that the fact that Credit Suisse did not participate in the underlying understanding and in the occasional instances of coordination did not alter the finding concerning Credit Suisse’s participation in exchanges of information.
367The Commission disputes the applicants’ arguments.
368As a preliminary point, it should be noted that, in the second part of the third plea, the applicants reiterate, in a circumscribed manner and by reference to the arguments set out in the first plea, their claim that, in essence, the exchanges of information are not independent of the underlying understanding. That claim must be rejected for the same reasons as those set out in paragraphs 48 to 59 above.
369That said, it should be recalled that the Commission correctly established that the underlying understanding and the exchanges of information constituted separate aspects of the single and continuous infringement, which was implemented by means of the same modus operandi, namely discussions concerning commercially sensitive information on a daily and frequent basis within the chatroom at issue (see paragraphs 334 to 342 above). Moreover, it is clear from the evidence cited in the contested decision that the Credit Suisse trader participated in the exchanges of information which took place in the chatroom at issue, which the applicants expressly acknowledge in their pleadings, if only for the vast majority of those exchanges.
370Thus, in view of the extensive and recurrent exchanges of current or forward-looking commercially sensitive information analysed in the contested decision, the Commission’s error in classifying the discussion of 9 May 2012, analysed in recital 276 of the contested decision (see paragraphs 107 to 117 above), is irrelevant to the Commission’s finding that Credit Suisse participated in the single and continuous infringement.
371It should also be observed that, in their written pleadings before the Court, the applicants do not dispute the continuous nature of the single infringement. During the hearing, they nevertheless stated that, in view of the fact that certain exchanges of information, on which the Commission relied in the contested decision, do not serve to prove the existence of the single and continuous infringement, such a finding should also lead the Court to examine the continuous nature of the overall cartel of which those exchanges of information allegedly form part. However, as the Commission argued, in essence, at the hearing, in the absence of having been submitted in the context of the application, such an argument put forward by the applicants, which is not based on matters of fact or law which have come to light during the proceedings and which cannot be understood as constituting the amplification of their argument put forward in support of the third plea, is out of time and therefore inadmissible, pursuant to Article 84(1) of the Rules of Procedure, which essentially prohibits the submission of new arguments in the course of proceedings (see, to that effect, judgment of 16 December 2020, Fakro v Commission, T‑515/18, not published, EU:T:2020:620, paragraph 92).
372The Commission’s finding that Credit Suisse participated in the single and continuous infringement must therefore be upheld in so far as the exchanges of information are concerned.
373It follows from the foregoing that the second part of the third plea must be rejected and that that plea must therefore be rejected in its entirety.
4.The fifth plea, alleging infringement of the principle of sound administration
374In their fifth plea, the applicants complain that the Commission essentially infringed the principle of sound administration. They submit that the Commission failed to fulfil its obligation to conduct an impartial and diligent investigation of the evidence in the file, by relying selectively on that evidence and the statements of third parties to the hybrid proceedings, which the Commission cannot justify, it is argued, by its claim that it has dealt with numerous financial cases.
375Moreover, they argue that the principle of unfettered evaluation of evidence should not be understood as meaning that the Commission’s interpretation of the evidence is determinative, which would be manifestly inconsistent with Credit Suisse’s rights of defence. They claim that it follows from the principle that the presumption of innocence confers on the applicant the benefit of the doubt that, in the presence of unreliable evidence, the burden on the applicant to adduce evidence refuting the Commission’s allegations is itself also low.
376Furthermore, the applicants claim that the Commission, first, failed to comply with its obligations to examine critically the information provided in the context of the settlement procedure and, second, did not conduct a ‘de novo review of the evidence at its disposal, in accordance with the “tabula rasa” principle’ in the context of the standard procedure. The Commission also did not properly exercise its discretion in determining whether it was appropriate to adopt additional investigatory measures.
377The Commission disputes the applicants’ arguments.
378Article 41 of the Charter of Fundamental Rights of the European Union, relating to the ‘right to good administration’, provides that every person has the right, inter alia, to have his or her affairs handled impartially by the institutions of the European Union. That requirement of impartiality encompasses, on the one hand, subjective impartiality, in so far as no member of the institution concerned who is responsible for the matter may show bias or personal prejudice, and, on the other hand, objective impartiality, in so far as there must be sufficient guarantees to exclude any legitimate doubt as to bias on the part of the institution concerned (see judgment of 11 July 2013, Ziegler v Commission, C‑439/11 P, EU:C:2013:513, paragraph 155 and the case-law cited).
379The guarantees afforded by EU law in administrative proceedings, relating to the principle of sound administration, include the duty of the competent institution to examine carefully and impartially all the relevant aspects of the individual case (see judgment of 27 September 2012, Shell Petroleum and Others v Commission, T‑343/06, EU:T:2012:478, paragraph 170 and the case-law cited).
380As a preliminary point, it should be noted that, although the applicants claim an infringement of the rights of the defence in the context of the present plea, they have failed to put forward any specific arguments in that regard. Rather, their arguments appear to be based on a confusion between, on the one hand, compliance by the Commission with the requirement of objective impartiality and, on the other hand, the possibility of mounting an effective defence when the Commission’s complaints were sent to Credit Suisse. The applicants do not dispute that Credit Suisse had the opportunity to exercise its rights of defence effectively in accordance with all the procedural guarantees relating to the effective exercise of those rights in the context of the standard administrative procedure before the adoption of the contested decision, both in writing and orally, and therefore to challenge the facts and evidence on which the Commission relies.
381That said, the applicants maintain that the ‘concerns’ and clarifications put forward by Credit Suisse during the administrative procedure were ‘ignored’ by the Commission, thus allegedly leading the latter to infringe its obligation of impartiality when conducting a ‘de novo review of the evidence at its disposal, in accordance with the “tabula rasa” principle’. However, none of the arguments put forward by the applicants establishes that the Commission did not offer, in the present case, all the guarantees to exclude any legitimate doubt as to its impartiality in the examination of the case concerning Credit Suisse, and in particular in the examination of the arguments and evidence which Credit Suisse was able to submit in the context of the exercise of its rights of defence during the standard administrative procedure. The fact that the Commission did not find those ‘concerns’ and clarifications which Credit Suisse put forward during the administrative procedure convincing falls within the scope of the examination of the merits of the assessments made by the Commission and not within the scope of the review of compliance with the guarantees conferred by EU law in the standard administrative procedure.
382Thus, the applicants cannot validly criticise the Commission for any failure to exercise its discretion as to whether it was appropriate to adopt additional investigative measures in order to verify the statements made by Credit Suisse during the administrative procedure. In addition to being general in nature, that claim, in so far as the applicants thereby merely state that the use of additional investigative powers by the Commission would have been ‘appropriate in a complex market’, is based on the erroneous premiss that competitors are permitted, on the market in question, to exchange commercially sensitive information in the normal course of their business.
383In any event, in so far as the Commission has discretion as to whether it is appropriate to adopt investigative measures (see, to that effect, judgment of 2 February 2022, Scania and Others v Commission, T‑799/17, EU:T:2022:48, paragraph 154), it was under no obligation to use its investigative powers to request additional data or to consult an expert in order to verify Credit Suisse’s assertions.
384Moreover, it must be held that the requirement of impartiality when conducting the de novo review of the case relative to the settlement procedure, having regard to the party which decided not to enter into a settlement, cannot preclude findings of fact or legal classifications which are similar to those made in the settlement decision, if they are supported by the evidence available to the Commission (see, by contrary inference, judgment of 2 February 2022, Scania and Others v Commission, T‑799/17, EU:T:2022:48, paragraph 149).
385In that connection, the applicants’ claims concerning the evidence relied on by the Commission in the contested decision do not establish any bias on the part of the latter. Indeed, the applicants merely claim that the Commission selected the evidence in accordance with its pre-established version of the case, without providing any specific information in that regard. Moreover, they confine themselves to arguing that the Commission disregarded the other evidence produced by Credit Suisse which allegedly called into question the Commission’s findings in the contested decision, repeating in a single sentence an argument identical to that rejected in the context of their first plea.
386Furthermore, in so far as the applicants reiterate their arguments, put forward in the context of the first plea, alleging that the evidence submitted in order to obtain leniency and the documents which led to the settlement raise doubts as to the rules of the underlying understanding identified by the Commission, it should be recalled that it is not necessary for the Court to examine them for the reasons set out in paragraph 58 above.
387In addition, since the Commission established, on the basis of a sufficient body of evidence which is specific to the present case, the existence of the infringement and Credit Suisse’s participation in it, the fact that some of the Commission’s allegations concerning the objective pursued by the exchanges of information were taken from a sentence in a statement made by a third party to the administrative procedure cannot reveal any infringement of the principle of sound administration. Finally, the applicants’ claim that, in essence, the presumption of innocence implies that, in the presence of unreliable evidence, the burden on the applicants to adduce evidence refuting the Commission’s allegations is low must be rejected, since the applicants have failed to demonstrate that the evidence on which the Commission relied was unreliable.
388For the foregoing reasons, the fifth plea must be dismissed in its entirety.
5.Conclusion on Article 1 of the contested decision
389It follows from an examination of the first, second, third and fifth pleas, relied on in support of the claims for annulment of Article 1 of the contested decision, pursuant to which the Commission found that Credit Suisse had infringed Article 101(1) TFEU by participating in a single and continuous infringement, that those pleas are unfounded and that, consequently, that head of claim must be rejected.
B.Article 2(a) of the contested decision
1.The fourth plea, alleging infringement of Article 23 of Regulation No 1/2003, the Guidelines on the method of setting fines, the principles of proportionality and equal treatment, and the obligation to state reasons
390In their fourth plea, the applicants claim that the Commission made several errors at various stages of calculating the amount of the fine imposed on Credit Suisse. The plea is divided into five parts. The first part alleges error in the calculation of the proxy. The second part alleges error relating to the reduction of the fine granted on account of mitigating circumstances. The third part alleges error relating to an overestimation of the fine on account of the factor relating to the gravity of the infringement. The fourth part alleges infringement of the principle of equal treatment. Lastly, the fifth part alleges failure to state sufficient reasons in the contested decision as regards the proportionality of the calculation of the fine imposed on the applicants by comparison with that imposed on the other participants in the infringement.
391As a preliminary point, it should be noted that the system of judicial review of Commission decisions relating to proceedings under Articles 101 and 102 TFEU consists in a review of the legality of the acts of the institutions provided for by Article 263 TFEU (see judgment of 26 September 2018, Infineon Technologies v Commission, C‑99/17 P, EU:C:2018:773, paragraph 47 and the case-law cited).
392As regards the scope of judicial review provided for in Article 263 TFEU, it extends to all the elements of Commission decisions relating to proceedings applying Articles 101 and 102 TFEU which are subject to in-depth review by the Courts of the European Union, in law and in fact, in the light of the pleas raised by the applicant and taking into account all the relevant evidence submitted by the latter (see judgment of 26 September 2018, Infineon Technologies v Commission, C‑99/17 P, EU:C:2018:773, paragraph 48 and the case-law cited).
393In carrying out such a review, the Courts cannot use the Commission’s margin of discretion – either as regards the choice of factors taken into account in the application of the criteria mentioned in the Guidelines on the method of setting fines or as regards the assessment of those factors – as a basis for dispensing with the conduct of an in-depth review of the law and of the facts (judgment of 8 December 2011, Chalkor v Commission, C‑386/10 P, EU:C:2011:815, paragraph 62).
394It should be borne in mind, however, that the Courts of the European Union cannot, in reviewing the legality of acts under Article 263 TFEU, substitute their own reasoning for that of the author of the contested act (see, to that effect, judgment of 24 January 2013, Frucona Košice v Commission, C‑73/11 P, EU:C:2013:32, paragraph 89 and the case-law cited).
395First, in recitals 574 to 660 of the contested decision, the Commission stated that Credit Suisse had intentionally infringed Article 101 TFEU and that, as a result, the Commission intended to impose on it a fine in accordance with the methodology set out in the Guidelines on the method of setting fines.
396Secondly, in determining the value of sales which serves as the starting point for the calculation of the basic amount of the fines, the Commission found that the relevant products, namely G10 currencies, did not generate sales in the usual sense of the term, and were therefore not directly traceable in the accounts of the participating undertakings. Consequently, it was appropriate in the present case to calculate a proxy. That proxy was based, first, on the annualised notional amounts corresponding to the most traded currency pairs, including one of the currencies of the European Economic Area (EEA) (the euro, the Danish krone, the British pound, the Norwegian krone and the krona), of the transactions which took place with counterparties located in the EEA during the months of the undertakings’ participation in the infringement; and, second, on an adjustment factor consisting of a bid-ask spread reflecting the value of revenues from market-making activities and the value of revenues from trading on own account.
397Thirdly, the Commission calculated the basic amount of the fine for Credit Suisse by taking a proportion of 16% of the proxy in respect of the gravity of the infringement at issue, multiplied by a duration coefficient specific to Credit Suisse, namely 0.42 years, and, finally, by adding an additional amount of 16% for deterrence purposes.
398Fourthly, the Commission adjusted the basic amount calculated for Credit Suisse. It did not find there to be any aggravating circumstances. By contrast, on account of mitigating circumstances, the Commission deducted 2% from the basic amount calculated for Credit Suisse due to the absence of its liability for the underlying understanding, on the one hand, and the absence of liability for the occasional instances of coordination, on the other hand, that is to say a total deduction of 4%. Finally, the Commission did not increase the fine for deterrence purposes.
(a)The first part of the fourth plea, alleging error in the calculation of the proxy
399The applicants dispute the determination of the value of sales in the contested decision, arguing that the proxy on which the Commission relied significantly and arbitrarily overstates Credit Suisse’s value of sales and the economic importance of the infringement at issue, thereby departing from the concept of ‘value of sales’ set out in the Guidelines on the method of setting fines.
400The Commission disputes those arguments.
401Under point 13 of the 2006 Guidelines on the method of setting fines: ‘in determining the basic amount of the fine to be imposed, the Commission will take the value of each undertaking’s sales of goods or services to which the infringement directly or indirectly … relates in the relevant geographic area within the EEA.’
402In the introductory section, the Guidelines on the method of setting fines provide, in point 6, that ‘the combination of the value of sales to which the infringement relates and of the duration of the infringement is regarded as providing an appropriate proxy to reflect the economic importance of the infringement as well as the relative weight of each undertaking in the infringement.’
403Thus, point 13 of the Guidelines on the method of setting fines pursues the objective of adopting as the starting point for the calculation of the fine imposed on an undertaking an amount which reflects the economic significance of the infringement and the size of the undertaking’s contribution to it. Consequently, while the concept of the ‘value of sales’ referred to in that point cannot extend to encompassing sales made by the undertaking in question which do not fall within the scope of the alleged cartel, it would however be contrary to the goal pursued by that provision if that concept were understood as applying only to turnover achieved by the sales in respect of which it is established that they were actually affected by that cartel (judgment of 11 July 2013, Team Relocations and Others v Commission, C‑444/11 P, not published, EU:C:2013:464, paragraph 76).
404It should also be recalled that, according to point 15 of the Guidelines on the method of setting fines, in determining the value of sales by an undertaking, the Commission will take that undertaking’s best available figures.
405In recitals 584 to 620 of the contested decision, the Commission determined the value of sales by means of a proxy in so far as foreign exchange spot trading in G10 currencies does not generate sales in the usual sense of the term.
406In that regard, the Commission deemed it appropriate to use as its proxy Credit Suisse’s annualised notional amounts corresponding to the most traded currency pairs, including one of the currencies of the EEA, namely the euro, the British pound, the Norwegian krone, the krona or the Danish krone, during the months of Credit Suisse’s participation in the infringement at issue, which took place with counterparties located in the EEA, those amounts next being multiplied by the uniform adjustment factor consisting of a bid-ask spread reflecting the value of revenues from, first, market-making activities (‘the adjustment factor related to market-making’) and, secondly, activities of trading on own account (‘the adjustment factor related to trading on own account’). The Commission therefore took the amount of EUR 381888991 as Credit Suisse’s value of sales.
(1)Credit Suisse’s annualised notional amounts
407In recital 628 of the contested decision, as regards Credit Suisse’s argument that the notional amounts relevant for the calculation of the proxy included trades in which Credit Suisse had bought liquidity and not those in which Credit Suisse had provided it, the Commission stated that it was irrelevant. It observed, in that regard, that the way traders chose to hedge and manage the risk linked to the open positions in their portfolio should not be taken into account and that seeking liquidity, holding, increasing or offloading open positions was the essence of a trader’s activity and were activities embedded in the bid and ask prices that traders submit to their clients.
408The applicants submit that the methodology adopted by the Commission in the contested decision wrongly includes notional amounts associated with transactions in which Credit Suisse obtains liquidity, in so far as such inclusion attributes to Credit Suisse the notional amounts of other banks. They claim that those amounts represent costs of providing the market-making service and, therefore, cannot be included in the calculation of the value of sales, which seeks to determine income. The Commission thus overstated the economic importance of the alleged infringement.
409The Commission disputes the applicants’ arguments.
410As a preliminary point, it should be observed that the applicants are not calling into question the very principle of annualising notional amounts, but are seeking to exclude Credit Suisse’s currency purchase transactions from them.
411That said, it should be noted that, when applying the methodology for calculating the proxy, in the absence of amounts directly traceable in the accounts of the participating undertakings, the Commission sought to proceed in a manner comparable to the calculation of the value of sales, in so far as the latter is determined by multiplying the volume of units sold by their selling price (recitals 585 and 586 of the contested decision). Accordingly, in the present case, that volume corresponded to the annualised notional amounts.
412It is true that, when calculating those notional amounts, the Commission did not differentiate between the notional amounts corresponding to transactions carried out by Credit Suisse for the purpose of providing or supplying liquidity. However, the absence of such differentiation when calculating the annualised notional amounts must be regarded as justified by the particularities of the foreign exchange spot trading market, which are not disputed by the applicants, whereby revenues depend on the trading of a currency pair as a combined transaction, namely a sale and a purchase. In particular, as the Commission rightly argues, since a trader on the foreign exchange spot trading market seeks to make a profit on both his or her purchase and his or her sale transactions, a transaction aimed at buying liquidity in the market forms an integral part of his or her normal activity on that market and, in particular, his or her revenues.
413Contrary to what the applicants claim, Credit Suisse’s currency purchase transactions cannot be excluded from the notional amounts for the purpose of determining the proxy on the ground, in essence, that they do not constitute income in the strict sense of the term, or on the ground that they attribute to that bank notional amounts of other banks.
414In that regard, the applicants cannot validly rely on a methodology set out in the note attached to Annex C.5 to the reply. In their reply, they merely make a general reference to that annex. It is not the task of the Court to search through all the matters in the file for those matters capable of supporting their claim (see, to that effect, judgment of 27 September 2006, Roquette Frères v Commission, T‑322/01, EU:T:2006:267, paragraph 209).
415Moreover, the applicants do not dispute that the transactions in which Credit Suisse, along with other banks, sought to obtain liquidity took place on the market affected by the cartel, on which, moreover, the search for liquidity is part of the management of risks linked to stocks at portfolio level. If it is not to minimise artificially the economic importance of the infringement committed by that bank and therefore lead to the Commission imposing fines with no real link to the scope of the cartel in question, the proxy – like the value of sales – cannot be calculated solely on the basis of the transactions which it is established were actually affected by that cartel, but may be calculated, as in the present case, on the basis of all the transactions falling within the scope of that cartel, as follows from the case-law cited in paragraph 403 above.
416The Commission was therefore entitled, in calculating the proxy, to use Credit Suisse’s annualised notional amounts relating both to currency purchase transactions and to currency sales transactions.
(2)The uniform adjustment factor used in the contested decision
417In the contested decision, in order to calculate the proxy, the Commission essentially applied a methodology consisting of multiplying the annualised notional amounts by an adjustment factor. As regards that adjustment factor, the Commission considered that it comprised, on the one hand, an adjustment factor related to market-making and, on the other hand, an adjustment factor related to trading on own account.
(i)The adjustment factor related to market-making
418In the contested decision, the Commission took the view that, as regards the value of revenues from market-making, the adjustment factor should be set at 50% of the bid-ask spread in order to take account of the fact that a full bid-ask spread depends on both parties to the transaction and that, consequently, the revenues made by a trader on a given transaction are half of that bid-ask spread. In order to reconstitute that spread, the Commission chose to use data from exchanges of information in the chatroom at issue during which bid-ask spreads concerning transactions of up to 500 million in the base currency were discussed, that is to say a sample of exchanges of information including bid-ask spreads covering a period from 2011 to 2012. It established a bid-ask spread range from 6.2 to 8.9 points, based on the methodology of setting (i) the lower end of the bid-ask spread range at the lowest of the average and the median of the minimum spreads, and (ii) the higher end of the range at the highest of the average and the median of the maximum spreads. That range was used to determine the appropriate bid-ask spread, which was estimated at 7 points since that level was close to the middle of the bid-ask spread resulting from those trades that involved any combination of G10 currencies including at least one EEA currency. Applying the reduction of half of the spread, it estimated that adjustment factor at 3.5 points (recitals 597, 598 and 604 to 609 and footnote 501).
419The applicants submit that the adjustment factor related to market-making is inappropriate and biased in that it uses higher bid-ask spreads which overstate the proxy for Credit Suisse’s value of sales and do not reflect the economic importance of the infringement. In their view, the sample selected by the Commission in the contested decision is limited and unrepresentative, all the more so since it does not cover the period of Credit Suisse’s participation in that infringement. In their view, the Commission could not conclude that its methodology was appropriate on the basis that the fine range was accepted by the parties to the settlement procedure. In any event, although the Commission relied on the sample referred to above, it could have weighted the spreads to take account of the distribution of trades, in order to remedy the unrepresentative nature of that sample.
420At the hearing, the applicants confirmed that, during the administrative procedure, they provided the Commission with the ‘best available figures’ as opposed to the sample used by the Commission in the contested decision, namely the Bloomberg BFIX data, for the relevant period. In response to a measure of organisation of procedure, the applicants produced those data before the Court.
421The Commission disputes the applicants’ arguments. In its view, the sample of exchanges of information used in the contested decision constitutes the most appropriate and representative database, since it most closely approximates the economic activity of Credit Suisse to which the infringement at issue relates. That sample covers all the discussions in the chatroom at issue in which the exchanges of information took place as well as the specific financial products concerned by the infringement, and comes from the undertakings which actually participated in that infringement. The Commission submits that, in order to ensure equal treatment, it was not required to develop a tailor-made methodology for Credit Suisse, but took into account evidence dating back to the material time which concerned the other participating undertakings. Moreover, it considers that taking into account only transactions of 500 million of the base currency or less makes the sample representative of the trading activities of the participating undertakings. In any event, the annualised notional amounts already take account of the individual period of Credit Suisse’s participation in the infringement and its weight in it. The Commission submits that the applicants’ arguments should be rejected.
422At the hearing, the Commission reiterated its position that the sample of exchanges of information used in the calculation of the adjustment factor related to market-making in the contested decision was more representative than the Bloomberg BFIX data, since it was specific to the infringement at issue. In response to a measure of organisation of procedure, the Commission produced that sample which, in its view, constituted the ‘best available figures’ for that calculation, thus enabling the best possible approximation of the economic importance of the infringement to be established.
423In that connection, it should be observed that the parties agree on the relevance, in the present case, of taking into account half the bid-ask spread for the purpose of calculating the adjustment factor related to market-making. However, the parties disagree, in essence, as to whether the sample of discussions on bid-ask spreads used in the contested decision to calculate the adjustment factor related to market-making constitutes the ‘best available figures’ within the meaning of point 15 of the Guidelines on the method of setting fines.
424First, it should be borne in mind that, according to point 15 of the Guidelines on the method of setting fines, ‘in determining the value of sales by an undertaking, the Commission will take that undertaking’s best available figures’. Thus, in implementing the methodology which it lays down, it is incumbent on it, inter alia, to ensure that it takes account of the ‘best available figures’, subject to detailed review, in law and in fact, by the Courts of the European Union, as is apparent from the case-law cited in paragraph 393 above.
425Secondly, according to the case-law of the Court of Justice, in adopting rules of conduct such as the Guidelines on the method of setting fines and announcing by publishing them that they will henceforth apply to the cases to which they relate, the Commission imposes a limit on the exercise of its discretion and cannot depart from those rules under pain of being found, where appropriate, to be in breach of the general principles of law, such as equal treatment or the protection of legitimate expectations (see, to that effect, judgment of 28 June 2005, 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, paragraph 211).
426In the present case, it is apparent from the contested decision that, for the purposes of calculating the adjustment factor related to market-making, the sample used by the Commission concerned 16 exchanges of information. It is also clear from that decision that the sample concerned bid-ask spreads from discussions which took place in 2011 (two discussions on 24 June, one discussion on 11 July, two discussions on 10 August, one discussion on 20 September, one discussion on 5 October and one discussion on 21 October) and in 2012 (one discussion on 5 January, one discussion on 11 January, one discussion on 31 January, one discussion on 17 February, one discussion on 1 March, one discussion on 5 April, one discussion on 18 April and one discussion on 18 June), during which traders disclosed bid-ask spreads for the following currency pairs: GBP/USD (six discussions), EUR/CHF (one discussion), EUR/GBP (seven discussions), GBP/JPY (one discussion) and EUR/JPY (one discussion).
427In that regard, in the first place, it is appropriate to examine the applicants’ criticisms of the sample used by the Commission.
428First, as regards the period concerned by the bid-ask spreads derived from the sample used in the contested decision, it should be pointed out, as the applicants state, that it cannot be regarded as appropriate to take that period into account since it exceeds the relevant period.
429The taking into account of bid-ask spreads resulting from several discussions dating, in particular, from 2011, prior to the period of Credit Suisse’s participation in the infringement, is not relevant in determining the proxy resulting from transactions in the various currency pairs for which the notional annualised amounts relate to 2012.
430In the light of the volatility and liquidity of the foreign exchange spot trading market, exchange rates, of which the bid-ask spread is a component and which are influenced by those particular features of the market, change rapidly, as is also reflected in the sample used in the contested decision. In particular, it is apparent that the bid-ask spreads for trades involving the same amount and the same currency pair differed depending on the date. Accordingly, it is clear that the application of bid-ask spreads reported by traders in 2011 cannot, from an objective point of view, provide the Commission with precise information on the actual value of the revenues derived from market making by Credit Suisse in 2012, during the infringement period.
431In those circumstances, the taking into account of bid-ask spreads relating to various dates in 2011 must be considered to be inadequate and inappropriate for the purpose of reflecting the actual economic situation of Credit Suisse during the infringement period.
432Secondly, the sample of 16 exchanges used to calculate the adjustment factor related to market-making is limited by comparison with the number of transactions corresponding to annualised notional amounts taken into account by the Commission in order objectively to represent, in particular, the economic importance of the alleged infringement.
433In that regard, although that number of transactions was not indicated in the contested decision, given that the Commission relied solely on transactions of amounts less than, or equal to, 500 million of the base currency (recital 604) and that the notional amounts which it took into account total EUR 1060802752528 (recital 592), if those notional amounts are divided by the maximum volume of the transactions on which it relied (500000000), it must be concluded that the number of transactions corresponding to those notional amounts is at least 2121. Moreover, the applicants state that Credit Suisse executed 22000 transactions during the period of its participation in the alleged infringement, without this being challenged by the Commission.
434Thirdly, it should be recalled that, in arriving at the proxy, the Commission used the annualised notional amounts corresponding to the most traded currency pairs involving one of the EEA currencies, namely the euro, the British pound, the Norwegian krone, the krona or the Danish krone, during the months of the banks’ participation in the infringement at issue (recital 589 of the contested decision). However, as is apparent from the evidence relating to the sample of discussions on bid-ask spreads used in the contested decision, that sample included just five currency pairs out of forty possible combinations.
435That lack of representativeness of the currency pairs concerned by the sample of bid-ask spreads used in the contested decision is likely to have a significant effect on the assessment of the ‘best figures’ available when the contested decision was adopted, for the purposes of calculating the proxy for Credit Suisse’s value of sales. As is apparent from that sample, the various currency pairs concerned in the present case differ in terms of liquidity and the bid-ask spreads applied (for example, for 200 million of a EUR/GBP currency pair, the minimum bid-ask spread was between 10 and 12 whereas the maximum bid-ask spread was between 12 and 14, while for the same volume of a GBP/USD currency pair, the minimum bid-ask spread was 18 whereas the maximum was 20).
436In the light of the foregoing, it must be held that the small number of elements in the sample used by the Commission to calculate the adjustment factor in question, more than half of which do not relate to the relevant period (see paragraph 426 above), do not have a level of detail specific to each currency pair selected as relevant by the Commission in that calculation. Consequently, that sample cannot be regarded as providing data that were sufficient to ensure that all the relevant currency pairs were proportionately represented in the light of the logic of the adjustment factor related to market-making, that is to say the taking into account of the fact that trading revenues are embedded in the bid-ask spread applied to a trade in a currency pair involving an EEA currency.
437In the second place, the applicants must nevertheless show that, within the framework of the methodology which the Commission has lawfully determined, there were in fact better data than those used by the Commission and that those data are in fact available (see, to that effect and by analogy, judgment of 24 September 2019, HSBC Holdings and Others v Commission, T‑105/17, EU:T:2019:675, paragraph 324). Consequently, it must be determined whether the Bloomberg BFIX data proposed by the applicants to the Commission for the purpose of determining the adjustment factor related to market-making constituted the ‘best available figures’ for implementation of the methodology defined by the Commission in the contested decision.
438In that regard, first, it should be noted that, as the applicants stated at the hearing, the use of the Bloomberg BFIX data would have enabled the Commission to take into consideration thousands of data points relating to the relevant period and therefore examples of bid-ask spreads concerning the period of Credit Suisse’s participation in the infringement that were far more numerous and, consequently, more representative in the light of the significant volume of transactions which were taken into account by the Commission in calculating the proxy (see paragraphs 432 and 433 above). Taking those data into account would therefore have been such as to establish a proportionate correlation between, on the one hand, those spreads and the annualised notional amounts covering the same period, in accordance with the underlying logic of the calculation of the value of sales in the present case, and, on the other hand, the currency pairs included in the scope of the calculation of the adjustment factor in question.
439Secondly, as it argued at the hearing, the Commission disputes the appropriateness of the Bloomberg BFIX data. In its view, those data have limitations, in so far as, inter alia, they do not make it possible to filter out the bid-ask spreads actually applied by the banks involved in the infringement. The data are also, as the Commission argued at the hearing and in response to a measure of organisation of procedure decided upon by the Court, not specific to the infringement at issue, in particular inasmuch as they are not from the place where that infringement occurred and do not relate to the transactions executed by the banks involved in the infringement.
440However, while it is true that the Guidelines on the method of setting fines are based on taking into account the value of sales of the products concerned to which the infringement relates in determining the basic amount of fines to be imposed (judgment of 10 July 2019, Commission v Icap and Others, C‑39/18 P, EU:C:2019:584, paragraph 26), they also provide, in point 16, that, where the figures made available by an undertaking are incomplete or not reliable, the Commission may determine that value on the basis, inter alia, of any other information which it regards as relevant and appropriate. Thus, it does not follow from those guidelines that the data taken into account by the Commission must necessarily be specific to the infringement.
441The concept of ‘best available figures’, within the meaning of point 15 of the Guidelines on the method of setting fines, does not entail finding data capable of providing a perfectly accurate reflection of the economic importance of the infringement and of the relative weight of an undertaking in it, but consists in finding data which are of a better quality than the other data available, in the sense, in particular, that they are more consistent, complete and reliable for the purposes of reflecting that importance and weight (see, to that effect, judgment of 11 July 2014, RWE and RWE Dea v Commission, T‑543/08, EU:T:2014:627, paragraphs 230, 234 and 237).
442However, in view of the fact that the Bloomberg BFIX data make it possible to take into account all the currency pairs included among the currencies concerned by the calculation of the adjustment factor in question, provide a larger sample so as to reflect proportionately the annualised notional amounts to which that factor was applied, and relate to the period of Credit Suisse’s participation in the infringement, they must be regarded as more consistent, complete and reliable than those on which the Commission relied in order to reflect the economic importance of the infringement and Credit Suisse’s relative weight in it.
443It follows from the foregoing that the Bloomberg BFIX data proposed to the Commission during the administrative procedure and produced by the applicants before the Court constitute, contrary to what the Commission claims, the ‘best available figures’, capable of reflecting more accurately the proxy for Credit Suisse’s value of sales.
444The Commission cannot validly justify taking into account data which are not representative of the actual value of the revenues from market making on the basis of the requirement to ensure equal treatment of the participants in the infringement, since the method applied should be identical for all the parties to the infringement.
445It should be recalled that the principle of equal treatment requires that comparable situations must not be treated differently and that different situations must not be treated in the same way unless such treatment is objectively justified (see judgment of 12 November 2014, Guardian Industries and Guardian Europe v Commission, C‑580/12 P, EU:C:2014:2363, paragraph 51 and the case-law cited). However, the application of the principle of equal treatment cannot preclude the taking into account, in the present case, of the individualised data of the undertakings participating in the infringement, while applying the same methodology for the calculation of the proxy. The use, in the present case, of individualised bid-ask spread data relating to the period of Credit Suisse’s participation in the infringement is merely one means of implementing the methodology developed by the Commission and applicable to all the parties to that infringement, which does not alter it in substance but is capable only of increasing the accuracy of the proxy for that bank’s value of sales as regards the scale and extent of the bank’s activities during the period of its participation in the infringement.
446Moreover, the Court has already held that the Commission’s approach of taking into account, in calculating the value of sales, data from different periods in the light of the different periods of participation in the infringement by the undertakings concerned, in order, in particular, to take account of the true economic situation of those undertakings during the infringement period and to avoid the value of sales from underestimating the economic significance of the infringement, is consistent with the principle of equal treatment (judgment of 29 September 2021, Tokin v Commission, T‑343/18, EU:T:2021:636, paragraphs 110 to 114 and 118).
447Furthermore, contrary to what the Commission claims, nor can the fact that the settling parties agreed to the use of uniform data for the purpose of calculating the adjustment factor applied within the framework of the methodology developed by the Commission invalidate the conclusion that it is inappropriate and inadequate in relation to Credit Suisse.
448Consequently, in not using the ‘best available figures’ to calculate the adjustment factor to be applied to Credit Suisse for the purpose of calculating the proxy, the Commission failed to comply with the Guidelines on the method of setting fines and miscalculated the basic amount of the fine which it imposed on Credit Suisse.
(ii)The adjustment factor related to trading on own account
449In the contested decision, as regards the adjustment factor related to trading on own account, in so far as the traders exchanged information relating to their open risk positions, the Commission, after noting that none of the banks concerned was able to provide the data needed to estimate those commercial revenues and in view of the technical challenges inherent in calculating the proxy value of the revenues earned from activities of trading on own account, decided to rely, for the purposes of determining the bid-ask spread, on public sources relating to trades at the interdealer level (see paragraph 24 above). Thus, in order not to overstate those revenues, the Commission decided to calculate them by multiplying the notional amount by 10% of an appropriate bid-ask spread, the latter being, at its discretion, assessed at a level of between 0.6 and 1.5 points. On the basis of those considerations, it took into account a bid-ask spread set at 1 point, which it divided by the aforementioned 10%. It therefore estimated that adjustment factor at 0.1 points.
450The applicants complain that the Commission arbitrarily used the adjustment factor to take into account the revenue that the participating banks derived from holding open risk positions and that it applied 10% of an inter-trader bid-ask spread to represent this. They claim that the revenue generated by holding open risk positions is based on the potential increase in the value of a trader’s position and not on the spread that a trader ultimately has to pay. Consequently, they argue that using the bid-ask spread to take into account the alleged revenue does not reflect the revenue that Credit Suisse could have derived from that activity. They claim that, in any event, the Commission failed to demonstrate that the average revenue earned from activities of trading on own account is proportional to the bid-ask spread.
451The Commission disputes the applicants’ arguments.
452In that connection, as a preliminary point, it should be noted that the applicants do not dispute the statement in recital 610 of the contested decision that, first, the banks involved in the infringement at issue entered into trades not only as market makers but also on their own account and, second, all those trades generated revenues. The finding in recital 630 of that decision that ‘bid-ask spreads reflect the banks’ expectation of the risk associated with an open exposure’ is also not disputed.
453That said, it should be recalled that, on the foreign exchange spot trading market, the bid-ask spread constitutes the compensation which the trader receives, in particular for the risk that he or she bears by having a given currency in his or her trading book, as a result of the exposure to open risk positions on account of the velocity of exchange rate fluctuations in the market (see paragraph 21 above and recitals 35 and 36 of the contested decision, which are not contested by the applicants).
454Thus, it must be held that the approach adopted by the Commission to calculate the adjustment factor related to trading on own account, based on a bid-ask spread, must be regarded as being consistent with the reality of the foreign exchange spot trading market. The aforementioned approach taken by the Commission is also consistent with the logic underlying the choice of the proxy, in that it takes into account, in particular, the revenue generated by Credit Suisse’s activities of trading on own account.
455In those circumstances, the applicants, who merely argue that the bid-ask spread is not relevant for calculating the revenues from activities of trading on own account without, however, proposing a more appropriate alternative method, cannot validly rely on the Commission’s alleged misuse of the bid-ask spread to determine the revenues which Credit Suisse could have derived from activities of trading on own account.
456More generally, the applicants’ claim that, in the context of activities of trading on own account, traders pay, and do not earn, a spread must be regarded as counter-intuitive from the point of view of the commercial logic and economic rationale of trading on own account and, in particular, as stated in the contested decision, which is not disputed on this point by the applicants, ‘they can very well increase, decrease or close open risk positions by “earning the spread” from other traders … This is indeed the essence of being a good market maker and trader in general and shows that market making and proprietary trading are intertwined’ (recital 630).
457Finally, while the applicants formally challenge the proportionality of the average revenue earned from activities of trading on own account to the bid-ask spread on the basis of the ‘high frequency spreads’ in one study, they have not put forward any specific explanation in support of that challenge. Indeed, they merely claim that ‘high frequency trading is not relevant in the present case’. In any event, the applicants cannot validly complain of any reversal of the burden of proof on the part of the Commission, in so far as it has a margin of discretion when calculating the fine. Moreover, the applicants’ argument that it was arbitrary and inappropriate to apply 10% of an inter-trader bid-ask spread to determine the adjustment factor in question cannot succeed; that argument is based on a mere assertion which is in no way substantiated.
458It follows from the foregoing that the Commission correctly determined the adjustment factor related to trading on own account. The applicants’ arguments in that regard must therefore be rejected as unfounded.
(3)The alternative methodology proposed by the applicants for the calculation of the proxy
459The applicants take the view that the alternative methodology proposed by Credit Suisse during the administrative proceedings constitutes a more appropriate and reliable basis for determining the value of its sales. They argue that that methodology consists, in essence, in calculating the notional amounts for each currency pair and trading month combination taken from transactions in which Credit Suisse provided liquidity, which are multiplied by an adjustment factor relating to a bid-ask spread corresponding to half the bid-ask spread for each specific currency pair and trading month combination, on the basis of data from an independent third party.
460According to the applicants, that methodology makes it possible to obtain a more accurate approximation of the proxy. However, in the applicants’ submission, the Commission did not seriously examine the arguments and alternative methodology proposed by Credit Suisse. The applicants claim that the Commission failed to comply with its obligation to state reasons in that regard.
461The Commission disputes the applicants’ arguments.
462As a preliminary point, it should be recalled that the methodology proposed by the applicants differs from that applied in the contested decision, in that it excludes from the calculation the notional amounts relating to purchase transactions and the adjustment factor related to trading on own account, but uses a weighting of the bid-ask spreads relating to market-making activities.
463As is clear from paragraphs 410 to 416 above, the applicants wrongly seek to establish that the calculation of annualised notional amounts should take account only of currency sales transactions. Such a calculation would not take into account all the relevant transactions carried out by Credit Suisse, even if it were calculated on the basis of the specific currency pair and trading month combination. In any event, as is apparent from the expert report to which the applicants refer in support of their methodology, the alternative calculation of the notional amounts is not based on precise data which would make it possible to determine the transactions in which Credit Suisse had provided liquidity.
464Moreover, as regards the fact that the methodology for calculating the proxy excludes the value of revenues from proprietary activities, while not taking those revenues into account would make it possible to reduce the allegedly ‘inflated’ proxy, such exclusion does not take into account the economic importance of the infringement and the relative weight of Credit Suisse in that infringement, in so far as such revenues are a source of earnings for the banks and concern the very essence of their activities.
465As to the argument concerning the failure to state sufficient reasons based on the lack of explanation, on the part of the Commission, for not taking into account the alternative method proposed by the applicants, it should be noted that, as is apparent from recitals 626 to 633 of the contested decision, the Commission set out the reasons why it did not consider the alternative methodology proposed by the applicants to be appropriate. As to the remainder, it is clear that that argument is to a large extent indissociable from the examination of the merits of the assessment of that methodology. In any event, it should be recalled that, according to the case-law, the Commission cannot be required to explain in its decisions the reasons why, in relation to calculation of the amount of the fine, it did not adopt alternative approaches to the one in fact adopted in the contested decision (judgment of 15 July 2015, SLM and Ori Martin v Commission, T‑389/10 and T‑419/10, EU:T:2015:513, paragraph 206).
466The first part of the fourth plea must therefore be upheld solely in so far as the Commission failed to use the best available figures, in relation to Credit Suisse, in determining the adjustment factor related to market-making.
467That said, the Court considers it appropriate to continue its examination of the fourth plea in the application by addressing the third part and then the second, fourth and fifth parts.
(b)The third part of the fourth plea, alleging that the fine was overestimated as a result of the factor relating to the gravity of the infringement
468According to the Guidelines on the method of setting fines, the basic amount of the fine is to be related to a proportion of the value of sales, depending on the degree of gravity of the infringement, multiplied by the number of years of infringement (point 19). The assessment of the gravity of the infringement is to be made on a case-by-case basis for all types of infringement, taking account of all the circumstances of the case (point 20). As a general rule, the proportion of the value of sales taken into account will be set at a level of up to 30% of the value of sales (point 21). In order to decide on the level of the proportion of the value of sales to be considered in a given case, the Commission is to have regard to a number of factors, such as the nature of the infringement, the combined market share of all the undertakings concerned, the geographic scope of the infringement and whether or not the infringement has been implemented (point 22). In addition, since the most harmful restrictions of competition, such as horizontal price-fixing agreements, must be heavily fined, the proportion of the value of sales taken into account for such infringements will generally be set at the higher end of the scale (point 23).
469In recitals 634 to 637 of the contested decision, the Commission, in the light of the rules set out in paragraph 468 above, considered, in essence, that, having regard to the nature and the geographic scope of the infringement, the gravity factor should be set at 16%.
470The applicants claim that the Commission erred in setting the gravity factor for the infringement at 16%. They argue, first, that the Commission did not take into account the traders’ negligible market shares, or did not even explain why those market shares were not taken into account, even though those market shares are relevant for determining the economic strength of the participating banks and the gravity of the infringement. Secondly, when assessing the gravity factor for the infringement, the Commission failed to take account of the legitimate and pro-competitive object of the exchanges of information. Thirdly, they claim that the factors capable of affecting the gravity of an infringement, namely the number of instances and the intensity of the conduct, should reasonably have led the Commission to apply a lower gravity factor.
471The Commission disputes the applicants’ arguments.
472In that regard, it should be recalled that, as regards the non-exhaustive factors, referred to in point 22 of the Guidelines on the method of setting fines, taken into account by the Commission in order to determine the proportion of the value of sales which must be considered in a given case, the Commission correctly established that the infringement at issue covered the whole of the EEA and took the form of agreements and/or concerted practices designed, in essence, to reduce or remove the normal uncertainties inherent in the foreign exchange spot trading market, which enabled the banks involved in the infringement to be comforted in their pricing and risk management decisions, although the latter are the parameters on the basis of which those banks had to compete on the foreign exchange spot trading market. The infringement at issue is therefore among the most harmful restrictions of competition. In accordance with point 23 of the Guidelines on the method of setting fines, the proportion of the value of sales taken into account for such infringements will generally be set at the higher end of the scale.
473The Commission cannot therefore be criticised for setting the gravity factor at 16%, a level which is among the lowest on the scale of penalties for such infringements under the Guidelines on the method of setting fines.
474None of the arguments put forward by the applicants invalidates that finding.
475In the first place, the Commission was not obliged to take account of the market shares of the traders involved in the single and continuous infringement when setting the gravity factor. That is not one of the factors listed in point 23 of the Guidelines on the method of setting fines. Moreover, while it is true that, in order to determine the gravity of the infringement, it is necessary to take account of all the factors capable of affecting the assessment of its gravity, inter alia the size of the participating undertakings, the Court of Justice has also recalled that there is no binding or exhaustive list of the criteria to be taken into account when assessing the gravity of an infringement (see, to that effect, judgment of 26 September 2018, Infineon Technologies v Commission, C‑99/17 P, EU:C:2018:773, paragraphs 196 to 198 and the case-law cited).
476In the second place, the number and intensity of the incidents of anticompetitive conduct are also not among the factors listed in point 23 of the Guidelines on the method of setting fines. Although it is also true that, according to the case-law, such factors are generally relevant in determining the amount of the fines (see judgment of 26 September 2018, Infineon Technologies v Commission, C‑99/17 P, EU:C:2018:773, paragraph 197 and the case-law cited), the fact remains that it is open to the Commission to take into account the relative gravity of the participation of an undertaking in an infringement and the particular circumstances of the case when assessing the gravity of the infringement or when adjusting the basic amount according to the mitigating and/or aggravating circumstances (judgment of 11 July 2013, Team Relocations and Others v Commission, C‑444/11 P, not published, EU:C:2013:464, paragraph 104).
477In those circumstances, it must be held that the Commission was justified in setting a gravity factor by taking into account the infringement as a whole and that the relative gravity of Credit Suisse’s participation in that infringement could be examined, in particular, in the context of the assessment of mitigating circumstances and not at the stage of the gravity factor.
478In the third place and lastly, as regards the Commission’s alleged failure to take into account the legitimate and pro-competitive object of the exchanges of information, it should be recalled, as was pointed out in the examination of the second part of the second plea, that the Commission rightly rejected those arguments. Accordingly, it had no reason to consider them at the stage of setting the gravity factor.
479It follows that the third part of the fourth plea must be rejected as unfounded.
(c)The second part of the fourth plea in law, alleging an error in that the reduction in the fine granted on account of mitigating circumstances is disproportionately low
480In recitals 650 to 656 of the contested decision, the Commission found, in essence, that, pursuant to point 29 of the Guidelines on the method of setting fines, in the light of the particular circumstances of the case, namely, on the one hand, the absence of liability on the part of Credit Suisse for the underlying understanding and, on the other hand, the occasional instances of coordination, it was appropriate to grant that undertaking a 2% reduction in respect of each of those forms of conduct and therefore a total reduction of 4% in the fine on account of mitigating circumstances.
481The applicants claim that the reduction granted by the Commission on account of mitigating circumstances does not take sufficient account of the fact that Credit Suisse was held liable for only one of the three forms of conduct constituting the single and continuous infringement, especially since that conduct was the least serious of the three. In their view, in order to reflect that fact, the Commission should have applied a reduction of approximately 75% of the fine or, at the very least, two thirds, namely 66%. Moreover, they criticise the Commission for failing to take account of the mitigating circumstances relating, first, to the allegedly very limited involvement of Credit Suisse and, secondly, to the fact that it refrained from engaging in the anticompetitive conduct at issue, by sharing information obtained in the chatroom at issue with customers, sales offices and other traders.
482The Commission disputes the applicants’ arguments.
483In that connection, in the first place, it should be recalled that, as is clear from paragraph 335 above, all the forms of conduct constituting the single and continuous infringement had a common objective consisting, in essence, in mitigating the normal uncertainties inherent in the foreign exchange spot trading market, and it is not possible to identify any alleged difference between them as regards their gravity. Accordingly, the applicants’ general and unsubstantiated argument that the exchanges of information constitute the ‘least serious’ of the three forms of conduct constituting the infringement, which should be taken into account as a mitigating circumstance, must be rejected.
484In the second place, the arguments put forward by the applicants seeking to establish the existence of mitigating circumstances other than those accepted by the Commission in the contested decision cannot be upheld.
485In that regard, first, since Credit Suisse was held liable for the single and continuous infringement in so far as concerns the exchanges of information, it should be recalled that, as is apparent from over 100 discussions analysed in the contested decision, Credit Suisse’s conduct was very similar to that of the other participating banks during the same period. Accordingly, its involvement in the infringement was not substantially limited, which could have constituted a mitigating circumstance within the meaning of point 29 of the Guidelines on the method of setting fines.
486Secondly, as regards the alleged competitive conduct of Credit Suisse, it is sufficient to note that, in so far as that argument is based, in essence, on the arguments rejected at the stage of the Court’s examination of the second and third pleas, the Commission was in no way required to accept that element as a mitigating circumstance.
487Thirdly and lastly, it is necessary to analyse the effect of the Commission’s error found in the examination of the first plea in the application concerning the classification of the discussion of 9 May 2012 (see recital 276 of the contested decision) in so far as that classification forms part of the assessment of the relative gravity of Credit Suisse’s participation in the infringement at issue, capable of reducing that gravity on account of mitigating circumstances (see paragraphs 107 to 117 above).
488In that regard, however, it should be noted that the error of classification committed by the Commission does not have a bearing on the proportionality of the amount of the fine in the light of the particular circumstances of the present case. Having regard to the more than 100 discussions examined in the contested decision, some preceding and others following the discussion of 9 May 2012, which neither interrupted nor affected the course of Credit Suisse’s conduct, the fact that that single discussion was incorrectly relied upon against Credit Suisse does not make it possible to reduce the gravity of the infringement imputed to it in the light of the large number of incidences and the seriousness of Credit Suisse’s conduct.
489In the light of those elements, the second part of the fourth plea must be rejected.
(d)The fourth part of the fourth plea, alleging infringement of the principle of equal treatment
490The applicants claim that the Commission infringed the principle of equal treatment by failing to take into account, first, the fact that Credit Suisse is held liable for a single constituent element of the single and continuous infringement, for which the 4% reduction applied by the Commission does not take sufficient account, and, secondly, Credit Suisse’s limited participation in the alleged infringement, whereas the other banks involved in the alleged actions obtained a reduction for temporal overlap. Thus, the applicants claim that the Commission failed to take into account Credit Suisse’s individual liability. Finally, the applicants complain that the Commission did not properly exercise its discretion when setting the fines in the present case.
491The Commission disputes the applicants’ arguments.
492As a preliminary point, it is recalled that, according to the case‑law cited in paragraph 445 above, the principle of equal treatment is infringed only where comparable situations are treated differently or different situations are treated in the same way, unless such treatment is objectively justified.
493It should also be recalled that, when calculating fines imposed on undertakings which have participated in a cartel, differentiated treatment of the undertakings concerned is inherent in the exercise of the Commission’s powers in that area. In exercising its discretion, the Commission is required to fit the penalty to the individual conduct and specific characteristics of those undertakings in order to ensure that, in each case, the rules of EU competition law are fully effective (see judgment of 5 December 2013, Caffaro v Commission, C‑447/11 P, not published, EU:C:2013:797, paragraph 50 and the case-law cited).
494In that respect, as regards the claim of unequal treatment on account of the alleged failure to take into consideration Credit Suisse’s limited participation in the infringement, it should be borne in mind that, contrary to what the applicants claim, the duration of its participation, in this case the shortest of all the participants in the infringement, was taken into account by the Commission when calculating the fine (recitals 643 to 645 of the contested decision). Moreover, the Commission took into consideration the fact that Credit Suisse participated in only one single and continuous infringement, so far as the exchanges of information are concerned, covered by the contested decision in the present case, unlike other banks which took part in several infringements on various chatrooms. The different treatment of the situations of Credit Suisse and of other banks must, therefore, be regarded as objectively justified, within the meaning of the case-law referred to in paragraphs 492 and 493 above, by the differences in the duration and number of infringements committed on various chatrooms.
495Furthermore, it is clear that, by their general and unsubstantiated line of argument relating to an alleged infringement by the Commission of the principle of equal treatment resulting from the refusal to grant Credit Suisse a reduction of 66 to 75% of the basic amount of the fine, as was granted to other parties which participated in multiple infringements on that market in order to take the temporal overlap into consideration, the applicants are seeking, in essence, to rely on a mitigating circumstance deriving from the conduct of the other participants in the infringement.
496According to settled case-law, as a matter of principle, a participant in an infringement cannot allege a mitigating circumstance deriving from the conduct of the other participants in the infringement (see judgment of 11 July 2019, Huhtamäki and Huhtamaki Flexible Packaging Germany v Commission, T‑530/15, not published, EU:T:2019:498, paragraph 193 and the case-law cited). Accordingly, the fact that the other cartel members became involved in the cartel earlier, or participated in several infringements on various chatrooms, might well constitute an aggravating circumstance in relation to them but not a mitigating circumstance in favour of Credit Suisse (see, to that effect, judgment of 11 July 2019, Huhtamäki and Huhtamaki Flexible Packaging Germany v Commission, T‑530/15, not published, EU:T:2019:498, paragraph 194 and the case-law cited).
497In any event, it is clear that the reduction granted to the other participants in the settlement decision was applied not to the basic amount in respect of which the applicants claim that reduction, but in the context of the calculation of the proxy. The reduction sought is thus not comparable to that which the other participants in the infringement benefited from.
498Accordingly, the fourth part of the fourth plea must be rejected.
(e)The fifth part of the fourth plea, alleging failure to state sufficient reasons in the contested decision as regards the proportionality of the calculation of the fine imposed on the applicants compared to that imposed on the other participants in the infringement
499The applicants complain that the Commission failed to state sufficient reasons in relation to the factors which are relevant to the calculation of the fines imposed on the settling parties. They take the view that they are therefore deprived of the opportunity to assess the scope of the infringement of Credit Suisse’s right to equal treatment, just as the Court is deprived of the opportunity to assess the proportionality of the reduction granted to Credit Suisse on account of its limited participation.
500The Commission disputes the applicants’ arguments.
501In that regard, it is noted that the applicants’ line of argument relating to the alleged failure to state sufficient reasons as regards the proportionality of the amount of the fine, as set out in the application, is based, in essence, on the impossibility of comparing the reduction of Credit Suisse’s fine with the reduction granted to the other settling parties, on the ground that the applicants did not have the settlement decision when the application was lodged. However, once that decision had been made available during the proceedings pending before the Court, the applicants did not present any further argument concerning that failure to state sufficient reasons at the reply stage.
502When questioned at the hearing, the applicants confirmed that, in essence, they no longer maintained their arguments set out in the fifth part of the fourth plea in so far as concerns the failure to state sufficient reasons in the contested decision on the ‘factors that are relevant for the calculation of the fines imposed on the settling parties’. However, they took the view that the second part of their line of argument, relating to the proportionality of the fine, was still relevant for the purposes of the present action.
503In that regard, it is clear that the applicants’ explanation, put forward at the hearing, concerning the proportionality of the fine, is in no way supported by the arguments contained in the fifth part of the fourth plea, as summarised in paragraphs 499 and 501 above. The claim relating to the proportionality of the reduction of the fine, which is general and unsubstantiated, and addressed in two sentences of the application, the second of which is a consequence which the applicants draw from the unavailability of the settlement decision, is in no way formulated independently of the alleged failure to state reasons. Accordingly, since that claim relates to an argument which was abandoned in the course of the proceedings, there is no longer any need to examine it.
504In any event, the reduction of the fine imposed on Credit Suisse on account of mitigating circumstances, on the basis of the reductions granted to other participants in the infringement, cannot constitute an appropriate factor in assessing the possible lack of proportionality of the fine (see, by analogy, judgment of 4 July 2006, Hoek Loos v Commission, T‑304/02, EU:T:2006:184, paragraph 85). Those reductions are, in particular, the consequence of various circumstances relating to the individual conduct of Credit Suisse and to that of the other participants in the infringement, and the taking into account of those circumstances in the calculation of the fine is justified by the principle that penalties must be specific to the offender and the offence. However, as is clear from paragraphs 492 to 498 above, in the light of the fact that the other banks involved in the infringement at issue also participated in other infringements which took place in other chatrooms and were therefore in objectively different situations from Credit Suisse, the Commission was entitled to apply a different reduction. Consequently, that line of argument cannot succeed.
505It follows from the foregoing that the fourth plea is well founded in part, in so far as concerns the first part thereof in respect of the adjustment factor related to market-making activities. However, the plea must be rejected as to the remainder.
2.Conclusion on Article 2(a) of the contested decision
506It follows from all the foregoing that the Commission’s error in relation to the data used to determine the adjustment factor to be applied for the calculation of the proxy leads to the annulment of Article 2(a) of the contested decision.
C.The application for a reduction of the amount of the fine
507By their third head of claim, the applicants ask the Court to reduce the amount of the fine imposed on Credit Suisse, irrespective of the Court’s findings on their first head of claim seeking annulment of the contested decision.
508Accordingly, it is necessary to rule on the applicants’ application for a reduction of the amount of the fine, pursuant to the unlimited jurisdiction conferred on the Court by Article 261 TFEU and Article 31 of Regulation No 1/2003.
509In that regard, it should be recalled that, although, according to the case-law of the Court of Justice, the unlimited jurisdiction conferred on the EU judicature by Article 31 of Regulation No 1/2003 in accordance with Article 261 TFEU, empowers the competent Court, in addition to carrying out a mere review of legality with regard to the penalty, to substitute its own appraisal for the Commission’s and, consequently, to cancel, reduce or increase the fine or penalty payment imposed, the exercise of that jurisdiction does not amount to a review of the Court’s own motion, and proceedings are inter partes. It is, in principle, for the applicant to raise pleas in law against the decision at issue and to adduce evidence in support of those pleas (see, to that effect, judgment of 26 September 2018, Infineon Technologies v Commission, C‑99/17 P, EU:C:2018:773, paragraphs 193 and 194 and the case-law cited).
510The exercise by the Court of its unlimited jurisdiction involves, in accordance with Article 23(3) of Regulation No 1/2003, taking into consideration the seriousness and duration of the infringement at issue, in compliance with the principles of, inter alia, adequate reasoning, proportionality, the individualisation of penalties and equal treatment, and without it being bound by the indicative rules defined by the Commission in its Guidelines on the method of setting fines, even where the latter may give guidance to the EU Courts when they exercise their unlimited jurisdiction (see judgment of 21 January 2016, Galp Energía España and Others v Commission, C‑603/13 P, EU:C:2016:38, paragraph 90 and the case-law cited).
511It should also be observed that, by its nature, the fixing of a fine by the Court is not an arithmetically precise exercise. Furthermore, when it adjudicates in the exercise of its unlimited jurisdiction, the Court must make its own appraisal, taking account of all the circumstances of the case (see judgment of 15 July 2015, Trafilerie Meridionali v Commission, T‑422/10, EU:T:2015:512, paragraph 398 and the case-law cited).
512In the context of its obligation to state reasons, it is for the Court to set out in detail the factors which it takes into account when setting the amount of the fine (see, to that effect, judgment of 14 September 2016, Trafilerie Meridionali v Commission, C‑519/15 P, EU:C:2016:682, paragraph 52).
513In the present case, in order to determine the amount of the fine to be imposed as a sanction on Credit Suisse for its infringing conduct, as results from the examination of the five pleas raised in the present action, it is appropriate to take into account the following factors.
514In the first place, as far as concerns the gravity and duration of the infringement, the following observations must be made.
515It is appropriate to use a methodology which – much like the methodology used by the Commission in the present case – first identifies the basic amount of the fine which can subsequently be adjusted according to the specific circumstances of the case.
516First of all, with regard to the value of sales as initial data, it is appropriate to use, as a proxy, Credit Suisse’s annualised notional amounts relating to the purchase transactions and sales transactions for the currency-pairs during the relevant period. Those annualised notional amounts should next be adjusted by applying a factor comprising the sum, first, of half the bid-ask spread relating to the revenues from market-making activities and, secondly, 10% of the bid-ask spread relating to the revenues from activities of trading on own account.
517As is clear from paragraphs 428 to 436 above, the Commission incorrectly considered that the data from a sample of the exchanges of information which took place in the chatroom at issue constituted the ‘best available figures’ for the purpose of calculating the adjustment factor related to market-making, thereby vitiating the calculation of the proxy for Credit Suisse’s value of sales, even though it is a component of the calculation of the basic amount of the fine and, consequently, the final amount of the fine imposed on Credit Suisse. The contested decision should therefore be altered to take account of the Bloomberg BFIX data in the determination of the proxy (see paragraphs 437 to 448 above).
518On the one hand, with regard to the adjustment factor related to market-making, the applicants submit that, on the basis of that data, the bid-ask spread amounted to 1.47 basis points and that, consequently, that adjustment factor, constituting 50% of that spread, amounted to 0.73 basis points. The applicants calculated that spread of 1.47 basis points by using a weighted average of the spreads derived from the Bloomberg BFIX data for G10 currency pairs comprising at least one EEA currency for the relevant period, and applying those spreads to Credit Suisse’s notional amounts for each of those currency pairs. The applicants take the view that the frequently traded currency pairs are more liquid and therefore have lower bid-ask spreads. Accordingly, the application of a simple average significantly inflates the value of Credit Suisse’s sales.
519In its response to a measure of organisation of procedure decided upon by the Court, the Commission took the view, for its part, that, on the basis of the Bloomberg BFIX data, the bid-ask spread amounted to 2.3 basis points and that, consequently, the adjustment factor related to market-making amounted to 1.15 basis points. The Commission calculated that bid-ask spread on the basis of a simple average of the bid-ask spreads derived from the Bloomberg BFIX data for G10 currency pairs comprising at least one EEA currency for the relevant period. The Commission submits that the method of applying a simple average to the bid-ask spreads is the method which is closest to the methodology used in the contested decision.
520In that regard, it should be noted that Credit Suisse’s notional amounts have already been taken into account by the Commission in the calculation of the annualised notional amounts. As the Commission observed, in essence, in its rejoinder, while those amounts are intended to reflect Credit Suisse’s activity, the bid-ask spreads, for the purpose of calculating the adjustment factor, are intended in particular to reflect the economic importance of the infringement and the actual structure of the participating undertakings’ transactions relating to the financial products concerned by the infringement. That would not be the case if the spreads were weighted according to Credit Suisse’s notional amounts for each currency pair. Moreover, since the bid-ask spreads derived from the Bloomberg BFIX data do not relate solely to transactions effected by Credit Suisse, weighting those spreads by reference to Credit Suisse’s notional amounts for each currency pair may, on the contrary, significantly understate the value of Credit Suisse’s sales. It must therefore be held that the method of calculating the spread proposed by the applicants is less appropriate than that proposed by the Commission for the purpose of calculating the adjustment factor related to market-making and that the factor proposed by the Commission, namely 1.15 basis points, should therefore be accepted.
521On the other hand, as regards the adjustment factor related to trading on own account, the applicants have not succeeded in calling into question the Commission’s assessment that it amounts to 0.1 basis points (see paragraphs 449 to 458 above). Consequently, the total adjustment factor relating to the proxy, calculated on the basis of the Bloomberg BFIX data, amounts to 1.25 basis points (or 0.0125%).
522Accordingly, since the applicants have not succeeded in calling into question the figure of EUR 1060802752528 for the annual notional amounts of Credit Suisse arrived at in the contested decision (see paragraphs 407 to 416 above), after multiplying the value of those amounts by the adjustment factor of 1.25 basis points, the proxy for the value of sales of Credit Suisse is EUR 132600344, as was calculated by the Commission in its response to a measure of organisation of procedure decided upon by the Court.
523Next, the Court takes the view that it is appropriate to take into consideration the nature of the infringement as well as its geographical scope and use 16% of the value of sales on account of its gravity for the reasons set out in paragraphs 472 and 473 above.
524In addition, it is appropriate to rely on the duration of the applicants’ participation in the infringement as found in the contested decision, as it was not disputed by the applicants and since it is not affected by the conclusion, set out in paragraph 117 above, relating to the discussion of 9 May 2012. Consequently, it is appropriate to use 0.42 years in respect of the duration of participation in the infringement.
525Finally, it is necessary to add, as the Commission did, an additional amount of 16% of the value of sales in order to meet the need to impose a fine of an amount that has a deterrent effect irrespective of the duration of the infringement.
526In the second place, it is appropriate, in line with the considerations set out in the contested decision, to reduce the basic amount of the fine, on account of mitigating circumstances, by 2% on account of the absence of liability on the part of Credit Suisse for the occasional instances of coordination, and by 2% on account of the absence of its liability for the underlying understanding.
527In the third place, the amount of the fine determined by the Court takes due account of the need to impose a fine of an amount that has a deterrent effect, as is clear from paragraph 525 above.
528In the light of all of the foregoing considerations, the Court finds that, on a fair assessment of the circumstances of the case, having regard to the principle that penalties should be specific to the offender and the offence and to the principle of the proportionality of that fine, the amount of the fine for which UBS Group – the successor in law to Credit Suisse Group –, UBS – the successor in law to Credit Suisse – and Credit Suisse Securities (Europe) are jointly and severally liable should be set at EUR 28920000.
IV.Costs
529Under Article 134(3) of the Rules of Procedure of the General Court, where each party succeeds on some and fails on other heads, the parties are to bear their own costs.
530In the present case, the applicants have been successful as regards their head of claim seeking the annulment of the contested decision, in so far as Article 2(a) thereof is concerned, and in their application for a reduction of the amount of the fine, but have been unsuccessful as regards their head of claim seeking the annulment of Article 1 of the contested decision. In those circumstances, the applicants and the Commission should be ordered to bear their own costs.
Operative part
On those grounds,THE GENERAL COURT (Seventh Chamber)hereby:
1.Annuls Article 2(a) of Commission Decision C(2021) 8612 final of 2 December 2021 relating to a proceeding under Article 101 TFEU and Article 53 of the EEA Agreement (Case AT.40135 – FOREX (Sterling Lads));
2.Sets the amount of the fine for which UBS Group AG – the successor in law to Credit Suisse Group AG –, UBS AG – the successor in law to Credit Suisse AG – and Credit Suisse Securities (Europe) Ltd are jointly and severally liable at EUR 28920000;
3.Dismisses the action as to the remainder;
4.Orders each party to bear its own costs.
Delivered in open court in Luxembourg on 23 July 2025.
Source: EUR-Lex, © European Union, 1998-2026. Reuse authorised under Decision 2011/833/EU.