CJEU, 8th chamber, September 16, 2013, No T-482/07

THE GENERAL COURT (Eighth Chamber),

1 By Decision C(2007) 4441 final of 3 October 2007 relating to a proceeding under Article 81 [EC] (Case COMP/38.710 – Bitumen Spain), a summary of which is published in the Official Journal of the European Union of 29 December 2009 (OJ 2009 C 321, p. 15; ‘the contested decision’), the Commission of the European Communities found that the 13 companies to which that decision was addressed had participated in a complex of market-sharing and price-coordinating agreements in the road construction penetration bitumen business in Spain (excluding the Canary Islands) and imposed fines of between EUR 6 435 000 and EUR 83 850 000 on 10 of those companies.

2 Those 10 companies brought actions against that decision, by applications lodged at the Court Registry between 18 and 20 December 2007:

– Repsol Lubricantes y Especialidades, SA, formerly Repsol Lubricantes YPF y Especialidades, SA, Repsol Petróleo, SA and Repsol, SA, formerly Repsol YPF, SA (together ‘Repsol’), in Case T‑496/07;

– Productos Asfálticos (PROAS), SA, in Case T‑495/07;

– Compañía Española de Petróleos (CEPSA), SA, in Case T‑497/07;

– Nynäs Petroleum AB and Nynas Petróleo, SA (together ‘Nynäs’ or ‘the applicants’), in the present case;

– Galp Energía España, SA, Petróleos de Portugal, SA and Galp Energia, SGPS, SA (together ‘Petrogal’ or ‘Galp’), in Case T‑462/07.

A – Relevant market

3 The product concerned by the infringement is penetration bitumen used for road construction. Bitumen is a by-product produced during the distillation of specific heavy crude oils. Around 85% of the bitumen produced in the European Union is used for road construction and maintenance, as an adhesive in the production of asphalt where it is used to bind the stones together. The remaining 15% is used in other fields of construction, for example in the construction of airport runways and car parks, and in industrial applications, such as roofing and pipe coating.

4 Approximately 80% of the bitumen used for road construction and maintenance is not subject to further processing: this is called penetration bitumen. The remaining 20% of bitumen used in road construction and maintenance is accounted for by bitumen which is further processed, such as bitumen emulsions, which are produced by mixing penetration bitumen with water using an emulsifying agent (used in road maintenance more than in construction), and modified bitumen, which is obtained by mixing penetration bitumen with a chemical product, usually polymers, in order to enhance its performance (polymer modified bitumen or PMB) (recitals 4, 9 et seq. to the contested decision).

5 Recital 15 to the contested decision defines the relevant market as being penetration bitumen, without any further processing, used for road construction and maintenance (‘penetration bitumen’ or ‘bitumen’).

6 That definition of the relevant market is confirmed as follows by recital 513 to the contested decision:

‘… this case concerns a cartel between sellers of the same product in the same business area, namely penetration bitumen in Spain …’.

7 The value of the Spanish penetration bitumen market is estimated at EUR 286 400 000 for 2001, the last full year of the infringement (recital 67 to the contested decision).

B – Undertakings in question

8 In Spain there are, on the one hand, three bitumen producers, Repsol, CEPSA‑PROAS and the BP group, of which BP plc is the holding company and of which BP España, SA and BP Oil España, SA are subsidiaries operating in Spain (together ‘BP’) and, on the other hand, a number of importers, including Nynäs and Petrogal (recitals 63 and 64 to the contested decision).

1. Repsol Group

9 Repsol Productos Asfálticos, SA (RPA) became Repsol YPF Lubricantes y Especialidades on 12 December 2001 (‘RPA/Rylesa’). RPA/Rylesa was owned 99.99% from 1991 to 2002 by Repsol Petróleo, itself a subsidiary owned 99.97% by Repsol YPF, the ultimate parent company of the Repsol group. That international group of oil and gas companies is present mainly in Spain and Latin America.

10 RPA/Rylesa produces and markets bitumen products. One of Repsol Petróleo’s activities is the production of penetration bitumen and the sale thereof to RPA/Rylesa in order to be marketed.

11 Two other companies of the Repsol Group, Petróleos del Norte, SA (‘Petronor’) and Asfalnor, SA, carry out a penetration bitumen-related activity in Spain.

12 In 1991, Petronor was owned 56.19% by Repsol YPF, and this shareholding increased to 85.98% on 31 December 1992. Asfalnor was owned 60% by Petronor in 1991 and, in April 1992, companies of the Repsol group owned 80% of it (60% by Petronor and 20% by Repsol YPF) (recital 395 to the contested decision).

13 Petronor produces bitumen which, between 1990 and 1998, it sold to Asfalnor and, on occasions, to RPA/Rylesa, in order to be marketed. Since 1999, Petronor has been selling bitumen directly to third parties.

14 Asfalnor marketed bitumen between 1990 and 1998. It bought bitumen from Petronor and, on occasions, from RPA/Rylesa. Since 1999, Asfalnor has been acting as Petronor’s agent.

15 RPA/Rylesa and Petronor achieved a turnover of EUR 97 500 000 in Spain from their sales of penetration bitumen to third parties in 2001, that is 34.04% of the relevant market. The consolidated total turnover of the Repsol Group was EUR 51 355 000 000 in 2006, the year preceding the adoption of the contested decision (recitals 16 to 26 and 67 to the contested decision).

2. CEPSA-PROAS

16 CEPSA is an international group of companies in the energy sector whose shares are publicly listed and is present in several countries. PROAS, which has been a wholly-owned subsidiary of CEPSA since 1 March 1991, markets bitumen produced by CEPSA and produces and markets other bitumen products (recital 31 to the contested decision).

17 PROAS achieved a turnover in Spain of EUR 90 700 000 from its sales of penetration bitumen to third parties during the business year 2001, that is 31.67% of the relevant market. The consolidated total turnover of CEPSA in 2006 was EUR 18 474 000 000 (recitals 44 and 67 to the contested decision).

3. BP

18 BP Oil España achieved a turnover in Spain of EUR 43 500 000 from its sales of penetration bitumen to third parties during the business year 2001, that is 15.19% of the relevant market. The consolidated total turnover of BP in 2006 was EUR 211 776 000 000 (recitals 35, 42 and 43 to the contested decision).

4. Nynäs Group

19 The Nynäs group, of which Nynäs Petroleum AB, a Swedish company, is the ultimate holding company, produces and sells bitumen internationally. Nynas Petróleo markets bitumen in Spain (recitals 46 and 53 to the contested decision).

20 From 22 May 1991 until 1999 Nynas Petróleo was wholly owned by the holding company Nynäs International BV. During the same period Nynäs International BV was itself a wholly-owned subsidiary of Nynäs Petroleum (recital 438 to the contested decision).

21 In 1999 Nynäs Petroleum acquired from Nynäs International the entire issued share capital of Nynas Petróleo, which remained a wholly-owned subsidiary of Nynäs Petroleum until 2003 (recital 439 to the contested decision). Nynäs International was wound up on 12 June 2003. Following its dissolution its share capital was repaid to Nynäs Petroleum, which thus became its economic successor and took over liability for the infringement previously committed by Nynäs International, which had ceased to exist as a separate legal entity (recital 440 to the contested decision).

22 The Nynäs group has no production facilities in Spain but has one bitumen depot in Villagarcía de Arosa, in Galicia (Spain). Nynas Petróleo has its headquarters in Madrid (Spain) and its business activity is the sale and marketing of bitumen in Spain (recital 53 to the contested decision).

23 Nynas Petróleo achieved a turnover in Spain of between EUR 14 000 000 and EUR 15 000 000 from its sales of penetration bitumen to third parties during the business year 2001, that is 4.89% to 5.24% of the relevant market. The consolidated total turnover of the Nynäs group in 2006 was EUR 1 941 000 000 (recitals 54 and 67 to the contested decision).

5. Petrogal Group

24 Between 1990 and 2003 the shares in Galp Energía España (formerly Petrogal Española, SA) were held as to 89.29% by Petróleos de Portugal and as to 10.71% by Tagus, RE, an insurance company which was itself controlled as to 98% by Petróleos de Portugal. Since 2003 Galp Energía España has been a wholly-owned subsidiary of Petróleos de Portugal, which has been a wholly-owned subsidiary of Galp Energia, SGPS since 22 April 1999 (recitals 56, 57, 59, 456, 458 to the contested decision).

25 Galp Energía España’s business activity is the sale and marketing of bitumen in Spain. Its turnover for bitumen sales to unrelated parties in Spain was EUR 13 000 000 in 2001, the last full year of the infringement, that is 4.54% of the relevant market. The consolidated total turnover of Galp Energia, SGPS amounted to EUR 12 576 000 000 in 2006 (recitals 61 and 67 to the contested decision).

C – Administrative procedure

26 By letter of 20 June 2002 BP informed the Commission of an alleged cartel relating to the road bitumen market and submitted an application for immunity from fines pursuant to the Commission Notice on immunity from fines and reduction of fines in cartel cases (OJ 2002 C 45, p. 3; ‘the 2002 Leniency Notice’). The application consisted of a letter and eight annexes describing the alleged anti-competitive activities on the Spanish bitumen market (recital 79 to the contested decision).

27 At a meeting with the Commission’s services on 25 June 2002 BP enlarged upon its application for immunity and produced supporting documents. BP submitted further information on 4, 8 and 10 July 2002 (recital 80 to the contested decision).

28 On 19 July 2002, the Commission granted BP conditional immunity from fines in accordance with point 8(a) of the 2002 Leniency Notice (recital 81 to the contested decision).

29 Pursuant to Article 14(3) of Council Regulation No 17 of 6 February 1962: first Regulation implementing Articles [81 EC] and [82 EC] (OJ English Special Edition 1959-1962, p. 87), as subsequently amended, inspections were carried out on 1 and 2 October 2002. Repsol, PROAS, BP, Nynäs and Petrogal were concerned (recital 82 to the contested decision).

30 BP supplied further information on 21 October 2002 concerning the alleged anti-competitive activities on the relevant market (recital 83 to the contested decision).

31 On 5 November 2003 the Commission’s services interviewed Mr A.T., manager of the ‘Bitumen’ Department of BP España, pursuant to BP’s obligation to cooperate. Following that interview, BP supplied technical information concerning bitumen on 1 December 2003 (recital 84 to the contested decision).

32 On 6 February 2004 the Commission sent the undertakings concerned a first round of requests for information pursuant to Article 11(3) of Regulation No 17 and also sent BP an informal request for information (recital 85 to the contested decision).

33 Repsol submitted most of its response on 6 April 2004 and the remainder on 20 April 2004. PROAS provided its responses concerning the Spanish market on 21 April 2004 (recital 86 to the contested decision).

34 By fax of 31 March 2004 Repsol submitted a leniency application to the Commission pursuant to the 2002 Leniency Notice, together with a corporate statement (recital 87 to the contested decision).

35 On 2 April 2004 an additional request for information was sent to Repsol concerning the documents found during the inspection carried out at its premises, to which Repsol replied on 22 April 2004 (recital 88 to the contested decision).

36 By fax of 5 April 2004 PROAS submitted an application to the Commission pursuant to the 2002 Leniency Notice, together with a corporate statement (recital 89 to the contested decision).

37 On 20 April 2004 Repsol filed two bundles of documents further to its application pursuant to the 2002 Leniency Notice (recital 90 to the contested decision).

38 On 24 October 2005 the Commission sent the companies concerned a second round of requests for information, pursuant to Article 18(2) of Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles [81 EC] and [82 EC] (OJ 2003 L 1, p. 1) (recital 91 to the contested decision).

39 Repsol and PROAS replied on 8 November and 18 November 2005, respectively (recital 92 to the contested decision).

40 On 29 March 2006 the Commission sent Repsol and PROAS a third request for information and also sent BP an informal request for information. Repsol replied on 5 April 2006, BP on 6 April 2006 and PROAS on 7 April 2006 (recitals 93 and 94 to the contested decision).

41 In order to clarify the extent to which BP, Nynäs and Petrogal had been involved in the cartel, the Commission sent a fourth request for information to Repsol and PROAS on 26 April 2006 to which Repsol and PROAS replied on 9 May 2006 (recitals 95 and 96 to the contested decision).

42 On 22 May 2006 the Commission sent Repsol, PROAS and Petrogal a fifth request for information, concerning liability issues. PROAS replied on 29 May 2006, and Repsol and Petrogal on 30 May 2006 (recitals 98 and 99 to the contested decision)

43 By letters of 2 August 2006 the Commission informed Repsol and PROAS, pursuant to point 26 of the 2002 Leniency Notice, that it intended to apply to them, pursuant to point 23(b) of that notice, a reduction of any fine imposed of between 30 and 50% for Repsol and between 20 and 30% for PROAS (recitals 100 and 101 to the contested decision).

44 On 22 August 2006, the Commission took the decision to initiate the procedure in this case (third citation of the contested decision).

45 Between 24 and 28 August 2006 the Commission notified to BP, Repsol, CEPSA-PROAS, Nynäs and Petrogal the statement of objections adopted on 22 August 2006 (recital 102 to the contested decision and third citation of that decision).

46 The Guidelines on the method of setting fines imposed pursuant to Article 23(2)(a) of Regulation No 1/2003 (OJ 2006 C 210, p. 2) replaced, as from 1 September 2006, the Guidelines on the method of setting fines imposed pursuant to Article 15(2) of Regulation No 17 and Article 65(5) [CS] (OJ 1998 C 9, p. 3; ‘the 1998 Guidelines’).

47 BP, Repsol, PROAS (apart from CEPSA), Nynäs and Petrogal exercised their right of access to the parts of the Commission’s file that were available only at the Commission’s premises (recital 103 to the contested decision).

48 BP, Repsol, CEPSA-PROAS, Nynäs and Petrogal responded in writing within the prescribed period to the objections raised against them (recital 104 to the contested decision).

49 All the addressees of the contested decision, with the exception of Repsol Petróleo, Repsol YPF and CEPSA, availed themselves of their right to an oral hearing. The hearing was held on 12 December 2006 (recital 105 to the contested decision).

50 On 16 February 2007 the Commission sent all the undertakings concerned a request for information in order to obtain confirmation or corrected figures relating to penetration bitumen sales previously supplied and also information relating to the turnover of each group for the business year 2006 (recital 106 to the contested decision).

D – Contested Decision

1. Finding of the infringement

51 The contested decision finds that the 13 companies to which it is addressed infringed Article 81 EC by participating in a complex of agreements and concerted practices in the marketing of penetration bitumen on the Spanish territory (excluding the Canary Islands).

52 In the infringement, the Commission identified two components: first, market-sharing; and, second, price coordination, which consisted in agreements to increase or decrease bitumen prices by an equivalent amount and to implement them at the same time (recital 366 to the contested decision).

53 The various types or components of unlawful conduct which were identified are as follows:

– the establishment of sales quotas;

– the allocation of volumes of the product and customers between all participants in the cartel, on the basis of those quotas;

– the monitoring of the implementation of the market-sharing and customer-sharing arrangements, through the exchange of information on sales volumes;

– the establishment of a compensation mechanism to correct deviations from the market-sharing and customer-sharing arrangements;

– the agreement on the variation of bitumen prices and the time at which the new prices would apply;

– participation in regular meetings and other contacts in order to agree on the above restrictions of competition and to implement or modify them as required (recital 373 to the contested decision).

54 In the first place, the Commission described the market-sharing activities on the basis of the statements submitted by BP, Repsol and PROAS in their applications pursuant to the 2002 Leniency Notice and in their replies to the Commission’s requests for information (recital 122 to the contested decision).

55 The Commission considered that the existence of those activities was confirmed by contemporaneous evidence, namely documents obtained during the inspections and other contemporaneous documents submitted in applications for immunity from fines, that is to say applications pursuant to the 2002 Leniency Notice, or in replies to requests for information (recital 123 to the contested decision).

56 It is apparent from BP’s application for immunity from fines that when that undertaking began to produce penetration bitumen in Spain in July 1991 it found that Repsol and PROAS were involved in an ongoing market-sharing agreement and that BP was required to participate in that agreement in order to enter the market with some degree of success (recital 119 to the contested decision).

57 Other bitumen suppliers present on the Spanish market coordinated their sales with Repsol, PROAS and BP: Nynäs and Petrogal participated in the cartel, Nynäs at least from 1991, and Petrogal at least from 1995 (recital 120 to the contested decision).

58 According to the statements submitted by Repsol and PROAS, the parties to the cartel effected contacts for the purpose of market-sharing around a negotiating table known as the ‘asphalt table’, composed of the companies of the Repsol group (RPA/Rylesa, Asfalnor and Petronor), PROAS and BP, but also of Nynäs and Petrogal, although the latter two companies participated only in the discussions concerning their areas of influence and did so on a bilateral basis with Repsol or PROAS, and not with other members of the cartel (recitals 124 and 129 to the contested decision).

59 The Commission identified the following phases in the market-sharing mechanism in question implemented within the cartel that was found to have existed:

(a) an in-house market analysis, carried out around September, during which each producer separately prepared a market study for the following year estimating bitumen consumption in Spain;

(b) an in-house pre-distribution of the market, carried out around October, with each bitumen producer preparing a draft market distribution document to be submitted in the negotiations with its competitors;

(c) an agreement on the size of the market, that is to say, total bitumen consumption forecast for the following year, concluded around November between Repsol, PROAS and BP;

(d) negotiations on the provisional distribution of the market thus agreed, carried out in December/January;

(e) the annual market-sharing agreement: between 1994 and 2000 the competent sales managers of Repsol and PROAS normally held closing discussions in December/January in order to resolve any remaining issues concerning the distribution of the market. The document containing the market-sharing agreement for a given year was called ‘PTT’ or ‘Petete’;

(f) communication of information to and negotiations with Nynäs and Petrogal; once the distribution of the market had been agreed by the three bitumen producers, Repsol or PROAS held a meeting with Nynäs and another meeting with Petrogal in order to inform them of and negotiate the sales volumes and customers that would be allocated to each of them in their respective areas of influence (recital 130 to the contested decision).

60 In the second place, the Commission found that the price-coordination activities provided the necessary support for the market-sharing arrangements by ensuring that the agreed allocation of volumes and customers would not be affected by the application of independent pricing policies by suppliers (recital 290 to the contested decision).

61 The Commission described the price arrangements on the basis of the voluntary statements made by BP, Repsol and PROAS and on their replies to the requests for information. It then presented a chronological overview of the contemporaneous documents in its possession that confirmed the price arrangements described in those statements (recital 291 to the contested decision).

62 The price variations and the date on which they were to be implemented were generally decided on between Repsol and PROAS, which then informed BP, Nynäs and Petrogal of their conclusions (recital 354 to the contested decision).

63 The Commission then found that the complex of agreements or concerted practices had the object of restricting competition in Spain (excluding the Canary Islands), a substantial part of the internal market (recital 371 to the contested decision).

64 The Commission recalled that, according to the case-law, there is no need to take into account the actual effects of an agreement when it has as its object the prevention, restriction or distortion of competition within the common market. Consequently, it is not necessary to show actual anti-competitive effects where the anti-competitive object of the conduct in question is proved (recital 375 to the contested decision).

65 The same applies to concerted practices. Although, according to Article 81 EC, the concept of a concerted practice requires not only concertation between undertakings but also conduct on the market resulting from the concertation and having a causal connection with it, it may be presumed, subject to proof to the contrary, that undertakings taking part in such concertation and remaining active in the market will take account of the information exchanged with competitors in determining their own conduct on the market, all the more so when the concertation occurs on a regular basis and over a long period. Such a concerted practice is then caught by Article 81 EC even in the absence of anti-competitive effects on the market (recital 331 to the contested decision).

66 None the less, in the present case, the Commission took the view that, on the basis of the elements presented in the contested decision, it had also proved that the cartel agreements had been implemented and that they had probably produced actual anti-competitive effects (recital 376 to the contested decision).

67 Moreover, the Commission considered it established that the staff of Nynas Petróleo had participated in the cartel. In the light of the case‑law on the presumption that a parent company actually exercises decisive influence over its wholly-owned subsidiary, and given the shareholding relationship between Nynas Petróleo, Nynäs International and Nynäs Petroleum, the Commission considered that Nynäs Petroleum, Nynäs International and Nynas Petróleo constituted a single undertaking between 22 May 1991 and 1 October 2002 for the purposes of the application of Article 81(1) EC. The Commission also considered that Nynäs Petroleum was the economic successor of Nynäs International (recital 459 to the contested decision).

2. Calculation of the amount of the fines

68 The Commission recalled that, under Article 23(2) of Regulation No 1/2003, it may by decision impose fines on undertakings where, either intentionally or negligently, they infringe Article 81 EC. The Commission also stated that, under Article 15(2) of Regulation No 17, which was applicable at the time of the infringement, the fine for each undertaking participating in the infringement could not exceed 10% of its total turnover in the preceding business year and that Article 23(2) of Regulation No 1/2003 lays down the same limitation (recital 496 to the contested decision).

69 The Commission considered that each of the two restrictions of competition established, namely the horizontal market-sharing arrangements and the price-coordination, was by its nature among the most serious types of infringements of Article 81 EC, which, according to the case-law, are capable of warranting the classification of ‘very serious’ infringements solely on the basis of their nature, without there being any need for such conduct to cover a particular geographic area or have a particular impact (recital 500 to the contested decision).

70 The Commission took the view that it was not possible to measure the actual impact of the cartel on the market, due inter alia to insufficient information on likely bitumen net price developments in Spain in the absence of the arrangements. The Commission took the view that it was not required to demonstrate precisely the actual impact of the cartel on the market or to quantify it, but that it could confine itself to estimates of the probability of such an impact. In any event, the Commission considered that the cartel agreements were effectively implemented and that it was likely that the cartel arrangements produced actual anticompetitive effects (recital 501 to the contested decision).

71 In view of the nature of the infringement, the Commission considered that Repsol, PROAS, BP, Nynäs and Petrogal had committed a very serious infringement of Article 81 EC and stated that this conclusion was irrespective of whether the cartel had had a measurable impact on the market. The Commission added that it took account of the fact that the collusion concerned only the Spanish market (recital 509 to the contested decision).

a) Determination and adjustment of the ‘starting amount’ of the fine

72 The Commission set the ‘starting amount’ of the fines to be imposed, taking into account the gravity of the infringement, the estimated value of the relevant market, namely EUR 286 400 000 in 2001, the last full year of the infringement, and the fact that the infringement was limited to sales of bitumen in one Member State. In the light of those factors, the Commission set the starting amount of the fines at EUR 40 000 000 (recital 510 to the contested decision).

73 The Commission then placed the undertakings to which the contested decision was addressed in different categories, defined by reference to their relative importance on the relevant market, for the purposes of applying differential treatment, in order to take account of their effective economic capacity to cause significant damage to competition. In order to do so, the Commission relied on their shares, expressed in values of sales, of the Spanish market for penetration bitumen in 2001 (recitals 511 and 512 to the contested decision).

74 Repsol and PROAS, whose shares of the relevant market amounted to 34.04% and 31.67%, respectively, were placed in the first category, BP, with a market share of 15.19%, in the second category, and Nynäs and Petrogal, whose market shares were between 4.54 and 5.24%, in the third category. On that basis, the starting amounts of the fines to be imposed were adjusted as follows (recitals 514 and 515 to the contested decision):

– category 1, for Repsol and PROAS: EUR 40 000 000;

– category 2, for BP: EUR 18 000 000;

– category 3, for Nynäs and Petrogal: EUR 5 500 000.

75 In order to set the amount of the fines at a level that would ensure a sufficient deterrent effect, the Commission considered it appropriate to apply a multiplier of 1.8 and 1.2 respectively to the fines to be imposed on BP and Repsol, by reference to their total turnover for 2006, the last year preceding the adoption of the contested decision, but not to apply a multiplier to the fines to be imposed on PROAS, Nynäs and Petrogal (recital 521 to the contested decision).

76 The starting amounts of the fines were therefore adjusted as follows: (recital 522 to the contested decision):

– Repsol: EUR 48 000 000;

– PROAS: EUR 40 000 000;

– BP: EUR 32 400 000;

– Nynäs: EUR 5 500 000;

– Petrogal: EUR 5 500 000.

b) Duration of the infringement

77 The Commission considered that Repsol and PROAS should be held liable for their participation in the infringement from 1 March 1991 to 1 October 2002, namely a period of 11 years and 7 months.

78 The Commission considered that BP should be held liable for its participation in the infringement from 1 August 1991 to 20 June 2002, namely a period of 10 years and 10 months.

79 The Commission considered that Nynas Petróleo should be held liable for its participation in the infringement from 1 March 1991 to 1 October 2002, namely a period of 11 years and 7 months, and that Nynäs Petroleum should be held liable for its participation in the infringement from 22 May 1991 to 1 October 2002, namely a period of 11 years and 4 months.

80 Lastly, the Commission considered that Galp Energía España (formerly Petrogal Española) and Petróleos de Portugal should be held liable for their participation in the infringement from 31 January 1995 until 1 October 2002, namely a period of 7 years and 8 months, and Galp Energia, SGPS, from 22 April 1999 to 1 October 2002, namely a period of 3 years and 5 months (recital 523 to the contested decision).

81 The Commission increased the starting amount of the fines by 10% for each full year of the infringement, and by 5% for each additional period of at least six months but less than a year. The increases to be applied to the starting amount of the fines were therefore as follows (recitals 524 and 525 to the contested decision):

– Repsol: 115%;

– PROAS: 115%;

– BP: 105%;

– Nynäs:

– Nynas Petróleo: 115%;

– Nynäs Petroleum: 110%;

– Petrogal:

– Galp Energía España and Petróleos de Portugal: 75%;

– Galp Energia, SGPS: 30%.

82 The amounts of the fines to be imposed on each undertaking were therefore as follows (recital 526 to the contested decision):

– Repsol: EUR 103 200 000;

– PROAS: EUR 86 000 000;

– BP: EUR 66 420 000;

– Nynäs:

– Nynas Petróleo: EUR 11 825 000;

– Nynäs Petroleum: EUR 11 550 000;

– Petrogal:

– Galp Energía España and Petróleos de Portugal: EUR 9 625 000;

– Galp Energia, SGPS: EUR 7 150 000.

c) Attenuating circumstances

83 The Commission compared the applicants’ role with the roles of Repsol, PROAS and BP and considered whether a reduction of the amount of the fines was justified. The Commission considered it appropriate to distinguish the different role played by the applicants by taking into account their more limited involvement in the infringement and decided to reduce the amount of the fines by 10% (recitals 566 and 567 to the contested decision).

84 The final amount of the fine imposed on the applicants thus came to EUR 10 642 500 for Nynas Petróleo, of which Nynäs Petroleum was held jointly and severally liable for EUR 10 395 000 (recital 568).

3. Operative part of the contested decision

85 The operative part of the contested decision is worded as follows:

‘Article 1

The following undertakings infringed Article 81[(1) EC] by participating, during the periods indicated, in a complex of agreements and concerted practices in the penetration bitumen business which covered the territory of Spain (excluding the Canary Islands) and which consisted in market-sharing arrangements and price coordination:

…

[Nynas Petróleo], from 1 March 1991 to 1 October 2002; [Nynäs Petroleum], from 22 May 1991 to 1 October 2002.

…

Article 2

For the infringement referred to in Article 1, the following fines are imposed:

…

[Nynas Petróleo]: EUR 10 642 500; of which, [Nynäs Petroleum], jointly and severally liable for the payment of EUR 10 395 000; …

…

Article 4

This Decision is addressed to:

…

[Nynas Petróleo]

[Nynäs Petroleum]

…’

Procedure and forms of order sought

86 By application lodged at the Court Registry on 20 December 2007, the applicants brought the present action.

87 The applicants claim that the Court should:

– require the Commission, first, to produce a formal statement on its reason for initially refusing the applicants access to the recording of the interview of Mr A.T., the manager of BP España’s ‘Bitumen’ department, with the Commission’s investigating agents, and, second, to clarify what policy changes on access to tapes had been made which would have justified such a refusal.

– annul Article 1 of the contested decision in so far as it applies to the applicants for the period 1991 to 1996;

– annul Article 1 of the contested decision in so far as it applies to the applicants in respect of price coordination;

– annul Article 2 of the contested decision in so far as it imposes fines of EUR 10 642 500 on Nynas Petróleo and EUR 10 395 000 on Nynäs Petroleum or, in the alternative, reduce those fines appropriately;

88 The Commission contends that the Court should:

– dismiss the action;

– order the applicants to pay the costs.

Law

89 It is necessary to examine, in the first place, the request for measures of organisation of procedure and, in the second place, the claims seeking annulment of Article 1 of the contested decision, which are put forward principally.

90 It is necessary, in the third place, to examine together the claims seeking annulment of Article 2 of the contested decision, which are also put forward principally, and the claims seeking a reduction of the amount of the fine, which are put forward in the alternative.

91 The applicants put forward the same pleas in support of those last two heads of claim.

A – The request for measures of organisation of procedure

1. Arguments of the parties

92 The applicants claim that when they had access to the file at the Commission’s premises on 8 September 2006 they were able to consult the statements made by BP, Repsol and PROAS in the context of their leniency applications. While examining the typed transcript of the interview of Mr A.T. by the Commission’s staff on 5 November 2003 the applicants asked to hear extracts from the original recording of that interview. Contrary to the conditions of access to the file expressly agreed the previous week, access to the original recording was refused, on the ground that the Commission had changed its policy on access to the file and that the probative document was the transcript of the interview. The official concerned eventually produced the tape recording and the applicants were permitted to listen to a number of short extracts.

93 Three of the four extracts to which they listened revealed inaccuracies in the typed transcript of the interview and the applicants sought and were subsequently granted access to the full recording; they then found that the words: ‘But according to you, don’t worry, it’s not whether it is true or not’ reproduced in the following extract had been omitted from the transcript:

‘Commission: So the only importer that you are aware of …

A.T.: Sorry. There was another one which was Nynäs. Nynäs had some installations. They built installations, some depots, in Villagarcía De Arosa and they were importing products from – I think it was – Antwerp, and that was already and they were another part which was part of the similar size as Petrogal.

Commission: But were they also then part of the discussions or their share was also deducted from the total volumes? So Nynäs during [blank] were not involved?

A.T.: I don’t think they were part of the discussions.

Commission: But according to you, don’t worry, it’s not whether it is true or not.

Commission: And Petrogal?

A. T.: I don’t know if they were told or whether they were having discussions.’

94 The words omitted from the transcript of the interview deserve consideration. At a crucial point in a formal interview by the Commission where the employee of a leniency applicant was being questioned on the possible involvement of other participants in the cartel, a member of the Commission’s staff told him that he could make allegations without being concerned about their veracity. More particularly, the official’s intervention came after the employee, who was being interviewed because of his supposed special direct knowledge of the arrangements, had suggested that the applicants were not involved in discussions pertaining to the cartel. The official’s comments therefore appear, at the very least, to encourage that employee to reconsider his original response. The Commission and the Court should therefore not be able to rely on the evidence provided by the interviewee, particularly following the official’s intervention.

95 Furthermore, the very fact that one of the officials handling the case made such a statement during the interview would clearly give rise to a concern in the mind of a neutral observer as to whether that method is indicative of the method used by the Commission to obtain evidence from leniency applicants and whether that official’s intervention may not have been the only inappropriate indication given by the Commission to the leniency applicants. That intervention took place during a formal recorded interview, whereas the leniency applicants and the Commission also had numerous discussions which were not recorded.

96 An impartial observer would be even more concerned to learn that, for a completely obscure reason, the italicised words were omitted from the typed transcript made available to the applicants as a record of the evidence obtained from a leniency applicant. Indeed, that omission might give rise to the suspicion that the Commission sought to avoid disclosing a sensitive and telling piece of information, namely a statement betraying the Commission’s approach to the evidence supplied. Any such suspicion is reinforced by the fact that the initial refusal to grant access to the recording of the interview was based on an alleged change in policy on access to the file. In addition, the official who made the inappropriate statement was the same official who attempted to prevent access to the recording of the hearing. In the absence of a satisfactory explanation of that intervention and its omission from the typed transcript of the interview, the official’s intervention might be thought to reveal a distinct lack of impartiality and an undeniable disregard for any rigour in the treatment of evidence, contrary to the fundamental principles of the investigation.

97 The Commission appears to believe that it can resolve the problem by not relying on that part of the interview. That approach is insufficient, since the impression that it creates in the mind of an impartial observer affects the way in which the Commission’s conduct must be considered more generally in the present case.

98 The applicants further observe that Mr A.T.’s response was favourable to their defence. Consequently, if it had not been improperly dealt with by the official, Mr A.T.’s statement might well have constituted exculpatory evidence, which the official’s intervention prevented.

99 The defence does not satisfactorily explain how the Commission was able, apparently arbitrarily, to change its policy on access to transcripts and tapes and why crucial elements were omitted from the first version of the transcript provided to the applicants.

100 The fact that the full tape was eventually made available to the applicants and that the Commission, quite correctly, did not rely on evidence which was edited from the transcript provided is not a sufficient answer. If the transcript was wrongly withheld or edited before being given to the applicants and if the Commission’s officials described those events inaccurately, then, even if the Commission has taken the steps described, those are significant matters which are relevant to the applicants’ and the Court’s assessment of whether the Commission properly and critically analysed the information provided to it. The fact that the applicants have not presented their argument as an autonomous plea in law in their action does not reduce the gravity of the problem, nor does it render it unworthy of examination.

101 The applicants are entitled to ask the Commission to explain why they were initially refused access to the recording of Mr A.T.’s hearing and, in particular, whether any, and if so what, changes were made to the policy on access to tapes between 28 August 2006 and 8 September 2006.

102 The applicants therefore request the Court to require the Commission to produce a formal statement on its reason for initially refusing access to the recording of the interview in question and to explain in particular any changes in its policy on access to tapes that may have been made between 28 August 2006 and 8 September 2006 that would have justified that refusal.

103 The Commission contends that the request for measures of organisation of procedure should be rejected.

2. Findings of the Court

104 By their request for measures of organisation of procedure, the applicants seek to obtain, first, a formal statement from the Commission setting out the reason for the initial refusal to grant them access to the recording of the interview of Mr A.T., the manager of BP España’s ‘Bitumen’ department, with the Commission’s staff and, second, clarification regarding the policy changes on access to tapes supposedly relied on in support of that refusal.

105 However, the applicants do not place the Court in a position to know how the elements requested might make it possible to establish that one of the pleas on which they rely in support of their claims for annulment or variation is well founded.

106 In the first place, the applicants’ request does not seek to enable them to have access to parts of the Commission’s file that were not communicated to them during the administrative procedure.

107 In that regard, even if it is accepted that the member of the Commission’s staff criticised by them initially refused them access to the recording of the interview of Mr A.T., that recording was eventually produced and the applicants were permitted to listen to a number of short extracts from it. In addition, at their request, the applicants were subsequently granted access to the entire recording.

108 In the second place, the applicants base their request on the words of a member of the Commission’s staff during the interview of 5 November 2003 (But according to you, don’t worry, it’s not whether it is true or not), on the omission of the extract containing those words in the typed transcript of the interview and on the initial refusal to allow them access to the recording of that interview.

109 However, as the applicants agree in their reply, they do not present the matters set out in the previous paragraph as an autonomous plea in law for annulment, by which they might submit, for example, that those matters are such as to cast doubt on the impartiality and rigour displayed by the Commission in the investigation of the case.

110 It is apparent from the foregoing that the applicants’ arguments do not enable the Court to determine how the statement and explanation whose communication they request may have a bearing on the result of the case (see, to that effect, Case C‑185/95 P Baustahlgewebe v Commission [1998] ECR I‑8417, paragraph 70, and Joined Cases C‑189/02 P, C‑202/02 P, C‑205/02 P to C‑208/02 P and C‑213/02 P Dansk Rørindustri and Others v Commission [2005] ECR I‑5425, paragraph 68).

111 The applicants’ request for measures of organisation of procedure should not therefore be granted.

112 In any event, the words in issue used by the member of the Commission’s staff during the interview of 5 November 2003 do not constitute in themselves a sufficient basis on which to conclude that the Commission displayed a lack of impartiality.

113 In addition, Mr A.T.’s reply, the probative value of which the applicants contest owing to the fact that it was preceded by the intervention in issue of the member of the Commission’s staff, was not used as evidence in the Statement of Objections because of its lack of certainty and precision (recital 113 to the contested decision), something which the applicants do not contest.

114 With respect to the other factors on which the applicants rely, that is to say, first, the absence of the extract containing the words in issue of the member of the Commission’s staff from the typed transcript of the interview of 5 November 2003 and, second, the initial refusal to grant the applicants access to the recording of that interview, they are not sufficient to demonstrate a lack of impartiality.

115 Indeed, as the Commission observes at recital 113 to the contested decision, the typed transcript of the interview of Mr A.T. was provided only for the convenience of the undertakings concerned and the only document which the Commission considers to have evidentiary value are the tapes on which the interview was recorded in full. That is why certain words, sentences or expressions are not included in the transcript. Accordingly, the absence of the extract containing the words in issue of the member of the Commission’s staff from the transcript of the interview does not constitute evidence of a lack of impartiality by the Commission.

116 Furthermore, the applicants were able to have access on the same day to the requested extracts of the recording and, subsequently, to the whole of that recording. In that context, the initial refusal to grant the applicants access to the recording would appear to be a limited incident on the basis of which it is not possible to establish a lack of impartiality.

117 Thus, even if a plea alleging breach of the principle of impartiality had been put forward by the applicants, it would be necessary to reject such a plea without there being any need to grant the measures of organisation of procedure requested.

B – The claims for annulment of Article 1 of the contested decision

118 In support of their claims for annulment of Article 1 of the contested decision, the applicants raise two pleas in law relying, first, on the illegality affecting the finding that they participated in the annual market-sharing arrangements between 1991 and 1996 and, second, on the illegality affecting the finding that they participated in price coordination between 1991 and 2002.

1. First plea for annulment, alleging illegality affecting the finding that the applicants participated in the market-sharing arrangements between 1991 and 1996

a) Arguments of the parties

119 The applicants maintain that the material on which the Commission relied in order to establish that they participated in the annual market-sharing arrangements between 1991 and 1996 does not constitute evidence that demonstrates to the requisite legal standard that there was a ‘meeting of minds’ between the applicants, Repsol, PROAS and BP during the relevant period.

120 In the first place, the applicants infer from the case-law that the Commission must conduct the administrative procedure with due care and examine carefully and impartially all the relevant aspects of the case. Undertakings which, like Repsol and PROAS, are only the second or third to seek leniency are under pressure to provide new and additional material in order to secure the highest possible reduction of the fine. They may thus have an incentive to make exaggerated or partial statements with the understandable aim of securing leniency or better leniency treatment. It cannot therefore be presumed that an undertaking will go against its own interests in its leniency statements concerning the involvement of a third party in the cartel.

121 Those statements should therefore be examined with particular care and the Commission should show its awareness by treating such evidence with due caution and analysing it seriously in order to ensure that it is robust. However, the Commission did not obtain sufficient evidence relating, in particular, to the period 1991 to 1996 and the applicants’ alleged involvement and it therefore attempted to rely on questionable interpretations of the documents available. It also appears to have accepted the statements of the leniency applicants with excessive zeal and without the slightest critical analysis. It follows that significant parts of its findings are unreliable and unsustainable.

122 There is very little documentary material on which the Commission relies or can rely, although contemporaneous documentary material is the best available evidence in the present case. The Commission relies on statements based on the recollections of certain individuals relating to a period more than 10 years ago, and concerning matters in respect of which those concerned have not been shown to have had direct knowledge and of which in most cases they could have had no knowledge. There is no reason to think that those providing witness evidence relied on information provided by employees of the company who have direct knowledge of the facts complained of. The evidence from the period 1991 to 1996 referring to the applicants was wrongly interpreted as implicating them in the infringement, although it merely describes their situation and often the situation of other Spanish undertakings. The Commission did not test the evidence adduced or make further inquiries. The allegations made on behalf of BP, PROAS and Repsol are not consistent and in reality do not implicate the applicants.

123 Repsol’s leniency application and the documents which it subsequently supplied are ambiguous about the time when the applicants became involved in the unlawful arrangements. Repsol provides no contemporaneous documentation specifically implicating the applicants.

124 The only specific references made by PROAS to the applicants’ involvement relate to the period after 1996 and to Mr T.S.B., an employee of the applicants from mid-1996. The absence of any detail relating to 1991 to 1996 suggests that PROAS is able to make only the most vague and unsubstantiated allegations: PROAS initially stated in its leniency application that the cartel between PROAS, Repsol, BP, the applicants and other undertakings had begun in 1987, then maintained that the applicants had become involved in 1991. That impromptu revised statement, made at the hearing on 12 December 2006, was made by Mr J.L.F., an employee of PROAS, who himself acknowledged shortly afterwards that he had direct knowledge only of matters relating to the period beginning in 1997.

125 As for BP, it is not clear on which material the Commission may be relying in order to claim that the applicants participated in the unlawful arrangements in 1991. In so far as the material comes from the interview with Mr A.T., the Commission acknowledges that it cannot be relied on, given the approach which it adopted during that interview. The very fact that the interview was conducted irregularly reinforces the impression that the leniency applicants’ statements were not examined with a sufficiently critical eye.

126 It is true that the Commission should take into consideration the body of evidence as a whole, the absence of contemporaneous evidence for a given year is not sufficient to preclude the applicants’ participation in the infringement and a gap between manifestations of the cartel does not undermine the finding of a continuous infringement, depending on the nature of the case. The applicants observe that they also acknowledge that where a participant does not withdraw from the cartel the Commission can regard the infringement as not having been terminated. The applicants add that it is true that they did not distance themselves from the alleged cartel before 1996.

127 Those arguments are wholly irrelevant, however. The Commission did not have sufficient evidence to conclude that the applicants participated in the infringement before 1996. The possibility of withdrawal is therefore irrelevant. The suggestion that the applicants could otherwise be presumed to have participated in a continuous infringement is simply a misapplication of the case-law.

128 Furthermore, the Commission is ‘plainly wrong’ if it takes ‘considering the evidence as a whole’ to mean interpreting the pre‑1996 evidence in the light of the post-1996 evidence. It must necessarily have sufficient evidence of the pre-1996 infringement and cannot merely find a subsequent infringement and then ‘read it back’ into previous inadequate evidence.

129 The Commission also has a tendency to ‘double count’ the documentary evidence. It thus cites a document from 1992, a document from 1993 and a document from 1994 as evidence of activity in 1991, 1992 and 1993 respectively. Citing documents more frequently does not increase their probative value.

130 In the second place, the applicants examine, year by year, the evidence adduced by the Commission, according to the method followed in the contested decision.

Evidence relating to 1991

– BP España internal note of 30 October 1991 (recitals 201 to 204 to the contested decision)

131 The applicants state that the principal evidence of their involvement in the cartel arrangements from 1991 put forward by the Commission is an internal note of 30 October 1991 from BP España, a copy of which is attached as Annex A.8 to the application, and which is worded as follows:

‘ … Repsol negotiated with Nynäs the latter’s official entry into the Spanish market (without consulting Petromed or BP Med), allocating to them no less than 3.74% of the Spanish market (4% in real terms) in exchange for Nynäs helping Repsol to enter the Belgian market officially …’

132 That note does not support the finding that the applicants participated in the bitumen cartel in 1991. In the first place, it deals with an alleged agreement on the applicants’ entry into the market by referring to the rumoured conclusion of an agreement to exchange bitumen between Repsol and the applicants. In fact such an agreement is a common and legitimate arrangement providing mutual logistical and other advantages between suppliers with supply points in different geographic locations. The author of the note had no contact with the applicants. As is clear from Annex A.9 to the application, BP stated at the hearing of 12 December 2006 that the note merely reported ‘rumours’ in the market and did not clearly show the description of any unlawful conduct. Last, in spite of the importance of that document, the Commission failed to obtain clarification of it at any stage in its investigation.

133 The applicants note that the Commission takes issue with them for not adducing any evidence of the exchange agreement. In fact, the Commission does not rebut the argument based by the applicants on the existence of an exchange agreement bearing no relation to the cartel and merely claims that that argument is irrelevant. In any event, such a rebuttal is not plausible, since such exchanges clearly took place, there are written agreements relating to the years after 1991 and their existence is also confirmed by the witness statement attached in Annex A.11 to the application of Mr L.C., then sales director for Eastern France and the Belgian market with the joint venture Esha-Nynäs Bitumen NV, responsible for marketing bitumen from the applicants’ Antwerp refinery through Smid & Hollander.

134 The existence of bitumen exchanges in 1992-1993 has never before been questioned. The applicants enclose in Annex C.1 to the reply a document apparently drafted in 1992, since it states that ‘no continuation is foreseen for 1993’ and refers to actual exchanges and to possible opportunities for exchanging bitumen. The author of the document observes that ‘Nynäs [has] an exchange agreement with Repsol also for 1992 …’, that the agreement is ‘pure cost driven’ and that it dates from the time when Repsol was importing bitumen ‘from Spain to Antwerp’. That does not constitute direct evidence of the legitimate exchange agreement between Repsol and the applicants in 1992 but it confirms the unchallenged fact of its existence.

135 In the second place, the account of the author of the note in issue contains significant errors, showing that he did not know the date and the circumstances of the creation of Asfaltos Europeos. While the numerous factual flaws in the note are briefly noted by the Commission at recital 203 to the contested decision, they further reinforce the conclusion that the note, in so far as it proves anything, establishes the existence of mere speculation and not of facts.

136 Furthermore, the rumours reported by the author of the note suggest that the cartel arrangements were put in place as part of a deal between the applicants and Repsol to enable the applicants to enter the Spanish market and Repsol to enter the Belgian market. Such a proposal does not make sense and is contradicted by the fact that the applicants had already successfully entered the Spanish market and by the imports of 76 000 metric tonnes (‘MT’) achieved on that market in 1990. Recognising that contradiction, the author draws a distinction between the applicants’ actual entry and their ‘official’ entry. However, the idea of paying a competitor to achieve a new entry at a lower rate of penetration than that already achieved is prima facie implausible, as is such a high price as that consisting in ‘paying’ Repsol by reserving 21 000 MT for it on the Belgian market, on the ground, in particular, that that payment would have constituted consideration for a reduction of around 25% of the applicants’ actual sales to 55 000 MT, or their level of sales in 1989.

137 Nor does the note clearly indicate how the author considered that the Belgian side of the supposed deal would have been executed. There is no suggestion or any reason to believe that Smid & Hollander were willing to give up 6% of the Belgian market to allow Repsol to enter that market. Following the dissolution of the Esha-Nynäs Bitumen joint venture, the applicants pursued an aggressive strategy of acquiring market share in Belgium and in 1992 their sales on the former joint venture’s territory exceeded those achieved by the joint venture in the previous year. In his statement, Mr L.C. clearly rejects the existence of such an asymmetrical arrangement between the applicants and Repsol.

138 In fact, the author of the note confuses with an unlawful cartel arrangement the legitimate exchange and purchase agreements concluded in 1992 by Repsol and the applicants, on the one hand, and with an ‘allocated’ market share, that is to say, a market share agreed between Repsol and the applicants, which he knew from the applicants’ former market share in Spain, on the other hand. According to the most plausible interpretation of the note, its author, an employee of a member of the cartel, presumed that the exchange agreement was the means whereby the applicants were brought into the cartel. BP confirmed at the hearing of 12 December 2006, moreover, that that is what actually happened, but the Commission ignored it in the contested decision.

139 That presumption did not correspond to reality. The exchange agreements, which were lawfully concluded between the applicants and Repsol, were implemented shortly after the note in question was drafted (Annex A.8 to the application) and are still in force today in an amended form. Although the applicants no longer had copies of the agreement with Repsol going back to 1992, the date of the agreement, they informed the Commission of its existence and in their response of 10 May 2004 to a request for information (pages 109 and 118) they referred the Commission to copies of written agreements for later years.

140 Contrary to the terms described in the penultimate paragraph of the note, there was never an agreement or suggestion between Repsol and the applicants for exchanges of volumes approaching 40 000 MT. In fact, the amount exchanged or purchased by the applicants never exceeded 24 000 MT. The volume of 40 000 MT can therefore only have been a rumour.

141 In addition, the ‘allocation’ by Repsol of 55 000 MT to the applicants indicated in the note is not evidence of any agreement between Repsol and the applicants under which those 55 000 MT would represent the applicants’ market share. It is more sensible to interpret it as a report by Repsol of the actual level of supply to the Spanish market by the applicants, based on their actual sales in a recent year, 1989.

142 The Commission’s conclusion that the note is good evidence of a cartel arrangement between Repsol, PROAS, BP and the applicants is therefore manifestly wrong. It follows that the assertion at recital 203 to the contested decision that the note confirms the statements submitted by BP, Repsol and PROAS in the context of their leniency applications is unfounded.

– Statements made in the context of the leniency applications

143 The applicants maintain that the statements and other material provided by BP, the employer of the author of the note of 30 October 1991, do not support either the Commission’s analysis of the note or the finding that the applicants participated in the cartel from 1991. Furthermore, BP’s representative stated at the hearing of 12 December 2006 that the author of the note and its addressee no longer worked for BP, that BP had no knowledge of any meeting with the applicants and that even the reference to Mr T.S.B., an employee of the applicants, was not based on direct knowledge but had been inferred from other documents. When providing the material relating to the elementary issue of who represented the applicants for the purpose of entering into the unlawful arrangements, BP mentioned only Mr T.S.B., who had been employed by the applicants only since 1996 (Annex A.12 to the application). All other references to the applicants are vague and in any event relate to post-1996 developments.

144 All that can be inferred from the interview of 5 November 2003 is that what was said by the BP employees who were interviewed does not shed any light on the extent or the duration of the applicants’ role in the cartel and that the applicants did not participate in any discussion of market allocation.

145 Repsol has not provided any contemporaneous document specifically implicating the applicants. Vague statements are not a sound basis in support of serious allegations of competition law infringements. In addition, the Commission clearly failed to attempt to clarify the matters pertaining to the applicants and, in particular, the precise dates of their alleged involvement.

146 For example, Repsol stated in its leniency application as follows:

‘During the whole period from 1991 the [asphalt] table was formed by PROAS, [Repsol] and BP … Later PROAS and [Repsol] [met] up with Nynäs to negotiate the share that corresponded to each’ (Annex A. 13 to the application).

147 PROAS, for its part, did not provide any contemporaneous document implicating the applicants during the period 1991 to 1996. The only specific references to their involvement made on behalf of PROAS related to the period after 1996 and to Mr T.S.B, who was employed by the applicants after mid-1996. The absence of any detail relating to the period 1991 to 1996 suggests that PROAS is able to make only the most vague and unsubstantiated allegations.

148 The reference to the February 1992 BP internal memorandum in Annex A.14 to the application adds nothing new. That document cannot render reliable otherwise unreliable evidence relating to 1991. In any event it is not sound evidence, for the reasons set out below.

149 Last, no undertaking has alleged that the applicants were subject to the monitoring arrangements mentioned at recital 204 to the contested decision, which the Commission describes as a key part of the cartel arrangements between BP, Repsol and PROAS.

Evidence relating to 1992

150 The applicants refer to a BP internal memorandum from February 1992 entitled ‘Bitumen Market Analysis 1992’, referred to at recital 206 to the contested decision and attached as Annex A.14 to the application. That document states as follows:

‘[The internal market] also includes a quota of 3.74% for Asfaltos Europeos (Nynäs) by express decision [por designio expreso] of Repsol and CEPSA, which means that the resulting internal market is (1 481 100) – 55 393 = 1 425 707 [MT].

That market may be summarised thus:

Repsol (50%): 712 853 [MT]

CEPSA (40%): 570 282 [MT]

BP (10%): 142 570 [MT]

…’

151 The Commission appears to maintain at recital 212 to the contested decision that that memorandum shows that ‘market‑allocation discussions [between Repsol, PROAS and the applicants] concerning the market share that should correspond to [the applicants] must have taken place already in 1992’ and that ‘BP was aware of it’. Even if that interpretation were correct, the fact that the parties to the unlawful arrangements were discussing what they would like to do does not show that the applicants supported that conduct.

152 The reference to the applicants’ market share is descriptive and does not imply that they were involved in the cartel: the memorandum states that the Spanish market amounted to 1 481 100 MT of bitumen, although that figure was subject to the condition that the applicants supplied 3.74%. In other words, the amount available for the purpose of sharing the market between BP, Repsol and PROAS came to 1 425 707 MT. That market was shared as follows: 50% for Repsol, 40% for PROAS and 10% for BP; and it was that sharing of the market that constituted the market-allocation cartel.

153 If the applicants had been a party to the cartel, the members of the cartel would not have deducted the applicants’ portion of the theoretical market from the total amount of the national market before sharing it among themselves. Repsol and PROAS merely recognised the fact that the applicants had entered the market in the late 1980s, that they had a constant presence on the market and that they had quickly built up a market share of 55 000 MT. BP considered that it was entitled to have a larger market share and the author of the memorandum naturally compared the share allocated to BP by the cartel to the applicants’ sales volume. That does not mean, however, that the applicants agreed with Repsol, PROAS or BP that they would have a market share of 3.74%. Indeed, the applicants’ actual sales in Spain in 1992, which amounted to 72 500 MT (Annex A.10 to the application), strongly indicate that there was no agreement of that type.

154 In addition, as stated at recital 210 to the contested decision, the memorandum shows that BP wanted compensation and market valuation mechanisms to be set up with Repsol and PROAS. If the applicants had been members of the cartel and had been of such concern to BP, BP would clearly have wished to make sure that they were included in any arrangement of that type.

155 The expression ‘designio expreso’ might simply mean ‘expressly planned’, in the sense of ‘anticipated’, and in any event does not mean ‘granted’. Likewise, the word ‘dispone’ means simply ‘has’ in this context. The wording of the memorandum is therefore wholly consistent with the applicants’ argument that the memorandum indicates a description or recognition of their position, as the remainder of the memorandum confirms, and does not indicate that their market share was granted, agreed or mandated.

156 The applicants therefore maintain that the assertion at recital 212 to the contested decision that the memorandum confirms the leniency applications submitted by BP, Repsol and PROAS, which do not in themselves provide good evidence of the applicants’ involvement in the cartel in 1992, is unfounded.

157 Last, the assertion at recital 212 to the contested decision that the figure of 55 393 MT cited in the memorandum is the same as the figure of 54 000 MT allegedly agreed for the applicants as their share of the relevant market is entirely consistent with the purely descriptive nature of that document. At the same recital, the Commission refers to the handwritten comment ‘Nynäs 55/56’ in a ‘Repsol document of 1993’, which is none the less clearly dated December 1994. That reference cannot render unreliable evidence reliable and in any event, for the reasons set out below, is not in itself sound evidence on which the Commission could rely.

Evidence relating to 1993

158 The applicants claim that the main piece of evidence put forward by the Commission of their participation in the market-sharing agreements is the ‘Repsol 1993 Table’ drawn up by Repsol and setting out accumulated sales data for 1993 (Annex A.15 to the application). The applicants are not mentioned in that table and it cannot therefore have any probative value so far as they are concerned, other than that their absence from the table implies that they were not part of the cartel.

159 The Commission also relies on two ‘Repsol 1994 Tables’ setting out accumulated sales data for December 1994 (Annex A.15 to the application) and also, in particular, on the handwritten annotations on the first of those two tables. It is somewhat strange that the Commission should seek to rely in relation to 1993 on a table relating to a later period but, in any event, the Repsol 1994 Tables do not alter the analysis of the Repsol 1993 Table. Nor do they reveal any involvement by the applicants in the unlawful activity in 1994.

160 Like the 1993 Table, the Repsol 1994 Tables include two sets of three columns headed ‘Actual’, ‘Theoretical’ and ‘Difference’. The applicants are not mentioned in either of the 1994 tables, which cannot therefore have any probative value so far as they are concerned, other than that they were not part of the cartel.

161 The handwritten notes placed on the first Repsol 1994 Table read as follows:

‘Petrogal 48

Nynäs 55/56

Ferrovial 75

Shell España – 28/30

Esso 55/56

-------

265’

162 Those notes reveal that the applicants were not involved in any unlawful arrangements at that time, since they appear at most to be a description of the applicants’ position on the market.

163 The Commission has acknowledged that Ferrovial, Shell España and Esso were not parties to the unlawful arrangements at any time and that Petrogal was not party to the unlawful arrangements in 1993 and 1994. Yet Repsol’s leniency application is not used against Petrogal in relation to 1994, although it is cited as evidence at recital 213 to the contested decision. The Commission therefore sought to rely solely on the reference to the applicants in that column in order to infer that they were involved in the unlawful arrangements with BP, Repsol and PROAS. That interpretation of the handwritten note is clearly contradictory: the applicants are cited there with the other undertakings which were not members of the cartel and the numbers alongside each handwritten name simply describe their market shares and are not indicative of any market-sharing agreement with the three members of the cartel.

164 In addition, the subsequent addition of handwritten annotations to the first 1994 Repsol Table also suggests that the applicants were not involved in any unlawful arrangements at that time. There is no indication of when those handwritten annotations were made or why they were made. No one has claimed that they were made prior to or following a meeting with the applicants or with any other company having a similar name.

165 Another piece of evidence relied on in connection with the arrangements in 1993 concerns the compensation mechanism which, according to the Commission, was being operated by the cartel (Annex A.16 to the application). The Commission does not contend that the applicants participated in that mechanism, since that document does not as such implicate them. Their absence from the mechanism is however consistent with the fact that they did not participate in the unlawful arrangements between 1991 and 1996.

Evidence relating to 1994

166 The applicants observe that the Commission appears to rely on a single document, the ‘PROAS 1994 List’, an internal list prepared by PROAS dated 28 March 1994 (Annex A.17 to the application), cited at recital 218 to the contested decision. That list contains columns stating the year (1994); Spanish provinces; presumably details of the relevant PROAS district office and department; certain specific addresses and locations (which the Commission suggests are projects); a column headed ‘Consumo no considerar’ about which PROAS was unsure, but which may be a forecast tonnage for the work; a tonnage column; a column apparently indicating the construction company to which the contract to carry out the work had been awarded; a column headed ‘Tipo’ under which the initials ‘GO’ (signifying a major work - Grande Obra) appear in every row; and, lastly, a column headed ‘A’ under which various numbers appear.

167 In the light of certain information provided by PROAS, the Commission asserts that the numbers in the final column correspond to the following producers or importers in the relevant market: 1 – Repsol; 2 – PROAS; 4 – BP; 5 – Nynäs; 7 – Petrogal; 9 – Imports. However, that document certainly does not indicate that the applicants were involved in any unlawful activity in 1994, but shows, on the contrary, that they were not parties to the unlawful arrangements that allegedly existed between Repsol, PROAS and BP.

168 In the first place, the document contains in column ‘A’ numbers which, according to the Commission itself, relate to suppliers, such as Petrogal and other importers, which are not alleged by the Commission to have participated in the unlawful arrangements. The applicants do not maintain that the companies in column ‘A’ are not parties to the arrangement, but merely state that a number of companies, including the applicants, mentioned in column ‘A’ did not participate in the alleged arrangements.

169 In the second place, in so far as that document is intended to show that the applicants were an important player in the unlawful arrangements, it succeeds in showing precisely the opposite, since the number that is supposed to represent the applicants appears only four times in the entire list.

170 In the third place, it follows from internal inquiries concerning the four ‘projects’, carried out after service of the contested decision, that, as the document in Annex A.18 to the application shows, the applicants supplied only Covsa, mentioned in connection with a supposed project at Ordenes Boixaca, in the province of Coruña (Spain) (row 7). However, the applicants supplied Covsa with around 9 500 MT at the municipality of Porrino, in Pontevedra province (Spain), and not 5 000 MT in the province of Coruña, as stated in the table. The applicants also supplied that company just under 2 000 MT at Meson Vento, a district in the city of Coruña. There is no obvious link to Ordenes Boixaca on the PROAS List. The applicants did not make supplies to the other three projects attributed to them. The PROAS 1994 List is clearly highly inaccurate and it cannot therefore be inferred from that list that works were assigned to the applicants under the market-allocation arrangement.

171 The fact that certain numbers used in that document correspond to those used in other documents does not enhance its importance or that of the documents to which it refers. Indeed, owing to its unreliability and to the fact that it refers to suppliers not involved in any unlawful arrangements, that document is evidence against the Commission’s case.

172 None of the other contemporaneous evidence on which the Commission relies in relation to 1994 implicates the applicants in the unlawful arrangements. In addition, the document referred to at recital 223 to the contested decision and attached as Annex A.19 to the application indicates, on the contrary, that the applicants were not parties to any unlawful arrangements in 1994. That document, entitled ‘Total market summary 1994’, provides various figures relating to various bitumen suppliers listed in the left part of the table: ‘RPA (Repsol), PROAS, Asfaltos Europeos, Ferrovial, Petrogal, other importers, non-assigned’. While accepting that Ferrovial and the other importers were not parties to the unlawful arrangements at any time and that Petrogal was not a party to those arrangements in 1994, the Commission none the less sought to rely on the reference to the applicants in that column to substantiate their participation in the unlawful arrangements of BP, Repsol and PROAS.

173 Consequently, none of the documentary material used relating to 1994 is capable of supporting the Commission’s conclusions. The only reference to contemporaneous evidence is to the handwritten annotations which have been commented on above.

Evidence relating to 1995

174 The applicants maintain that, according to the Commission, the document entitled ‘Study Sheet – Forecast 1995 Nynäs’ provided by Repsol (Annex A.20 to the application) contains details of volumes which the applicants had negotiated at a joint meeting with Repsol and PROAS and which they were to supply to particular customers during the forthcoming year. The applicants had no knowledge of the Study Sheet, whereas if it were a document which had been negotiated with the applicants, they would have expected to know about and have a copy of such a document. In so far as they have been able to ascertain, the applicants did not take part in any meeting with Repsol and PROAS in 1995 at which that document could have been discussed. The applicants have never heard of any meeting of that type, they have no document relating to such a meeting or the outcome thereof and, in particular, they do not have the Study Sheet in question.

175 The fact that no such meeting took place is implicitly confirmed, moreover, by PROAS, in so far as that undertaking made specific statements about meetings held from 1997, while there is no mention of any meeting with the applicants before that date. PROAS’s statements concerning meetings with an employee of the applicants mention only one person whom the applicants recruited only in 1996. The fact that one of the supposed participants in the meeting at which the Study Sheet was drawn up is unable to confirm that that meeting, the existence of which the applicants deny, took place therefore fundamentally undermines that allegation.

176 When the Commission examined the 1995 Study Sheet it did not recognise that that document might have been a purely internal document prepared by Repsol for its own purposes. Furthermore, PROAS, which, according to Repsol, originated the format of those study sheets, stated that they were designed for internal use and for subsequent discussion with Repsol. That document may well have been drawn up for use by the other members of the cartel, PROAS and BP. However, no agreement of that type involving the applicants was concluded. That document is not evidence of such an agreement and the applicants deny that any such agreement existed.

177 Furthermore, the Commission accepted all of Repsol’s assertions regarding the document uncritically even where they did not appear to make sense. Thus, when asked whether the sales volumes on the Study Sheet were put together by other suppliers and subsequently communicated to the applicants, Repsol replied, as stated in Annex A.21 to the application (p. 664 (f)), as follows:

‘The volumes in the “OK” column were put together in a joint meeting [of] PROAS, [Repsol] and Nynäs.’

178 That statement appears to be nothing more than a mechanical repetition of a similar comment made in respect of a document concerning BP for the same year. Repsol therefore appears to have answered the Commission’s question concerning the applicants without even looking at the document. In fact, the document relating to the applicants does not contain any ‘OK’ column. Repsol therefore made a statement implicating the applicants which, on the face of the document, is untrue, yet none of those points was remarked upon by the Commission.

Evidence relating to 1996

179 The applicants maintain that there is no contemporaneous evidence relating to them for 1996.

180 The Commission contends that the present plea should be rejected. More specifically, with respect to the evidence relating to 1993 and 1994, the Commission states, first, that it did not rely on the tables submitted by Repsol to establish that the applicants had participated in the 1993 market-allocation arrangements, as may be seen from recitals 213 to 215 to the contested decision and, second, that it did not rely on the PROAS 1994 list to prove the applicants’ participation in 1994. Recital 218 to the contested decision clearly states that that list was drawn up on the basis of market information and contains PROAS’s estimates.

b) Findings of the Court

181 Where, as in the present case, it hears an action for the annulment of a decision applying Article 81(1) EC, the General Court must undertake in general a comprehensive review of the question whether or not the conditions for applying that provision are met (see Case T‑214/06 Imperial Chemical Industries v Commission [2012] ECR II‑0000, paragraph 63 and the case-law cited).

182 According to settled case-law, it is incumbent on the Commission to adduce evidence capable of demonstrating to the requisite legal standard the existence of circumstances constituting an infringement of Article 81(1) EC (Baustahlgewebe v Commission, paragraph 58). It must produce sufficiently precise and consistent evidence to support the firm conviction that the alleged infringement took place (see Imperial Chemical Industries v Commission, paragraph 53 and the case-law cited).

183 It is for the undertaking invoking the benefit of a defence against a finding of an infringement to demonstrate that the conditions for applying such defence are satisfied, so that the Commission will then have to resort to other evidence (Joined Cases 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 Aalborg Portland and Others v Commission [2004] ECR I‑123, paragraph 78).

184 Although according to those principles the legal burden of proof is borne either by the Commission or by the undertaking concerned, the factual evidence on which a party relies may be of such a kind as to require the other party to provide an explanation or justification, failing which it is permissible to conclude that the burden of proof has been discharged (Aalborg Portland and Others v Commission, paragraph 79).

185 It is not necessary for every item of evidence produced by the Commission to satisfy those criteria in relation to every aspect of the infringement. It is sufficient if the body of evidence relied on by the institution, viewed as a whole, meets that requirement (see Imperial Chemical Industries v Commission, paragraph 54 and the case-law cited).

186 It is also necessary to take account of the fact that anti-competitive activities take place clandestinely, and accordingly, in most cases, the existence of an anti-competitive practice or agreement must be inferred from a number of coincidences and indicia which, taken together, may, in the absence of another plausible explanation, constitute evidence of an infringement of the competition rules (see Imperial Chemical Industries v Commission, paragraph 56 and the case-law cited).

187 Moreover, the fact that evidence of the existence of an infringement was not adduced for certain specific periods does not preclude the infringement from being regarded as having been established during a more extensive overall period than those periods, provided that such a finding is based on objective and consistent indicia (see, to that effect, Case C‑113/04 P Technische Unie v Commission [2006] ECR I‑8831, paragraph 169).

188 Lastly, where the Court is in doubt the undertaking which is the addressee of the decision finding an infringement must have the benefit of that doubt, in keeping with the principle of the presumption of innocence which, as a general principle of European Union law, applies in particular to the procedures relating to infringements of the competition rules applicable to undertakings that may result in the imposition of fines or periodic penalty payments (see Imperial Chemical Industries v Commission, paragraph 64 and the case-law cited).

189 With respect to the probative value of statements by leniency applicants, it is settled case-law that there is no general provision or principle of European Union law which prohibits the Commission from using statements against an undertaking which have been provided by other undertakings involved in the infringement. Statements made pursuant to the Leniency Notice cannot therefore be regarded as devoid of probative value on that ground alone (see Imperial Chemical Industries v Commission, paragraph 58 and the case-law cited).

190 Some caution as to the evidence provided voluntarily by the main participants in an unlawful agreement is however understandable, since they might tend to play down the importance of their contribution to the infringement and maximise that of others. None the less, given the inherent logic of the procedure provided for in the 2002 Leniency Notice, the fact of seeking to benefit from the application of that notice in order to obtain a reduction in the fine does not necessarily create an incentive for the other participants in the cartel in question to submit distorted evidence (see Imperial Chemical Industries v Commission, paragraph 59 and the case-law cited).

191 Indeed, any attempt to mislead the Commission could call into question the sincerity and the completeness of cooperation of the undertaking seeking to benefit, and thereby jeopardise its chances of benefiting fully under the Leniency Notice (see Imperial Chemical Industries v Commission, paragraph 59 and the case-law cited).

192 On that point, it should be borne in mind that the Commission may not compel an undertaking to provide it with answers which might involve an admission on its part of the existence of an infringement which it is incumbent upon the Commission to prove (Case 374/87 Orkem v Commission [1989] ECR 3283, paragraph 35).

193 None the less, the risk that the undertaking concerned might not benefit fully under the Leniency Notice, a risk which, as was stated at paragraph 191 above, encourages it to cooperate sincerely with the Commission, including by the submission of evidence or statements running counter to its interests, cannot be placed on the same footing as a coercive measure implemented by the Commission requiring that undertaking to admit to the existence of an infringement. The application of the Leniency Notice stems initially from an initiative by the undertaking in question which seeks to benefit from the provisions of that notice and not from unilateral action taken by the Commission which is imposed on that undertaking.

194 Accordingly, in the absence of any coercive measure requiring the undertaking to incriminate itself, statements by that undertaking admitting the existence of an infringement are not devoid of probative value.

195 Thus, when a person admits that he committed an infringement and therefore admitted the existence of facts other than those whose existence could be directly inferred from the available information, that implies, a priori, in the absence of special circumstances indicating otherwise, that that person had resolved to tell the truth. Statements which run counter to the interests of the declarant must therefore in principle be regarded as particularly reliable evidence (see Imperial Chemical Industries v Commission, paragraph 60 and the case-law cited).

196 However, according to well‑established case-law, an admission by one undertaking accused of having participated in a cartel, the accuracy of which is contested by several other undertakings similarly accused, cannot be regarded as constituting adequate proof of an infringement committed by the latter unless it is supported by other evidence (see Imperial Chemical Industries v Commission, paragraph 61 and the case-law cited).

197 For the purpose of examining the probative value of statements by leniency applicants, the Court takes into account inter alia the strength of consistent evidence supporting the relevance of those statements and the absence of evidence demonstrating that they have tended to play down the importance of their contribution to the infringement and maximise that of other undertakings (see Imperial Chemical Industries v Commission, paragraph 62 and the case-law cited).

198 It is in the light of those general observations that it is necessary to examine the evidence gathered by the Commission in support of the applicants’ participation in the infringement, whether in the context of the present plea or in the context of the other pleas by which the applicants’ participation in the infringement is contested.

199 In the context of the present plea, it should be stated at the outset that the applicants do not submit that they never participated in the cartel; they dispute only that they participated in it between 1991 and 1996.

200 Furthermore, the applicants do not put forward any specific argument or complaint relating to their participation in price coordination, which will in any event be examined in the context of the second plea for annulment.

201 Accordingly, the applicants merely contest the delimitation of the period during which they allegedly participated in the cartel, and in the case only of a part of the unlawful conduct found in respect of that cartel.

202 In that regard, according to recital 125 to the contested decision, Repsol, PROAS, BP and the applicants were members of the ‘asphalt table’ from the beginning of the cartel, that is from 1 March 1991. That recital relies, in particular, on Repsol’s leniency application (footnote 88 to the contested decision).

203 Repsol stated as follows in its leniency application:

‘During the whole period from 1991 the [asphalt] table was formed by PROAS, [Repsol] and BP … Later PROAS and [Repsol] [met] up with Nynäs to negotiate the share that corresponded to each.’

204 By stating that PROAS and Repsol met the applicants later in order to negotiate the share corresponding to them, Repsol did not intend to assert that the applicants started participating in the annual market-sharing arrangements at a point after 1991. That statement signifies that, once the distribution of the market had been agreed between Repsol, PROAS and BP, Repsol or PROAS held a meeting with the applicants in order to communicate to them and negotiate with them the sales volumes and customers to be allocated to them in their area of influence.

205 It is also apparent from Repsol’s leniency application that the applicants were represented between 1991 and 1996 in the market‑sharing meetings by Mr L.M.T. According to the statement of 20 December 2007 by the President of Nynas Petróleo, which is attached as Annex A.23 to the application, Mr L.M.T. was an employee of the applicants until 1996.

206 Repsol’s statements concur therefore in showing that the applicants did indeed participate between 1991 and 1996 in the market‑sharing and customer allocation between the members of the cartel, by means of bilateral meetings with Repsol or PROAS, and that they were represented throughout that period by Mr L.M.T., an employee of the applicants until 1996.

207 Those statements are moreover corroborated by the body of evidence consisting of some of the documents contemporaneous with the material events relied on by the Commission in the contested decision.

208 In the first place, it is common ground that in BP España’s internal note of 30 October 1991, which is mentioned at recital 201 to the contested decision, it is stated as follows:

‘ … Repsol negotiated with Nynäs the latter’s official entry into the Spanish market (without consulting Petromed or BP Med), allocating to them no less than 3.74% of the Spanish market (4% in real terms) in exchange for Nynäs helping Repsol to enter the Belgian market officially …’.

209 In stating that Repsol ‘allocated’ to the applicants, after ‘negotiation’, approximately 3.74% of the Spanish market, the note in issue refers to a sharing of the market.

210 Furthermore, the reference, in the extract cited previously, to consulting Petromed or BP Med, two companies in the BP Group, reveals that, for BP, Repsol failed to comply with an obligation on it to inform BP prior to the allocation to the applicants of 3.74% of the relevant market. Such an obligation to inform a competitor of the market share allocated to another competitor is conceivable only in the context of a market‑sharing arrangement.

211 In the second place, in its internal memorandum of February 1992 entitled ‘Bitumen Market Analysis 1992’ cited at recital 206 to the contested decision, BP evaluates the volume of the relevant market at 1 481 100 MT, of which a ‘quota’ of 3.74% is ‘allocated’ to Nynäs ‘by express decision of Repsol and CEPSA’, which corresponds to a volume of approximately 55 393 MT.

212 The applicants cannot validly maintain that the content of that memorandum means that they supplied 3.74% of the relevant market.

213 Indeed, the statement that the national market ‘includes a [quota of] 3.74% for Asfaltos Europeos (Nynäs) by express decision [por designio expreso] of Repsol and CEPSA’ cannot be regarded as merely descriptive of the share of the market held by the applicants, but, by the wording used, which attests to an allocation of market share decided upon deliberately, refers to a sharing of the market.

214 In addition, the 3.74% share allocated to Nynäs, which appears in BP’s internal memorandum of February 1992, corresponds to that mentioned in the aforementioned internal note of BP España of 30 October 1991.

215 Moreover, the volume of 55 393 MT mentioned in the memorandum of February 1992 is confirmed, virtually identically, in Repsol’s leniency application, where it is specified as follows:

‘The market shares were as follows: [Repsol] 50.3% … , PROAS 39.7%, BP 10%. The market shares were calculated according to the available market, that is to say the total market minus imports … Nynäs was allocated 54 000 MT …’.

216 The virtually identical evaluation, by Repsol and BP, of the share of the relevant market allocated to the applicants confirms that BP España’s internal note of 30 October 1991 and BP’s internal memorandum of February 1992 both relate to the market‑sharing arrangements described in Repsol’s leniency application.

217 It follows from the foregoing that BP España’s internal note of 30 October 1991 and BP’s internal memorandum of February 1992 constitute two contemporaneous items of evidence which corroborate the applicants’ participation in the annual market-sharing arrangements.

218 It is true that the applicants put forward an alternative explanation to account for BP España’s internal note of 30 October 1991. However, that explanation, which is based on the conclusion by Repsol and the applicants of an agreement to exchange bitumen intended to enable them to enter the Spanish market and Repsol to enter the Belgian market, does not appear to be credible in the light of the wording of BP España’s internal note of 30 October 1991 and of the consistency between the information which emerges from that note and the information from the memorandum of February 1992 and from Repsol’s leniency application.

219 In addition, the explanations provided by the applicants concerning the exchange agreement allegedly concluded with Repsol lack detail.

220 Lastly, although the applicants should be able to provide precise and convincing evidence relating to the existence of that agreement to which they claim to have been party, they merely produce, first, the witness statement of one of their employees referring, without providing any other clarification, to the existence of an agreement with Repsol starting in 1992 and, second, an undated and unsigned document, whose origin does not appear, which mentions an exchange agreement with Repsol in 1992. That evidence is not sufficient to demonstrate the existence of an exchange agreement and, a fortiori, to demonstrate that BP España’s internal note of 30 October 1991 concerns such an agreement.

221 In the third place, the document entitled ‘Study Sheet – Forecast 1995 Nynäs’, which was communicated by Repsol and mentioned at recital 225 to the contested decision, contains a column corresponding to provinces, a column corresponding to customers and several other columns displaying, in the heading, a number and specifying the sales volumes relating to each customer. Thus, one of those columns bears the heading ‘5’. The total volume which appears at the bottom of that column is 56 000. In its response of 5 April 2006 to a request for information, Repsol specified that the number 5 referred to the applicants, that the volumes indicated had been negotiated in a joint meeting between Repsol, PROAS and the applicants and that the applicants had been represented by Mr L.M.T., who, as is common ground, was an employee of the applicants until 1996 (see paragraph 205 above).

222 Such a document constitutes a contemporaneous item of evidence which corroborates the applicants’ participation in the annual market-sharing and customer‑allocation arrangements.

223 In that regard, the applicants’ complaint relating to the error allegedly made by Repsol, in its response of 5 April 2006 to a request for information, in referring to a column, headed ‘OK’, which does not appear in the document entitled ‘Study Sheet – Forecast 1995 Nynäs’ cannot be upheld. As is apparent from the documents in the case‑file, the response of 5 April 2006 is composed of several parts, each corresponding to a question by the Commission. Those parts are each referenced by a letter, those letters being sorted in alphabetical order. Repsol referred to the column headed ‘OK’ in part (f) of the response of 5 April 2006. However, the Commission’s questions relating to the document entitled ‘Study Sheet – Forecast 1995 Nynäs’ only start afterwards, from part (g) of that response. Accordingly, the applicants’ complaint relates to a document other than that entitled ‘Study Sheet – Forecast 1995 Nynäs’.

224 In addition, as the Commission observes at recital 225 to the contested decision, the total sales volume of 56 000 MT which appears in the document entitled ‘Study Sheet – Forecast 1995 Nynäs’ is very close to both the theoretical volume of 54 000 MT allocated to the applicants, according to Repsol’s leniency application, and the volume of 55 393 MT indicated in BP España’s internal memorandum of February 1992. It also corresponds to the 3.74% share of the relevant market estimated by BP in its internal note of 30 October 1991.

225 In that regard, it must be observed that it is not contested that the quota allocated by the asphalt table to Nynäs remained stable throughout the entire duration of the cartel (recital 156 to the contested decision).

226 In view of the foregoing and in the light of the case‑law that the fact that evidence of the existence of an infringement was not adduced for certain specific periods does not preclude the infringement from being regarded as having been established during a more extensive overall period than those periods, provided that such a finding is based on objective and consistent indicia (see, to that effect, Technische Unie v Commission, paragraph 169), it must be concluded that the Commission was right to consider that it had sufficient evidence to demonstrate that the applicants participated in the annual market-sharing and customer‑allocation arrangements between 1991 and 1996.

227 Consequently, the Commission was right to find in the contested decision that the applicants had participated between 1991 and 1996 in the annual market-sharing and customer‑allocation arrangements.

228 None of the applicants’ arguments invalidates the foregoing conclusion.

229 In the first place, the applicants rely on having marketed 72 500 MT of bitumen on the Spanish market in 1992.

230 In that regard, it should be recalled that at recital 336 to the contested decision it is stated that the ‘overall scheme’ involving all aspects of the infringement ‘qualifies as an agreement between undertakings within the meaning of Article 81 [EC] in the sense that, during the bilateral and multilateral meetings of the cartel members, the undertakings concerned expressed their joint intention to conduct themselves on the market in a specific way’. The Commission added at the same recital that ‘[t]his behaviour consisted essentially in following a jointly preconceived volume and customer allocation system, price coordination and refraining from competition with regard to customers allocated to the other participating competitors’.

231 Accordingly and even though, elsewhere, the Commission found that certain conduct, like the regular exchange of sales volume information between the undertakings and the direct or indirect contacts to coordinate prices, fell rather within the category of ‘concerted practices’ within the meaning of Article 81 EC (recital 339 to the contested decision), none the less it considered that the cartel could be characterised as an ‘agreement’, within the meaning of Article 81 EC, inasmuch as it had involved, inter alia, bilateral meetings on volume and customer allocation.

232 The characterisation as an ‘agreement’ of the bilateral meetings sharing the market and, therefore, fixing the annual bitumen sales quota is not contested as such by the applicants.

233 As regards agreements, it is sufficient for the Commission to show that the undertaking concerned participated in meetings at which anti-competitive agreements were concluded, without manifestly opposing them, to prove to the requisite standard that the undertaking participated in the cartel. Where participation in such meetings has been established, it is for that undertaking to put forward evidence to establish that its participation in those meetings was without any anti-competitive intention by demonstrating that it had indicated to its competitors that it was participating in those meetings in a spirit that was different from theirs (see Aalborg Portland and Others v Commission, paragraph 81 and the case-law cited).

234 In the present case, it must be concluded, contrary to the applicants’ submission, that the fact that the applicants marketed 72 500 MT of bitumen on the Spanish market in 1992, that is a significantly higher quantity than the sales volume allocated by Repsol and PROAS, does not permit the conclusion that they did not participate in the market‑sharing arrangements. The applicants’ argument does not make it possible to establish that they indicated to their competitors that they were participating in the market‑sharing meetings in a different spirit and that they were opposed to the resulting agreements.

235 Even on the assumption that the bilateral meetings concerning volume allocation should be characterised as concerted practices rather than agreements, the applicants’ argument does not establish that their conduct on the market might have been adopted independently of the contacts that they had with their competitors, contacts whose anti‑competitive object has been established.

236 In the second place, the applicants rely on the lack of probative value of the leniency applicants’ statements. However, they merely claim in a vague and general manner such a lack of probative value on the part of the leniency applicants’ statements, without substantiating their claims.

237 In addition, the material in the case‑file does not permit the conclusion that Repsol and PROAS might have made exaggerated or partial statements with the aim of securing leniency or better leniency treatment.

238 As can be seen at recital 126 to the contested decision, PROAS took the view in its response of 18 November 2005 to a request for information that Repsol, PROAS and BP were permanent members of the ‘asphalt table’, while the applicants participated at the table less frequently.

239 Similarly, it follows from Repsol’s leniency application, which is confirmed by PROAS’s leniency application, that the applicants participated only in the last of the six successive phases relating to the creation of the annual market-sharing arrangements and that, during that last phase, once the sharing of the market between the three bitumen producers, Repsol, PROAS and BP had been agreed upon, Repsol or PROAS met with the applicants to inform them of and negotiate with them the sales volumes and customers corresponding to them in their area of influence (recital 130(f), footnote 95, recital 139, footnote 106, recital 143, and footnote 109 to the contested decision).

240 In the third place, the applicants rely on the Commission’s lack of distance and its uncritical approach with respect to the leniency applicants’ statements.

241 However, by submitting that the Commission ‘appears’ to have accepted the statements of the leniency applicants with excessive zeal and without the slightest critical analysis and that ‘significant parts of its findings are unreliable and unsustainable’, the applicants merely resort to pure allegations which are unsupported by evidence.

242 It should also be noted that the Commission based its finding that the applicants participated in the annual market-sharing arrangements between 1991 and 1996 not only on the statements of the three leniency applicants but also on the responses of the undertakings concerned to its requests for information, as is clear from footnotes 88, 184, 202 and 204 to the contested decision.

243 Those responses would appear a priori to be reliable since, if they were untrue or distorted, their author would be exposed to fines under Article 15(1)(a) and (b) of Regulation No 17 and Article 23(1)(a) of Regulation No 1/2003.

244 In the fourth place, as is apparent from recital 125 to the contested decision, Repsol’s leniency application, BP’s memorandum of 10 July 2002, PROAS’s leniency application and its response of 18 November 2005 to a request for information show that the bilateral meetings between the applicants and, jointly or separately, Repsol and PROAS, for the purposes of negotiation of the applicants’ market share were held without BP’s being present. Consequently, neither BP’s alleged lack of knowledge of any meeting with the applicants nor the ambiguity as regards the extent or duration of the applicants’ role which supposedly affected the comments made by BP’s staff during the interview of 5 November 2003 are such as to call into question the Commission’s finding that the applicants participated in the annual market-sharing arrangements between 1991 and 1996.

245 It follows from all the foregoing that the first plea for annulment must be rejected.

2. Second plea for annulment, alleging illegality affecting the finding that the applicants were involved in price coordination

a) Arguments of the parties

246 In the first place, the applicants take issue with the Commission for not having demonstrated, with respect to the alleged illegality affecting the finding that they were involved in price coordination, all the elements of the infringement or their total duration; they refer, so far as the period 1991 to 1996 is concerned, to the argument which they developed above in response to the finding that they were involved in the market-sharing arrangements. Indeed, their position on the alleged price-fixing arrangements is a fortiori the same as their position on market‑sharing. The supposed references to the applicants in the context of the pricing practices are merely repetitions of the references made in the context of the allegations relating to market-sharing.

247 In the second place, the applicants maintain that there is no reliable evidence of their involvement in the cartel’s price coordination for the infringement period after 1996. The most significant evidence on which the Commission relies for that period consists in a series of price increases or reductions applied by the applicants which, in the Commission’s submission, closely followed those applied by the participants in the unlawful price-fixing arrangements. However, the contemporaneous evidence does not specifically demonstrate the existence of such arrangements and is wholly consistent with rational strategies involving no unlawful behaviour.

248 The Commission refers at recital 316 to the contested decision to certain price changes applied by the applicants and compares them with the price changes applied by Repsol on 3 April 2000, 20 October 2000, 16 November 2001 (a price reduction), 1 April 2002, 27 May 2002 and 3 June 2002. The Commission concludes that owing to the proximity and similarity of Repsol’s price changes and the applicants’, there is good evidence to support the conclusion that the applicants set their prices unlawfully.

249 Before considering those particular examples, the applicants note that the simple fact of parallel pricing is not good evidence of unlawful price‑fixing behaviour. The Commission ignores the documents which show that the applicants adopted their own pricing policy. The self-serving statements of the leniency applicants do not enhance the value of those documents or provide a sufficient basis for any such finding.

250 The Commission states at recital 300 to the contested decision that the market operators requested price increases; it refers to the statement of Mr A.T., the manager of BP España’s ‘Bitumen’ department, reproduced at recital 293, that ‘everyone would say, “we need to increase the prices”’. Given BP’s self‑acknowledged lack of information and the way in which the Commission dealt with BP during the interviews, that is not in any event sufficient credible evidence.

251 It appears that the Commission took that statement out of context, since Mr A.T. said:

‘The fact is it was not that kind of discussion because normally everyone would have seen that the crude oil for instance was rising, [there] was [an] increase in prices, and our cost of product was also rising so our margins were decreasing, so in … normal circumstances everyone would say, “we need to increase the prices”.’

252 The interview with Mr A.T. therefore does not clearly establish whether he was talking about actual meetings or about a feeling on the market that bitumen prices should rise following the increase in raw material prices. Mr A.T. also appears to be speaking about Repsol and PROAS, but the Commission used that quotation out of context in order to suggest the involvement of ‘everyone’. In those circumstances, that statement cannot be admitted as good evidence against the applicants.

253 The BP internal e-mail of 1994 referred to at recital 310 to the contested decision and attached as Annex A.26 to the application, in which BP considers that it is unable to increase prices ‘owing to Repsol’s negative response’, seems to indicate only Repsol’s price leadership and is not evidence of collusion involving the applicants. According to another possible and obvious interpretation of that message, Repsol’s ‘negative response’ could mean that it did not follow BP’s increase in the way in which BP envisaged. The applicants also note that that episode is not in keeping with the Commission’s attempted description of the way in which prices were established on the market.

254 The only other evidence which the Commission appears to put forward against the applicants is to be found in footnote 347 to the contested decision, which refers to the hearing of 12 December 2006 (Annex A.9 to the application). However, that evidence also depends upon a tendentious reading and interpretation of a statement by Mr J.B., a Repsol employee, who stated that he had had meetings with Mr T.S.B., an employee of the applicants, ‘when costs were tight and Nynäs needed to increase prices on the market’ (‘cuando los costes estaban ajustados y Nynäs necesitaba subir los precios en el mercado’).

255 According to the Commission, ‘necesitaba’, the imperfect of the Spanish verb ‘necesitar’, means that the applicants needed to increase prices on the market. However, the phrase could equally mean ‘when costs were tight and I [Mr J.B.] needed to increase prices on the market’. The Commission therefore quite improperly opts for the translation which is the most damaging to the applicants.

256 The Commission was not entitled, at paragraph 105 of the defence, merely to present that translation as ‘totally coherent with its context’. The translation put forward by the applicants is at least as valid.

257 The price of oil accounts for around 80% of the total costs of the production of bitumen, the price of which should therefore follow the price of oil quite closely on a competitive market. Nynäs purchased bitumen for delivery to Villagarcía de Arosa from its parent company at a price based on the previous month’s published oil price, plus a mark-up of approximately USD 20. The prices charged by the applicants closely followed the price of oil, according to the laws of competition. Nynäs charged its customers a margin over the price of imported bitumen in order to recover the costs of its storage depot, administration and operations in Spain. The applicants kept a fairly steady margin until heavy fuel oil prices rose rapidly in 1999 and bitumen price increases were unable to respond as rapidly.

258 The Commission refuses to accept that the applicants may have entirely sound reasons for adopting the pricing strategies which they did adopt. Well-respected economic theory recognises that it may be a rational strategy for a fringe player simply to follow price leaders. In the present case, Repsol and PROAS were clearly the price leaders on the market. There were therefore clearly coherent and rational economic explanations for pricing strategies that involved following the prices set by the leaders, none of which relied on unlawful arrangements.

259 An undertaking is not required to follow that single strategy, however, and may thus consider it appropriate, depending on all the circumstances, either to gain market share by reducing prices or to gain profit by increasing its prices to above the market leaders’ prices. It is clear that, at least until 1999, the applicants followed the prices of the leaders, Repsol and PROAS, as is indicated in an internal document of the applicants which states that they were for the first time the price leader on the market in September 1999, when they applied an increase of USD 20. They subsequently attempted on certain occasions to diverge from the price leaders’ prices. Both of those strategies were legitimate. The Commission erred in concluding that applying more than one pricing strategy over time was contradictory. More generally, the Commission ought to have recognised that the strategies adopted had a rational and lawful explanation.

260 The Commission’s approach suggests a tendency to elide the differences between market-allocation and price-fixing arrangements. That approach is unjustified, as it avoids ensuring that the requisite standard of proof is met. The Commission’s approach enables it, on the basis of ‘guilt by association’, to make allegations without sufficient evidence and therefore does not constitute a sound basis for such a serious finding.

261 That is particularly so in the present case, since, according to the Commission’s own argument, the three main players in the alleged unlawful arrangements were subject to very different conditions from those to which the fringe players were subject, in that they applied monitoring and compensation mechanisms between them in the context of the market-allocation arrangements. There are, therefore, good reasons to suppose that even if fringe players had been implicated in market‑allocation arrangements the same conditions would not have applied to them as to the main players.

262 Likewise, even on the assumption that it could be shown that the three main players entered into price-fixing arrangements, that would not mean that the fringe players were included. Indeed, it does not appear that it would be necessary to agree prices with the fringe players, given the market shares of the three main players: either the fringe players would automatically follow the prices, or their attempts to compete would have little adverse effect on the main players.

Price increase on 3 April 2000

263 The applicants observe that they changed their prices on 3 April 2000, one week before Repsol changed its prices. That completely contradicts the account provided by Repsol and PROAS on which the Commission relied, namely that it was Repsol and PROAS that took the decision and the initiative on implementing price changes. Moreover, the BP e-mail referred to at recital 313 to the contested decision clearly indicates that the other bitumen suppliers were the first to apply a price rise and that they were forced to adjust their prices downwards when the usual price leader, Repsol, did not follow. The lack of free choice to which the Commission refers simply means that an undertaking cannot price a commodity product at a level considerably higher than that applied by the largest supplier on the market and expect customers to accept that higher price.

Price change on 20 October 2000

264 The applicants observe that the Commission relies on the fact that they changed their price on 20 October 2000 and that Repsol changed its price on 25 October 2000. Once again, the dates do not support the Commission’s case and contradict the account on which it relied, namely that Repsol and PROAS took the decision and the initiative to apply the price changes. The Commission also ignores the fact that Repsol’s price change was PTE 4 000, whereas the applicants’ price change was only PTE 3 400, a difference of more than 15% (equivalent to EUR 4). In the applicants’ experience, customers will change supplier where there is a price differential for bitumen of EUR 1 or 2. The existence of that differential shows therefore that the applicants did not participate in any price-fixing arrangements.

Price reduction in November 2001

265 In the applicants’ submission, the Commission maintains that, since an internal document discovered during the inspections (Annex A.24 to the application, footnote 372 to the contested decision) refers to a price reduction of PTE 2 400 on 16 November 2001, it shows that the applicants had prior contact with Repsol and PROAS. However, that document is dated 20 November 2001 and more than four days after the price reduction. It is not, therefore, good evidence of any prior contact, but merely demonstrates that there has been a price reduction. Furthermore, the price reduction referred to is clearly that applied by the applicants (PTE 2 400) and not that applied by Repsol (PTE 2 500).

Price change in April 2002

266 The applicants observe that the price changes implemented in April 2002 took effect on 1 April 2002. The applicants made three different price changes, depending on the customer concerned. According to recital 317 to the contested decision, although the price variations are generally for absolute amounts, the April 2002 variation shows that that was not always the case for the applicants. The Commission appears to draw a negative inference from the fact that PROAS made a similar price change, but almost two months later. That inference is manifestly unfounded.

Price changes in May and June 2002

267 In the applicants’ opinion, the price changes in May and June 2002 are not evidence of collusion on prices and the Commission does not state the amount of the changes. At recital 318 to the contested decision the Commission compares the dates of price changes by BP and by the applicants, but the conclusion which it draws is unclear. The reference to the BP internal table entitled ‘Variation Prices Paving’ shows only that BP was not interested in the applicants’ price rises, since the table does not include those price rises.

268 There is a broad coincidence of dates between some of the price changes applied by BP and by the applicants, but again, that is not sufficient evidence of unlawful arrangements between the parties. It is a phenomenon that would be expected on the market for a commodity product such as bitumen. Furthermore, the discrepancy between the dates of the price rises in December 1999 and the EUR 3 price difference in the April 2002 price rises clearly indicate that there was no collusion on prices between the applicants and BP.

Price change in September 1999

269 The applicants take issue with the Commission for having omitted to refer to the price variation applied in September 1999, which shows that the applicants followed an independent pricing policy and were not aware of how the other undertakings on the market would react to their price changes. That incident also categorically contradicts the self‑serving general statements made by PROAS and by Repsol as regards the ‘equal’ involvement of the applicants and shows that Repsol was clearly regarded as the price leader.

270 An internal document of the applicants attached as Annex A.25 to the application contains the following:

‘Nynäs in September for the first time has been leader in market price, [applying an] increase ([of USD] +20). Will Repsol follow?’

271 A hand-written annotation made more than two weeks later on the same document states as follows:

‘The rest of [the] main players (Repsol, PROAS and BP) followed two weeks later with a smaller increase ([of USD] + 13 dollars)’.

272 The applicants clearly acted alone in taking the decision to increase their bitumen prices by USD 20. The other suppliers did not follow that increase, but increased their prices by only USD 13. The applicants subsequently changed their prices in order to offer competitive prices and their price increase was eventually PTE 2 000, or EUR 12. In fact, the applicants were unable to maintain their initial price rise and were therefore forced to adjust to the market norm.

273 That incident shows as follows:

– the applicants tried to fix their prices competitively and did not simply follow the prices fixed by the other undertakings;

– the dominance exercised over prices by Repsol and PROAS had a significant effect and it was difficult for the applicants, in their capacity as a fringe player, to challenge that dominance;

– the incident is therefore inconsistent with agreements being reached on prices between Repsol and PROAS, on the one hand, and the applicants, on the other.

274 At paragraphs 99 and 100 of the defence, the Commission failed to acknowledge that it ignored the price increase applied by the applicants in September 1999. That omission is not insignificant, as it is symptomatic of the Commission’s unjustifiably selective approach to evidence. Furthermore, the reason stated for rejecting that evidence (‘[i]n any case, it has to be recalled that, even in this particular circumstance, the parties took only two weeks to align their prices’) simply shows the Commission’s failure to understand the natural and lawful movement of prices on that commodity market.

275 The Commission contends that the present plea should be rejected.

b) Findings of the Court

276 The applicants allege first of all the illegality of the finding that they were involved in the price‑coordination arrangements between 1991 and 1996 by confining themselves to a general reference to the argument which they developed in support of the first plea for annulment, alleging illegality of the finding that they were involved in the market-sharing arrangements during that period.

277 However, as stated above, in the context of the first plea for annulment the applicants did not put forward any argument relating to price coordination.

278 It must therefore be considered that the applicants’ participation in the price coordination between 1991 and 1996 is established in the absence of any arguments by the applicants seeking to call in question that participation.

279 In any event, that participation is established (see paragraph 308 below).

280 As regards the applicants’ participation in the price‑coordination activities during the period after 1996, it is necessary to determine whether the consistency, on the one hand, of the leniency applicants’ statements presented both in their leniency applications and their responses to the Commission’s requests for information with, on the other, the documents contemporaneous with the material events makes it possible to find the existence of that participation.

281 With respect to the leniency applicants’ statements, first, as is apparent from recitals 294 and 295 to the contested decision, both Repsol and PROAS acknowledged in their leniency applications the existence of price‑coordination activities on the Spanish bitumen market and admitted to having agreed uninterruptedly from 1991 to 2002 with each other and, to a lesser extent, with BP, bitumen price variations and the time of their implementation.

282 Second, as is apparent from recitals 296 and 305 to 307 to the contested decision, Repsol and PROAS stated that the agreements on price variation and the time of their implementation were communicated to the other undertakings and that those undertakings more or less followed the price variations thus agreed.

283 As footnote 339 to the contested decision states, the description of the price‑coordination activities set out in that decision is based, inter alia, on Repsol’s leniency application, on its response of 5 April 2006 to a request for information and on the leniency application of PROAS and its responses of 7 April and 9 May 2006 to requests for information. Those documents, to which the contested decision refers on numerous occasions (inter alia, at recitals 296 and 305 to 307 to the contested decision which were mentioned at paragraph 282 above), concur in establishing the applicants’ participation in the price coordination.

284 It is apparent from Repsol’s leniency application that the price variations were usually decided upon during bilateral meetings between the heads of the commercial management of Repsol and PROAS, and were subsequently brought to the attention of the other market actors, during bilateral meetings or by telephone, but not in writing.

285 In addition, in its response of 5 April 2006 to a request for information, Repsol replied as follows to the question how the other market actors were informed of the price variation:

‘The information was communicated during bilateral meetings or by telephone, but not in writing.

…

Once adopted, the decision to vary prices was communicated to the other operators by PROAS’s and Repsol’s sales managers. Over the years, that communication was made by different means according to the operators active at the time and the staff of each undertaking, but in general it was made during meetings or by telephone, but not in writing.’

286 In the same response, Repsol went on to state that the price variations were communicated to the operators active at the time from 1991 to 2002 and cited, inter alia, Mr L.M.T. and Mr T.S.B., employees of the applicants, as being amongst the sales managers to whom those communications were addressed.

287 Thus, where Repsol refers, in its leniency application or in its response of 5 April 2006, to the other market actors or operators, it is referring, inter alia, to the applicants.

288 With respect to PROAS, in its response of 7 April 2006 to a request for information, it states that the other market actors were informed of the price variations from the inception of the cartel until 2002 and that they followed the price variation agreed.

289 PROAS also stated, in its response of 7 April 2006, that the bilateral agreements between Repsol and PROAS or between Repsol and BP represented a rapid and flexible method for assisting the operation of the asphalt table, given that the other members of the asphalt table were informed of those agreements and aligned themselves with the agreements reached between Repsol and PROAS once the agreements had been communicated to those members by telephone.

290 Also in its response of 7 April 2006, PROAS, just like Repsol in its own statements, cited Mr T.S.B., an employee of the applicants, as one of the employees of the other market actors to whom the price variations were communicated.

291 Thus, when PROAS refers, in its response of 7 April 2006, to the other market actors or the other members of the asphalt table, it is referring, inter alia, to the applicants.

292 Third, Repsol explained in its response of 5 April 2006 to a request for information that the commercial managers of the other bitumen suppliers with a smaller market share, including the applicants, usually contacted Repsol and PROAS whenever an important increase of oil prices occurred to ask for an agreement to increase prices, given that their inferior logistical position made their margins more sensitive to cost variations.

293 Similarly, in its responses of 7 April and of 9 May 2006 to requests for information, PROAS confirmed that all the market operators, including the applicants, took the initiative on different occasions to request price increases.

294 In its response of 9 May 2006 mentioned above, PROAS added that, in the context of the discussions relating to price increases, the applicants assumed exactly the same role as Repsol and PROAS as regards the initiatives taken with a view to price variation, even though Repsol had the final say.

295 The various self-incriminating statements mentioned just above are sufficiently precise and consistent to establish that the applicants participated, sometimes actively, in the price‑coordination arrangements between 1991 and 2002 and thus, necessarily, between 1996 and 2002.

296 With respect to the documents contemporaneous with the material events, the existence of price coordination is demonstrated, inter alia, by contemporaneous documents produced by BP (recitals 310 and 313 to the contested decision).

297 Above all, as the Commission indicates at recital 303 to the contested decision, Repsol stated in the extract from its leniency application of 31 March 2004 that Mr J.H., its representative, and Mr J.T.G., PROAS’s representative, held, inter alia, two bitumen price consultation meetings on the following dates:

– 24 October 2001: (price change on 16 November 2001);

– 21 March 2002: (price change on 1 April 2002).

298 According to recital 316 to the contested decision, a document was discovered at the applicants’ premises during the inspection of 1 and 2 October 2002. That document consists of a price list bearing the following entry:

‘Last decrease prices Spain = 16.11.01 (-2,400)

New prices as from 16.11.01’

299 That document reveals the existence of a price variation which took effect on 16 November 2001. That date corresponds exactly to the date on which one of the price variations decided upon in advance by Repsol and PROAS took effect (see paragraph 297 above). In addition, as can be seen from recital 316 to the contested decision, the size of the reduction announced by the applicants (PTE ‑2 400) was only EUR 1 less than that planned for the same date by Repsol (PTE ‑2 500).

300 It is apparent moreover, also on reading recital 316 to the contested decision, that a number of faxes by the applicants, discovered at their premises during the inspection of 1 and 2 October 2002, announce to some of their customers for 1 April 2002, a date which coincides with the other date mentioned in paragraph 297 above, a price increase that is either identical (+ EUR 16), or virtually identical (+ EUR 17 or 17.5), to the increase effected by Repsol for the same date. In addition, recital 318 to the contested decision refers to a BP internal table, according to which BP introduced a price increase of EUR 20 on 2 April 2002.

301 It follows from those elements, whose existence is not called in question by the applicants, both that the dates on which the price changes introduced by Repsol, the applicants and, in one of the cases, BP were implemented were the same or, at the very least, virtually the same and that those changes were of an identical or similar size.

302 The larger size of the price change applied by BP on 2 April 2002 in comparison with that introduced by Repsol and the applicants can be explained by the fact that, according to what PROAS states in its leniency application, BP generally aligned its prices whilst increasing them in relation to those agreed between Repsol and PROAS.

303 In so far as the applicants submit that they were forced to align with the price-leaders’ price changes, they indirectly confirm the existence of the close similarity in terms of size and timing of their price changes in relation to Repsol’s and PROAS’s.

304 Moreover, the applicants cannot validly maintain that that closeness in timing is explained by the fact that, on a competitive market, the price of bitumen should follow the price of oil quite closely.

305 The date on which the two price increases just examined were implemented by the applicants had been decided upon in advance in the context of bilateral meetings between Repsol and PROAS (recital 303 to the contested decision). Although the operation of a competitive market might possibly be the cause of price changes by different undertakings on relatively close dates, it is difficult to explain by the sole operation of a competitive market that the applicants applied price changes exactly on the date fixed in advance by Repsol and PROAS. That is all the more true given that the documents relating to the price change implemented on 1 April 2002 predate that change.

306 In addition, according to the concurring statements of Repsol and PROAS, coordinated bitumen price changes were decided upon in particular when oil prices increased and decreased, which indicates that, far from being automatic, the adjustment of the prices of bitumen to take account of oil price variations required the implementation of concerted action.

307 Those factors confirm that the adjustment of bitumen prices to take account of price variations of raw materials was not the result of independent decisions adopted by the economic operators, but of price coordination between those operators.

308 Consequently, in the light of the case‑law cited above according to which the fact that evidence of the existence of an infringement was not adduced for certain specific periods does not preclude the infringement from being regarded as having been established during a more extensive overall period than those periods, provided that such a finding is based on objective and consistent indicia (see, to that effect, Technische Unie v Commission, paragraph 169), it follows from the leniency applicants’ self-incriminating and concurring statements, which are corroborated by the documents examined above contemporaneous with the material events, that the applicants participated, sometimes actively, in the price‑coordination arrangements between 1991 and 2002 and thus, necessarily, between 1996 and 2002.

309 The applicants’ argument relating to the independent nature of their pricing behaviour does not invalidate the foregoing conclusion.

310 In that regard, it should be borne in mind that, at recital 339 to the contested decision, the Commission considered that direct or indirect contacts to coordinate prices can be classified as ‘concerted practices’ within the meaning of Article 81 EC.

311 That characterisation is not contested as such by the applicants.

312 As regards concerted practices, subject to proof to the contrary, which the economic operators concerned must adduce, it must be presumed that the undertakings taking part in the concerted action and remaining active on the market take account of the information exchanged with their competitors in determining their conduct on that market (Case C‑199/92 P Hüls v Commission [1999] ECR I‑4287, paragraph 162, and Case C‑49/92 P Commission v Anic Partecipazioni [1999] ECR I‑4125, paragraph 121).

313 In the present case, in the light of the precise and consistent evidence examined previously, not only has the applicants’ participation in the concerted action been established, but the fact that the applicants took account of the information exchanged in determining their conduct has been as well.

314 Accordingly, the applicants’ argument relating to the independent nature of their pricing behaviour is not well founded.

315 Moreover, the fact that, in September 1999, the applicants were the first to introduce a bitumen price increase which was followed two weeks later by a smaller increase implemented by Repsol, PROAS and BP is not sufficient, in itself, to demonstrate the existence of independent conduct on the market.

316 As regards the document on which the applicants rely in order to establish the existence of such conduct, its probative value is limited, since it is an undated and unsigned document, whose origin does not appear and which, moreover, is internal to Nynäs. Whilst admittedly the applicants state that that document is part of the file of the procedure followed by the Commission, they do not specify whether they provided that document themselves or whether it was gathered during the inspections of 1 and 2 October 2002.

317 Lastly, even on the assumption that the direct or indirect contacts relating to price coordination should be classified as agreements and not as concerted practices, the applicants’ argument relating to the independent nature of their pricing behaviour is not sufficient to establish that they indicated to their competitors that they were participating in those agreements in a different spirit and that they were opposed to them.

318 It follows from all the foregoing that the Commission cannot be regarded as having found unlawfully, in the contested decision, that the applicants participated in the price coordination between the members of the cartel in respect of the period between 1991 and 2002.

319 In those circumstances, the second plea for annulment cannot be upheld and, accordingly, the claims seeking annulment of Article 1 of the contested decision must be rejected.

C – The claims for annulment of Article 2 of the contested decision and, in the alternative, for variation

320 The applicants request that the Court annul Article 2 of the contested decision, which relates to the fine and, in the alternative, that it reduce the amount of the fine imposed on them.

321 It should be recalled that the Courts of the European Union are empowered to exercise their unlimited jurisdiction where the question of the amount of the fine is before them (Case C‑3/06 P Groupe Danone v Commission [2007] ECR I‑1331, paragraph 62).

322 Accordingly, the applicants, which specifically challenge the fine imposed on them and do not challenge the contested decision in its entirety or only the articles of the operative part of that decision which do not relate to the fine, must be regarded as putting forward claims for variation, despite the use in their pleadings of the term ‘annulment’.

323 In that regard, it should be pointed out that, in Case C‑441/11 P Commission v Verhuizingen Coppens [2012] ECR I‑0000, even though the applicant in that case claimed principally that Articles 1 and 2 of the decision that it was contesting should be annulled, articles relating to the finding of the infringement and to the fine respectively, and, only in the alternative, that that fine should be reduced (paragraph 10 of the judgment), the Court of Justice partially annulled Article 1 and, without annulling Article 2, reduced the amount of the fine in the exercise of its unlimited jurisdiction.

324 In support of their claims for variation, the applicants put forward three pleas in law, the first alleging illegality affecting the finding that they participated in the cartel, the second alleging illegality affecting the characterisation of the annual market-sharing arrangements and the price‑coordination activities as a ‘very serious infringement’, and the third alleging errors in the assessment relating to their degree of involvement in the infringement.

1. First plea for variation of the fine, based on the illegality affecting the finding that the applicants participated in the cartel

a) Arguments of the parties

325 The applicants maintain that the fine imposed on them must be substantially reduced, in the light of their claims for annulment set out above. The amount of the fine does not take account of the failure to show that they participated in any unlawful arrangements between 1991 and 1996 or in any price-fixing arrangements after 1996.

326 The Commission does not specifically answer that argument.

b) Findings of the Court

327 It should be noted at the outset that the review of legality by the Courts of the European Union is supplemented by the unlimited jurisdiction conferred upon them by Article 17 of Regulation No 17, in accordance with Article 172 of the EEC Treaty, and subsequently by Article 31 of Regulation No 1/2003, in accordance with Article 229 EC (now Article 261 TFEU). That jurisdiction empowers the Courts, in addition to carrying out a mere review of the lawfulness of the penalty, to substitute their own appraisal for the Commission’s and, consequently, to cancel, reduce or increase the fine imposed by the Commission (Case C‑272/09 P KME and Others v Commission [2011] ECR I‑0000, paragraph 103, and Case C‑386/10 P Chalkor v Commission [2011] ECR I‑0000, paragraph 63).

328 It is in the light of those general observations that it is necessary to examine the applicants’ complaints, whether in the context of the present plea or in the context of the other pleas seeking variation of the fine.

329 With respect to the present plea, since examination of the claims for annulment of the contested decision has not disclosed any illegality or error, the plea, which is not based on any specific arguments, cannot succeed.

330 It must therefore be rejected.

2. Second plea for variation of the fine, alleging illegality affecting the characterisation of the market-sharing arrangements and the price‑coordination activities as a ‘very serious infringement’

a) Arguments of the parties

331 The applicants take issue with the Commission for having, at recital 509 to the contested decision, characterised the cartel in question as a ‘very serious’ infringement of Article 81(1) EC. In previous decisions the Commission has characterised an infringement as ‘serious’ where it was limited to a national market and to a sector of medium economic importance and where it did not directly affect consumers. The Commission ought therefore to have concluded, in accordance with its own analysis and its practice in previous decisions, that the present arrangements were ‘serious’ and to have reduced the fine imposed on the applicants accordingly.

332 The Commission contends that the present plea should be rejected.

b) Findings of the Court

333 At recital 500 to the contested decision, the Commission observed that each of the two groups of anti‑competitive conduct established, horizontal market‑sharing arrangements and price coordination, was, by its very nature, among the most serious types of infringement of Article 81 EC and that the case‑law has confirmed that this type of restrictions may warrant the classification of infringements as ‘very serious’ solely on the basis of their nature, without it being necessary for such conduct to have a particular impact.

334 The Commission reaffirmed at recital 509 to the contested decision that, in view of the nature of the unlawful conduct found, the undertakings to which the contested decision was addressed had committed a very serious infringement of Article 81 EC and stated that this conclusion was irrespective of whether the cartel had had a measurable impact on the market.

335 In that regard, it follows from the indicative description of very serious infringements set out in Section 1 A of the 1998 Guidelines, which the Commission applied in the contested decision, that agreements or concerted practices involving, as in the present case, market‑sharing and price‑fixing may be classified as very serious on the basis of their nature alone, without there being any need to characterise such conduct by reference to a particular geographic area or by the impact of the infringement on the market.

336 That conclusion is supported by the fact that, while the indicative description given by Section 1 A of the 1998 Guidelines of the types of infringement liable to be considered as serious mentions that they comprise infringements of the same type as those defined as minor ‘but more rigorously applied, with a wider market impact, and with effects in extensive areas of the common market’, the description of very serious infringements makes, by contrast, no mention of a requirement other than the nature of the infringement (Joined Cases C‑125/07 P, C‑133/07 P, C‑135/07 P and C‑137/07 P Erste Group Bank and Others v Commission [2009] ECR I‑8681, paragraph 103, and Case T‑38/02 Groupe Danone v Commission [2005] ECR II‑4407, paragraph 150).

337 It follows that the market‑sharing arrangements and price‑coordination activities in which the applicants participated could be rightly considered by the Commission to be very serious infringements on the basis of their nature alone.

338 It must moreover be recalled that the Court of Justice has held that the Commission’s practice in previous decisions does not itself serve as a legal framework for the fines imposed in competition matters and that decisions in other cases can give only an indication for the purpose of determining whether there is discrimination (Case C‑549/10 P Tomra Systems and Others v Commission [2012] ECR I‑0000, paragraphs 104 to 108). Accordingly, the applicants cannot usefully rely on such practice.

339 At recital 509 to the contested decision, the Commission did not therefore erroneously classify the conduct in issue as a very serious infringement of Article 81(1) EC.

340 In the light of the foregoing, since the Commission did not commit any error, the present plea must be rejected as unfounded.

3. Third plea for variation of the fine, alleging errors of assessment relating to the degree of the applicants’ involvement in the infringement

a) Arguments of the parties

341 In the first place, the applicants claim that the Commission established the degree of their involvement in the infringement throughout its entire duration on the basis of evidence relating to 2001 and 2002. In so doing the Commission failed properly to recognise that, even on its own analysis of the evidence, their involvement during the earlier period of the infringement was much more limited.

342 None of the leniency applicants is in a position to provide the slightest specific detail of the applicants’ alleged contacts, except for 2001 and 2002. Furthermore, even if those contacts could give the impression that the applicants played any active role in the alleged unlawful arrangements, which the applicants deny, the Commission erred in relying on those late events as a basis for concluding that that degree of involvement had obtained throughout the entire period of the infringement. There is no good reason why the Commission should rely on such a presumption, in breach of the principle that the applicants should have the benefit of the doubt in relation to such an issue.

343 The Commission therefore made at least one error in failing properly to consider the duration over which the applicants were, on its own analysis, actively involved in the alleged unlawful arrangements, in particular at recital 543 to the contested decision.

344 In the second place, the applicants maintain that when setting the amount of the fine the Commission ought to have taken account of their limited involvement in the cartel, which can be established on the basis of a number of indicia.

The applicants’ size on the Spanish bitumen market

345 The applicants emphasise that they were and still are a small supplier on the Spanish bitumen market. They are not a major oil company and their group cannot be compared with the groups to which the other undertakings to which the contested decision is addressed belong.

346 In addition, they held at most only 5% by volume of the Spanish bitumen market throughout the entire period of the infringement. The amount of any fine should therefore be limited in order to reflect their relative size and should be at the lower end of any bracket which the Commission considers to be justified.

The applicants’ presence was limited to a single province of Spain

347 The applicants assert that their involvement in the cartel was limited to contacts with Repsol and PROAS and related only to their activities in Galicia. When the Commission assessed the gravity of the infringement, it was not therefore entitled to take as a starting point for the calculation of the fine the extent of an infringement committed in a cartel having national reach and influence. The operation and the impact of the asphalt table did not depend on the applicants’ contribution and there is no reliable evidence that they were involved in an infringement that was national in nature.

348 It is insufficient and erroneous to contend, as the Commission does at recitals 347 and 542 to the contested decision, that the applicants participated in arrangements with a national dimension on the basis that they ‘should have known’ that the discussions between Repsol, PROAS and BP concerned the whole of Spain. In the absence of any evidence, the Commission found that the applicants had knowledge of arrangements covering the whole of Spain and that they consented to such arrangements. The Commission therefore exaggerated the nature and extent of the applicants’ involvement in the infringement. Even if the infringement covered the whole of Spain, the applicants’ involvement did not. The Commission did not properly take that factor into account when setting the amount of the fine.

The applicants’ limited and passive or follow-my-leader role in the infringement

349 The applicants observe that in setting the amount of a fine the Commission is required to follow a coherent and non‑discriminatory policy, in accordance with the 1998 Guidelines, section 3 of which provides, in the first indent, that the basic amount of the fine is to be reduced where there are attenuating circumstances, such as an ‘exclusively passive or “follow-my-leader” role in the infringement’.

350 In setting the fines, the Commission distinguishes between the participants in a cartel according to whether they have been ringleaders, active participants or passive participants. The evidence, as analysed by the Commission itself, appears to indicate that Repsol and PROAS were the ringleaders, BP an active participant and the applicants (and perhaps Petrogal), contrary to the self-serving statements of the leniency applicants, passive participants. The amount of any fine imposed ought therefore to have been reduced in accordance with the Guidelines and the case-law, in order to reflect that passivity, yet the reduction of the amount of the fine granted by the Commission is disproportionately low in the light of the number of factors militating in favour of such a reduction.

351 In particular, the Commission did not find that the applicants were involved in the in-house market analysis, the in-house pre‑distribution of the market, the market distribution negotiations and the drafting of the annual market-sharing agreements, within the meaning of recital 130(a) to (e) to the contested decision, that is to say, in the key stages in ‘the creation of any anti-competitive agreements’. Indeed, the applicants did not attend any of the meetings which, within the meaning of recital 234 to Commission Decision 2002/271/EC of 18 July 2001 relating to a proceeding pursuant to Article 81 [EC] and Article 53 of the EEA Agreement (COMP/36.490 – Graphite Electrodes) (OJ 2002 L 100, p. 1), may be described as ‘Top Guy’ or ‘Working Level’ meetings, at which Repsol, PROAS and BP agreed on how the relevant market would be shared. The Commission considered that the applicants were involved only after the ‘Top Guys’ had adopted sensitive decisions on the sharing of the relevant market. However, the Commission granted a reduction of 40% of the amount of the fine in Decision 2002/271.

352 The evidence on which the Commission found that the applicants had been actively involved in the cartel meetings was supplied by PROAS and Repsol. BP, the main leniency applicant, acknowledged that it had never met the applicants. It is clear that both PROAS and Repsol have strong interests in playing down their roles as ring leaders and implicating the applicants as an active participant. The evidence which they supplied ought therefore to have been treated with the greatest prudence, since it is not reliable evidence. It follows that the analysis at recital 543 to the contested decision is flawed. Furthermore, the applicants’ sporadic attendance at meetings confirms that they had no involvement whatsoever in convening, organising, chairing or funding those meetings.

353 Furthermore, the Commission, in particular at recital 543 to the contested decision, incorrectly assimilated the applicants’ active involvement at certain times to ‘full involvement’. On the assumption that, contrary to the applicants’ contention, the Commission is correct to assert that they were, at certain times, actively involved in the unlawful arrangements, it is clear from the Commission’s own analysis that that did not amount to full involvement. Apart from the fact that they did not attend the meetings, play any role in the decision‑making process and were generally more passive, there are also particular aspects of the arrangements in which the Commission itself does not maintain that the applicants participated, for example in the compensation mechanism which is supposed to have existed between Repsol, PROAS and BP (recitals 348 and 543 to the contested decision).

354 Nor is there any other basis on which it might be concluded that the applicants were parties to any of the monitoring arrangements alleged to have been implemented by the three main members of the cartel (recital 348 to the contested decision). The applicants did not participate in the meetings which their competitors held twice every month in order to assess the application of the PTT, or provide fortnightly sales volumes, or prepare summary tables. Last, the contested decision does not maintain that the applicants ever received the PTT and the applicants categorically deny ever having received such a document.

Impact of the agreements – the applicants’ involvement

355 The applicants claim that, as demonstrated by the table showing the discrepancies between their actual sales and their alleged quota of 54 000 MT, they clearly disregarded the alleged volume allocation, with respect both to sales volumes and to customers.

356 The applicants were not parties to the monitoring arrangements and were free to sell whatever volumes they wished to customers of their choice, because Repsol and PROAS did not attempt to prevent them from doing so. The applicants’ approach cannot be regarded as mere cheating, since there was nothing ‘underhand’ about their behaviour. Nor did they disguise the volumes to be reported, as no such reporting took place. They merely endeavoured to maximise their sales spontaneously. Furthermore, Repsol and PROAS never took any steps to limit the applicants’ imports through Villagarcía de Arosa.

357 The opening by the applicants of a sales depot at Bayonne (France) in 2002, which enabled them to import additional volumes potentially amounting to 10 000 MT, apparently did not give rise to any comment on the part of the members of the cartel, although it would have constituted a disruption of an alleged quota-fixing arrangement. The internal e-mail attached as Annex A.28 to the application shows, moreover, that the members of the cartel, in this instance BP, knew about the depot and did nothing.

358 There is no evidence whatsoever that the applicants complied with the wishes of Repsol and PROAS concerning the customers which they could supply. The applicants regularly supplied customers of Repsol and PROAS and attracted new customers by seeking to be the first to develop a business relationship and to win their confidence. Last, there is no evidence of any monitoring or compensation, in spite of the fact that the applicants considerably exceeded their alleged quota. They were not constrained and did not agree to be constrained in their operations in Spain.

Competition between the different Spanish regions

359 The applicants maintain that throughout the duration of the alleged infringement they had only one supply point, an import terminal in Villagarcía de Arosa, where they achieved 76% of their sales in Spain in 2001. The great majority of their other sales were in the two provinces adjoining Galicia, namely León and Zamora, and less than 1% of their other sales were made further away, in Palancia.

360 The area supplied by the applicants is one of the most competitive in Spain. The impact of their involvement in any alleged arrangements was, therefore, minimal on customers and consumers. However, the Commission did not take that factor into account when setting the amount of fine to be imposed on them.

Prices in Spain and Portugal

361 The applicants observe that they sold bitumen in Portugal and Spain and that the evidence shows that they charged similar prices in both countries. However, the Commission has not claimed that there was any cartel in Portugal. In most years prices in Spain tended to be lower than in Portugal. The average prices were higher in Spain in only one year and the difference was less than 4%.

362 Thus, the available material would therefore appear to indicate that the alleged infringement had no effect whatsoever on the price at which the applicants sold bitumen. The Commission was therefore not entitled to assume or to estimate without any basis for doing so that the applicants’ alleged involvement in the infringement produced actual anti‑competitive effects. That circumstance is a relevant factor which the Commission ought to have taken into account when it set and reduced the fine imposed on the applicants.

363 The Commission therefore did not, or did not properly, take the elements of assessment set out above into account when it set the amount of the fine. The reduction of 10% granted at recital 567 to the contested decision is therefore manifestly inadequate and disproportionately low.

364 The Commission contends that the present plea should be rejected.

b) Findings of the Court

365 In the context of this plea, the applicants essentially put forward the following complaints:

– absence of evidence establishing their participation in the cartel for periods other than 2001 and 2002;

– modest position on the market;

– involvement in the cartel limited to their activity in Galicia;

– limited and passive or follow-my-leader role;

– non-implementation in practice of the arrangements;

– absence of any actual impact of the applicants’ participation in the cartel;

– absence of participation in the system for monitoring the market-sharing and customer-allocation arrangements.

366 It is necessary to examine each of those complaints in turn.

Absence of evidence establishing the applicants’ participation in the cartel for periods other than 2001 and 2002

367 It must be stated at the outset that the applicants’ arguments in this respect lack detail and are based more on a general assertion than a detailed examination of the contested decision, which, as regards the description of the events in issue, is based on over 200 recitals referring to numerous documents gathered by the Commission during the investigation.

368 Next, as is clear from the examination of the first plea for annulment, the finding that the applicants participated in the annual market-sharing and customer-allocation arrangements during the period after 1996 is based on sufficient evidence.

369 The same is true, as is apparent from the examination of the second plea for annulment, of the finding that they participated in the price‑coordination activities, not only during the period before 1996, but more broadly from 1991 to 2002 (see paragraphs 308 and 318 above).

370 Thus, the absence of evidence establishing that the applicants participated in the cartel for periods other than 2001 and 2002 has not been demonstrated and, consequently, it is not established that the Commission failed to correctly examine the period during which the applicants were involved in the unlawful arrangements.

371 With respect to the passive nature – alleged by the applicants – of their involvement in the cartel, this will be examined in paragraph 388 et seq. of this judgment.

Modest position on the market

372 The applicants’ argument that the Commission ought, when setting the amount of the fine, to have taken account of their modest position on the market has no factual basis. It follows on the contrary from recitals 510 to 515 to the contested decision that the Commission took that factor into consideration in respect of differential treatment of the undertakings concerned, the purpose of which is precisely to take account, in the light of their relative importance on the relevant market, of the extent of their respective economic capacity to cause significant damage to competition.

373 The Commission first of all determined a starting amount, which it set at EUR 40 000 000, in view of the fact that it classified the infringements found as ‘very serious’, of the geographic scope of the market concerned and of its value (recital 510 to the contested decision).

374 By way of differential treatment of those undertakings, the Commission then adjusted that general starting amount to take account of the economic importance of each member of the cartel by using, at recital 512 to the contested decision, its market share expressed by the value of its bitumen sales on the relevant market in 2001, the last full year of the infringement. By relying on the market share of 4.54 to 5.24% held by the applicants during that year, the Commission placed them in the third and last category of the undertakings by order of importance and the individual starting amount of their fine was set at EUR 5 500 000.

375 The amount of the fine ultimately imposed on the applicants therefore reflects well the relatively low volume of their sales on the Spanish market and their relative importance within the cartel.

376 Moreover, as can be seen from recitals 516 to 522 to the contested decision, the Commission also took account of the total turnover of the undertakings concerned, thus deciding, on the basis of that turnover, to apply a multiplier to BP and to Repsol, but not to the other undertakings, including Nynäs.

Involvement in the cartel limited to their activity in Galicia

377 The applicants have not established or even claimed that the Commission ought to have distinguished different infringements, which ought to have been the subject of separate decisions. In that regard, it will be recalled that the Commission identified, in the contested decision (recital 130 to that decision), a sharing of the market in which it took the view that all the undertakings involved in the cartel participated to varying degrees. Similarly, the Commission identified a complex of price‑coordination activities in which it took the view that all the undertakings involved in the cartel participated to varying degrees (recitals 305 and 359 to the contested decision). Lastly, the Commission concluded that there was a single and continuous infringement of Article 81 EC (recitals 364 to 370 to the contested decision).

378 It is not contested by the applicants, with respect to the geographic scope of that single and continuous infringement, that that infringement related to penetration bitumen sold in Spain (excluding the Canary Islands).

379 Accordingly, the Commission was entitled, when determining the starting amount of the fine, to take account of the fact that the infringement concerned the territory of Spain, excluding the Canary Islands (recital 510 to the contested decision).

380 Next, it is apparent from the examination of the two pleas for annulment that the applicants participated in the market-sharing and customer-allocation arrangements, as well as in the price coordination, and that they sometimes took an active part in the price coordination.

381 Thus, although they exercised their activity in a limited area of the territory of Spain and were allocated bitumen volumes and customers only for that area, the applicants contributed to the operation of a cartel which was liable to restrict competition on the entire Spanish market (excluding the Canary Islands).

382 The Commission was therefore entitled to take account of the whole of that territory in order to calculate the starting amount of the applicants’ fine.

383 Lastly, the applicants have failed to establish that the 1998 Guidelines required the Commission to reduce the amount of the fine in the light of the limited area of the Spanish territory in which they exercised their activity and in respect of which they were allocated bitumen volumes and customers.

384 Nor have the applicants demonstrated that the Commission infringed the principle of proportionality in that regard.

385 As stated above, the Commission adjusted the starting amount of the fine applied to the applicants in proportion to Nynäs’s small share in 2001 of the Spanish bitumen market.

386 The market share of each of the undertakings concerned in the market which formed the subject-matter of a restrictive practice constitutes an objective factor which gives a fair measure of the responsibility of each of them as regards the potential harmfulness of that practice for the normal operation of competition (see Case T‑127/04 KME Germany and Others v Commission [2009] ECR II‑1167, paragraphs 61 and 62 and the case-law cited).

387 Having thus already taken into account an indicator giving a fair measure of the applicants’ involvement, the Commission was not required, contrary to the applicants’ submission, to take account also, when determining a fine level corresponding to their involvement in the cartel, of the limited area of the Spanish territory in which they exercised their activity and in respect of which they were allocated bitumen volumes and customers.

Limited and passive or follow-my-leader role

388 It should be noted at the outset that, in the context of the present complaint, the applicants rely, at paragraph 8.14 of the application, on the provisions of the first indent of Section 3 of the 1998 Guidelines.

389 Those provisions provide for a reduction of the amount of the fine by way of attenuating circumstances in the event of ‘an exclusively passive or “follow-my-leader” role in the infringement’, which implies that the undertaking will adopt a ‘low profile’, that is to say not actively participate in the creation of any anti‑competitive agreements (Case T‑220/00 Cheil Jedang v Commission [2003] ECR II‑2473, paragraph 167).

390 In that regard, it is clear from the case-law that the factors which may indicate that an undertaking has played a passive role in a cartel include where its participation in cartel meetings is significantly more sporadic than that of the other members of the cartel, where it enters the market affected by the infringement at a late stage, regardless of the length of its involvement in the infringement, or where a representative of another undertaking which has participated in the infringement makes an express declaration to that effect (see Case T‑43/02 Jungbunzlauer v Commission [2006] ECR II‑3435, paragraph 252 and the case-law cited).

391 However, as the first indent of Section 3 of the Guidelines confers entitlement to a reduction of the amount of the fine by way of an attenuating circumstance only in the event of an ‘exclusively’ passive or ‘follow-my leader’ role by the undertaking in the infringement, it is not sufficient that that undertaking adopted a ‘low profile’ only during certain periods of the cartel or in relation to certain cartel agreements (Jungbunzlauer v Commission, paragraph 254).

392 On the contrary, in the light of the wording of the first indent of Section 3 of the Guidelines, recognition of a passive role implies that the undertaking in question has confined itself, throughout its participation in the infringement, to adopting a ‘low profile’.

393 Thus, the fact that other undertakings may have been more active does not automatically mean that an undertaking had an exclusively passive or follow-my-leader role, since only total passivity on its part could be taken into account as a factor (Joined Cases T‑109/02, T‑118/02, T‑122/02, T‑125/02, T‑126/02, T‑128/02, T‑129/02, T‑132/02 and T‑136/02 Bolloré and Others v Commission [2007] ECR II‑947, paragraph 611).

394 In the present case, first, it is apparent from the examination of the first plea for annulment that the applicants participated in the market-sharing and customer-allocation arrangements.

395 In that regard, it can be seen from the concurring responses of Repsol and PROAS of 9 May 2006 to a request for information that the applicants negotiated actively with them the volumes and customers corresponding to the applicants in their area of influence on the basis of a proposal previously agreed upon by Repsol and PROAS and subsequently adjusted to take account of the applicants’ requests (recitals 139 and 140 to the contested decision).

396 Second, it is apparent from the examination of the second plea for annulment that the applicants participated, sometimes actively, in the price coordination with other members of the cartel.

397 Third, as can be seen from pages 16 and 53 of their response to the statement of objections attached as Annex A. 5 to the application, the applicants admitted that one of their employees, Mr T.S.B., had attended a brief meeting at the Hotel Orense in April 2002 during which he had commented on the estimates prepared by Repsol and PROAS in order, inter alia, to protect the applicants’ position on the market.

398 The applicants also acknowledged at page 53 of their response to the statement of objections that Mr T.S.B. had confirmed that subsequently he did complain about PROAS supplying bitumen to Shell, which, in turn, supplied Bitumat, which subsequently supplied one of Mr T.S.B.’s customers. The applicants stated, in the same passage of their response to the statement of objections, that an internal PROAS e‑mail of 6 September 2002 suggests that Mr T.S.B. had tried to use PROAS’s access to the market to his advantage.

399 As the Commission states at recital 279 to the contested decision, without being challenged in that regard by the applicants, PROAS explained in its response of 18 November 2005 to a request for information that Mr T.S.B. thought that, in its capacity as supplier of Shell, PROAS could have an influence on Shell’s sales policy.

400 The conduct described in the foregoing paragraphs cannot be classified as ‘passive’, since one of the applicants’ employees attempted to use the relations that he had with one of the undertakings participating in the cartel in order that that undertaking impede the delivery, by a competitor of the applicants, of bitumen to one of the applicants’ customers.

401 It follows from the foregoing that the applicants are not justified in complaining that the Commission found that they had not played an exclusively passive role in the infringement or, consequently, that the Commission did not grant them any reduction of the fine under the first indent of Section 3 of the 1998 Guidelines.

402 The foregoing conclusion cannot be called into question by the additional arguments essentially put forward by the applicants.

403 In particular, the limitation of the applicants’ participation to the last of the phases relating to the creation of the annual market-sharing arrangements, which is referred to at recital 130(f) to the contested decision, is not such as to establish the existence of an ‘exclusively’ passive or follow‑my‑leader role in the infringement within the meaning of the first indent of Section 3 of the 1998 Guidelines.

404 Similarly, by merely observing that ‘there were particular aspects of the alleged arrangements to which even the Commission does not allege Nynäs was a party, for example, the compensation mechanism which is alleged to have existed between Repsol, PROAS and BP’, the applicants cannot establish that they had an ‘exclusively’ passive or follow‑my‑leader role in the infringement.

405 It should be noted, more generally, that, given that the applicants’ arguments concerning the compensation mechanism lack detail in the context of the present plea as a whole, the applicants cannot be regarded as having put forward a separate complaint in this respect.

Non-implementation in practice of the cartel

406 It should be borne in mind that the Commission is not required to recognise the existence of an attenuating circumstance consisting of non-implementation of a cartel unless the undertaking relying on that circumstance is able to show that it clearly and substantially opposed the implementation of the cartel, to the point of disrupting the very functioning of it, and that it did not give the appearance of adhering to the agreement and thereby encourage other undertakings to implement the cartel in question. It would be too easy for undertakings to reduce the risk of being required to pay a heavy fine if they were able to take advantage of an unlawful agreement and then benefit from a reduction in the fine on the ground that they had played only a limited role in implementing the infringement, when their attitude encouraged other undertakings to act in a way that was more harmful to competition (Case T‑44/00 Mannesmannröhren-Werke v Commission [2004] ECR II‑2223, paragraph 277, and Joined Cases T‑259/02 to T‑264/02 and T‑271/02 Raiffeisen Zentralbank Österreich and Others v Commission [2006] ECR II‑5169, paragraph 491).

407 In the present case, it follows from the examination of the first plea for annulment that the applicants participated between 1991 and 1996 in the annual market‑sharing and customer-allocation arrangements. With respect to the period after 1996, it is not established or even claimed by the applicants that they did not participate in those arrangements.

408 Against that background, the fact that the applicants exceeded, as they submit, the quota allocated to them in the context of the cartel, that they opened a sales depot in the south west of France in order to import additional volumes of bitumen into Spain and that they sold bitumen to customers of Repsol, PROAS or to new customers cannot be regarded as non‑implementation in practice of the cartel for the purposes of the case‑law cited in paragraph 406 above.

409 In particular, it is common ground that the quota allocated to the applicants accounted for a small percentage of the relevant market; it is apparent from the documents examined in the context of the first plea for annulment that BP evaluated that quota at 4% or less of that market. Accordingly, even on the assumption that the discrepancies relied on by the applicants between their actual sales and the quota are established, those discrepancies, notwithstanding the fact that they represented a significant proportion of the quota (in general, 30 to 45% thereof), are not capable of showing that the applicants clearly and substantially opposed the implementation of the cartel, to the point of disrupting the very functioning of it, and that they did not give the appearance of adhering to the arrangements and thereby encourage other undertakings to implement the cartel in question.

410 It is even less possible to regard the applicants’ conduct as opposition to the implementation of the cartel given that, as is apparent from the examination of the second plea for annulment, until 2002, the applicants moreover continued to participate, sometimes actively, in another aspect of the cartel, namely price coordination.

Absence of any actual impact of the applicants’ participation in the cartel

411 The applicants submit that their involvement in the cartel produced a limited impact on customers and consumers and that the infringement, in so far as it concerns the applicants, had no effect whatsoever on the price at which the applicants sold bitumen.

412 They submit that, in the light of those factors, the amount of the fine is disproportionate.

413 Such an argument must be rejected.

414 The infringement in question was classified as ‘very serious’ by the Commission and, as is apparent from the examination of the second plea for variation of the fine, the Commission was able to lawfully find that the infringement established constituted a restriction of competition which, on account of its nature, was very serious.

415 Next, it is not contested by the applicants, with respect to the geographic scope of the infringement, that the infringement relates to penetration bitumen sold in Spain (excluding the Canary Islands).

416 It is clear from case‑law that a geographical market of national dimension corresponds to a substantial part of the common market (Case 322/81 Nederlandsche Banden-Industrie-Michelin v Commission [1983] ECR 3461, paragraph 28; Joined Cases T‑49/02 to T‑51/02 Brasserie nationale and Others v Commission [2005] ECR II‑3033, paragraph 176; and judgment of 25 October 2005 in Groupe Danone v Commission, paragraph 150).

417 Lastly, as is apparent from recital 67 to the contested decision, it is common ground that the total value of the market for penetration bitumen in Spain amounted to EUR 286 400 000 in 2001, the last full year of the infringement. In addition, it also apparent from that recital that Nynäs’s sales of bitumen in Spain amounted to between EUR 14 000 000 and EUR 15 000 000 in 2001.

418 In the light of those amounts, the gravity of the infringement and the extent of the geographic market concerned, and in view also of the duration of the applicants’ participation in the infringement, namely more than 11 years, the amounts of the fines applied to the applicants, which were set at EUR 10 642 500 for Nynas Petróleo and at EUR 10 395 000 for Nynäs Petroleum, do not appear to be disproportionate even though the actual impact of the cartel on the relevant market was not taken into account by the Commission.

419 Accordingly, the Commission was entitled to adopt the fines mentioned in the previous paragraph without having to take account of the actual impact of the applicants’ participation in the cartel.

Absence of participation in the system for monitoring the market‑sharing and customer‑allocation arrangements

420 The applicants put forward, in paragraphs 8.23 to 8.25 of their application, an independent line of reasoning by which they seek a reduction of the amount of the fine on the ground that they did not participate in the system for monitoring the market‑sharing and customer‑allocation arrangements.

421 In that regard, it should be noted that the applicants stated in paragraph 8.23 of the application, without being challenged in that regard by the Commission, that there is no basis for concluding that Nynäs was party to any of the monitoring arrangements.

422 In addition, the contested decision does not reveal an express reference to the applicants in the various parts of the statements of the three leniency applicants analysed at recitals 174 to 189 to the contested decision, which deal with that system. In particular, none of the applicants’ employees appears among the participants in the monitoring meetings mentioned in table 3, which is reproduced at recital 180 to the contested decision.

423 Lastly, as regards the contemporaneous documents, at recital 189 to the contested decision, the Commission refers to some of the recitals to that decision containing evidence establishing, in its view, the existence of the monitoring system. Recital 189 refers, inter alia, to recital 213 to that decision, in which reference is made to the applicants. In addition, at recital 213 to that decision there is a footnote (181) in which reference is also made to the applicants.

424 It is necessary to examine the various items of evidence which, according to the Commission, supposedly demonstrate the applicants’ involvement in the monitoring system, in particular, the items set out in the recitals mentioned in the previous paragraph.

425 First, the Commission states, in the fourth paragraph of recital 213 to the contested decision, with respect to the sales chart relating to the first fortnight of December 1994, as follows:

‘Also worth noting is the handwritten scribble on the page which contains the sales chart for the first fortnight of December 1994:

“Petrogal 48

Nynäs 55/56” (…)

The Commission considers that the above-mentioned amounts refer to the theoretical volumes allocated by the cartel to Petrogal and Nynäs, as indicated by Repsol in its leniency application (see recital 154):

Petrogal: 48 000 tonnes;

Nynäs: 54 000 tonnes.

In response to a request for information, Repsol confirmed that the phrases “Petrogal 48” and “Nynäs 55/56” refer to the thousands of tonnes negotiated with Petrogal and Nynäs respectively as their participation in the Spanish market.’

426 As is clear from the extract cited in the previous paragraph, the handwritten scribble on that chart merely reproduces the theoretical volumes allocated to the applicants following the final stage of the annual market‑sharing agreements which is referred to at recital 130(f) to the contested decision. Thus, those annotations do not relate to the applicants’ actual sales. They therefore have no probative value with respect to the applicants’ participation in the monitoring system.

427 Second, the ‘monitoring chart for 1995’, mentioned in footnote 181 (in limine) to the contested decision, which is a summary of theoretical sales, actual sales and the difference between the two, mentions the numbers, 3+1, 2 and 4, which are associated with Repsol, PROAS and BP, respectively. That chart does not mention the applicants. Mention is also made later in that footnote of the ‘chart for 2001’ in which theoretical quotas and actual sales are compared. That chart relates only to ‘the three suppliers’, which necessarily means Repsol, PROAS and BP, to which reference was made just before in the footnote.

428 The applicants’ absence from those follow‑up charts does not make it possible to consider those documents to be evidence of the applicants’ participation in the monitoring system.

429 Although footnote 181 (in fine) states that ‘Repsol explains the meaning of the numbers used to designate each supplier in the “charts”: 1: [Repsol], 2: PROAS, 3: Petronor, 4: BP, 5: Nynäs, 7: Petrogal, 9: Imports’, that clarification, which merely confirms the link between those numbers and the participants in the infringement, does not permit the conclusion that the applicants were mentioned in the follow-up charts referred to earlier in the footnote.

430 On examination, it is therefore apparent that the contemporaneous documents by which the Commission, in the contested decision, links the applicants with the monitoring system do not constitute indicia of their participation in that system.

431 It follows from the foregoing that it has not been established to the requisite legal standard in the contested decision that the applicants participated in the system for monitoring the market‑sharing and customer‑allocation arrangements.

432 In addition, there is nothing else in the case‑file capable of establishing that the applicants participated in that aspect of the infringement.

433 It is in the light of the foregoing conclusions that it is necessary to determine whether and to what extent the amounts of the fines imposed on the applicants should be varied.

434 It must be borne in mind in this respect that the unlimited jurisdiction conferred on the Courts of the European Union by Article 31 of Regulation No 1/2003 empowers them, in addition to carrying out a mere review of the lawfulness of the penalty, to substitute their own appraisal for the Commission’s and, consequently, to cancel, reduce or increase the fine or penalty payment imposed where the question of their amount is before them (judgment of 8 February 2007 in Groupe Danone v Commission, paragraphs 61 and 62). In this context, it should be pointed out that the Courts of the European Union are free to determine the amount of the fine that they consider appropriate according to the method of their choice, provided that that does not result in discrimination between undertakings which have participated in an agreement or concerted practice contrary to Article 101(1) EC (see, to that effect, Case C‑534/07 P Prym and Prym Consumer v Commission [2009] ECR I‑7415, paragraph 112; Erste Group Bank and Others v Commission, paragraph 255; and judgment of 12 November 2009 in Case C‑554/08 P Carbone-Lorraine v Commission, not published in the ECR, paragraph 72).

435 In the present case, the Court considers that it is appropriate to apply the method followed by the Commission in the contested decision.

436 It should be noted at the outset that, by the present complaint, the applicants seek only to contest the fine and not to obtain the annulment of Article 1 of the contested decision, which holds them liable, inter alia, in respect of the monitoring system.

437 The finding that the applicants did not participate in the monitoring system is capable, on its own, of enabling the Court to reduce the fine, and the Court is not required to ensure, of its own motion, that the Commission acted unlawfully by holding the applicants liable, in Article 1 of the operative part of the contested decision, in respect of that aspect of the infringement.

438 Moreover, the exercise of unlimited jurisdiction does not amount to a review of the Court’s own motion, and proceedings before the Courts of the European Union are inter partes. With the exception of pleas involving matters of public policy which the Courts are required to raise of their own motion, such as the failure to state reasons for a contested decision, it is for the applicant to raise pleas in law against the contested decision and to adduce evidence in support of those pleas (KME and Others v Commission, paragraph 104, and Chalkor v Commission, paragraph 64).

439 In the context of a single and continuous infringement, the case‑law draws a distinction between the concepts of liability and participation in the infringement; an undertaking can be held liable for the infringement as a whole, including for unlawful conduct in which that undertaking did not participate (see Commission v Verhuizingen Coppens, paragraph 45).

440 It follows from the foregoing that it is not for the Court to adjudicate on the applicants’ liability in respect of the monitoring system.

441 Accordingly, there is no need to vary the starting amount of the fine, which takes account of all the forms of unlawful conduct found in the contested decision (see paragraph 72 above).

442 Next, in the absence of any errors relating to the assessment of the applicants’ effective economic capacity to cause significant damage to competition, and to the duration of their participation in the infringement, the amounts of the fines calculated, before taking into account the reduction in respect of attenuating circumstances, will remain unchanged, that is EUR 11 825 000 for Nynas Petróleo and EUR 11 550 000 for Nynäs Petroleum.

443 By contrast, it is necessary to the increase the reduction of the fine applied to the applicants by the Commission in respect of attenuating circumstances.

444 In that regard, it should be borne in mind that, at recital 567 to the contested decision, the Commission relied, in particular, on the applicants’ less regular or active participation in the monitoring system in order to grant them a 10% reduction of the fine.

445 Accordingly, the illegality which the Commission committed by finding that the applicants participated in the monitoring system results in a further 2% reduction of the amount of the fine, that reduction thus being added to the reduction of 10% already granted in the contested decision.

446 The Court therefore upholds the fifth plea in part and, in in the light of the foregoing considerations and all the circumstances of the case, decides that it is appropriate to grant a larger reduction of the fine, namely by a total of 12% of its amount.

447 Accordingly, the amount of the fine imposed on Nynas Petróleo is set at EUR 10 406 000 instead of EUR 10 642 500 and the amount of the fine imposed on Nynäs Petroleum is set at EUR 10 164 000 instead of EUR 10 395 000.

Costs

448 Under Article 87(3) of the Rules of Procedure of the General Court, where each party succeeds on some and fails on other heads, the Court may order that the costs be shared or that each party bear its own costs. In the circumstances of the present case it is appropriate to order the parties to bear their own costs.

On those grounds,

THE GENERAL COURT (Eighth Chamber)

hereby:

1. Sets the amount of the fine imposed on Nynas Petróleo, SA in Article 2 of Commission Decision C(2007) 4441 final of 3 October 2007 relating to a proceeding under Article 81 [EC] (Case COMP/38.710 – Bitumen (Spain)) at EUR 10 406 000, and the amount of the fine imposed on Nynäs Petroleum, AB in Article 2 of that decision at EUR 10 164 000;

2. Dismisses the remaining heads of claim in the application;

3. Orders each party to bear its own costs.

Truchot

Martins Ribeiro

Popescu

Delivered in open court in Luxembourg on 16 September 2013.

[Signatures]

Table of contents

Background to the dispute

A – Relevant market

B – Undertakings in question

1. Repsol Group

2. CEPSA-PROAS

3. BP

4. Nynäs Group

5. Petrogal Group

C – Administrative procedure

D – Contested Decision

1. Finding of the infringement

2. Calculation of the amount of the fines

a) Determination and adjustment of the ‘starting amount’ of the fine

b) Duration of the infringement

c) Attenuating circumstances

3. Operative part of the contested decision

Procedure and forms of order sought

Law

A – The request for measures of organisation of procedure

1. Arguments of the parties

2. Findings of the Court

B – The claims for annulment of Article 1 of the contested decision

1. First plea for annulment, alleging illegality affecting the finding that the applicants participated in the market-sharing arrangements between 1991 and 1996

a) Arguments of the parties

Evidence relating to 1991

– BP España internal note of 30 October 1991 (recitals 201 to 204 to the contested decision)

– Statements made in the context of the leniency applications

Evidence relating to 1992

Evidence relating to 1993

Evidence relating to 1994

Evidence relating to 1995

Evidence relating to 1996

b) Findings of the Court

2. Second plea for annulment, alleging illegality affecting the finding that the applicants were involved in price coordination

a) Arguments of the parties

Price increase on 3 April 2000

Price change on 20 October 2000

Price reduction in November 2001

Price change in April 2002

Price changes in May and June 2002

Price change in September 1999

b) Findings of the Court

C – The claims for annulment of Article 2 of the contested decision and, in the alternative, for variation

1. First plea for variation of the fine, based on the illegality affecting the finding that the applicants participated in the cartel

a) Arguments of the parties

b) Findings of the Court

2. Second plea for variation of the fine, alleging illegality affecting the characterisation of the market-sharing arrangements and the price‑coordination activities as a ‘very serious infringement’

a) Arguments of the parties

b) Findings of the Court

3. Third plea for variation of the fine, alleging errors of assessment relating to the degree of the applicants’ involvement in the infringement

a) Arguments of the parties

The applicants’ size on the Spanish bitumen market

The applicants’ presence was limited to a single province of Spain

The applicants’ limited and passive or follow-my-leader role in the infringement

Impact of the agreements – the applicants’ involvement

Competition between the different Spanish regions

Prices in Spain and Portugal

b) Findings of the Court

Absence of evidence establishing the applicants’ participation in the cartel for periods other than 2001 and 2002

Modest position on the market

Involvement in the cartel limited to their activity in Galicia

Limited and passive or follow-my-leader role

Non-implementation in practice of the cartel

Absence of any actual impact of the applicants’ participation in the cartel

Absence of participation in the system for monitoring the market‑sharing and customer‑allocation arrangements costs