EC, September 3, 2004, No 2006-485

THE COMMISSION OF THE EUROPEAN COMMUNITIES,

Having regard to the Treaty establishing the European Community, Having regard to the Agreement on the European Economic Area, Having regard to Council Regulation (EC) No 1-2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty (1) and, in particular, Articles 7(1) and 23(2) thereof, Having regard to the Commission Decision of 29 August 2003 to initiate proceedings in this case, Having given the undertakings concerned the opportunity to make known their views on the objections raised by the Commission pursuant to Article 19(1) of Council Regulation No 17 of 6 February 1962, First Regulation implementing Articles 85 and 86 of the Treaty (2), Article 27(1) of Regulation (EC) No 1-2003 and Commission Regulation (EC) No 2842-98 of 22 December 1998 on the hearing of parties in certain proceedings under Articles 85 and 86 of the EC Treaty (3), After consulting the Advisory Committee on Restrictive Practices and Dominant Positions, Having regard to the final report of the Hearing Officer in this case (4), Whereas:

A - INTRODUCTION

(1) This Decision is addressed to the following companies:

- Boliden AB, Boliden Fabrication AB and Boliden Cuivre & Zinc SA ("Boliden" or "BCZ"),

- Austria Buntmetall AG and Buntmetall Amstetten GesmbH,

- Halcor SA,

- HME Nederland BV,

- IMI plc, IMI Kynoch Ltd. and Yorkshire Copper Tube Ltd. (formerly: IMI Yorkshire Copper Tube Ltd.),

- KM Europa Metal AG, Tréfimétaux SA and Europa Metalli SpA ("KME-group"),

- Mueller Industries, Inc., WTC Holding Company, Inc., Mueller Europe Ltd., DENO Holding Company, Inc. and DENO Acquisition EURL,

- Outokumpu Oyj and Outokumpu Copper Products Oy,

- Wieland Werke AG ("Wieland")

(2) The addressees of this Decision participated in a single, continuous, complex and, as far as Boliden, the KME group and Wieland are concerned, multiform infringement of Article 81 of the Treaty and Article 53 of the EEA Agreement. The behaviour started in June 1988 and finished in March 2001. Different companies were involved during different time periods.

B - THE COPPER PLUMBING TUBES INDUSTRY

1. THE PRODUCT

1.1. (Plain and plastic-insulated) copper plumbing tubes

(3) Copper tubes are generally divided into two product groups: (i) industrial tubes which are segregated in sub-groups based on the end use (air-conditioning and refrigeration, fittings, gas heater, filter dryer and telecommunications), and (ii) plumbing tubes (also called sanitary tubes, water tubes or installation tubes), which are used for water, oil, gas and heating installations in the construction industry (5).

(4) For the purpose of this cartel investigation, copper plumbing tubes are to be assessed as a product distinct from industrial tubes, which are not the subject of this Decision. Such approach derives from the following considerations.

(5) First, the main customers for plumbing tubes are distributors, wholesalers and retailers that sell the plumbing tubes to installers and other end consumers, whereas industrial tubes are usually used by and directly sold to industrial customers, original equipment manufacturers or part manufacturers (6). Second, end use and technical specifications for plumbing tubes are usually different from those for industrial tubes. The assessment of plumbing and industrial tubes as different products is supported by the Commission's Decision in Case No COMP/M.3284 - Boliden/Outokumpu, in which the Commission considered copper plumbing tubes and industrial copper tubes to constitute different product markets (7). Third, and most importantly in the context of this Decision, the arrangements pertaining to plumbing tubes on the one hand and those relating to industrial tubes on the other hand involved different companies (and employees), and were organised in a different way.

(6) Turning to plumbing tubes, traditionally, they were mainly made of copper, that is to say, recycled copper, newly refined copper (cathode copper) or copper ingots (8), and, to some extent, of steel (9). Since the early 1990's, plumbing tubes have increasingly been produced from plastic or compounds (plastic with layers of aluminium) (10).

(7) The substitution process was enhanced by public discussion on quality standards for drinking water and the subsequent adoption of Council Directive 98-83-EC of 3 November 1998 on the quality of water intended for human consumption ("the European Drinking Water Directive") (11). Occasionally, after the adoption of that Directive, public subsidies were granted in case of installation of plastic or compound tubes.

(8) In its recent Decision in Case No COMP/M.3284 - Boliden/Outokumpu, the Commission recognised a certain amount of competitive pressure arising from sanitary tubes made especially of plastic or multi-layer tubes over plumbing tubes made of copper. Furthermore, the Commission indicated that plastic and multi-layer sanitary tubes appear to be gaining some market shares on the sanitary tubes market, although this trend seems to be slow (12). Certain evidence submitted by the parties, although not unambiguous, appears to confirm this slow trend (13).

(9) Substitution of copper plumbing tubes in certain European countries by plastic and compound tubes was at least partially compensated by growth of demand for copper plumbing tubes in Eastern Europe and, during the 1990s, general substitution of steel tubes by plastic, compound and copper tubes (14). According to the "European Copper Plumbing Promotion Campaign" ("ECPPC") (15) in the period from 1989 to 1999, European demand for copper tubes for plumbing and heating applications has continuously increased from ~ 875 million metres in 1989 to ~ 1.175 million metres in 1999 (increase of ~ 3,4%/year) (16), whereas from 1992 until 2001, Western European demand seems to have decreased from 746,9 to 721,2 million metres with a peak of 755,4 million metres in 1998 (17). These figures confirm explanations by Outokumpu that there was one big demand boom in the segment of copper plumbing tubes in Western Europe due to the German reunification between 1990 and 1997 (or probably more correctly 1999 (18)) (19).

(10) Turning to the copper plumbing tubes, which are the subject of this Decision, they are manufactured and sold as "hard", as "medium" or "half-hard" and as "soft" tubes, in form of straight tubes (5 m) and coils (25 or 50 m) with diameters from 6 mm to 267 mm and a thickness between 0,7 mm and 3 mm (20). Compared to hard tubes, "half-hard" tubes have the advantage of being bendable and of a higher corrosion resistance.

(11) Copper plumbing tubes are to be looked at as a product group comprising two sub-families of products: plain copper plumbing tubes on the one hand, and insulated plumbing tubes on the other hand. Indeed, certain applications (for example transport of warm water) require insulation of the copper plumbing tube and it appears that around 20 % of the copper plumbing tubes are coated with PVC (21). Copper plumbing tubes are either insulated by the copper plumbing tube producer during the production process, in the course of their installation (for example, by wrapping them in insulation material) or by specialised companies (so-called insulators) (22). Depending on the application and on whether copper plumbing tubes are plastic-coated (23), producers market their copper plumbing tubes under different brands.

(12) Three brands are of particular importance in these proceedings. "SANCO" (see recital (116)) is the brandname for a plain copper plumbing tube (produced by the KME Group, Wieland and BCZ). "WICU" (produced by Wieland, KME and until 1998 by BCZ) and "Cuprotherm" (produced by Wieland and KME) are brand names for plastic-insulated copper plumbing tubes (24). As will be shown in this Decision, these brands were critical in the organisation and implementation of the cartel as they provided a forum for the concerned undertakings to meet and a framework for them to organise and implement anti-competitive arrangements including the exchange of information.

(13) Plain and plastic-coated copper plumbing tubes are to be looked at as one product group for the purposes of this Decision because the arrangements pertaining to both sub-families of products involved essentially the same companies (and employees) and were organised in a similar way (in particular through the framework provided by the aforementioned brands).

(14) Some companies submitted that customers have considerable buyer power. Often, smaller customers form purchasing associations (25). Multi-sourcing of large customers is common.

(15) Outokumpu indicated that the copper tube industry is a capital intensive industry with high fixed cost (26). At least some of the companies active in the copper plumbing tube industry have suffered from low profitability.

(16) This Decision deals with arrangements concerning plain copper plumbing tubes and, as far as two producers are concerned, with arrangements concerning "WICU"- and "Cuprotherm" plastic-insulated copper plumbing tubes.

1.2. Geographic scope of the business

(17) The geographic scope of business of the copper plumbing tube suppliers is essentially Europe including the Community/EEA (27). Export volumes of copper plumbing tubes and the ventilation of turnover of companies concerned with copper plumbing tubes (based on the turnover figures provided by the parties) show that there are limited exports outside Europe (28). Based on the copper plumbing tube price, transportation costs are estimated to be mainly below 5% within Europe, and below 8% in the rest of the world (29). Most European suppliers are in a position to supply the entire European market regardless of factory location. There are no significant regulatory barriers to entry within Europe. Although a European copper tube standard "EN 1057" has been adopted, the sale of copper plumbing tubes in national markets regularly requires certification for some applications, which leads to a certain administrative burden (30).

1.3. Pricing in the copper plumbing tube industry

(18) Prices for copper plumbing tubes are mainly established on the basis of two factors: the copper price (variable) and the conversion margin. The conversion margin is the added value of the copper plumbing tube manufacturer of transforming copper into a plumbing tube.

(19) Prices are negotiated on the basis of price lists (mostly containing price lines ("indexed price lists") (31)) and rebates (32). Parties indicated that plumbing tubes are generally sold on an ad hoc basis and each contract is negotiated individually with the customer (33).

1.3.1. Indexed price lists

(20) Starting point for the negotiation of copper plumbing tube prices are price lists (that is to say the end price of the tube comprising the copper price and the conversion margin). In the European copper plumbing tube industry, price lists are usually issued in the form of indexed price lists containing price lines. Each indexed price line corresponds to a pre-defined price-range of copper, generally based on the value of copper pursuant to indices such as the London Metal Exchange ("LME"). When the copper price crosses a predetermined threshold, a different price line is applied (34). The system of indexed price lists takes account of the fact that the copper price is volatile and changes daily (35). Copper accounts for between approx. 50-65% of the sales value of copper plumbing tubes. It appears that price lists are widely dispersed and readily available in the market. The applicable price line of the indexed price list only serves as a basis for negotiating the final copper plumbing tube price.

1.3.2. Rebates

(21) The second major factor for determining the copper plumbing tube price is rebates. A rebate is always negotiated from the list price. Depending on the country, customer group and producer, rebates may vary between 30 % to more than 60 % of the total list price. The "total rebate" may be composed of a number of rebates granted on different terms (36).

1.4. Size, value and market shares - plain and plastic-coated copper plumbing tubes

(22) The parties estimated the market size of plain copper plumbing tubes to be between ~ 252 000 and ~ 263 000 tonnes in 1989 and between ~ 337 000 and ~ 365 000 tonnes in 2000. The Commission's market enquiry showed a market size in 2000 of 350 267 tonnes. The market size of plastic coated copper plumbing tubes is estimated at ~ 42 184 tonnes in 2000. In its Decision in Case No COMP/M.3284 . Boliden/Outokumpu, the Commission stated that in 2000, the EEA market of copper plumbing tubes (including plastic-coated copper plumbing tubes) amounted to ~ 370 000 tonnes.

(23) On the basis of the turnover information provided by the undertakings active in the copper plumbing tube sector, the EEA market value of plain copper plumbing tubes is estimated to be approximately EUR 970,1 million in 2000 (37), the EEA market value of plastic-coated copper plumbing tubes at approximately EUR 180,9 million in 2000 (38). The aggregated market value of plain and plastic-coated copper plumbing tubes amounts to EUR 1.151 million (EEA in 2000).

(24) Between 1989 and 2001, market shares amongst the parties to these proceedings remained relatively stable in the EEA with respect to plain copper plumbing tubes. The market shares in volume are indicated in the Annex.

(25) The combined share of IMI (only plain tubes), KME, Mueller (only plain tubes), Outokumpu (only plain tubes) and the Wieland-group ("group of the five", see recital (216)) of the total plain and plastic-coated copper plumbing tubes market in the EEA in 2000 was approximately 64,6% (value and volume). The combined share together with Halcor, HME and Boliden (and Buntmetall (39)) ("group of the nine", see recital (216); with the exception of KME and Wieland only plain tubes) in the total EEA market in 2000 was approximately 78,7 % (value) and 79,2 % (volume). In 1998 and 1999 it was around 2% higher.

1.5. Trade between the Member States

(26) The copper plumbing tube market has been characterised by important trade flows between Member States and Contracting Parties to the EEA. The majority of the copper plumbing tube manufacturers produce in one, two or three European home markets, and sell tubes throughout Europe. Production sites of different producers are spread across Europe. During the period, which is the subject of this investigation, Outokumpu, for example, had relevant production facilities in Finland, Spain and Sweden, Wieland in Germany and Austria, KME group in Germany, France and Italy, Mueller in the United Kingdom and France, IMI in the United Kingdom, Boliden in Belgium and Sweden. From these units, copper plumbing tube manufacturers supply the entire EEA and the rest of Europe. Large parts of the total copper tube consumption in different Member States stems from imports from other Member States (40).

2. THE MARKET PLAYERS

2.1. Producers subject to these proceedings

2.1.1. Boliden AB and Boliden Cuivre & Zinc SA

(27) Boliden Cuivre & Zinc SA is a wholly owned subsidiary of Boliden Fabrication AB (Sweden), which is wholly owned by Boliden AB. Boliden AB is a corporation incorporated under the laws of Sweden with operations in Europe and Canada. Boliden AB is engaged in the mining, processing and selling of metals and mineral products, principally copper and zinc (41).

(28) Boliden produced to a minor extent copper plumbing tubes in its subsidiary Boliden Gusum AB before its acquisition of BCZ in 1988 (42).

(29) BCZ sells its plumbing tubes in the European copper plumbing tube market through a number of European subsidiaries: Boliden Cuivre & Zinc (Deutschland) GmbH, Boliden Cuivre & Zinc (France) / CDC France EURL, Boliden Cuivre & Zinc (España) SA, Boliden Cuivre & Zinc (Polska) Sp. Z o.o., Copper Distribution Center ("CDC"), Boliden Metal Supplies Ltd. ("BMSL"), United Kingdom (43), and HME France SA (44).

2.1.2. Buntmetall Amstetten GesmbH and Austria Buntmetall AG

(30) Buntmetall Amstetten GesmbH ("Buntmetall" or "BMA") of Austria has been wholly owned by a holding company, Austria Buntmetall AG, since December 1989. Wieland Werke AG acquired 75,1% of the latter on 9 July 1999, and the ownership was brought to 82,8% on 1 October 1999 and to 83,3 % on 30 November 2000. Until the acquisition, Buntmetall was not affiliated to Wieland (45).

(31) Buntmetall is a manufacturing company of semi-finished and special products in copper and copper alloys. Its main business is the manufacture of copper and copper alloys tubes for industrial and sanitary application. Its manufacturing plant is in Amtstetten, Austria.

(32) Buntmetall holds a number of service subsidiaries: Metallwerk Möllersdorf Handelsges.m.b.H., Austria; Buntmetall France SAR.L., France (liquidated in October 2001); Caro-Supersan Installationstechnik GmbH (formerly Carobronze Eisleben GmbH), Germany.

(33) With the transfer of the copper plumbing tube business to Wieland, all relevant employees of Buntmetall participating in the contested behaviour either stayed in their functions in Buntmetall or became employees of Wieland.

2.1.3. Halcor SA

(34) Halcor SA ("Halcor") was established in Greece in 1977. It manufactures and trades rolled and extruded copper and copper alloy (brass) products. The main extruded products are copper tubes and brass rods.

(35) Through different holding companies, Viohalco SA holds 65,55% of Halcor SA's shares.

(36) Halcor consolidates three small wholesalers or distributors: Metal Agencies Ltd., Surrey, United Kingdom (since 1994 a 67% shareholding); TePro Metall AG, Düsseldorf, Germany (with a 1 % interest until September 2002, and since then 37,5%); MKC GmbH, Huerth, Germany.

2.1.4. HME Nederland BV

(37) HME Nederland BV ("HME") was founded in 1988 as a result of a management buyout. HME went into bankruptcy in November 2001. Its activity was the production and sale of copper plumbing tubes. HME had a sales office in France (HME France SA).

(38) In January 2002, BCZ acquired certain assets of HME. The assets were brought into Boliden HME, a newly created legal entity and subsidiary of BCZ. Boliden HME continued HME's activities as a producer of copper plumbing tubes (46).

2.1.5. IMI plc.

(39) IMI plc. ("IMI group") is an international engineering business with activities in two main business areas: Fluid Controls (pneumatics, severe service valves and indoor climate) and Retail Dispense (beverage dispense and merchandising systems) (47). IMI plc is a corporation incorporated under English law. IMI as undertaking (including its subsidiaries) will be referred to as "IMI".

(40) The following legal entities of the IMI group were involved in the production and sale of copper plumbing tubes, and were part of IMI's Building Products/Hydronic Controls division, which was dissolved after a restructuring of IMI in 2001 (48): IMI Yorkshire Copper Tube Ltd. ("YCT"), Liverpool, manufacturing company (sold to KME); IMI Yorkshire Copper Tube (Exports) Ltd., Liverpool, export sales company (sold to KME); and Irish Metal Industries Ltd, Ireland (distribution, sold to KME); Raccord Orléanais SA, France (distribution, sold to Aalberts Industries NV on 9 September 2002); R Woeste & Co .Yorkshire. GmbH, Germany (distribution, sold to Aalberts Industries NV on 9 September 2002); YIM Scandinavia AB, Sweden (commercial agent, sold to KME); IMI Refiners Ltd, United Kingdom, supplier of billet to YCT, still belonging to IMI. Outside the EEA, IMI has fourteen subsidiaries that were involved in the distribution of copper plumbing tubes. IMI sold copper plumbing tubes in the EEA market through these distribution subsidiaries and independent distributors.

(41) YCT, the main subsidiary active in copper plumbing tubes, was a fully owned subsidiary of IMI Kynoch Ltd., which was a 100 % subsidiary of IMI plc.

(42) On 2 October 2002, IMI sold its Copper Tube business to KM Europa Metal AG (KME). The transaction was concluded on 29 November 2002 (49).

2.1.6. KME group

2.1.6.1. The relevant entities

Societa Metallurgica Italiana SpA

(43) Societa Metallurgica Italiana SpA("SMI"), whose largest shareholder is the Italian Orlando family, is the Italian holding company of the KME-group, to which Europa Metalli SpA ("EM" or .EM/LMI. or "Europa Metalli") and Tréfimétaux SA ("Tréfimétaux" or .TMX") belong. As a holding company, its purpose is limited to the acquisition of shareholdings and financial activities. SMI itself never took part in any of the contacts or meetings described below.

(44) To the extent relevant for copper plumbing tubes, the SMI-group was formed in the following sequence: In 1976, SMI created EM which was held 84% by SMI and 16 % by Pechiney. In 1986, EM acquired - through SMI - 100 % control of TMX. In 1990, SMI acquired 76,9% of Kabelmetal AG ("KM") from M.A.N. In 1995, SMI group was restructured and its shareholdings in TMX and EM were transferred to KM, whereby TMX and EM became wholly owned subsidiaries of KM. The name of the latter was changed to KM Europa Metal AG (KME). In 1999, SMI increased its shareholding in KME to 98,6%. The management of KME, TMX and EM was centralised.

KM Europa Metal AG

(45) KM Europa Metal AG ("KME"), formerly Kabelmetal AG ("KM"), currently has four main operating companies, Europa Metalli SpA in Italy, Tréfimétaux SA in France, Sociedad Industrial Asturiana SA (SIA) in Spain and KME Metal GmbH in Germany.

(46) KME group is the largest processor of copper and copper alloys in the world with 18 production locations in Europe and Asia. It has four product divisions, Brass Rods, Tube Systems, Special Products and Rolled Products. The KME group has a worldwide full service sales network.

Tréfimétaux SA

(47) Tréfimétaux SA ("Tréfimétaux" or "TMX") of France was integrated into the SMI group in 1986 through Europa Metalli, of which it was a wholly owned subsidiary until 1995. Upon restructuring of the SMI-group in 1995 it became a wholly owned subsidiary of KME (50). It has four industrial production sites in France.

Europa Metalli SpA. (51)

(48) Europa Metalli SpA(EM) is the Italian production company of KME. With its 2.000 employees and annual production of over 270.000 tonnes, Europa Metalli is the largest Italian producer of copper and copper alloy semi-finished products.

(49) EM has its headquarters in Florence. It runs three production plants in central and northern Italy. The Head Sales Office is located in Milan, and its commercial network of branch offices and warehouses covers the entire country.

2.1.6.2. Legal and economic links within the SMI group

(50) Both EM and TMX have belonged to the SMI group since 1976 and 1986, respectively. Within the SMI holding structure, TMX was a wholly owned subsidiary of EM during the period 1986-1995 (52). In 1987, TMX's business plan and commercial strategies were aligned with those of EM, and Italian managers were introduced to TMX's organisation at board level (53). A common sales organisation, EMT, was established for TMX and EM on 1 January 1993, and Mr. [...] (TMX) was appointed EM's Commercial Director for Industrial Tubes (54), while in 1993, Mr. [...] was appointed Commercial Director of the EM Plumbing Tube Business (55). From 1993 to 1998, Mr. [...] reported to Mr. [...], who was TMX's Managing Director and Head of EM's Tube Business (56). From 1990 to 1995, the Vice President of EM was also the Managing Director of the holding company SMI (57).

(51) SMI acquired 76,9 % control of KM in 1990. Since then, KM, EM and TMX have thus all belonged to the same holding. As a result of the restructuring of the SMI group in 1995, EM and TMX became KM's (the name of which was changed to KME) wholly owned subsidiaries. SMI's shareholding in KME was brought to 98,6 % in 1999. In 1999, the management of KME, TMX and EM was also centralised, and Mr. [...] (KME) became responsible for the industrial tubes business unit, while from 1999 to 2001 Mr. [...] became responsible for the plumbing tubes business unit of the KME Group, together with Messrs. [...] and [...] (58).

(52) SMI's, EM's and TMX's board members were appointed by their shareholders in the general shareholders' meeting, as required by Italian and French law, respectively. KME's board members were appointed by its supervisory board (59).

(53) During the period 1986-1995, while KM's management board was different from that of SMI, EM and TMX, there were partial overlaps and interlocking directorships between SMI, EM and TMX management boards as follows (60):

- Mr. [...] was simultaneously SMI's President (1986-2001) and EM's President (1986-1995),

- Mr. [...] was simultaneously SMI's Director General (1986-1996) and EM's Board Member (1986-1995),

- Mr. [...] was simultaneously SMI's Board Member (1986-1990) and Vice- President (1991-1995), EM's Vice-President (1986-1990) and Board Member (1991- 1995), and TMX's Board Member (1987-1995),

- Mr. [...] was simultaneously EM's Board Member (1986-1995) and TMX's Board Member (1986-1992),

- Mr. [...] was simultaneously EM's Director-General (1986-1995) and TMX's Board Member, Vice-President and Director-General (1987-1991),

- Mr. [...] was simultaneously EM's Board Member (1986-1989) and TMX's President (1988-September 1990),

- Mr. [...] was simultaneously EM's Board Member (1988-1995) and TMX's Board Member (1988-2000).

(54) During the period 1995-2001 following the restructuring of the group, after which KME controlled 100 % of the capital of both EM and TMX, KME's management board was also interlocked to that of SMI, EM and TMX through the following links (61):

- Mr. [...] was simultaneously SMI's Director-General (until May 1996), KME's Board Member (1995-2001) and EM's Vice-President (1995) and President (1996-2001),

- Mr. [...] was simultaneously KME's Chairman (1995-2001) and EM's Board Member (1996-2001),

- Mr. [...] was simultaneously KME's Board Member (June 1995 -December 1999) and TMX's Director-General (1995- March 2001).

(55) During the period 1995-2001 there were further links between SMI's, TMX's and EM's management boards as follows (62):

- Mr. [...] Jr. was simultaneously SMI's Board Member (1996-2000), EM's Board Member (1996-2001) and TMX's President (1995- March 2000),

- Mr. [...] was simultaneously SMI's Board Member (1995-2001) and EM's Board Member (1996-2001).

(56) With regard to the operational management, KME's "Business distribution plan" dated 25 June 1995, states that "Mr. [...] [Chairman of the KME board] has the responsibility of the global business; Mr. [...] [commercial manager of TMX] will closely co-operate with Mr. [...] and will have the responsibility of managing the EMT division." (63) Similarly, the "Business distribution plan" dated 19 March 1997 confirms that "Mr [...] has the responsibility of the tubes division" (64).

2.1.7. Mueller Industries Inc.

(57) Mueller Industries, Inc., ("Mueller") is one of the largest producers of copper tubes and other copper products in the world seated in Memphis (United States). It has operations in the United States, Canada, Mexico, the United Kingdom, and France.

(58) In 1997, Mueller acquired Wednesbury Tube & Fittings Company Ltd. and Desnoyers SA, thereby entering the European copper tube market.

Wednesbury Tube & Fittings Company Ltd. ("Wednesbury") / Mueller Europe Ltd. ("Mueller Ltd.")

(59) [Description of restructurings within the Mueller group] (65)[...] (66).[...] (67)[...].(68)

Desnoyers SA ("Desnoyers") / Mueller Europe SA ("Mueller SA")

(60) [Description of restructurings within the Mueller group] (69).[...] (70)[...].(71)[...] (72)[...] (73).

(61) During the period from May 1995 through May 1997 - prior to its acquisition by Mueller - Desnoyers' head office was located at one of its production facilities in Laigneville, France. Desnoyers had another production facility in Longueville, France. While both production facilities were involved in the production and sale of copper plumbing tubes, only the production facility of Longueville was maintained. The operations at Laigneville were discontinued in December 1998 (74).

2.1.8. Outokumpu Copper Products Oy ("Outokumpu")

(62) Outokumpu Oyj is a Finnish group operating world-wide (Outokumpu group). Its headquarters are located in Espoo, Finland. Outokumpu Oyj focuses on base metal production, stainless steel, copper products and technology. The copper tube business was originally carried out by Pori Tube Mill under Outokumpu Copper Products- Division. Neither Pori Tube Mill nor Outokumpu Copper was a separate legal entity but both were part of Outokumpu Oyj. Outokumpu Copper Products Oy was registered as a separate corporate entity, although under a different corporate name, in May 1988 and took over this activity fully by December 1988. Its name was changed to Outokumpu Copper Products Oy ("OCP") in 1996. Outokumpu as undertaking (including its subsidiaries) will be referred to as "OTK" or "Outokumpu" (75).

(63) OCP has been wholly owned by Outokumpu Oyj for the whole period of its existence. OCP itself is divided into divisions, each with their own business lines and manufacturing units. The European copper plumbing tube production is now concentrated in Pori in Finland (Outokumpu Poricopper OY), Zaratamo in Spain (Outokumpu Copper Tubes SA) and a small facility in Västerås in Sweden (Outokumpu Copper Products AB) (76). OCP sells copper tubes in the EEA market through sales personnel and sales offices belonging to Outokumpu.

2.1'9. Wieland Werke AG

(64) Wieland Werke AG (hereinafter .WW., "Wieland" or "Wieland Werke") is a German company the main activity of which is in the production, sale and distribution of semi-finished and special products in copper and copper alloys. Apart from its own manufacturing activities, Wieland is the holding company of 45 other corporate entities. Wieland owns at least 50 % of the share capital in 39 of the latter.

(65) Wieland has several plants in Germany, in Ulm, Verbert-Langenberg, Villingen-Schwenningen and Vöhringen. The company's headquarters are in Ulm. Other companies of the group are seated in Austria, United Kingdom, Spain, and Belgium.

(66) On 9 July 1999, Wieland Werke AG acquired 75,1% of Austria Buntmetall AG which is the holding company of Buntmetall Amstetten Ges.m.b.H ("Buntmetall" or "BMA"). The ownership was brought to 82,8% in October 1999 and to 83,3% in November 2000.

(67) Two legal entities within the Wieland group currently sell copper plumbing tubes, the parent company itself, Wieland Werke AG, and Buntmetall. Wieland owns five service companies: Wieland SAS (France), Semimetais Lda. (Portugal), Pando Metales SA (Spain), Caro-Supersan Installationstechnik GmbH (Germany) and Metallwerk Möllersdorf (Austria). Furthermore, Wieland holds seven distribution and trading companies, which are active in Belgium/the Netherlands, Denmark/Finland/Norway/Sweden, United Kingdom/Ireland, Italy, Austria, and Portugal.

2.2. Other producers

(68) Besides the companies described above, there exist a number of smaller producers in Europe such as Silmet (Torbole, Italy), Feinrohren (Lumezzane, Italy), La Farga Lacambra (Spain), Hutmen (Wroclaw, Poland), Czepel (Budapest, Hungary), and Slobodan and Majdanpek (former Yugoslavia). MKM Mansfelder Kupfer und Messing GmbH (Hettstedt, Germany) ("MKM"), a producer of former Eastern Germany, (re-) entered the market beginning of the 90's (77). Furthermore there exist other large producers world-wide that are not, or are only to a minor extent, active in Europe.

2.3. Trade associations and certification organisations

(69) There exist a number of trade associations in the copper plumbing tube industry on a national and international level, some of which shall be explained below.

(70) The International Wrought Copper Council ("IWCC") (78) is a trade association founded in 1953 and based in London representing the interests of the copper industry worldwide. It has several subcommittees and groups that collect certain information. (79)

(71) The European Copper Institute ("ECI") (80) is based in Brussels (formerly in London) and was created as a joint venture between the International Copper Association ("ICA") and the IWCC. The ECI represents the interests of the copper industry in Europe, the Middle East, and Africa (81). The ICA represents copper producers and is seated in New York. (82)

(72) The ECPPC (83) is administrated by the Secretariat of the IWCC and is supported by the ICA and the European Fittings Manufacturers Association ("EFMA").

(73) Amongst the German associations are Deutsches Kupfer Institut e.V. (DKI), Düsseldorf (84), Wirtschaftsvereinigung Metalle e.V. (WV Metalle), Düsseldorf (85), and Gesamtverband der Buntmetallindustrie (GV BMI), Düsseldorf.

(74) IMI explained that the .EMCI Steering Committee Meetings. were meetings of representatives of Eurométaux Copper Industry, Brussels, and the ECI.

(75) Amongst the certification organisations in the Community are the Deutsche Vereinigung des Gas- und Wasserfaches e.V. (DVGW) (86) (Germany), Kitemark (United Kingdom), Afnor (France) (87), UNI (Italy) (88) and AENOR (Spain).

C - PROCEDURE

3. INVESTIGATION AND APPLICATIONS FOR LENIENCY

(76) On 9 January 2001 Mueller Industries Inc. ("Mueller") informed the Commission about the existence of a cartel in the copper tube market and expressed its willingness to co-operate with the Commission pursuant to the Commission Notice on the non-imposition or reduction of fines in cartel cases ("the 1996 Leniency Notice") (89). [...] (90).

(77) On 22 and 23 March 2001 the Commission carried out unannounced inspections at the premises of OTK, Wieland Werke, KM Europa Metal, TMX, Europa Metalli LMI, Lawton and IMI pursuant to Article 14 of Regulation 17 (91).

(78) On 7 April 2001, the Commission conducted an interview with a person formerly active in the copper plumbing tube sector (92).

(79) On 9 April 2001, a further inspection was carried out at the premises of Outokumpu. On that date, Outokumpu Oyj informed the Commission about its willingness to cooperate within the framework of the 1996 Leniency Notice both concerning industrial tubes and copper plumbing tubes (93).

(80) On 10 April 2001, a further inspection was carried out at the premises of KM Europa Metal AG (KME) pursuant to Article 14 of Regulation 17. After the inspections, the investigation in the copper tube market was separated into three different proceedings, Case COMP/E-1-38.069 (Copper Plumbing Tubes), [pending case] and Case COMP/E-1-38.240 (Industrial Tubes) as there were three different cartels involving different organisational framework, partially different participants and different, albeit neighbouring, products.

(81) By letter dated 30 May 2001, OTK submitted a Memorandum with a number of Annexes, describing the arrangements (94).

(82) [...]

(83) On 5 June 2002, the Commission interviewed representatives of OTK for the purposes of Case COMP/E-1-38.240 (Industrial Tubes). The interviews were conducted at the Commission's initiative within the framework of OTK's offer to co-operate with the Commission. Outokumpu indicated its willingness for the Commission also to conduct interviews with its employees that were involved in the arrangements in the copper plumbing tubes, which are described below.

(84) In July 2002, concerning Case COMP/E-1-38.240 (Industrial Tubes), the Commission sent requests for information under Article 11 of Regulation 17 to Wieland Werke and KME, including their subsidiaries, and asked OTK to provide further information.

(85) On 15 October 2002, KME replied to the request for information of July 2002. A company statement concerning copper plumbing tubes was included in that submission on the basis of voluntary cooperation. KME requested application of the 1996 Leniency Notice in the Copper Plumbing Tubes Case (95).

(86) On 5 November 2002, KME's representatives met with the Commission services to discuss their submission and the procedural steps. The meeting took place at KME's initiative.

(87) On 19 December 2002, Wieland Werke's representatives met with the Commission services to discuss their submission in reply to the information request of July 2002 concerning Case COMP/E-1-38.240 (Industrial Tubes) and the procedural steps. The meeting took place at Wieland Werke's initiative. Wieland announced its willingness to co-operate under the 1996 Leniency Notice in the Copper Plumbing Tubes Case.

(88) On 23 January 2003, Wieland, also in the name of its subsidiary Buntmetall, submitted a company statement including its request for application of the 1996 Leniency Notice (96).

(89) On 4 February 2003 and 11 February 2003, the Commission interviewed representatives of OTK. The interview was originally planned for June 2002 but was postponed at the request of the Commission. OTK's representatives gave oral explanations on the functioning of the alleged behaviour.

(90) On 10 February 2003, the Commission asked Outokumpu to provide certain additional information.

(91) On 17 February 2003, KME submitted a second memorandum.

(92) On 3 March 2003, the Commission sent requests for information under Article 11 of Regulation 17 to Boliden, Boliden LDM Nederland BV/HME and Halcor.

(93) On 20 March 2003, the Commission sent requests for information under Article 11 of Regulation 17 to IMI.

(94) Between March and May 2003 the Commission sent a number of information requests and received various replies from the parties.

(95) On 9 April 2003, Halcor's representatives met with the Commission services and applied for leniency. The meeting took place at Halcor's initiative.

(96) On 8 August 2003, the Commission conducted a second interview with a person formerly active in the copper plumbing tube sector (97).

4. THE ADOPTION OF THE STATEMENT OF OBJECTIONS AND SUBSEQUENT PROCEDURE

(97) On 29 August 2003, the Commission initiated proceedings in this case and adopted a Statement of Objections against the undertakings to which this Decision is addressed.

(98) Parties were granted access to the file in the form of two CD-ROMs and one complementary CD-ROM containing a full copy, excluding business secrets, other confidential information and internal documents, of all documents in the Commission's file on this case.

(99) In accordance with the provision of Article 19 (1) of Regulation No 17 and Regulation (EC) No 2842-98, the Parties were entitled to submit their views on the objections within a time limit of eight weeks from the date of receipt of the Statement of Objections. At the request of a number of parties, the Hearing Officer extended the time limit to 7 and 10 November 2003, respectively. All parties replied within the time limit.

(100) Having replied in writing to the Statement of Objections, all addressees of this Decision, with the exception of HME, took part in the Oral Hearing on the case, which was held on 28 November 2003. None of the parties substantially contested the facts or the anti-competitive infringements identified in this Decision, with the exception of HME that did not comment on the infringements.

(101) In preparation of the Oral Hearing, and to provide each party with the opportunity to comment on any additional evidence that it concerns, the Commission gave the parties access to summaries of certain arguments, allegations and evidence contained in the parties. responses to the Statement of Objections that could lead to a new interpretation of facts contained in the Statement of Objections. These arguments, allegations and evidence did not raise any new charges or objections. The parties submitted their observations by 5 December 2003.

D - DESCRIPTION OF THE EVENTS

5. DESCRIPTION OF THE CONTESTED BEHAVIOUR IN THE COPPER PLUMBING TUBES INDUSTRY ON THE EUROPEAN LEVEL FROM JUNE 1988 UNTIL MARCH 2001

(102) This Decision concerns behaviour of copper plumbing tube producers on the European level from June 1988 until March 2001. The presentation of facts will start with introductory remarks (see Section 5.1). The description of competitor meetings and other contacts in the copper plumbing tube industry on a European level is divided in two parts: (i) arrangements amongst the SANCO, WICU and/or Cuprotherm producers on the European level (starting from (at least) June 1988, see Sections 6 and 7), and

(ii) arrangements amongst a larger group of participants on the European level (starting from (at least) September 1989, see Sections 8 and 9).

(103) In both cases, the factual description starts with a general presentation of the nature and content of the contested behaviour (see Sections 6 and 8), and will be followed by a chronological summary of meetings (see Sections 7 and 9).

5.1. Introduction

(104) Meetings and competitor contacts in the European copper plumbing tube industry first occurred at a national level (98), bilaterally and in the context of "SANCO" meetings ("SANCO meetings") amongst the SANCO producers.

(105) Although competitor contacts at a national level, bilateral level (99) and European level (including SANCO meetings) involved different competitors, BCZ, IMI, the KME (-group), OTK, Wieland and, for a certain time, Mueller are considered to be the core participants of the arrangements.

5.2. National contacts

(106) National arrangements are not as such subject of this Decision. [Information not relevant for the EU proceedings] (100)[...] (101)

(107) Unless national arrangements can be considered to be part of the described arrangements on the European level or have certain relevance for them, they will not be described in this Decision (102). The Commission notes that meetings and contacts concerning national arrangements also involved employees, which were at the same time representatives in meetings on the European level. A number of contacts relating to national markets occurred on a bilateral level; occasionally, such contacts emerged from contacts on a European level. In such circumstances, contacts appear to have supplemented arrangements on the European level (103).

5.3. Contacts on a European level including the contacts among SANCO, WICU and Cuprotherm producers

(108) SANCO meetings involved the SANCO producers KME (including TMX and EM), BCZ and Wieland. Meetings were held at least since 1988, most likely already earlier in the beginning of the 1980's, and covered a number of Member States. KME and Wieland also had contacts concering WICU and Cuprotherm tubes.

(109) A description of arrangements among SANCO, WICU and Cuprotherm producers (see Sections 6 and 7) will precede the presentation of arrangements among the larger group of participants (see Sections 8 and 9).

(110) The described behaviour of the larger group of participants on the European level is better understood with knowledge of the activities of the SANCO producers. There is a certain parallelism with respect to organisation and content of the arrangements. With respect to discussions on prices and rebates, SANCO producers were treated as one group within the larger group.

(111) While occasional contacts on a European wide level date back until 1987, European copper plumbing producers first met in an organised form in September 1989, involving the majority of the main European producers. The core group of participants (BCZ, KME (-group), Wieland, OTK, IMI and, for a limited time period, Mueller) had continuous contacts from September 1989 until the Commission's inspections at the end of March 2001, whereas Buntmetall, Halcor and HME participated only temporarily.

5.4. Industry association meetings

(112) Competitors involved in the arrangements frequently met on the occasion of various industry association meetings such as IWCC, ECI and ECPPC-meetings. The parties agree that anti-competitive practices did not occur in association meetings themselves (104). The Commission has no evidence suggesting that association meetings were used as a vehicle for anti-competitive practices. However, meetings were often organised on an ad hoc basis (105), possibly on the occasion of official events (106). It appears that sensitive information was exchanged and discussed in such cases at such meetings (107).

6. MEETINGS OF SANCO PRODUCERS AND AMONG THE WICU AND CUPROTHERM PLUMBING TUBE PRODUCERS, AND COMPETITOR CONTACTS WITHIN

(113) As far as meetings and other competitor contacts concerning SANCO copper plumbing tubes that went beyond the technology and trademark transfer licence relationship, they usually involved all SANCO producers (KME-group, BCZ (until around mid 1995) and Wieland).

(114) Contacts relating to WICU and Cuprotherm plastic-insulated copper plumbing tubes that went beyond the technology and trademark transfer licence relationship involved only two WICU producers (KME and Wieland) and the Cuprotherm producers (KME and Wieland).

6.1. The SANCO organisation and the SANCO technology transfer licence agreements

(115) The SANCO organisation ("SANCO club" (108)) was founded either in the context of the establishment and licensing of the SANCO(r) trade mark ("SANCO") around 1981 (109), or some time later in the course of the 1980's (110). While the SANCO club was not organised in the form of an association, it had a 'secretariat. for the collection of information, which was run, at least since 1988, by KM(E) (111).

(116) "SANCO" is the trademark for copper plumbing tubes that are produced according to a specific technical process which was patented in 1980 by the Usines à Cuivre et à Zinc (UCZ) ("SANCO patent") (112). The technology was developed to produce a premium anti-corrosive plumbing tube. UCZ (and its predecessor BCZ (113)) held the original patent for the production process until its expiry on 7 February 2000 (114), but it did not own the SANCO trademark for a number of European countries. The trademark, as far as it was not owned by UCZ, was owned by its German competitor KM that had filed and obtained a trademark registration in its own name for SANCO in Germany (as of January 14, 1981) and a number of other European countries (115). Subsequently, KM(E) patented a number of improvements/developments of the original SANCO patent (116). UCZ and KM(E) were cross-licensees for each other's patents and trademarks. For their respective licences, they did not have to pay any royalties to each other (117).

(117) In 1987, BCZ acquired certain assets of UCZ. At the time, UCZ was in bankruptcy proceedings. Amongst the assets purchased was the SANCO patent. Since BCZ was not the legal successor of UCZ, it did not legally inherit into UCZ's SANCO trademark licensee position. Thus, BCZ had to renegotiate with KM certain aspects of the joint use of the SANCO trademark (owned by KM(E)). (118)

(118) Besides KM(E) and BCZ, members of the SANCO club included their licensees Wieland (licensee since 1981), EM/LMI (licensee since 1983) and TMX (licensee since 1981) (119). Wieland, EM/LMI and TMX were obliged to pay royalties to KM(E) and BCZ. Royalties were calculated on the basis of volumes sold (120). Accordingly, for the calculation of royalties, the licensees Wieland, EM and TMX had to provide the licensors (KM(E), BCZ) on a monthly basis with sales volume information (121). While certain patent licence agreements have expired, the trademark licence agreements, partly renewed, are still in force (122).

(119) [Summary of the Marketing strategy of SANCO trademark] (123) (124).(125)

(126)

(120) SANCO tubes are sold in most Member States, Switzerland and Norway, and have a significant market position in Germany, Belgium, France and the Netherlands.

6.2. WICU and Cuprotherm cooperation in the framework of the WICU and Cuprotherm technology transfer licence agreements

(121) WICU(r) patents and the trademark ("WICU") belong to KM(E). Wieland has been a patent and trademark licensee from 19 April 1979, and UCZ/BCZ was a licensee from 16 October 1980 until 31 December 1998 (127). KM(E) has been a Cuprotherm(r) patent and trademark ("Cuprotherm") licensee of Wieland since 21 January 1985 (128). WICU tubes are distributed in Belgium, Germany, France, the Netherlands and Austria, Cuprotherm tubes in Germany, Austria and Switzerland (129).

6.3. Participants in the arrangements of SANCO producers and in the arrangements of WICU- and Cuprotherm producers

(122) Until mid 1995, SANCO producers including BCZ, KM(E) (EM and TMX) and Wieland met, or had contact, in the framework of the SANCO club (130).

(123) After 1995, meetings and contacts continued only between the KME-group and Wieland, with the exception of one meeting on 21 June 1999, in which BCZ also participated (131).

(124) KM(E) and Wieland had continuous contact and a number of bilateral meetings concerning SANCO copper plumbing tubes, WICU and Cuprotherm tubes (132). With respect to WICU and Cuprotherm tubes, the contacts started at least in 1991.

6.4. Arrangements involving (at least) the SANCO producers (BCZ, KME, EM, TMX and Wieland)

(125) Although SANCO arrangements going beyond the mere licence relationship between the SANCO producers possibly existed before 1987 (133), sufficient evidence only exists for the time after June 1988 (134).

(126) The general description that will follow (recitals (128)-(149)) only refers to the period from 1988 until at least mid-1994, possibly until mid-1995. It is proven that the cooperation described below (recitals (128)-(149)) occurred between, and involved, the SANCO producers at least until mid-1994. The description of the arrangements from a person who formerly worked as an executive in the copper plumbing tube industry, which was confirmed by BCZ and contested by KME, corresponds almost entirely to the description of the broader European arrangements of KME (see recitals (203)-(207)) and of Wieland (see recital (210)). Due to these different recollections, it could not be established, whether the cooperation described in recitals (128)-(149) was part of the .SANCO club. and occurred only between SANCO producers, or if it also involved other European producers (and would therefore have to be considered as part of the broader European arrangements).

(127) For the period after 1995, while the volume information exchange involving all SANCO producers continued, only a less intense cooperation among SANCO producers occurred, principally just on a bilateral level between KME and Wieland, and mainly concerning Germany. The extent of this cooperation is described below (recitals (151)-(153)).

6.4.1. Arrangements involving SANCO producers between June 1988 and mid-1994, possibly until mid-1995

(128) Agreements until mid-1994, possibly until 1995 concerned at least Belgium, France, Germany, Italy, the Netherlands and Switzerland, and possibly also other countries (135). At least until 1995, the enforcement of the agreements was facilitated by the segmentation of national markets due to certification requirements (136).

(129) For this period, price, discount and volume agreements were always concluded orally, either during meetings or by phone (137). After instructions had been passed on, written notes were destroyed (138). If anyone sent notes to one of the participants, this person was reminded by phone to destroy the paper (139).

6.4.1.1. Agreeing on target prices and discount rates

(130) SANCO copper plumbing tube producers co-ordinated and agreed upon lists, price line increases and decreases, and a discount scheme. Prices were not based on each producer's own calculations (140).

(131) Indexed price lists (including price lines) were agreed upon at the suggestion of the respective market leader (141). The market leader (for example KM(E) for Germany, BCZ for Belgium) issued the national SANCO reference price list (informally) applicable for all SANCO producers, while each producer formally had his own price list (142). KM(E) issued the reference index price list for Germany, BCZ for Belgium, and TMX for France (143). The SANCO producers fixed prices for different countries at different levels (144).

(132) The SANCO producers developed a sophisticated system of joint price increases on the basis of such indexed price lists.

(133) Prices were indexed based on the copper price of the LME (145). For different countries, different index numbers were used in order to hide the fact that the price lists were coordinated amongst the SANCO producers for different Member States (146).

(134) With respect to Germany, KM(E) was the price leader that decided to modify the index. KM(E) informed the other SANCO producers - usually by telephone conference - that the index would be modified (147). Propositions for price changes, especially with respect to other domestic markets, came from national producers (148).

(135) In regular meetings, SANCO producers fixed the discounts for customer groups based on quantities sold (149). The SANCO producers determined for each of these customer groups three groups of copper plumbing tube producers (local SANCO producers, SANCO importers and non-SANCO producers) that had to apply different discount levels. Domestic SANCO producers granted the lowest discounts. SANCO importers usually applied an additional discount of up to 2%. The discount for non-SANCO producers was set below the one of SANCO importers. However, it could not be established whether SANCO producers proposed the discount for non-SANCO producers (150). Discount schemes usually remained the same, while price lines often changed.

(136) Following the agreements, SANCO producers sold their copper plumbing tubes on the basis of identical price lists and a co-ordinated discount scheme. They synchronised the change of price lines (151).

6.4.1.2. Allocation of volumes and market shares

(137) SANCO producers allocated a share of the respective estimated national demand to each SANCO producer (152). In certain markets they fixed "extremely precise quantities" to be sold to each distributor. Thus, customers were allocated (like distributors, for example, in the Netherlands). In the case of allocation of customers (distributors), the demand of certain customers was either reserved for one producer or split between several producers. In the Netherlands, for instance, BCZ was required to retain the Dutch company ROBA as a distributor for SANCO-products in the Netherlands. ROBA participated in this arrangement (153). It is possible that other European producers such as HME also participated in these arrangements.

(138) Volume allocations were regularly co-ordinated with non-SANCO producers on the European level (154). SANCO producers that did not participate were informed of the results of the meetings with non-SANCO producers (155).

6.4.1.3. Monitoring

(139) Initially, after the joining of BCZ until around 1990, the SANCO producers met very frequently to effectively monitor the arrangements. With particular regard to the meetings held in Zurich, they might have also involved other European producers, and would therefore have to be considered part of the broader European arrangements. After 1990, meetings occurred less frequently but at regular intervals (156).

(140) Monitoring was based on four factors: (i) the report of sales figures to the SANCO 'secretariat., (ii) a control of the indicated sales figures through import statistics, (iii) the calculation of royalties and (iv) advertising fees.

(141) First, SANCO manufacturers (including the licensors) reported their production and sales volume figures often on a monthly basis to the secretariat. The figures were compiled and circulated among the SANCO club members (157). Second, until the beginning of the 1990s, import statistics allowed the control of sales figures, since they were a reliable source for data on production volumes, at least for countries like Belgium, where only one national producer (BCZ) existed (158). BCZ still reports figures to INTRASTAT (159). Third, royalties for the SANCO patent and trademark licence were calculated on the basis of production volume (160). Last, SANCO Club members shared costs for SANCO advertising campaigns and related activities for Benelux, France, Germany and Italy. Expenses for advertising were split between the SANCO producers on the basis of sold tonnage in the respective Member State (161).

(142) Similar agreements for the calculation of royalties and the sharing of joint advertising exist for WICU- and Cuprotherm licensees (162).

(143) The report of sales volume figures during meetings or in the framework of the licence and advertising agreements allowed the control of sales volume. According to SANCO licence agreements, principally, SANCO manufacturers sold plain copper plumbing tubes exclusively under the SANCO brand (163), just as WICU and Cuprotherm producers sold the respective type of plumbing tubes only under the respective brand.

(144) At least until 1995, monitoring was facilitated by national certification procedures. Copper plumbing tubes had to be certified in each Member State. Each Member State had its own certification label. Certification organisations, such as the Deutsche Vereinigung des Gas- und Wasserfaches e.V. (DVGW), prohibited producers at least until 1995 to indicate different national certifications on plumbing tubes (164). This segmentation of national markets led to different price levels in different Member States. Between different Member States, margins were up to twice as high for the same copper plumbing tube (165).

(145) If one of the SANCO members did not respect the agreed quantities, producers tried to reconcile the different positions. Concessions as to the quantities to be sold by SANCO producers were made in order to maintain price levels and profit margins (166).

(146) Whenever a participant made an offer to the client allocated to another producer and/or did not respect the agreed discount scheme, he was immediately called and asked for justification (167).

6.4.2. Relationship between SANCO and non-SANCO producers

(147) Insofar as meetings among the SANCO producers were part of the broader European arrangements, all SANCO producers also participated in these broader arrangements. Furthermore, SANCO producers were treated as one group at the broader European level.

(148) "Outsiders" like OTK perceived the SANCO producers as an inner circle with a tighter Cooperation (168). KM(E) approached OTK twice in 1990 and in 1992 to join the SANCO arrangements, which OTK rejected (169). Also IMI was approached.

6.5. Arrangements and discussions between KME and Wieland concerning WICU and Cuprotherm tubes

(149) The WICU and Cuprotherm producers KME (not including EM and TMX) and Wieland had anti-competitive contacts involving an exchange of sensitive information and co-ordination of volumes and prices with respect to plastic-coated copper plumbing tubes in particular concerning Germany. KME submitted an economic study confirming the finding of the Commission, the results of which were also presented in the Oral Hearing. The study found that .prices charged by KME AG for WICU and Cuprotherm tubes increased as a result of their contacts with competitors.(170). Wieland has neither contested its involvement in these arrangements nor rebutted the finding of the study submitted by KME (171).

7. MEETINGS AND OTHER CONTACTS AMONGST THE SANCO PRODUCERS IN CHRONOLOGICAL ORDER

(150) Certain competitor contacts repeatedly occurred in the same form over a period of time. In such cases, the same kind of evidence often exists for a number of different dates. To facilitate understanding, such evidence will be presented at the first relevant point in time.

7.1. Continuous contacts and exchange of information between SANCO producers from 1987 to 2000

7.1.1. Contacts, in particular between KME and Wieland

(151) Following explanations given by KME, the Commision found that Wieland and KME had contacts and met on a bilateral basis from time to time at least since the late 1980s in, for example, Düsseldorf, Hanover, Stuttgart and Frankfurt. KME provided evidence in the form of travel expenses for around 25 meetings that it was able to recall between January 1992 and February 2000 during which, based on the recollection of KME employees, it was reasonably likely that commercially sensitive information was exchanged. Also BCZ recalls its participation in a number of meetings (172). "Many of the discussions with WW took place on an ad hoc basis on the occasion of, for example, factory visits or technical meetings" (173).

(152) Participants at the meetings included "business unit leaders, commercial employees, and technical/market advisors" (174). Issues discussed at the meetings were mainly "routine commercial matters" concerning KME's and Wieland's business relationships for example as licensees of certain plumbing tube technologies such as "advertising campaigns for plumbing tubes, technical developments, and environmental or regulatory issues" (175). On the other hand, some meetings also included discussions on "prices, customers, and market shares. Neither production nor capacity was discussed. Discussions related mainly to the German market" (176).

(153) In addition to the meetings, KM(E) and Wieland occasionally exchanged information on certain customers, usually over the phone "when the need to discuss a (usually large) customer arose" (177).

7.1.2. Institutionalised information exchange

(154) Evidence listed below demonstrates that a regular and institutionalised information exchange with a high degree of detail was established between the SANCO producers. SANCO producers exchanged sales volume statistics at least since 1987 for Germany, Italy, the Netherlands, Belgium, and at least since 1990 for France.

Concerning SANCO tubes

(155) On 27 January 2000, KME sent to Wieland SANCO statistics (tables) of 27 January 2000 for Germany containing sales volumes of plumbing tubes (178) of KM Europa Metal, Wieland, BCZ, TMX and Europa Metalli from 1991-1998 broken down on a yearly basis and, for January-December 1999, on a monthly basis (179).

(156) A written comment of a Wieland employee of 29 January 2000 noted on the table refers to potential complaints by KME about the development of sales of Wieland. Hence, the exchanged sales volume information was subject to discussions amongst the SANCO producers (180).

(157) According to instructions on the table, sales volume figures were to be provided by the SANCO producers to KME by the 5th of each month. Contrary to the majority of tables, the one dated 27 January 2000 does not list BCZ as one of the recipients of the table.(181)

(158) Wieland possessed tables with SANCO figures for the following countries and years:

- Germany: 1987 (182), 1988 (183), 1989 (184), 1990 (185), 1991 (186), 25.01.1994 (187), 24.01.1995 (188), 21.01.1997 (189), 26.01'98 (190), 18.01'99 (191), 20.04.2000 (192);

- France (193): 1990, 1991, 1992, 25.01.1994, 24.01.1995, 21.01.1997, 18.01'99, 20.04.2000 (194);

- Italy: 1987, 1988, 1989, 1990, 1991, 1992, 25.01.1994, 24.01.1995, 21.09.1995, 21.01.1997, 26.01'98, 18.01'99, 20.04.2000 (195); some of the tables concerning Italy also include sales volumes of SANCO producers of non-SANCO plumbing tubes;

- Belgium: 1987, 1988, 1989, 1991, 1992, 25.01.1994, 24.01.1995, 21.01.1997, 26.01'98, 18.01'99, 20.04.2000 (196);

- the Netherlands: 1987, 1988, 1989, 1991, 1992, 25.01.1994, 24.01.1995, 21.01.1997, 26.01'98, 18.01'99, 20.04.2000; (197)

- Switzerland: 23.02.2001 (198).

Concerning WICU tubes

(159) Furthermore, Wieland was in the possession of statistics for WICU plumbing tubes for the years 1982-2000 (199) for Germany (200). Statistics concerning Germany were generally exchanged between KM(E) and Wieland. According to the distribution list on the tables, BCZ received copies of the tables between 1986 and 1995 (201).

(160) Wieland also had in its possession WICU tables dated 14 April 2000 containing volume figures for 1990 (or 1991, 1992, 1993, 1997) until March 2000 for the following countries: Belgium, France, Germany, Greece, Italy, Austria, the Netherlands and Portugal (202).

Concerning Cuprotherm tubes

(161) According to the distribution list on the statistics, Wieland and KM(E) exchanged order and sales data of Cuprotherm tubes for the years 1988 until March 2000 (203).

(162) An employee of Wieland wrote an internal note on one of the tables asking "why don't we meet the 60 % anymore?" (204). The figures contained in the tables suggest that Wieland had a market share of Cuprotherm tubes that was slightly below 60 %.

7.2. Further events between 1988 and 2001

7.2.1. Meeting of 3 June 1988

(163) According to notes found in the office of an employee of KME, there was a SANCO meeting on 3 June 1988 in which at least KM and BCZ participated. Volumes were discussed.

(164) BCZ "wished" the following quantities: 2000 t/year in Belgium, 1800 t/year in France, 2400 t/year in Spain/Portugal, and 3000 t/year in the United Kingdom. The minutes also summarise BCZ's planning in terms of volume output for Germany, France, the Netherlands, the United Kingdom, Belgium and Spain/Portugal (205).

7.2.2. Liège, 4 November 1988

(165) Notes found at the above-mentioned office of the employee of KM, dated 4 November 1988, report of a meeting between KM's and BCZ's employees in Liège. According to the notes, the strategy for "canalisation" was to "fill capacities" (rationalise production), then to control quantities and prices.

(166) Rebates for the German market were discussed (27 % medium size commerce, "MC", 29% large commerce, "LC", 30-31% exceptions, Herwig 35-37%, additional discounts for foreign SANCO and non-SANCO producers) as well as customers (206) and the issuing of a new price list starting from "15/11/88".

(167) KM indicated that it had reduced its production of SANCO tubes for Germany by five shifts "in the common price interest" (207). BCZ confirmed that it had no interest to disturb markets (208). BCZ proposed that it would not deliver "WICU" tubes to Germany and in return, KM should not deliver any "WICU" tubes to Belgium. BCZ would produce WICU for KM in Belgium, while KM should do the same for BCZ in Germany (209).

(168) BCZ took over the SANCO price list of 1988, KM had presented the WICU price list 1988. The "right to sell to Sidal" was mentioned (see explanations given in recital (137)). Volume information concerning BCZ's sales in the Netherlands was exchanged and planned figures were "openly discussed" (210).

7.2.3. Zurich, 24 November 1988

(169) Notes found at the office of the above-mentioned KM employee suggest that a meeting took place on 24 November 1988 in Zurich between KM and BCZ (211) and maybe other SANCO producers (212).

- Discount rates for the German market were discussed: 27% (MC), 29% (LC) and 30-31% ("exceptions"). Foreign SANCO had an additional discount of 1-2%, and non-SANCO 2-3%. The notes refer to a talk with Herwig on limiting the discount to 30 %.

- It is stated that BCZ made DEM 1 million profits each month. For the Netherlands, "figures were exchanged". The Netherlands was historically an important market for BCZ and the target was 200 tonnes/month. Discount rates for Belgium were discussed (TMX 64%, KM 53%, other competitors 57%). BCZ requested more than 39% of the Belgian market, the target was 60 %. BCZ was to issue the new price list of SANCO, and KM of WICU. A visit was fixed for 29 and 30 November 1988 in Brussels (213).

7.2.4. Stuttgart, 26 November 1993

(170) On 26 November 1993, KM, TMX and Wieland met at the request of Wieland in Stuttgart to discuss pricing concerning France and Germany. Wieland was pricing slightly lower than TMX in France. The participants planned to reduce "differences". Wieland reported some "price"-examples that were to be examined (214).

7.2.5. January/February 1994

(171) A meeting among SANCO producers was planned for January/February 1994 (215).

7.2.6. Düsseldorf, 26 September 1995

(172) In a memo, Wieland reported a "working group meeting" "copper tubes" of 26 September 1995 in Düsseldorf (216). Besides KME, MKM-Hettstedt also participated in the meeting. KME employees discussed with Wieland market strategies for 1996. They indicated that it was not yet clear whether an emphasis would be put on quantities of sales or on price stability with the risk of further losses of quantities (217). KME also stated that its capacity would be fully used until end of the year (218).

7.2.7. 5 March 1997

(173) Price coordination with respect to price lines and price lists between KME and Wieland is evidenced by an internal memo of Wieland dated 5 March 1997 (219).

7.2.8. Frankfurt, 24 June 1997

(174) According to the agenda of a Wieland employee, he had a meeting with representatives of KME in Frankfurt (Kempinski Hotel Gravenbruch, Salon 4) on 24 June 1997 (220).

7.2'9. Frankfurt, 27 June 1997

(175) According to his agenda, a Wieland employee had a SANCO/WICU meeting in Frankfurt (Airport Center, Airport Club "Luxembourg", and/or in room "Zurich") on 27 June 1997 (221).

7.2.10. Monitoring by KME, 26 August 1997

(176) An internal note of KME dated 26 August 1997 shows that KME monitored whether Wieland respected the agreed price lines. The statement "Wieland sells out the copper positions" implies that Wieland did not respect the price line that both companies had agreed upon. KME considered that as an "attack" on a wide front (222).

7.2.11. Around 1997

(177) For WICU tubes, there was an agreement between KME and Wieland around 1997 that the market share of KME in the total of 9 markets was 70 % and of Wieland 30 %. Wieland noted that Wieland's sales decreased in Germany above average. As a remedy it recommended cutting back in foreign markets and catching up in the German market (223).

7.2.12. From 1998 until 2000

(178) According to notes, minutes and tables found at Wieland, regular "brand-meetings" were held between KME and Wieland. Such meetings took place at least on 9 June 1998 (224), 21 and/or 22 October 1999 (225), 19 November 1999 (226), 22 December 1999 (227) and 4 February 2000 (228), mostly in Düsseldorf, but also at other places. Regular participants at those meetings were employees of KME and Wieland.

(179) The subject of the meetings was the whole spectrum of strategic commercial information and behaviour, like, for example, product strategy, technical issues, competition by plastic tubes, licence contract and payments, planned investments, pricing especially for SANCO, WICU and Cuprotherm copper tubes, and market leader agreements (229).

7.2.13. Amsterdam, 9 June 1998

(180) According to a memo of a Wieland employee of 10 June 1998, that reports a meeting with KME in Amsterdam on 9 June 1998, in the first five months of 1998, the "additional profit" in the German market went up on average by DEM 30,- % kg (230). Accordingly, during that time period the KME group generated additional profits of DEM 3,9 million in Germany. Since July 1997, the "additional profit" in Germany increased by around DEM 70,- % kg so that "the additional profit is even substantially higher" (231). Calculated on the basis of "only DEM 30,- % kg", the additional profit of Wieland amounted to DEM 1,54 million.

(181) The profit situation, that is to say the "additional profit", concerning France and the United Kingdom was estimated to be more or less the same as in Germany, while it was slightly lower in the Netherlands and much lower in Spain (where Wieland indicated a need to catch up) (232). On the basis of improved prices, HME was estimated to generate additional profits on the basis of 15.000 tonnes/year of DEM 11 million in 1998 compared to 1997 (233).

7.2.14. 12 and 15 April 1999

(182) According to handwritten minutes of 12 April 1999 found at KME, the commercial cooperation between KME and Wieland with respect to WICU tubes dated back at least until 1979. In 1992, the WICU agreement was extended to other geographic markets. KME alerted that Wieland could form a "new club". Some volume statistics concerning WICU and SANCO tubes were exchanged on 15 April 1999 (234).

7.2.15. Düsseldorf, 21 October 1999

(183) According to minutes of Wieland, a tube meeting took place between KME and Wieland on 21 October 1999 in Düsseldorf (235). Both companies discussed their "two-brand" strategy and decided to maintain the status quo (236).

(184) KME's internal e-mails and notes, and its correspondence with Wieland concerning indirect imports of Wieland into Denmark illustrate the nature of the WICU and SANCO licence agreements (237). For example, KME reminds Wieland that it has to make sure that the licensor receives the "necessary information" so that it can better monitor the market developments in separate markets and if necessary control the market (238).

7.2.16. Düsseldorf, 14 January 2000

(185) According to their agendas, employees of KME and Wieland met in Düsseldorf (Arabella Airport Hotel, Room "Gebrüder Wright") on 14 January 2000 (239).

7.2.17. Düsseldorf, 9 February 2000

(186) According to his agenda, a KME employee met with representatives of Wieland in Düsseldorf on 9 February 2000 (Arabella Sheraton Airport Hotel, Conference Room "Gebrüder Wright") (240).

7.2.18. 28 April 2000

(187) On 28 April 2000, Wieland wrote an internal note indicating that Wieland had agreed with KME on the following: end of 2001, all Supersan deliveries would be stopped into all markets with the exception of Austria (241).

7.2.19. Düsseldorf, 30 May 2000

(188) According to minutes found at KME, on 30 May 2000 a "licence partner" meeting took place in Düsseldorf. Participants were KME and Wieland. Under the heading "market leadership", the minutes confirm that "internal papers" were exchanged (242). Participants also discussed timing of a price list ("only starting from 29.5.", "target +D[E)M 30,-"), flow of information (243), and the Water Directive Meetings, specifically "price targets" and "rebate structures" (244).

7.2.20. Düsseldorf, 21 June 2000

(189) According to his agenda, on 21 June 2000 a representative of Wieland travelled to a "brand conversation tubes" to Düsseldorf (Lindner Airport Hotel, Room "Landshut") (245).

7.2.21. 4 September 2000 and 22 November 2000

(190) According to minutes found at KME, a "license conversation" took place between KME, TMX and Wieland on 4 September 2000 and another between KME and Wieland on 22 November 2000 (246).

- Subject of the first "license conversation" were various business issues related to the German, Italian, French and Spanish market such as the introduction of the "half-hard" tube, brand strategies, pricing of WICU and SANCO tubes, and volume statistics (247). For plastic coated tubes, it was agreed that Wieland would set the target price and that a price reduction would be implemented together with EM (248).

- At the 22 November meeting, KME and Wieland mostly discussed business information related to the "WICU" tubes (the publication of a new price list of KME on 15.12.2000, cost structures of KME and Wieland, but also business and brand strategy of Buntmetall/KME related to SANCO tubes) (249). There were contact persons for Wieland at KME in the technical items sector, the sales handling sector and the price fixing sector (250).

7.2.22. 21 March 2001

(191) On 21 March 2001 (08:50), an E-mail was sent from one KME employee to another KME employee, confirming contact with Wieland. The e-mail suggests that Wieland and KME planned to reach an agreement on prices for WICU tubes: according to the e-mail an agreement for "WICU flex for Germany and for France" could be found "basing on actual market conditions". KME suggested basing price calculation for France that would be suggested to Wieland on an "average sales price" calculated by TMX (251).

8. DEVELOPMENT, STRUCTURE AND SUBJECT OF COMPETITOR CONTACTS ON THE EUROPEAN LEVEL

(192) Structure, organisation and operation of the contested behaviour was based upon exchange of sensitive business information, allocation of market shares, monitoring of sales volumes, a market leadership mechanism that involved a shared assessment of markets, and coordination of prices (including price-lists, the application of price-lines and rebates). Participants attempted to stabilise the market by using market shares of a (previous) reference year as a basis for fixing a target for future market shares.

(193) After a period of only occasional and rather unorganised contacts before September 1989, besides the SANCO contacts described under Section 6, European-wide contacts in the copper plumbing tube industry developed in three stages (September 1989 until at least July 1994, July 1994 until July 1997, and July 1997 until March 2001).

(194) In addition to certain information pertaining to specific meetings, Outokumpu, KME and Wieland in particular provided the Commission with general descriptions of the arrangements in the copper plumbing tube sector for the time period from 1987 until 2001. These general descriptions will be presented to a large extent before the information concerning specific meetings, as they complement the information concerning those specific meetings. For the period from 1987 until July 1997, due to differences in details, statements will be described separately. Mueller provided detailed information on the content of specific meetings, sometimes including general information.

(195) Instead of describing the content of specific meetings and contacts, KME and Wieland provided general descriptions of the content and organisation of meetings and contacts. Thus, their general descriptions will be relevant for each meeting listed in the chronology of events (Section 9) that they described respectively.

8.1. The period before September 1989

(196) Initial contacts between competitors of the copper plumbing tube industry, including those during the time before September 1989, are not the subject of this Decision, with the exception of the institutionalised contacts of the SANCO producers that are described under Section 6 (see also recital (102)).

(197) However, in order to understand the context of competitor contacts and the establishment of the information exchange that was organised on a European level starting from September 1989, evidence related to initial contacts of competitors on a European level before this period will be briefly described below (Section 9.1).

(198) According to OTK, in the initial period from 1987 until August 1989, European copper plumbing tube manufacturers had sporadic, unorganised contacts in connection with various international meetings (252).

8.2. Contacts from September 1989 until late 1994

(199) The Commission has found - following explanations given by OTK (see recitals (199)-(202)) - that in September 1989, an organised and systematic exchange of information (the so-called "[...] spreadsheet" (253)) was established (254). The exchange of information took place a few times a year. Cooperation included meetings at the top level (directors or presidents of the copper tube business) as well as at the operational level (mainly marketing directors or similar functions) (255).

(200) For the purpose of stabilising the market shares, the participating companies usually met around October/November to discuss the size of the relevant market and to agree on target volumes (256). Market shares and volume information were exchanged a few times a year and followed by meetings, usually at the operational level (257). Deviations from agreed volume targets had admittedly sometimes led to aggressive arguments, but no penalty scheme was adopted (258).

(201) Discussions on price levels started at some point at the operational level meetings (259). Market leaders informed other participants about the price level in their respective market. They presented their discount policy, which the participants should follow. However, no specific price agreements were reached (260).

(202) From around 1993 until August 1994, the information exchange on the basis of the "[...] spreadsheet" was discontinued, apparently because the figures were not considered reliable. Instead, IWCC data served as a basis for the meetings. Sales volume data of each main market was submitted to the market leader, normally on a monthly basis. Tables, including forecast volumes, of each main market were prepared (most likely by IMI). Forecasts were compared with actual volumes in meetings that took place in intervals of a few months. A few top and operational level meetings occurred in 1993 in connection with the official meetings of the IWCC and the ECI (261). The market leader recommended a discount scheme (discounts were determined based on the brand and/or provenance of the tube and on customer categories), while no specific agreements on prices were reached (262).

(203) Following explanations given by the KME-group (see this recital and recitals (204)-(207)), plumbing tube suppliers (263) had discussions from at least September 1989 until July 1994 involving "a general exchange of information on market trends and pricing patterns" ("Zurich" meetings) (264). KME initially submitted that between "1990 and December 4, 1992" no Zurich meetings took place. Although it might be true that no meetings were held in Zurich, competitor contacts and meetings continued during this period in other locations. In 1993, at least two meetings took place in Zurich, and at least three in 1994. Purpose of the meetings was "to (i) render the market more transparent by estimating demand and establishing volume targets; and (ii) establish price targets" (265). Issues discussed at the meetings were volume figures, price targets and discounts (266).

(204) With respect to price agreements, three principal elements for setting prices are to be distinguished: (i) pricelists, (ii) pricelines, and (iii) rebates (267).

(205) Price lists were not generally the subject of discussions. However, "pricelists were usually issued once a year (in autumn) by the leader in the respective national market. The market leader generally sent, or otherwise conveyed, the pricelist to certain competitors before its publication, to invite adaptation of their prices. The pricelist was, however, also freely available in the market (for example, from customers). The other producers would then generally issue a similar pricelist within one to two weeks. [...] other suppliers were expected to adapt their prices to the leader's pricelist" (268).

(206) In the event that a different "priceline" was applied due to a change in the copper price, "the market leader would increase or decrease its price and its competitors would follow. The other competitors would be informed of the price change of the market leader in advance by fax or telephone and would then decide whether to increase or decrease their prices accordingly. The purpose was to harmonise price changes, avoid 'price wars,' and render price changes more predictable and transparent for the producers." "Customers following the plumbing tube price development in relation to the LME [the London Metal Exchange] copper price would have been able to predict price changes based on historical data. In order to render price changes less transparent for the customers, the market leader would therefore occasionally deviate from its pricelines" (269).

(207) With respect to rebates, "in order to avoid price erosion through large rebates, the suppliers established guidelines on rebate levels. The plumbing tube manufacturers in the respective national markets divided customers into three or four groups according to size, without, however, discussing individual customers" (270).

(208) Wieland confirmed (see recitals (208)-(212)) that regular competitor contacts, which included discussions of sensitive information, began at least in 1993 (271). While KME, OTK, IMI, BCZ and Wieland participated regularly from the beginning, during certain time periods, also Halcor, HME, Wednesbury and Desnoyers took part (272). Meetings were organised on two levels, the operational and the top level.

(209) Invitations to operational level meetings were organised in a rotating system. The hosting company organised the meetings and was responsible for drafting the minutes. Minutes did not mention any of the anti-competitive arrangements. Discussions were only related to plain copper tubes (273).

(210) Market shares of a past reference year served as the basis for future volume agreements. Sanctions were considered but never agreed upon (274). The market leader reported on the market situation of copper plumbing tubes in the respective country and discussed these with the participants (275). There was an agreement that the market leader was responsible for price lists and the respective price line applicable (276). The companies agreed on rebates. Agreed rebate-schemes were circulated (277). Specific customers were not discussed (278). The joint price target was a conversion margin of DEM 200,- (later: EUR 100,-) for a standard tube of 15 mm x 1 mm (279).

(211) Top level meetings were held at longer intervals, sometimes on the occasion of international association meetings (280), in order to monitor the agreements reached on the operational level (281). Principal subjects of the top level meetings were market shares (especially monitoring and discussion) and the general price situation (282). Minutes were drafted on the basis of a rotating system (283). Occasionally, Wieland prepared graphics concerning the price development in a certain geographic area that it handed out to participants of the meetings (284).

(212) A participant of the meetings confirmed that meetings involved volume discussions, continuous price co-ordination through regular adaptation of price lines (mostly by phone) and agreements on rebates. These arrangements continued without any interruption from (at least) September 1989 until at least end of 1994 (285).

(213) The cartel arrangements were described as highly efficient by one top-level participant during the period from 1988 until mid-1995. Although it could not be established whether the arrangements were efficient for the whole period and all countries, the top-level participant indicated that there had been substantial price increases in the United Kingdom (see also recital (277)) (286).

8.3. Period of fewer contacts from late 1994 until July 1997

(214) The Commission has found, following the explanations of OTK (recitals (214)-(215)), that from late 1994 onwards, probably due to the German construction boom that lead to increased demand, there was a "lower intensity period in the cooperation and information exchange" (287) with fewer contacts and exchanges and possibly no meetings in 1995 (288).

(215) Contacts were re-intensified in 1996 as a result of the slow-down in the German economy. The construction boom right after the German reunification had slowed down by the mid 1990s. A few meetings occurred (289), during which participants tried to establish a system of cooperation proposed by KME based on 1991 market share levels (290). Meetings were held at two levels. Executives discussed broad market conditions, marketing managers the operational level. Information exchange concerned five Member States (291).

8.4. The period from July 1997 until March 2001

(216) From July 1997 until March 2001, KME, IMI, OTK and Wieland held meetings under the pretext of discussing the European Drinking Water Directive (EDWD) (292). After first approaches in July 1997, and at the invitation of IMI, Mueller attended meetings from October 1997 until 1998. The five companies together are referred to as "group of the five".(293) At a second stage, from August 1998 until September 1999 or March 2001, respectively, BCZ, Buntmetall, Halcor and HME participated in certain meetings and discussions (together with the group of the five: "group of the nine").

(217) The name "EDWD" was chosen to hide the purpose and true content of the meetings. Directive 98-83-EC was adopted on 3 November 1998. While the adoption of the new directive involved many discussions among competitors with associations (294) and with European institutions, it was only a marginal subject of the EDWD meetings that will be described in the chronological part (Section 9) (295). Minutes of the meetings .were made up. (296). While date, participants and places were accurate (297), the minutes usually referred to activities and discussions relating to the Drinking Water Directive and associations (298). However, participants were not the experts of the technical divisions (299).

(218) Other precautionary measures included the endeavour to limit the number of people involved in the cooperation and the fact that "instructions to the actual operating entities in the company to implement the targets and agreed principles" were given verbally (300).

(219) Meetings were organised on two levels: "elephant meetings" at the level of senior executives, and operational management meetings (or "sweepers meetings") (301) .

(220) "elephant meetings" had the function of establishing market shares and output in each company (302). Subjects were discussions on sales volumes, volume allocation and estimation of price increases (303). To the question of Mueller, why the "elephants" had decided to fix market shares rather than prices, KME replied that, "if there was allocation of volumes, prices would follow" (304).

(221) The second level was the level of sales and operational management (or sales representatives), also called "dung-movers" or "sweepers". Principal subject of the discussions of "sweepers' meetings" was the monitoring of agreements and "implementation of the elephants' discussions by the sales or operational managers of the elephants. Specifically, implementation of volume allocation and pricing". They occurred from early 1998 until 1999, and, according to Mueller, possibly continued (305). "Sweepers" often met in Zurich or in Brussels, and discussed also European prices (306).

(222) Each country's market leader (Germany, France, the Netherlands: KME; Spain: OTK; United Kingdom: IMI) presented his views about past, current and future price trends in the respective country. The leaders also agreed on discounts. Each participant was provided with an overview of sale volumes of each competitor. Formally, the meetings were not led by a single company. However, as a market leader in Europe, according to Mueller, KME generally initiated the process (307) .

(223) As of 1 January 1998, a data exchange took place initially on a monthly, later on a quarterly basis through the World Bureau of Metal Statistics (WBMS) ("World Euro Metal Statistics"). WBMS statistics only contained aggregated figures and no company specific information (308). Participating companies submitted orders and sales data to the WBMS on a per country basis (for Germany, the United Kingdom, Spain, France and the Netherlands) each month (309). Each participant estimated to have a market share in excess of [...] % in one of the five countries, provided on a quarterly basis an estimate of total consumption for the respective country for the previous quarter (310). For its participation, Mueller provided evidence from January 1998 onwards. It withdrew its participation by fax dated 10 December 1999 (311). At the meetings (especially "sweepers" meetings), market share allocations were policed and compared (312).

(224) Following explanations of OTK, the Commission has found that EDWD meetings on the operational level (313) were held about every second month and, starting from 1998, almost on a monthly basis. At the top level (314), they took place less frequently, perhaps a few times a year. OTK affirmed that the subjects discussed were market shares, volumes, national price levels and market developments (315). Over time, agreements became more detailed when the cooperation became more organised (316). Meetings and exchange of information concentrated on the main markets in Europe (Germany, France, Spain, the Netherlands and the United Kingdom) and included a market leader Mechanism (317). While a compensation scheme and retaliatory mechanism did not exist, OTK indicated that on a case-by-case basis, a competitor not following the agreed principles could face aggressive market conduct by other tube producers or an individual producer against that company (318).

(225) KME confirmed that the main topics of the EDWD meetings "concerned the setting of (i) market share targets, (ii) pricing guidelines, and (iii) rebates for customer groups" (319). Participants included a core group (Wieland, OTK, IMI, KME and, for a limited period, Mueller), and BCZ, Halcor, Buntmetall and HME (320). A representative of IMI confirmed that he was aware that Elephant meetings had taken place (321).

(226) Wieland explained that internal memos of its employees attending these meetings summarised the content of the meetings although they were drafted in a neutral way. Agreements and information exchanged with competitors were summarised as if they were internal knowledge or Wieland's own price policy (322). Wieland explained that in 2001, competitors exchanged sales volume information also for Poland (323).

9. CHRONOLOGY OF MEETINGS AND COMPETITOR CONTACTS

(227) Meetings and information exchange on a European level amongst all participating copper plumbing tube producers were structured in a similar way to the SANCO arrangements.

(228) Terminology relating to the contested behaviour that was found in documents copied during the inspections or used in different statements of the parties does not appear to be coherent. For reasons of clarity, the Decision uses a coherent terminology. Meetings on the executive level are referred to as "top level" or, in the period after July 1997, as "elephant" meetings (324). Meetings on the lower management level are referred to as "operational level" or, in the period after July 1997, as "sweepers" meetings (325). In addition, meetings occurring after July 1997, for the top level as well as for the operational level are coherently called "EDWD" meetings (326). Since the precise date of contacts could not always be reconstructed, occasionally contacts might have taken place the day before or after the indicated date.

9.1. Chronology of meetings from 1987 to September 1989

(229) According to OTK, first informal competitor contacts, mostly on a bilateral basis, and at association meetings, started as early as 1987. Although neither contacts nor subject of contacts on a general European (meaning not brand-related) level that occurred before 1989 are sufficiently evidenced, they included the occasional and limited exchange of confidential information, for example concerning market trends, market shares and volumes (327).

(230) Handwritten notes of an employee of OTK evidence that a limited number of bilateral contacts between OTK and KM or other competitors occurred that included the exchange of sensitive information (328). For example, undated notes of that employee mention "the control of the Spanish market: - market shares, - price lists, - channels" (329). In notes dated 9 January 1989, he states "agreed with Lett. France: they will follow the price list. If the price increases, they will reduce quantity." "Lett." appears to stand for an employee of TMX (330). His notes of 16 February 1989 mention the objectives "positive cooperation, no crash, no price war. market review United Kingdom, France, Spain, Scandinavia" (331). Notes stemming from 7 March 1989 in Madrid indicate that "it was agreed that volumes to Finland and Sweden remain the same - we will not increase in Germany" (332).

(231) Handwritten notes of KM dated 7 March 1989 report contact with OTK. According to the notes, investments that overlap are to be avoided as well as "overlapping war". The notes mention "workshops" "every 3 months" for "commercial" and "technical exchange/visits" (333).

(232) On 9 June 1989, KM sent a confirmation to OTK with details of a scheduled meeting in Hamburg on 14 (Fischereihafen-Restaurant) and 15 June 1989 (Hotel Elysee) with around 10 participants of KM and OTK. An employee of KM suggested for the agenda of the meeting, "as this is merely a beginning", that both parties would "just use a breakdown into product groups", including, as a separate product group, plumbing tubes (334). As of 13 June 1989 (335), KM possessed detailed statistics of Outokumpu Copper's export volumes (of both its plants in Finland and Sweden). A handwritten remark on the cover sheet of the statistics confirms discussions with OTK concerning sales volumes of OTK on 14/15.6.89 (336). The statistics contain a breakdown of volumes exported into the Community broken down by country and by product category (337), market shares for Italy, the United Kingdom, Germany, France, Portugal and Denmark for 1988 and sometimes also for 1987 (338).

(233) Notes from OTK dated 15 June 1989 in Hamburg summarise the meeting. During the meeting, market conditions of a general nature were discussed (for example overlaps between the two competitors). The goal of the meetings was to "improve [the] relationship", "avoid overlapping", "avoid war" and "technical cooperation". KM and OTK agreed to hold a cooperation profitability meeting once a year, a "high level meeting" (339) on 6 and 7 September 1989, and another meeting in October-November in Copenhagen (340).

(234) Undated notes of OTK, which seem to stem from a meeting of 15 June 1989 most likely with KM (341) in Hamburg, explain that the general situation in the sanitary tube sector is very turbulent and dangerous due to a number of factors. The notes point out that "the market should be ordered", shares should be frozen on the level of the year 1987, and afterwards adjusted in some way. Furthermore, it is "TMX['s] interest to creat[e] the system again" in France. Also, quantities and commercial terms were discussed (342).

(235) Notes of KM of 15 June 1989 indicate that, at the initiative of KM, a meeting was planned for September on the European level. OTK, EM, Wieland and KM were listed as participants. According to the notes, at the same time, domestic discussions should be "maintained" (343).

9.2. Chronology of meetings from September 1989 to March 2001

9.2.1. 1989

(236) According to OTK, the first European-wide meeting that led to an organised system of information exchange took place on 29 September 1989 in Zurich (Airport Forum). Participants included Chief Executives of the main producers of copper plumbing tubes. Present were employees of KM, Wieland, EM/LMI (see footnote 51), BCZ, IMI (344) and OTK (345). The meeting was initiated by KME that also acted as chairman (346). The meeting followed a SANCO club meeting held the previous day (347). OTK's notes specify the principles of cooperation and evidence the preparatory role of the SANCO arrangements for the general European cooperation amongst all major competitors (unofficial translation) (348):

"The objective is to keep the prices in the high price level countries high - if possible to increase even more

By following the development of the market, everybody is ready to decrease volumes maintaining his relative share, if the situation get[s] worse.

The agreement concerns water, heating and gas installation tubes. KMO was not ready to include [unreadable] tubes in these but considers them industrial tubes.

An external firm will not be used in the communication but we will operate in a closed circle - firms currently involved + TMX with open figures. The secretary will be Mr. [...], who had drafted an estimation on the consumption, imports, exports and production by country, based on IWCC's statistics on plumbing - the estimation does not correspond to reality. IMI? had also drafted statistics of all copper tubes.

It was agreed that [...] sends to everyone blank forms, which everyone fills out with country- and company-specific figures. Denmark is considered part of Scandinavia. [Employee of BCZ] and I will review the figures before they are sent. Also countryspecific figures have to be made available. The forms must be returned to [...] by mid- October. [...] will make a summary and distribute it immediately to the participants.

The next meeting will be held on 17.10. at 11h in the negotiation room of the Amsterdam airport. [Employee of OTK] will take care of the arrangements.

On the basis of the years 87, 88 and the first quarter of 1989, as well as the forecasts for 1990 we will try to define [unreadable] for 1990 at the meeting to be held on 29.11. at 10:30 (Zurich - to be arranged by IMI).

For the time being, information should not be given outside this group.

Other issues

- Based on the SANCO-club meeting (apparently not unanimous) yesterday, the market situation appeared to be the following: Italy + 2%, Holland +-0, Germany +-0, UK .1-? .2%, France eventually -1.

- The problem is Spain.

- Everybody should follow market leader's sales principles list - discounts.

- In Germany Outokumpu should get - 34- -35%

- C&L and Silmet were noted to be problems.

- Apparently [employee of BCZ] will negotiate on the acquisition of HME."

(237) In his personal notes, taken directly after the meeting most likely on his flight back to Finland, an OTK employee further reflected about the meaning and the content of the concluded arrangements as well as about future goals of the meetings:

"More discussions

1. Control principles

- limiting of quantities according to demand

- rising prices due to shortages

- tight co-operation

- continues meeting by country by main suppliers arranged by market leader

- control meeting of top management.

3. Freezing principles

- market shares following the market situation

- how to operate against the customers; how to co-operate when some is loosing" (349).

(238) According to OTK's recollection, in the light of the absence of reliable statistics with regard to copper plumbing tubes, one of the goals of this meeting was the organisation of an information exchange system concerning market shares and volumes (350). As a result, the parties established the so-called "[...] spreadsheet". The underlying system was the following: Each producer provided Mr. [...] with its volume figures of deliveries on a country-by-country basis on a monthly or quarterly basis. With these figures, Mr. [...] prepared a "spreadsheet" that contained the collected data. Subsequently, the data was discussed in the meetings. Such meetings often followed the collection of the data (351). Occasionally, data was provided only at the meetings (352).

(239) OTK explained that the basic goal of the meetings was to protect the main producers' home markets and to freeze the market shares on the basis of the volume figures of previous reference periods. A further objective was to avoid price wars (353). [...] spreadsheets, at least at a later stage, contained volume information (sales and targets) on all Member States and Switzerland (354).

(240) According to OTK, the cooperation organised by the meeting of 29 September 1989 lead to meetings on two levels, at the operational and at the top level (355). Further contacts were established between representatives of the companies in the international trade associations for the copper industry (356).

(241) OTK remembers that from September 1989 until around the end of 1992, top level meetings took place two or three times. Operational level meetings were organised more often, maybe three to four times a year until end of 1991, often in Zurich or in Frankfurt (357).

(242) Undated notes of OTK headed "Frankfurt" relate to the general principles of the arrangements:

"Main principles: 1. Freezing of market shares. it seems that the shares would be frozen at the beginning of 1988. A group of co-operation - marketing managers. price list arrangements - gross price arrangements - discounts?. Objective: To prevent price war - increase prices - by freezing the market shares. Timing of freezing. beginning 1988, according to shares 6/12. The domestic supplier should be guaranteed 65-80 %. The price list arrangement." (358).

(243) OTK explained that participants varied according to the content of the meeting (359). Although it was not able to reconstruct the participants of each meeting, it indicated that high level meetings included executives of EM, KM, OTK and Wieland (360). Participants of the operational level meetings generally included operational management of EM, KM, OTK, Wieland, BCZ, TMX, IMI and sometimes also smaller producers (361). Around 1992, also Wednesbury participated (362). The operational level meetings were normally chaired by KM (363).

(244) OTK explained that the most important meetings were the late autumn (around October-November) meetings. During these meetings, the yearly evaluation of the market took place. On this basis the participants would agree on plans and envisaged market shares for the following year (364). Operational level meetings focused on monitoring market shares and the development of the total market (365).

(245) According to OTK, meetings also involved some form of price co-ordination. The market leader in each country would co-ordinate the information exchange system and introduce a price list as a basis for the respective country (366). It was intended to grant common discounts from the price lists. Discounts and prices varied according to different distribution levels (meaning large importers, wholesalers) (367).

(246) Evidence suggests that at least two further meetings were organised in 1989, one on 17 October 1989 in Amsterdam Schiphol Airport (meeting room) organised by OTK, the other on 29 November 1989 in Zurich organised by IMI (368). OTK was responsible for the organisation of the first operational level meeting after the top level meeting in Zurich on 29 September 1989. It indicated that this meeting was likely the meeting of 17 October 1989 and that it was probably organised in Copenhagen and not in Amsterdam (369).

(247) According to Mueller, a meeting was set up in 1989 (or in 1990) in Rome by EM (370). Participants at the meeting included EM, KM, OTK, Wieland, IMI and Wednesbury (371). Subject of the meeting was the discussion of sales volume contained in the [...] spreadsheet. The [...] spreadsheet was a matrix in A3 format that included market data by company, which had previously been provided by the participants to EM (372). Subsequent to this meeting, Wednesbury and companies submitted data to Mr.[...]. Prices might also have been discussed (373). OTK confirmed that a meeting was held in Rome during this time period and that two representatives from OTK attended (374).

9.2.2. 1990

(248) On 17 December 1990, KM created an internal document, which summarised its decision to apply a new price list and/or price line and set the discount rates for German SANCO producers (KM and Wieland) at 27-30 % (3-5t), 32% (LC), for foreign SANCO-producers at a maximum of 35% (LC), and for non-SANCO producers at a maximum of 38% (375). It appears that KM had knowledge about future rebates that OTK would grant its largest customer. The document states that OTK would apply the same "price system". The Commission understands that the price system, which the document refers to, is the price system proposed by KM in the same document (376).

9.2.3. 1991

(249) According to correspondence between EM and participants in the information exchange system established in 1989, competitors exchanged sales volume data on the occasion of an IWCC meeting that took place in London on 12 March 1991 (377). EM addressed a fax (including tables) dated 5 April 1991 to the following participants of the information exchange system: BCZ, IMI, Glynwed Tubes, Wieland, KM, OTK, TMX and EM. The circulated table contains a breakdown of sales volume and market shares for the years 1989, 1990, and 1991 of EM, TMX, KM and Wieland (both combined), IMI and Wednesbury (both combined), OTK and BCZ for the following countries (sometimes combined): Italy, France, Germany, the United Kingdom/Ireland, Scandinavia, Spain, Austria, Switzerland, Benelux: Belgium/the Netherlands, Greece, and Portugal (378).

(250) According to explanations given by EM, the figures were calculated .according to the information received during the 12 March meeting in London and to the information subsequently received from other participants.. KM prepared and attached a new table containing a new hypothesis for the 1991 budget that could represent the base for a definite discussion and agreement at the meeting that was scheduled for 25 April 1991 (379).

(251) Notes of TMX of 25 September 1991 suggest that a meeting between competitors occurred in Frankfurt. It could not be established whether copper plumbing tubes were also discussed (380).

(252) Some time in 1991, KM warned participants that "there had been an improper disclosure of confidential information relating to the allocation system in Germany. He [one of KME's employees] advised participants that the allocation system ought to be terminated and all documents related to the system be destroyed" (381).

9.2.4. 1992

(253) Another table, including figures for 1991, 1992 and handwritten estimates for 1993 is structured in the same way as the table of the meeting of 12 March 1991 (volumes/market shares/countries) and contains figures of EM, TMX, KM, Wieland, BCZ, IMI, OTK, Desnoyers, "Austria Metal", Sill, Wednesbury and Hettstedt (382). Further tables, submitted by OTK, suggest that the information exchange continued throughout at least 1993 or 1994 (383). A table of the United Kingdom market of 1994 includes information concerning IMI, Wednesbury, TMX/EM, KM, OTK, BCZ, Wieland, HME, Halcor and Brazil (384). Handwritten tables of France, Germany, United Kingdom/Ireland, Benelux, Scandinavia and Spain contain actual and partly corrected figures until March 1993 (385).

(254) In 1992, KM created with OTK working groups ("sales", "production" and "structure") and held a number of meetings (386).

(255) TMX was in the possession of various tables and notes, some of which contain sales volume and other commercial information of competitors (387).

(256) On 26 March 1992, an employee of EM met with a representative of IMI at EM's offices in Milan. The day after EM's employee briefed executives of KM and TMX (388).

"[...], whilst in the meantime the pressure of the English producers in quantities and low prices in continental Europe and mainly in Germany, France and N[e]therlands, has been very heavy.

[Employee of EM] also expressed his opinion that such a situation cannot continue without a reaction from the group...

[Employee of EM] stressed that if we wanted to defend the prices levels in Europe against the attack of the industries coming from E[a]stern Europe or other countries, a qualified product through the trade mark SANCO has to be produced and sold. He insisted very much to make [.employee of IMI] understand the advantages deriving from joining SANCO Club. He declared in [the] name of the group - and also in the name of other SANCO producers, that during the interim period they will not attack the UK market. He expressed this conviction that at the moment IMI joins SANCO Club, also Outokumpu will be forced to do the same.

... The European producers could take the engagement not to make any offer for the acquisition of Wednesbury to make easier its failure." (389)

(257) A letter dated 1 April 1992 from IMI to EM evidences "discussions [...] regarding [...] metal purchases" and "IMI's strategy". Based on statistics that Mr. [...] had available, EM would "shortly be in a position to discuss a plan" (390). Suggested discussions between KM and EM concerned the "entering in the SANCO Club of IMI and Outokumpu" and "the purchase of Wednesbury for closing it down" (391).

(258) According to a letter from EM to TMX and KM, the KME group planned to arrange a meeting with OTK, BCZ, Wieland and IMI for:

- "examining the situation of the sector 'Water tubes' in Europe, under the point of view of both the consumer trend and the production capacities;

- discussing a possible strategy in order to limit the surplus of production capacities and to increase the producers/sellers of SANCO tubes" (392).

(259) A discussion with the purpose of examining the 'situation of the sector .Water Tubes. in Europe, under the point of view of both the consumer trend and the production capacities. and discussing .a possible strategy in order to limit the surplus of production capacities. took place between (at least) EM, KM, TMX, BCZ, Wieland and IMI "during" an IWCC meeting in Budapest on 13 May 1992 (393).

(260) In order to examine "consumption - production", "possible actions", for a "tonnage review" to compare prices, a meeting was set up between EM, KM, BCZ, Wieland and (at least) IMI on 7 July 1992 in Frankfurt (394).

(261) Furthermore, OTK submitted partially undated tables and notes of the period from 1992 to1994 stemming from competitors, competitor contacts and/or competitor meetings. The data contains sales volumes for the United Kingdom (for 1994), United Kingdom pricing information for various periods of 1994, volume data for the Netherlands (for 1993), notes on discussions and tables concerning the United Kingdom/Ireland, France, Scandinavia, Spain, Germany, Benelux (concerning 1992, 1993 and 1994) (395).

(262) A meeting between Wieland, BCZ, IMI and OTK took place on 16 September 1992 in Lausanne in the context of an IWCC meeting (396).

(263) OTK confirms that from the end of 1991 until 1994 meetings continued, although less systematically than between September 1989 and the end of 1991 (397).

(264) OTK remembered that its employees attended a meeting on 4 November 1992 in Zurich ("Airport Forum", room No. 4) concerning top level and operational level Representatives (398). The meeting was originally planned for 16 September (399). KM addressed the invitation to IMI, BCZ, Wieland, OTK, EM and TMX. (400) According to OTK, an employee of KM acted as de facto Chairman (401). OTK recalls that the [...] spreadsheet was presented. Current market shares, company by company, were compared with intended shares (402).

(265) According to KME, a "Working Group" meeting was held in Zurich (Regensdorf, Trend Hotel) on 4 December 1992. Participants included employees of KM, EM and Wieland (403). According to the general description of KME, the issues discussed at the meeting were market trends, pricing patterns, market share allocation and price targets of main European markets (Belgium, Spain, France, Germany and the Netherlands) (404). KME submitted that this meeting was the starting point of a number of meetings until July 1994 (405).

9.2.5. 1993

(266) OTK indicated that on 18 March 1993, OTK, KM, IMI and Wieland met in Zurich ("Zurich Airport Forum") with the purpose of reviewing how the market had developed at the beginning of 1993 (406). According to correspondence, the previous meeting took place on 25 and 26 January 1993 (407).

(267) Some time between 29 March and 4 April 1993, an IWCC meeting was held in Milan. Participants included employees of Wednesbury, EM and OTK. On the occasion of this meeting Wednesbury and EM agreed to a certain information exchange concerning sales to the United Kingdom market (408). Wednesbury and OTK arranged a meeting in Pori (Finland) to discuss sales volumes or discounts concerning, for example, the Scandinavian and German market (409).

(268) According to KME, a "sectoral ECPPC-Meeting" took place on 2 June 1993 in Zurich (Regensdorf, Trend Hotel). Participants included employees of KM (410). According to KME, the described meeting was part of the "Zurich"-meetings. Travel expenses suggest that there was also a SANCO meeting (411).

(269) An internal fax of OTK dated 1 September 1993 mentions Desnoyers' wish to form a non-SANCO club, which, according to OTK, was never set up (412).

(270) OTK indicated that OTK, KM, Wieland and IMI met in the Paris Airport Hilton Hotel on 9 September 1993 to generally review and discuss the market situation (413).

(271) OTK recalls that a meeting took place on 29 November 1993 in Frankfurt (Hotel Gravenburgh Kempinski) in order to discuss the German, French, United Kingdom and Spanish markets with executives of the following companies participating: KM, IMI and Wieland. Operative level representatives might also have been present because of the discussion of particular markets (414).

(272) KME explained that another Zurich meeting occurred on 17 December 1993. From KM(E) there was only one person participating (415). KME indicated that this meeting was part of the Zurich meetings (416).

(273) An undated internal memo of OTK discusses the market situation of copper plumbing tubes in Europe for the period 1992-1993. It illustrates that commercial terms and sales were discussed amongst the competitors involved (417). Discussed figures . although for different countries apparently not accurate - indicated that Germans and IMI had (in total) increased their shares in Europe; BCZ had not submitted its figures yet. IMI had aggressively increased its exports, whereas OTK had slightly lost its share in the domestic market and in Germany.

(274) The overview of the market situation by country shows that OTK is well aware of its competitors. strategies and suggests that one of the SANCO club's objectives is to keep prices high (unofficial translation):

Concerning Germany: "... Deg. [...] is a troublemaker within the SANCO-club - as everywhere, it is seeking customers for tubes that it used to deliver to WW. SANCO club has been able to maintain the price of SANCO-tubes surprisingly high but now it is also under pressure and coming down. The Germans say that the market is going to shrink this year by ~6% - this is not true!..".

Concerning France: "...IMI has increased its deliveries - apparently they co-operate with TMX and have agreed on the deliveries to Brosett and Wolseley (UK). Prices were increased in France at the beginning of this year, but everybody sold at old prices - including TMX although they do not admit it. The domestic competition is tough and TMX is pushing Desn. Deg. to export more and more ago. to Spain... Our strategy is to sell the planned volumes at the market price..."

Concerning Spain: "... Prices are now really low, and because of the low demand it is impossible to do business whatever the price is. The members of the club have reduced their deliveries [supplies], and the problem cannot be solved only with them. We try, however, to get sufficiently power behind the price increase."

Concerning the United Kingdom/Ireland: "... We increased our deliveries somewhat at the beginning of this year - we replaced non-European tubes and thus did not disturb the market. Obviously, we also sold according to the old list, as everybody else..."

Concerning Scandinavia: "...The Brits have imported a lot more than they have told which has stained the market. [...] Prices are dropping - although the Scandinavian club is trying to co-operate. Apparently, IMI and Gusum have made an agreement with Nordin, which is a group formed by four big wholesalers [...] Could you raise the issue? It has previously been agreed that no-one will make an agreement with the above-mentioned organisation. The Scandinavian club has to stand up also in difficult times..."

Concerning Benelux: "Holland is rather well taken care of. Volumes have remained the same and the prices are going up. TMX is trying to increase in Holland, because HME has started to export to France."

In the summary, the author notes that "however, responsible producers have to fight for prices."

9.2.6. 1994

(275) An internal memo of OTK dated 2 February 1994 reports about a visit to EM, where OTK was informed about KM's structure, organisation, management, staff, raw materials and production volume (418).

(276) According to KME, a Sectoral ECPPC Meeting took place on 23 February 1994 in Zurich. Participants included employees of KME (419). This meeting was part of "the Zurich meetings." (420)

(277) According to Mueller, a meeting was organised in Brussels ("Hotel Royal Windsor") in Spring 1994 following an official meeting of the Copper Water Tube Committee of the IWCC. Participating companies included IMI, Wednesbury, KM, Wieland, OTK, Halcor, HME and BCZ. Subjects discussed at the meeting were price list increases, discounts and limitation of sales preceding the application of price increases. According to Mueller, IMI and Wednesbury proposed price increases and related issues. (421) "IMI and Wednesbury jointly presented a proposal for the introduction of a price increase in the UK. The proposal was later successfully implemented by way of successive price increases between April and November 1994, which resulted in the price of copper tubes rising by approximately 60 % overall." (422) Volume allocation/quotas for the United Kingdom may have also been discussed. (423)

(278) In a different submission, Mueller explained that a meeting that included a discussion of actual and target sales volumes of the Irish market occurred "probably sometime in the mid-1990's" following an IWCC meeting (424). Handwritten notes summarise sales volume discussions concerning the Irish market. Since they are written on paper of the "Royal Windsor Hotel", it is likely that both references of Mueller refer to the same meeting. The notes contain a list of tonnage supplied to Ireland estimated by IMI, indicated by each company, as well as the proposed decrease of sales volume for each competitor, for example IMI estimated sales of "280" by TMX, TMX indicated "350", and the sales volume foreseen was "300". The list also contains figures for IMI, OTK, KM, Wieland, Halcor and Wednesbury (425). An OTK employee remembered that an IWCC Copper Water Tube committee meeting took place on 23 April 1994 at Brussels in the "Royal Windsor Hotel". He explained that before or after the official IWCC meeting, an informal meeting might have taken place (426). An IWCC invitation suggests that an IWCC meeting was held from 23-25 March 1994 in the Windsor Hotel Brussels. Handwritten notes found in the office of an employee of IMI suggest that discussions at least concerning the Irish market took place (427).

(279) Between March and December 1994, EM and OTK had correspondence with respect to several planned meetings (in April, May, June, September und December). Amongst the subjects discussed at the meetings was also pressure on sanitary tubes in Europe (428). During this period, OTK also had contacts with Wieland (429).

(280) According to KME, a Zurich Meeting took place on 25 May 1994. Participants included one representative of EM (430). Notes submitted by OTK confirm this meeting. It was an operational level meeting (431).

(281) On 30 May 1994, OTK received a fax from KM containing (most likely KM's) sales figures or the first four months of 1994 (432). Further tables dated 9 June 1994 and 15 June 1994 show sales volumes and market shares for the first five months and 1993 of the SANCO producers KM, Wieland, BCZ, TMX, EM, and the non-SANCO producers Buntmetall, "DG/DS", Halcor, HME, IMI, OTK, "WB" and "MKM" (433). On 3 June 1996, TMX faxed to an employee of Outokumpu an indication that price list No. 19 was applied as of 1 June, also pointing out the rebates for "French SANCO", "Imported SANCO", "Imports with Stock", and "Imports ex Works" (434).

(282) A travel expense report of an EM employee suggests that he participated in a Zurich meeting on 16 June 1994 (435). OTK confirmed that a meeting was held at the operational level. Participating companies included Wieland, IMI, Wednesbury, TMX, EM, HME, Halcor, Desnoyers, BCZ, MKM and OTK. KME did not confirm the participation of its employees, who participated according to OTK (436). According to notes of OTK, the meeting focused mainly on the German market, where prices had deteriorated and the reliability of market share information became increasingly doubtful. KM and Wieland informed others about their determination to jointly retaliate if the agreed rules were not respected. Various tube producers indicated that they would respect agreed volumes (437).

(283) Notes of OTK of the meeting of 16 June 1994 confirm that the previous meeting was held on 25 May. Information on agreements concluded during the meeting of 25 May leaked shortly thereafter to distributors. The leak of information on the agreements upset participants and they expressed their reluctance to participate in further meetings should information leak. Wieland and KM were prepared to support price increases in all of Europe despite their 15,8% market share loss (438). Wieland indicated that it granted rebates of ~ 26% to "M/DS/ZO Orders, 5-10 t p order", and ~ 28% to large merchants, BCZ respectively 29-31% and 32%, MKM respectively 37-38% and 36%. Rebates of other producers (Buntmetall, HME, EM and Desnoyers) were also discussed (439). After everybody had presented their sales situation in Germany, and agreed on volumes (for example IMI: "honour the spirit"), the employee of Outokumpu left the meeting after 51/2 hours with the comment: "I will stick to the proposed volume thru August. If this group does not come to its senses, I will gladly join KM&WW and open the floodgates" (440).

(284) On 6 May 1994, a fax was sent from Wieland to Kabelmetal, OTK, TMX, EM, IMI, and Wednesbury containing an invitation for the "Copper Water-Tube meeting on July 8, 1994, Zurich Airport Forum Tower" (441). It could not be recalled by any of the parties whether the meeting took place (442).

9.2.7. 1995

(285) During the period from July 1994 until the end of 1995, competitors met on various occasions in the framework of association meetings (443). It could not be established to what extent anti-competitive contacts occurred on the occasion of these meetings. However, according to a former executive of the copper plumbing tube industry, and confirmed by BCZ, contacts continued until mid 1995 and at least one anti-competitive meeting occurred on the European level around May or June 1995, possibly on the occasion of an association meeting. Participants included BCZ and other usual participants of meetings on the European level (444). Following explanations given by KME, it appears that such a contact did not occur on the occasion of the IWCC meeting at Tucson (15-17 May 1995). KME submitted that its employees are not aware of any anti-competitive contacts .during these [association] meetings. for the indicated time period (445). However, KME "notes that it cannot completely exclude anti-competitive contacts." (446) During the Oral Hearing, OTK explicitly did not want to exclude that anti-competitive contacts occurred but did not remember any contacts in 1995 (447). While the statement of the former executive, confirmed by BCZ, suggests that an anti-competitive meeting occurred, KME's and Wieland's explanations do not prove the contrary. However, in the absence of a confirmation by other participating parties of the meeting around May or June 1995, and due to a lack of a specific date of this meeting, the Commission considers that it is also not proven that the meeting between the parties on the European level did occur. Some time in 1995, Boliden stopped participating in cartel meetings.

9.2.8. 1996

(286) Wieland indicated that a meeting occurred on 1 February 1996 in Stuttgart ("Hotel Mövenpick"). Participants were Wieland, TMX and EM (448). It appears that the meeting only concerned Wieland and KME (449). It could not be verified whether this meeting concerned high executives, as Wieland indicated, or just the French market, as KME suggests (450).

(287) According to Wieland, meetings of sales representatives occurred on 5 March 1996 in Düsseldorf (Arabella Hotel) (451) and on 26 March 1996 in Stuttgart ("Mövenpick") (452). No other party has a recollection of these meetings. Considering the location, the meetings might have only concerned SANCO producers (excluding BCZ) (453).

(288) Wieland explained that a meeting of high executives was held on 11 April 1996 in Frankfurt Airport ("Airport Center") (454). OTK confirmed that this meeting was a top level meeting although it has no specific recollection with respect to the participants. According to OTK, they were probably the usual representatives of the KME-group, Wieland, IMI and OTK participated (455). OTK indicated that at this meeting, the participants discussed the possibility of initiating the market share monitoring scheme (456).

(289) According to the recollection of OTK, an operational level meeting was held on 3 May 1996 in Zurich following the previous top level meeting on 11 April 1996 (457). Participants included OTK, KME, EM, Wieland and IMI. (458) OTK explained that a representative of KME functioned as a de facto chairman. Referring to the top level meeting of April 11, this representative proposed to stabilise the German market, where demand had weakened in such a way that KME was forced to shut down its Osnabrück factory for seven days. Thus, KME demanded significant volume reductions to other tube producers' imports into Germany. Otherwise, German tube manufacturers would react in other geographic markets. Participants replied that similar volume agreements had to be made for other main markets as well to accept this proposal (459). KME confirmed a Zurich Meeting on 3 May 1996, the participation of its employees (460), and that the meeting was intended to create a platform for future discussions (461).

(290) Notes of OTK taken during the meeting reflect the content:

"[Employee of KME] referred to the previous meeting on 11 April where "the big boys" were present. He wanted to the following:

1. Volume monitoring

2. Price conversation

3. Exchange of date on purchase orders and supply volumes

4. Communication channel

[Employee of KME] presented through indexes the development of volume and prices of KM and WW of the years 1989-1995 [...]

The treat of the Germans: If the imports will not decrease respectively with the decrease of the demand, the Germans will react in all the other markets. The objective of this meeting is to verify if it is possible to find an unanimous plan among KME-WWIMI-OKC.

[Employee of KME] inquired whether the participants were interested to co-operate (=will you reduce the quantities?)

- [Employee of IMI]: OK if price increases in 2 or 3 different countries. Precondition for success is an overall-package, which also can be accepted by the others.

- [Employee of OTK]: if you are expecting co-operation, KM must behave in Spain and WW must behave in England.

<emplacement tableau> <emplacement tableau>

Above excludes Woeste/Babcock pricing. Max discounts to C&G, Thyssen Metal Merchants.

AGREE OR NOT AGREE? by 17 May 1996

EMT >Degond/Desnoyers, HME

WW >BMA

Niemand >MKM

KM >BCZ, Halcor

IMI >Wednesbury

OKC

If all yes, go on 20 May 1996

[...]

1. MKT Q2 down 10 % from 1995>imports, to be reduced by 10 % as well

2. Freeze discounts (RAL)

3. German SANCO discounts up (5%) in order to reduce price difference

4. Linienspiel 1-2 lines/month starting July (target recoup [5%] + D[E]M40 lost earlier 1992-1995)

[...]

The next meeting

23 May-Milan

Birmingham 25 June afternoon." (462)

(291) Wieland indicated that a meeting of top executives was held on 21 June 1996 in Paris Airport ("Holiday Inn"). Participants included Wieland and TMX (463). Since other parties have no recollection of this meeting, the meeting might have only concerned the KME group and Wieland (SANCO producers) or the French market (464).

(292) Wieland indicated that a sales representatives meeting was held on 25 June 1996 in Manchester ("Hilton") (465). OTK recalls such a meeting on the operational level. Most likely, this meeting was also a follow-up meeting of the top level meeting of 11 April comprising the same participants as the meeting of 3 May 1996 (466).

(293) OTK initially indicated that a meeting occurred in June 1996 in Frankfurt (467). It later specified that the meeting was an operational level meeting taking place on 19 July 1996 in the Frankfurt Sheraton Hotel (468). OTK recalled that KME and IMI participated, while Wieland was not present (469). During this meeting, German plumbing tube producers proposed to introduce a single price list for Europe. IMI and OTK rejected this idea. Delivery volumes and market shares were reviewed. It was concluded that the statistics were not accurate, prices were eroding and the proposed volume limitations for the German market were not respected. According to OTK, soon after the meeting German producers increased volumes to fill their production capacity and dropped their prices (470).

(294) In November 1996, OTK circulated an internal memo to OTK's Copper Products unit on legal limits to cooperation and information exchange between competitors (471).

9.2'9. 1997

(295) According to Wieland, a high executives meeting took place on 14 February 1997 in Stuttgart ("Hotel Mövenpick"). The meeting is evidenced by the agenda of a Wieland employee, which indicates that a room was reserved for six persons (472). Neither KME nor OTK have any recollection of this meeting (473).

(296) On 24 June 1997, according to an agenda entry, Wieland met KME in the Frankfurt ("Kempinski Hotel Gravenbruch") (474).

(297) Wieland indicated that a sales representative meeting took place on 27 June 1997 in Frankfurt (Airport Center) (475). OTK confirmed that that meeting took place but explained that it was a top level meeting. According to OTK, the subject of the meeting was to discuss the re-initiation of the market monitoring scheme; participants included OTK, KME, IMI and Wieland (476).

(298) Notes of a call between KME and OTK evidence the agenda for the meeting of 28 and 29 July 1997. It could not be established which party set the agenda (477). According to the notes, participants planned to agree on basic rules concerning pricing of copper plumbing tubes. Tubes should be priced per metre, and goals for 1997 should be discussed. A second subject of the meeting was the demand in Spain, France, Germany, the Netherlands and the United Kingdom. "Real figures" should be used. Thirdly, concerning the allocation of quantities the following points should be raised: "Control (what & how, Info)", "market leadership principle", "further contries, ~ Polen", "what happens if somebody does not follow". Fourthly, concerning "price raise starting from September (40-50 DEM)", the current system and transparency should be discussed. Lastly, a Euro price list for 1998 was considered (478).

(299) An internal fax of 28 July 1997 [...] reported a planned contact with a representative of KME involved in the infringement (479). A [...] employee had learned from IMI about the existence of a "table" comprised of certain copper tube manufacturers (480).

(300) OTK indicated that an EDWD meeting was held in Hamburg ("Elysee Hotel") on 28 and 29 July 1997 with both top level (29 July) and operational level participants (28 July). OTK recollects that top level participants of the following companies participated: KME, Wieland, IMI and OTK. The operational level was attended by OTK, IMI, KME, EM, TMX and Wieland (481). The aim of the meeting was to reorganise the competitor contacts by reinitiating an accurate information exchange. Participants agreed to provide accurate volume and sales figures (482). Main topics of the meeting concerned the setting of market share targets, pricing guidelines and rebates for customer groups. KME and Wieland confirmed that the meeting took place (483). The invitation to the meeting came from KME (484). According to KME, meetings after the July meeting in Hamburg were organised according to a rotation system established by the participants (485).

(301) OTK indicated that a joint top and operational level meeting took place on 9 September 1997 in Frankfurt ("Sheraton Airport Hotel"). Participants were Wieland, KME, EM, IMI and OTK (486). Long discussions on the top level led to an "agreement on the market shares for each participant as well as on targets for price increases. The operational level was instructed to implement the agreement. The meeting also included a decision on the establishment of a reliable information exchange arrangement. IMI was asked to establish the information exchange through an independent body. This was in fact started through the World Bureau of Metal Statistics from the beginning of 1998." (487)

(302) Handwritten notes of OTK of the meeting of 9 September evidence the discussion of market shares of France, Germany, the Netherlands, Spain and the United Kingdom, discounts, price increases, the contacting of other competitors (BCZ - "[an employee of KME] will try", "HME - KME"), and whether other countries should be included (a list includes "CH, IT, AT, Poland, B, SC, Hungary, PL+CIS") (488). Wieland and KME confirmed this meeting, while Wieland also indicated that it was an executives. Meeting (489).

(303) OTK recalls that an operational level meeting took place in Paris ("Hilton") on 16 September 1997. Participants included representatives of KME, TMX, EM, Wieland, IMI and OTK (490). The meeting continued with practical work concerning the agreements of 9 September (491).

(304) A memo of Wieland dated 17 September 1997 summarises the results of one meeting, most likely the one of 16 September 1997 (492).

"Copper installation tubes Europe: summary record of our meeting to assess the five main EU markets in the light of our 97/98 sales planning. Leading market participants and their likely market shares, calculated on the basis of the available statistics and estimated market volumes: the five main EU markets account for 75-80 % of the total EU market volume.

Assessment from a price point of view of the five main markets in the EU, for which we need for WW a processing earnings improvement of DEM 50,- /kg in the first instance.

France: According to information from WF, there are rumours that TMX wants to publish a new price list with an increase of FF 2,~ kg, corresponding to DEM 60.- /kg at a 54% discount rate. Several customers in France have already been informed confidentially by TMX that, under the new price list, (percentage) rebates will be fixed by delivery volume as follows (in the table below we also indicate on the basis of past experience the differences in rebates compared with other tube grades):

[TABLE]

As soon as the TMX list appears officially on the market, we, too, shall accept from that date new orders from France only in accordance with the new TMX list and shall ourselves draw up without delay a separate WW list as we have done in the past.

United Kingdom: In order to achieve the earnings improvements we are aiming at in the UK as well, we shall proceed on the basis of the following new rebate scenario:

[TABLE]

Germany: In our German market also, we can achieve a rapid price improvement only through suitable price indexing. By late November, or early December at the latest, we intend to publish a new SANCO(r) price list...

[TABLE]

Spain: As demand for copper installation tubes is very strong in Spain, Outokumpu is said, according to unconfirmed reports from WEBA, to want to publish a new price list with an increase of 40-50 ptas/kg, and further price adjustments in the light of the good market situation are clearly also being planned via price indices.

Major partners have already been informed confidentially in advance by Outokumpu of the following rebate scheme (before discount):

[TABLE]

In addition, there continue, of course, to be annual bonuses in Spain. The above rebate rates seem to be intended for full lorry loads. As soon as the new list is officially out, Outokumpu intends, so our informant says, to apply it immediately to all new contracts." (493)

(305) Mueller pointed out that from about late 1997, possibly until February 2001, .meetings among the so-called 'elephants'. took place ("elephants" were executives of KME, IMI, Outokumpu, Wieland and, from around October 1997 until late 1998, at the invitation of IMI, Mueller) (494). IMI had approached Mueller "with a market allocation 'offer' from the elephants" (495). The principal subject of discussions was a "forecast of sales volumes, volume allocation and estimation of price increases" (496).

(306) Mueller indicated that at least since 21 October 1997, it was involved in the establishment of the World Bureau of Metal Statistics (WBMS) information exchange system (497). It concerned the gathering of data from each of the five participating companies (KME, Wieland, IMI, Outokumpu and Mueller) on sales and orders in five countries (Germany, United Kingdom, Spain, France and the Netherlands). WBMS would administer the statistical system (498).

"Under the system, each of the participating companies would submit every month information on volumes of sales and orders expressed in tonnes in the five countries. In return, the WBMS would provide on a monthly basis total tonnage of sales and orders by the five participants in the five countries, enabling each individual participant to calculate his share of the business as a percentage of the total business of the participants.

Further, under the system, each participant estimated to have a share in excess of [...] % in one of the five countries would provide on a quarterly basis an estimate of total consumption for that country for the previous quarter. It is believed that such estimates were never provided to the WBMS.

At the so-called "Sweepers" meetings, the participants would disclose their individual figures, as reported to the WBMS. The WBMS figures were used to monitor the volumes sold by participants as reported at the .Sweepers. meetings.. (499)

(307) In a letter dated 22 October 1997, Wednesbury invited Desnoyers to provide it with "forecast sales volumes for October, November and December", its view on Wednesbury's estimates of the 1997 market size by country. Wednesbury indicated that, for the following week's meeting, it was "due to present some information on European price levels." It would "also appreciate any input Desnoyers" could give on this subject, "particularly with reference to France, Germany and Spain". Wednesbury attached volume and price information (500).

(308) The following faxes and/or letters evidence the period in which the WBMS information exchange system was set up: Faxes of 21 October 1997, 16 December 1997, 13 January 1998, 27 January 1998 and 2 February 1998 from IMI to KME, Wieland, OTK and Desnoyers, sometimes also Wednesbury, and internal correspondence of Mueller as well as correspondence with IMI (501). [...] Mueller aimed at joining the "Table V" (502). The WBMS information exchange system was finally established on a monthly basis, and Mueller resigned from its participation in the information exchange system on 10 December 1999 (503).

(309) According to OTK, an operational level meeting was held on 22 October 1997 in Düsseldorf ("Arabella") (504). The "record of results" was distributed by KME to IMI, EM, TMX, KME, OTK and Wieland (505). Wieland and KME confirmed this meeting (506). The minutes of the meeting did not reflect any anti-competitive discussions. However, according to OTK, in continuation of the September meeting in Paris, the "participants prepared for the newest information exchange system and the new endeavours to build a closer cooperation" (507).

(310) OTK remembers that on 21 November 1997, a meeting was called together by KME in Düsseldorf ("Arabella Hotel"). Participants included also BCZ. The purpose of the meeting was to find out whether BCZ would participate in the market share stabilisation. OTK explained that the delivery volume provided by BCZ on the occasion of the meeting was reviewed by the participants, but appeared inaccurate and not reliable (508). In its reply to the Statement of Objections, BCZ confirmed that this meeting took place. During the meeting, in compliance with a settlement agreement of KME and Boliden concerning WICU tubes, BCZ provided correct sales volumes for WICU and SANCO tubes for the period from 1992 until 1997 to KME. It appears that BCZ rejected the joint offer of KME and OTK to participate in EDWD meetings.

(311) According to OTK, a EDWD working group meeting was held on 1 December 1997 in London Heathrow ("Airport Hilton") (509). Wieland and KME confirmed this meeting (510). The memorandum of this meeting was distributed to IMI, EM, TMX, KME, OTK and Wieland (511).

9.2.10. 1998

(312) OTK indicated that an EDWD meeting took place on 30 January 1998 at Brussels Airport ("Sheraton Airport Hotel"). Participants included representatives of KME, EM, Wieland, IMI and OTK (512). According to OTK, "market share information was reviewed and discussed in connection with which it was noted that the participants had lost market shares in various territories, such as OCP in Germany and Spain. The volume information from BCZ was also reviewed but could not be reconciled ." Because of new importers, certain market leaders wanted to renegotiate market shares in their territories (513). KME and Wieland confirmed this meeting (514).

(313) Wieland, in an internal memo dated 2 February 1998, summarised the results of the meeting as follows (515):

BCZ sales figures: "We have carried out the assessment of BCZ on the European installation tube markets for 1996 in accordance with the internal memo of 2.12'97. For 1997 we can now assume that the figures corrected by BCZ in the SANCO(r) statistics are in order and we now intend to take these as a basis for our future observation of this competitor.. (516)

Germany: [...] As we heard, Outokumpu has been under pressure from Babcock, reportedly on account of HME quotations received by Babcock; this was, however, clearly misinformation on the part of Babcock. - Please note most carefully that there will be no more index reductions in future. We must educate our German customers accordingly.

France: Price-wise, the French market seems to be pretty much under control as far as the traditional suppliers of copper installation tubes there are concerned...

Spain: We intend to stick to the rebate table we drew up on 17'9'97. As we have since learned, Outokumpu still grants an individual annual bonus to each customer. Our current rebate of between 34 and 38% without an annual bonus is therefore OK. We now absolutely must stabilise this rebate category, [...] and we hope that in the spring Outokumpu will in its own interests further improve the price level in Spain in the direction of DEM 200.- /kg of processing earnings for the principal gauge. This would of course ideally occur using suitable price index tactics. ...

Netherlands: As we have jointly established, the new HME list with price indices, i.e. as in the case of the German price list, is being strictly applied by us during the second marketing stage.". (517)

(314) Attached to the memo is a table containing the deliveries by KME, Wieland, OTK and IMI in Spain, France, Germany, the Netherlands and the United Kingdom (518).

(315) According to Wieland, a sales representatives meeting was held on 6 March 1998 in Düsseldorf ("Arabella Hotel") with representatives of Wieland and KME Participating (519). KME and OTK confirmed that an EDWD meeting took place (520). According to the minutes, further participants were representatives of KME, IMI and OTK, while EM apologised for its absence (521).

(316) On 12 March 1998, according to Wieland, a top executive EDWD took place in Venice ("Hotel Danieli") (522). Whereas OTK does not have any specific recollection of any top level meeting on the occasion of the IWCC meeting, it confirmed its participation in the IWCC meeting (523).

(317) Mueller explained that a "sweepers" meeting was held in Brussels on 24 April 1998 (524). OTK, Wieland and KME confirmed that an EDWD meeting took place on 24 April 1998, but indicated that it was held in Amsterdam ("Conference Center Schiphol") (525). Participants included representatives of Wieland, KME, IMI, OTK and Mueller. Price information was exchanged and volumes allocated (526).

(318) A presentation faxed on 28 April 1998 found in the office of a TMX employee, recommends a "competitive policy" for the introduction of the half-hard tube in France: "Make WW, BCZ, DD and OTK join to avoid increased competition" (527).

(319) Wieland indicated that a sales representative meeting was held on 8 June 1998 in Amsterdam (Conference Center "Schiphol") (528), followed by executive meeting (529). KME confirmed that a meeting was held on 8 June. (530) OTK does not recall this meeting but noted that one of their employees was in Amsterdam on 8 June 1998 and might have attended a preparatory meeting for 9 June (531).

(320) On 9 June 1998, a top level EDWD (or "elephant") meeting took place at the Schiphol Airport, Amsterdam ("Skyport Corporate Meeting Centre"). Participants included representatives of Wieland, IMI, KME and OTK (532).

(321) An internal memo of Wieland dated 10 June 1998 summarises the results of the meeting. Demand, (market) shares and prices were discussed, mainly concerning five Member States (France, Germany, Spain, the Netherlands and the United Kingdom).

"Demand: There are some clear deviations from the forecast figures. After four months, the situation seems to be this: France -12.8%, Germany -6.3%, Spain +14.7%, Netherlands -10.3%, United Kingdom +1,8%...;

Price situation: Since our internal findings in June 97, processing prices have risen by DEM 70.- on average in the five target markets. A slight erosion of earnings achieved has already set in. Our objective must, however, be to stabilise the price basis or to Europeanise the level. The following individual points are important: The differences between Spain and the UK are too large. They are approx. DEM 100.-. A further adjustment is necessary in Spain. The price differences between domestic suppliers and importers vary too much especially in Germany, the UK and France. The spread must be reduced. It is especially the above-mentioned free-riders that are exerting pressure. There is an urgent need to prepare ourselves for a European list. Price structures must be organised outside the five target markets as well. This applies particularly to Poland, Hungary, Portugal, etc." (533)

(322) Mueller pointed out that a "sweepers" (or EDWD) meeting was held on 23 June 1998 in Brussels ("Sheraton. Airport Hotel) (534). Wieland and KME confirmed this meeting (535). Participants included representatives of KME, Wieland, IMI and Mueller (536). According to Mueller, OTK discontinued its presence at meetings as of that date. Mueller indicated that KME and Wieland took over the presentation of the pricing situation in Spain (537). OTK confirmed its discontinuation "for some months" (538). Parties agreed to report figures for 1997 to the WBMS. (539) Besides the discussion of prices and volumes (540), Mueller and Wieland confirmed that the subject was the same as in other "sweepers" or sales representatives meetings (541).

(323) KME indicated that a discussion on the EDWD took place on 23 July 1998 in Brussels ("Sheraton Airport Hotel") (542). Participants were representatives of IMI, Wednesbury, Wieland and KME (543).

(324) Another EDWD meeting was planned for 27 July 1998 at London Heathrow but Cancelled (544). OTK confirmed that one of their employees visited London in order to attend an IWCC and ECI meeting. Although OTK cannot exclude that a top level meeting might have taken place, it does not have any recollections (545).

(325) According to Mueller, a "sweepers" (EDWD) meeting was held on 27 and/or 28 and/or 29 August 1998 in Zurich ("Airport Forum") (546). Participants were representatives of KME, Wieland, IMI, Mueller, Halcor (invited by KME), BCZ (invited by KME), HME (invited by IMI) (547) and BMA (invited by Wieland). OTK's absence was apologised (548). The newcomers were invited by different participants (549). OTK, Wieland, KME and Halcor confirmed the meeting on 28 August, and BCZ a meeting on August 27 (550). Wieland indicated that the meeting started on 27 August (551), and travel expenses of KME suggest that another meeting was held on 31 August 1998 (552). According to Mueller, the meeting on the first day only concerned the core participants (KME, Wieland, IMI, Mueller and, although absent, OTK). Prices and volumes were discussed. The meeting on the following day involved the extended group, including Halcor, HME, BCZ and Buntmetall. The extended group only discussed aggregated and not individualised sales data. Possibly the four new members held an additional separate meeting (553).

(326) Halcor's employee who participated in the meeting took the following notes:

"1. It was reported from KME, Wieland, IMI, Möller (Outokumpu was not present but Mr. Carretti re-assured, that Outokumpu follows the other four), that starting from the beginning of 1996 prices deteriorated rapidly till June 1997.

The above five companies (we shall refer to these companies as the Five) managed to restore the prices to previous levels under the sacrifice of losing market shares mainly in following five countries: Germany, France, UK, Spain and Netherlands.

They say that the countries Austria, Benelux and Greece have increased their sales to above countries from 1996 to 1998. The Five do not accept to lose any further market shares.

Therefore they are willing to freeze in 1998 their sales at the levels of 1996 and expect this also from A/BNL/GR (that is the Four mills).

[...]

Their additional argument was that freezing the volumes sold at the level of 1996 would make possible to increase the prices, so to increase in the end the profit even at lower quantities sold...

2. ...They ask that we commit ourselves to go down to the quantities of 1996 that is for 1998 we should sell 7.740 t instead of the extrapolated value of 9'960 t.

Buntmetall accepted to stick to the quantity they sold in 1996.

HME and BCZ said that they had made a mistake when they declared the figures of 1996. The correct figure is 32.000 t and not 24.400 t.

It was a great surprise for us to see that all, especially [employee of Wieland] and [employee of KME-group] readily accepted this excuse, and accepted that BCZ/HME sell in 1998 the quantity of 32.000 t, that is, abt. 8000 t more, than the originally declared quantity of 24.400 t.

We stressed that not knowing their request when we came to the meeting we cannot commit ourselves and that we have to revert to our management. Anyway since they accept a 31% increase for BCZ/HME we claim the same right for us. In that case our figure for 1998 would be 7.740 X 1,31= 10.140 t. Both [employee of KME-group] and [employee of Wieland] reacted violently against my statement, not accepting my argument, saying that we have to accept a correction of a mistake declared by BCZ/HME.

Our argument that our sales are low compared to plants with similar or even lower production and also to the high imports to the Greek market was answered by [employee of KME-group] that their group has a spare, shut down capacity of 30.000 t They do not activate this capacity in order not to destroy the prices in the long run. On the other hand, they, as well as, the other four producers representing abt. 70 % of the European production cannot accept of losing further shares. Our loss of market share in 1997 was observed, but they cannot accept an increase in 1998 from us (or from others).

If we find no agreement, then we should imagine what will happen in Europe if they activate their spare capacity.

Although we said that we can give an answer after 2 weeks, [employee of KME-group] pressed that. We call him till 04.09'98, since the next follow up meeting will take place on 10.09'98. In this meeting the discussion country by country will take place, as well as the presentation of a method for a close control of the agreed quantities to be sold, so that no market share shift can take place. The idea is to make profit not by increasing market shares but by increasing the prices.

OUR EVALUATION OF THE SITUATION

... [employee of KME-group] told me on the pause, that it is certainly our decision to increase or not increase our production, but they in their turn cannot accept to lose market shares to us or to others.

We could see that the Five have developed close connections and that participants have- at least at this stage - friendly relations, which have been developed during numerous meetings (especially when sometimes they meet the evening before). We had the feeling that they look upon HALCOR as a 'foreign matter'. The special relation of the Five has been forged since mid/end 1997 when they started to declare their figures to the World Bureau of Metal Statistics (554) which is apparently a better instrument to monitor and control market shares. They meet every 2 or 3 months to exchange the data processed by the Bureau. [employee of IMI] stressed that this procedure works and has brought the participants closer increasing the trust among them. At the meeting of 10.09'98 they will explain how this system works, fixing at the same time the volumes to be sold to the various European countries by each participant.

We suppose that they would not care if we do not agree to a self-restriction to the volume of 1996 (as declared to the IWCC). They would probably welcome this as an opportunity to dump material into the Greek market at a catastrophic price. It is now even easier, as the Five can share the losses.

OUR PROPOSAL

1. We think that at hard confrontation we would lose more.

The hard confrontation would be not to agree to their request, not to participate at the next meeting on 10.09'98 and consequently bear their massive sales to Greece... (555)

(327) Halcor stresses that the notes of its employees (556) were purely personal suggestions. They did not reflect the company's view at the time, but "provided a basis for the company's ensuing conduct, consisting in passive to minimal participation in the meetings that followed" (557). Halcor claimed that it "did not change its sales and pricing policy in the four markets concerned... (558).

(328) Mueller pointed out that a "sweepers" (EDWD) meeting took place on 10 September 1998 in Zurich "Airport Forum" that was organised by Wieland (559). The meeting was confirmed by OTK, Wieland, KME and Halcor (560). The invitation of Wieland was addressed to KME, IMI, HME (561), BCZ, BMA, Halcor and Mueller (562). Apologies were received for OTK's absence (563).

(329) According to notes of Halcor, OTK "for a second time did not come. They said that they do not wish to participate in such meetings. This decision was taken at group level. Nevertheless, they intend to co-operate" (564). The notes indicate that it was discussed whether the participants were willing to form two groups, one of the five (KME, IMI, Mueller, Wieland and OTK) and one of the four (BCZ, Halcor, Buntmetall and HME). Market shares were proposed for each company and group (565). "It was discussed that in order for the system to operate, the companies should cover each other's deliveries, if one of the companies cannot deliver." (566) Discounts were proposed for a number of countries and further meetings planned (567).

(330) On 17 September 1998, IMI sent a fax to Wednesbury, to OTK and possibly others concerning the extension of the WBMS reporting system (568) to four new members (BCZ, Halcor, Buntmetall and HME) (569).

(331) According to the agenda of a representative of IMI, on 18 September 1998, on the occasion of an IWCC General Assembly and Steering Committee meeting, an "elephants" and/or 'sweepers. meeting took place in Bruges (570). While a complete participant list cannot be reconstructed, participants included representatives of IMI, Mueller, OTK, KME and Wieland (571). OTK recalls that there was "most likely" a top level meeting. (572)

(332) Halcor indicated that a meeting took place on 28 September 1998 and provided a number of tables that, according to Halcor, stem from the meeting (573). Halcor further explained that a meeting occurred on 9 October 1998 (the invitation indicated the Skyport Corporate Meeting Center at Amsterdam Schiphol airport) (574), and submitted material (tables) from or relating to the meeting that is dated 6, 8, 10, 12, 13 or 23 October 1998 (575).

(333) An "elephants" (EDWD) meeting took place on 20 October 1998 in London ("Hilton.; room "Hertford Suite"). According to the EDWD minutes and travel expenses, employees of KME, IMI, OTK, Wieland and Mueller participated (576). The meeting is confirmed as being an executives. meeting by Mueller (577), Wieland (578) and OTK (579). One of the subjects was the report of accurate figures to the WBMS (580). The content of the meeting was summarised by an internal memo of Wieland of 21 October 1998: "Spain: [...] By the end of the first quarter of 1999 at the latest, a European level of D[E]M 200,- must be achieved. [...] Europe list: Market prices will converge, and it certainly makes sense to think about a European price list in good time. Rebates, bonuses, payment terms, metal basis, etc. will have to be standardised or at least structured in the same way..." (581)

(334) According to Mueller, an "EDWD Meeting" on the operational level took place on 28 and 29 October 1998 at Zurich ("Airport Forum"). Participants included representatives of KME, Wieland, IMI, Mueller Europe, Halcor, BCZ, HME and Buntmetall. Outokump's absence was apologised (582). Wieland, KME, Halcor and BCZ confirmed this meeting (583). According to the travel expense report of KME and Wednesbury's notes, a preparatory EDWD meeting concerning the German point of view was held on 28 October 1998 amongst KME, IMI and Mueller (584).

(335) Notes of Mueller summarise subject and information exchange of the "group of five" (including KME, Wieland, IMI, OTK and Mueller) ("G5") and those of the "group of nine" (including in addition HME, Halcor, BCZ and Buntmetall) ("G9") (585).

"At the G5 meeting the following topics were addressed:

i) a review of copper water tubes sales statistics for the period January to September 1998;

(ii) KME's proposal for a revision of the allocation of volumes;

(iii) KME's proposal to publish a price list in the Euro currency for all countries;" (586)

(iv) "Margins, agree levels to be achieved by year end

[(v)] Agree terms structures by market." (587)

"At the G9 meeting, only items (iv) and (v) were addressed." (588)

"Margins and Price Statements

The margins show in D[E]M/100Kgs for 1

5x1 (14x1f, 15 x 0,7 UK)

Margins calculated after all discounts and rebates, except payment discount

Copper is LME + $100 (19D[E]M/100Kgs)

The resultant margin is meant to be the lowest available on the market

Germany

July 97 Oct 99 Y/E target Incr. German SANCO 145 200¹ 230 +30 Euro SANCO 75 170² 215 +45 RAL 60 140³ 200 +40 see below 1. Equivalent to a rebate of 52% 2. Equivalent to a rebate of 55% 3. Equivalent to a rebate of 58%* * This level was acknowledged to be the .Italian. level other RAM producers were at max 50 % to wholesalers. This is equivalent to 160 DEM not 140.

Terms structure agreed, before payment discount.

Metals Dealers Large Wholesalers Medium Small German SANCO 54 51 501/2 50 Euro SANCO 551/2 521/2 52 511/2 RAL 57 54 531/2 53 Payment terms: 3% 10 days, 2% 30 days, nett 45 days

When the copper price permits, the index will be adjusted by 2 price lines. This together with the above structure will deliver the target margin.

France

July 97 Oct. 98 Y/E target Incr. TMX 103 ? 92 187-177 200 +13-23 Euro SANCO 75 187-173 190 +3-17 Non SANCO 67 164-1411 180 +16 sec below

1. Equivalent to 62-64% rebate 64% was acknowledged to be the .Italian. level Terms structure agreed

20 to 10 to Small TMX 581/2 561/2 541/2 Other SANCO 591/2 571/2 551/2 Demo 60 58 56 Other 601/2 581/2 561/2

Payments terms: 60 day/ end of month + 10 days (average 85)

Margin increase to be achieved through price line adjustment

Spain

July 97 Oct. 98 Y/E Target Incr. KM 141 155-134 170 +15-36 OKC 81 155-128 180 +25-52 Other 65 120 165 +45

Terms structure on new list

Cat 1 Cat.2 Other 30 27/28 OKC 28 26

Payment terms: 2% 15 days or net 60 days

Adoption of new list + adherence to terms structure will deliver target margin.

Netherlands

July 97 Oct. 98 Y/E Target Incr KM 150 196 220 +24 Other 145 186-170 215 +29

New list published 1st October 1998

Terms structure on new list

Large Wholesalers Medium Small Metal Dealer 32% 30 % 28%

Payment terms 1% cash or net 30 days

UK

July 97 Oct. 98 Y/E Target Incre UK Producers 175 230-247 240 +10 Other 160 201-225 225 +24

New terms structure re-affirmed

Wholesalers Large Medium Small UK Producers 601/2 60 591/2 Imports 62.0 611/2 61

Payments terms 4% 15 days or 2% monthly

Price list to be re-structured on 1st December

New price list already published + effective from 1st Nov. with deliver target margin." (589)

(336) According to Mueller, "the price level/sales terms reflect for each country (i) the maximum discounts that the participants should apply to different categories of customers for various products; (ii) the payment terms; and (iii) price list or price index adjustments. It is recalled that the price level/sales terms were presented by each of the market leaders:

(i) KME for Germany and the Netherlands

(ii) Tréfimétaux for France;

(iii) KME for Spain (in the absence of Outokumpu [...]); and

(iv) IMI for the UK ([...] would usually discuss the UK presentation with IMI before the presentation was made)." (590)

(337) Discussions of the group of five, and possibly also of the group of nine, included volume allocations for the new members proposed by KME (591).

(338) Notes of Halcor confirm that price, rebate and volume discussions took place (592). Halcor noted, for example, concerning Germany that .the increase they refer to, e.g. for KME would be from 200 to 230 and would amount to an increase of 3 % or 3 lines up. They proposed to achieve this either by decreasing the discounts or increase the lines. Rather a combination of the two seems to be preferable, as distributors focus mainly on discounts, and according to Hospach's [KME] view, what is proposed as a target is something that can be achieved as this means a 2,5 % less discount and they will increase the line. For new orders, [we were told] to decrease the discounts from 56 % to 54 % and they will increase the line in two steps. They will decrease discounts by 1 % and change the line twice i.e. they will move to a max. rebate of 1 %. (593).

(339) Contained in the memo are rebate rates for the United Kingdom for imported and United Kingdom-produced tubes (594). Attached is a sales volume and sales target table for July, August, and September 1998 of KME, Wieland, OTK, IMI, and Mueller for France, Germany, the Netherlands, Spain and the United Kingdom (595).

(340) An internal, handwritten report [...] on the allocation and achieved volumes of the "group of five" (including KME, Wieland, IMI, OTK and Mueller) ("G5") of fall 1998 written after the meeting of 29 October further evidences the exchange of sales volume information and the allocation of volumes in tonnage and percentage points for five of Member States (Germany, Spain, France, the United Kingdom, the Netherlands) (596).

(341) BCZ indicated that an EDWD meeting was held on 25 November 1998 in Zurich (597).

(342) Wieland confirmed that competitors agreed on discounts for France, Germany, the Netherlands, Spain and the United Kingdom and distributed discount lists among themselves (598). For example, as of 11 December 1998 rebates were fixed as follows:

- Germany: SANCO producers 50,0 % (small commerce ("SC")), 50,5% (medium commerce ("MC")), 51,0 % (large commerce and final customer/consumer ("LC")), 54,0 % (distributors ("D")), for SANCO importers 50,5% (SC), 52,0 % (MC), 52,5% (LC), 55,5% (D), DVGW-producers (599) 53,0 % (SC), 53,5% (MC), 54,0 % (LC), 57,0 % (D);

- Spain: OTK: 26,0 % (2nd level of distribution), 28,0 % (1st level), importers 27,0 % (2nd level), 30,0 % (1st level);

- France: SANCO TMX 54,5% (SC), 56,5% (MC), 58,5% (LC), SANCO importers 55,5% (SC), 57,5% (MC), 59,5% (LC), .Deno. 56,5% (SC), 58,0 % (MC), 60,0 % (LC), .NF Imp. ab F.. 56,5%, (SC), 58,5% (MC), 60,5% (LC);

- United Kingdom: UK-Tubes 59,0 % (very SC), 59,5% (SC), 60,0 % (MC), 60,5% (LC), imports 60,5% (very SC), 61,0 % (SC), 61,5% (MC), 62,0 % (LC);

- Netherlands (all qualities): 28,0 % (B Commerce), 30,0 % (A Commerce), 32,0 % (exceptions), 35,0 % (distribution). (600)

(343) On 9 December 1998, IMI had contact with Mueller (601).

(344) Mueller indicated that on 10 December 1998, a "sweepers" (EDWD) meeting was held in Zurich ("Airport Forum") (602). Participants were representatives of KME, Wieland, IMI, Mueller Europe and HME (603), BMA and BCZ; OTK was absent. It was the last "sweepers" meeting attended by Mueller; BCZ only participated briefly to announce its resignation from the meetings, "apparently following a business dispute with KME" (604). Wieland, KME and Halcor confirmed that this meeting took place (605). The meeting included a group of five meeting on December 9 (from which OTK was absent) (606) and a group of nine meeting.

(345) Wieland's internal memo dated 11 December 1998 summarises results of the Meeting (607): The memo contains a detailed market analysis with volume and price information of main competitors concerning the five target Member States (France, Germany, the Netherlands, Spain and the United Kingdom) (608). Concerning Spain, it states that: "In this case it looks as if Outokumpu finally really takes on its market leader role. Therefore, we should absolutely hold the price and discount level in order to contribute in this way more to stabilisation. (609).

(346) Attached to the memo is a sales volume and sales target table for September, October and November 1998 of KME, Wieland, OTK, IMI, and Mueller for Germany, France, Spain, the Netherlands and the United Kingdom (610).

9.2.11. 1999

(347) OTK indicated that between the beginning of 1999 and February 2000, four to five meetings were organised at the copper plumbing tubes operating level (611). Usually present were representatives of Wieland, KME, IMI, BCZ and OTK, sometimes also a representative of Buntmetall (612).

(348) KME possessed "SANCO/WICU statistics" that also include figures of non-SANCO/WICU tubes of IMI and OTK dated 15 March 1999, 23 April 1999, 15 June 1999, 26 June 1999, 16 July 1999, 26 July 1999, 19 November 1999, 17 January 2000, 09 May 2000 and 14 July 2000 for the Member States at the time (sometimes combined), Switzerland, and sometimes Hungary and Poland. The tables contain price lines, rebates, payment conditions and "current added value" of each producer (613).

(349) In early February 1999, Wednesbury discussed with IMI Mueller's resignation from the "Zurich meetings" (614).

(350) A memo of Wieland dated 8 February 1999 and headed "Preparation Install.-Tubes Europe" describes the issues for the meeting of 10 February 1999:

"Installation tubes EU (D, E, F, GB, NL)...

During our internal talks R/H and RH/An in mid-December '98 we established that KME is dissatisfied with the aggregate European market share and therefore an attempt will be made to alter the position somewhat in this respect. It will be recalled that in autumn '97 WW made major market share concessions in these five key EU markets and was compelled as a result to fill its capacities outside the EU (eastern European markets) at considerable expense. Therefore :

Market shares must, as agreed, be fixed per country. [...]

Market shares are based on an estimated annual volume for 1997 and then for 1998. On the basis of a "comfortable" price, KME could not make progress in, say, Germany by as much as might be expected. Clearly, whoever has the biggest market share in a particular country not only always enjoys the best opportunities, but, earnings-wise, profits the most in absolute terms from the good price level that prevails (if we record an earnings improvement of just DEM 50/kg, as we certainly did in '98 compared with '97, then with its actual quantities in '98 KME has achieved an earnings increase of approximately DEM 30 million while we have achieved just under DEM 9 million. On top of that, the KME group would have achieved a bigger turnover with uncontrolled markets than it has already done with controlled markets).

A KME model for reallocation as per the Annex, which was leaked at the end of '98, is unacceptable to us, especially since it was we who made the biggest sacrifice of all back in '97. Our current market shares are the lowest we can bear.

Our proposal to the participants in future talks following the departure of Boliden:

- The five "old" participants would, as before, be the KME group, IMI, Outokumpu, Müller Industries and WW. It is a sine qua non for operating in the five markets that Outokumpu should return to the table at .commercial meetings., all the more so since Müller Industries [...] is no longer going to send anyone along, although it promises otherwise to co-operate fully as before.

- Of the four "new" participants, i.e. HME, BMA, Halcor and BCZ, BCZ quit, as everybody knows, after the WICU(r) commotion in December '98. HME and BMA are sending out signals that they still want to co-operate, and we have learnt through third parties that Halcor is willing to do the same. Our proposal as far as the new participants are concerned is therefore that we should continue with the three remaining ones, as in the five main EU markets BCZ accounted in 1998 for only about 7,5 % of the nine former participants. total deliveries of some 225.000 tonnes. This is definitely better than jeopardising everything we have achieved so far just because of the KME/BCZ WICU dispute and the resulting exit of BCZ; this should be clear to every participant given the millions earned, as mentioned above, especially to the KME group, which was and still is the one that stands to gain the most in absolute terms from the whole exercise...

Price movements/monitoring

STAND I T ; WW Interpretation D D E F GB NL Dtsch Sa. DVGW Goal: Proc. Pr. Basis hard Length 5 m D[E]M % kg 200 200 200 200 200 200 Reached 15.1'99 209 164 160 177 193 204 - half-hard -25 -25 -3m La¨ -10 Status 15.1'99 209 v 164 [-] 160 [-] 177 [-] 158 [-] 179 [-]

... (GB) The new price list has been drawn up, but a mere 2-3 weeks after it was introduced rebating has started to get out of control again, with the result that, from where we stand now, there is nothing left of the processing price increase." (615)

(351) Wieland indicated that on 10 February 1999, an "elephant" (EDWD) meeting was held in Brussels (Conrad Hotel) after an EMCI meeting. The next day, 11 February 1999, ECI meetings took place. According to the minutes, representatives of Wieland, IMI, KME and OTK participated (616), possibly also Desnoyers (617). OTK confirmed that this meeting was a top level meeting arranged by IMI (618).

(352) According to travel expenses, a EDWD operational level meeting was held on 21 and 22 February 1999 in Zurich-Regensdorf ("Mövenpick"). Participants included representatives of Wieland, IMI, HME (619), Buntmetall, EM and KME. (620) Wieland and KME confirmed that this meeting took place (621).

(353) An internal memo of Wieland dated 23 February 1999 summarises the results of the meeting (622):

"Price situation on the five markets: Recorded in the attached rebate survey are the target rebates that we decided on during our conversation on the basis of the information we have obtained from our internal and external field service and from trading partners. ...

Germany: Our earnings target for 15 x 1 mm remains DEM 230,- /kg, which we have unfortunately not yet achieved. This would produce for RAL/DVGW tubes with a 3 percentage point rebate difference DEM 200,-- /kg. We therefore intend to stick to the previous rebate matrix, i.e. for large customers + purchasing associations 53

% rebate on our SANCO(r) tubes, assuming that suppliers of RAL/DVGW tubes must continue to have a 3 percentage point lead if they are to be able to sell at all. The problem in recent weeks has been the rebate positions for EURO-SANCO tubes both of BCZ and of TMX and LMI, which unfortunately have been close to or at the RAL/DVGW level. This naturally brings RAL/DVGW suppliers under pressure, so that we hope in the coming weeks to see once more from EURO-SANCO(r) producers rebate differences at least one percentage point lower than RAL/DVGW and two percentage points higher than SANCO(r) tubes by KME and WW ... [concerning the new price list planned by KME for March] Unfortunately we, too, cannot, if KME so acts, swim against the KME current, although we did intend, and still consider it correct to do so, to specify in our SANCO(r) list different prices for half-hard and hard tubes according to the extra manufacturing costs. We shall, however, have to follow the market leader." (623)

(354) An elephant EDWD meeting took place on 11 March 1999 in Paris ("Hilton", Edmonton meeting room, Charles de Gaulle). According to agendas and travel expenses, participants included representatives of IMI, KME, and Wieland (624). OTK confirmed the participation of one of their employees in this meeting (625), and Wieland that the meeting took place (626).

(355) Wieland indicated that an operational level meeting was held on 16 March 1999 in Rome ("Lord Byron") in which (at least) one representative of Wieland and one of KME participated (627). Halcor confirmed the participation of one of their employees in this meeting (628). Halcor's notes show intense discussions on prices, rebates and volumes (629).

(356) The internal memo of Wieland dated 18 March 1999 summarises the results of the meetings as follows (630):

"Statistics: The World Bureau of Statistics figures are now available for January and recently had to be reported for February. As we have been able to find out from the World Bureau of Statistics, the following have reported: KME, IMI, WW, Halcor, HME, BMA, Outokumpu and Mueller. BCZ no longer contributes to these statistics, having withdrawn from the IWCC, the ECPPC and all other industry bodies...

[...]

Netherlands: We recently received a visit from Hamel/[employee] and have already told him that we intend in future to apply also vis-à-vis Hamel, i.e. vis-à-vis metal dealers, the HME price list with a maximum 37% rebate for the metal trade...

Spain: This is currently the only one of the five big markets in which chaos reigns in every respect, as Outokumpu is, as far as we can see, not at all performing its role of market leader.

Other markets: Before our next talks in April, we must without fail examine more closely in-house the markets in Austria, Belgium and Greece, and I therefore ask RH/An and RHV/Bo to work out beforehand proposals as to how we might envisage the stabilisation or improvement of the market situation on these three markets so that we might then discuss together by what measures we for our part might contribute thereto." (631)

(357) Attached to the memo is a table specifying for each of the national markets France, Germany, the Netherlands, Spain and the United Kingdom the rebates for SANCO plumbing tubes, imported SANCO tubes, and other plumbing tubes (according to quantities sold). A second table lists the deliveries in the respective markets of KME, Wieland, Outokumpu, IMI, Mueller, Buntmetall, HME, BCZ and Halcor in January and February 1999 (632).

(358) On 31 March 1999, KME faxed its SANCO copper plumbing tube price lists to IMI. Handwritten notes appear to indicate rebates, most likely those of KME (633).

(359) According to travel expenses of a KME employee, an EDWD meeting took place on 28 April 1999 in Brussels ("Novotel Airport") (634). Wieland classified the meeting as "sales representative meeting". KME and Halcor confirmed their participation (635). While Halcor did not announce its withdrawal from the meetings, the Commission has no evidence that Halcor attended further EDWD meetings.

(360) Wieland's Memo dated 29 April 1999 summarises the results of the meeting (636): The memo contains an analysis of the European market and the five "most important markets" (France, Germany, the Netherlands, Spain and the United Kingdom). Concerning Germany, the memo states the following:

"Rebates have unfortunately increased in Germany as well. In Germany we have one advantage over the UK, and that is that the applicable price list and price index are being complied with by almost all operators, i.e. the list and index are being recognised by German copper installation tube buyers. All that matters here in Germany is therefore the rebate level. The maximum 53% obtainable from KME and WW for German SANCO(r) tube is, however, no longer tenable as BCZ is already offering discounts of up to 59,5% for Belgian SANCO(r) tubes and of up to 60 % for non-SANCO(r) tubes - If these rebate differences cannot be reduced once more, we shall be under permanent price pressure on the German market and even KME and WW will in the long run be unable to withstand it.

The new price lists of KME and ourselves for SANCO(r) tubes are now on the market and we intend first of all to try to apply the rebates contained in the attached summary ...

Netherlands: Here we intend to stick without fail to the rebates we last fixed on 30.3.99 ..." (637)

(361) Attached to the memo is a table specifying for each of the national markets France, Germany, the Netherlands, Spain and the United Kingdom the rebates for SANCO plumbing tubes, imported SANCO tubes and other plumbing tubes (according to quantities sold). A second table lists the deliveries in the respective markets of KME, Wieland, OTK, IMI, Mueller, Buntmetall, HME, BCZ and Halcor in January, February and March 1999 (638).

(362) According to OTK, an operational level meeting was held on 3 May 1999 in Birmingham (Novotel Airport Hotel) (639). Participants were most likely the ones listed above (640).

(363) On 9 May 1999, an "elephants" (EDWD) meeting was held at the Vilamoura Marine Hotel (room "Taurus") in Vilamoura (Portugal) after an ECI meeting at 14:00 (641). According to agendas or travel expense reports, Wieland, IMI, KME and OTK participated (642). Wieland confirmed its participation; OTK was not able to recall, if a top level meeting was held (643).

(364) According to the agenda of an employee of Wieland, a number of representatives of Wieland met with KME on 16 June 1999 in Frankfurt Airport ("Flughafen Airport Conference Center", Room 18) (644).

(365) An "elephant" (EDWD) meeting was held on 17 June 1999 in Brussels ("Sheraton Airport Hotel", Room "Andromeda"). According to agendas or travel expense reports, IMI, Wieland and KME attended (645). Wieland confirmed this meeting and OTK its participation (646).

(366) KME was in the possession of a set of tables "Construction of Sales Curve 1998" . "Copper Plumbing Tubes (Total)", "Plumbing Tubes Plain", and "Plumbing Tubes Covered" dated 18 June 1999 and corresponding graphics "Industrial Result 1998 by Competitor" - "Copper Plumbing Tubes (Total)", "Plain", and "Covered" (647). The tables contain sales volume figures and the added value expressed in "Euro % kg" with the respective plumbing tubes of KME, OTK, Wieland, IMI, Mueller, BCZ, HME, Halcor, MKM, Lacambra, Feinrohren, Silmet, Foam, Czepel, Hutmen and others for the following markets: Austria, Belgium, Czech Republic, Denmark, France, Germany, Greece, Hungary, Ireland, Italy, the Netherlands, Poland, Portugal, Scandinavia, Spain, Switzerland, the United Kingdom. Furthermore there is an estimate of the costs expressed in "Euro % kg" and of the "Total Result in Mio Euro" of each of the companies mentioned (648). The industrial results with copper plumbing tubes are compiled in the graphics (649). The tables are very detailed and contain information that is publicly not available. For instance, according to the tables, OTK realised sales including all plumbing tubes ("plain" and "covered") in 1998 in Germany of 7,670 t with an added value of 86 "Euro % kg", in Italy of 1,400 t with an added value of 60 "Euro % kg", in France of 3,900 t with an added value of 95 "Euro % kg", in the United Kingdom of 8,460 t with an added value of 119 "Euro % kg", no sales in Ireland, in Spain of 16,580 t with an added value of 72 "Euro % kg" (650).

(367) According to travel expenses, an EDWD meeting was held on 21 June 1999 in Zurich ("Mövenpick"/Zurich Airport Forum). Participating were representatives of Wieland, IMI, BMA, EM and KME. (651) KME confirmed this meeting (652).

(368) An internal Memo of Wieland discusses the pricing and rebate situation in Germany and its meaning for the prices of SANCO tubes (653):

"Proceeding on the assumption that at least KME and ourselves currently have a fair backlog of orders for SANCO(r) tubes producing an average processing profit in the region of DEM 190-DEM 200/kg, which should be enough to keep us busy for 4-6 weeks before we again come under heavy pressure to secure new orders, but bearing in mind on the other hand that competitors such as Halcor, Outokumpu and, above all, BCZ have in the meantime increased the rebates for their DVGW tubes to 60-61%, we shall be discussing on 21 May whether it might not make sense to reduce immediately after Whit the price index from 32 as at present to 29 subject to strict application of the new price list...

If foreign competitors nevertheless book orders for these, to them, very low processing earnings categories, the processing earnings from German SANCO(r) tubes will also be seriously jeopardised, and - as we know from 1997 - millions are at stake as far as German SANCO(r) producers are concerned." (654)

(369) According to Wieland, sales representatives met on 25 June 1999 in Amsterdam ("World Trade Centre Schiphol"). Participants included representatives of Wieland (655), KME and OTK (656).

(370) Wieland's Memo dated 28 June 1999 summarises the results of one of the last meetings (657). The memo concerns discussions of rebates and prices concerning five Member States (France, Germany, the Netherlands, Spain and the United Kingdom). Attached to the memo is a table containing the rebate structures on the five Member States France, Germany, the Netherlands, Spain and the United Kingdom (658).

(371) Wieland indicated that a sales representatives meeting was held in Amsterdam on 22 July 1999 ("World Trade Centre Schiphol") (659). KME confirmed the meeting (660).

(372) A memo of Wieland dated 23 July 1999 summarises the discussions (661):

"Statistics: ... As the attached table shows, we can, for lack of individual statistics, as yet only quantify the three competitors Halcor, HME and Müller Industries as a whole by subtracting the individual figures for KME, WW, Outokumpu, IMI, BMA and BCZ from the overall figures...

Spain: Outokumpu now seems to be taking its market leadership more and more seriously. We therefore intend to maintain our rebate steps, 46% for the first marketing stage, 44% for the second marketing stage, as deals are possible at those levels. In the case of new orders, we are going strictly by the new price list issued by Outokumpu in Spain..." (662)

(373) Attached to the memo is a table specifying for each of the national markets France, Germany, the Netherlands, Spain and the United Kingdom the rebates for SANCO plumbing tubes, imported SANCO tubes and other plumbing tubes (according to quantities sold). A second table lists the deliveries in the respective markets of KME, Wieland, OTK, IMI, Mueller, Buntmetall, HME, BCZ and Halcor for March, April, May and June 1999 (663).

(374) On 3 August 1999, an "elephants" (EDWD) meeting was held in Paris ("Hilton Airport", room "Stockholm"). According to agendas or travel expense reports, employees of IMI, KME and Wieland participated (664). OTK confirmed the participation of one of their employees (665). Wieland indicated that a top level meeting in Paris ("Hilton") was held on 30 August 1999 (666).

(375) A memo of Wieland dated 18 August 1999 summarises results of the meeting as follows: "United Kingdom: - Outokumpu above all lost quantities in the first few months, which it tried to make up for in June. Considerable price pressure was unleashed by similar action on the part of KME, with the result that current prices have sagged appreciably (-25%). We shall at all events adhere to our internal plan of a price increase in several steps...

Spain: ... For the first time the market is more orderly than it was during the first few months. The new price lists are presumably being applied...

Netherlands: Consumption is in the region of 11 000 tonnes/year. KME has doubtless lost market share. IMI's over-reaction has led to a fall in prices. I reckon, though, that the situation can be put to rights once more by mid-September at the latest." (667)

(376) In August 1999, Halcor filed its last return with monthly figures of July 1999 to the WBMS (668).

(377) According to Wieland, on 7 September 1999, a sales representatives (EDWD) meeting took place in Düsseldorf ("Arabella") (669). KME confirmed that a meeting took place on 8 September 1999 with Wieland. According to the travel expenses of the Wieland employee, the meeting could have also taken place on 7 September (670). OTK was also present (671).

(378) Wieland's Memo dated 08 September 1999 summarises the results of the meeting (672):

"Germany: Unfortunately, rebates have increased once more in recent weeks, albeit not by much, which is why we agreed we would have to carry out a one percentage point rebate extension at all levels. The new rebates are set out in the attached table. We must, however, now apply these absolutely strictly to counter any further rebate erosion. Through the price indices we intend in the next 4-6 weeks to bring processing earnings for the reference measurement of 15 x 1 mm back to DEM 200,- /kg (calculated without any minus tolerance, as usual) for SANCO(r) tubes manufactured by us.

France: ... The price/rebate situation in France is fairly stable and processing earnings are currently more or less comparable with those in Germany...

United Kingdom: ... We shall ourselves apply to new contracts in the UK the rebates that are set out in the attached table, being two percentage points higher than those of 22.7'99.

Netherlands: ... What we have here is another typical example of price reductions bringing only temporary benefits, if any, to those who introduce them." (673)

(379) Attached to the Memo is a table containing the rebate structures of the five Member States France, Germany, the Netherlands, Spain and the United Kingdom (674).

(380) A top level meeting might have been organised on the occasion of an ECI and ECMI Steering Group meeting on 9 September 1999 in Brussels in the Sheraton Airport Hotel. Participants might have included representatives of IMI and OTK (675).

(381) On 28 September 1999, an "elephant" (EDWD) meeting took place in London. According to agendas or travel expense reports, representatives of IMI, Wieland, and KME participated (676). Wieland and OTK confirmed that a top level meeting took place, OTK indicated that it was held at the premises of IMI (677).

(382) An operational level meeting took place on 8 October 1999 in Amsterdam ("Schiphol Hilton"). Participants included Wieland (678), OTK (679) and KME (680). BCZ confirmed the participation of two of their employees (681).

(383) Wieland indicated that a top executive meeting took place on 17 October 1999 in London at IMI. It appears that OTK participated (682), although it has no recollection of this meeting (683).

(384) Travel expenses of Wieland suggest that an EDWD meeting on the operational level occurred on 25 November 1999 in Zurich ("Airport Forum"). Wieland, KME, OTK and BCZ confirmed their participation (684).

(385) According to agendas and travel expense reports, an "elephant" (EDWD) meeting was held on 29 November 1999 in Brussels ("Sheraton Airport-Hotel"). Participants included IMI, Wieland, KME (685) and OTK (686).

9.2.12. 2000

(386) In January or February 2000, IMI confirmed to Wednesbury by phone that "the members of the Zurich meetings had continued contacts" (687).

(387) According to Wieland and confirmed by KME, an EDWD meeting took place on the operating level on 27 January 2000 in Zurich ("Airport Forum"). Participants included Wieland, Buntmetall, EM, KME and IMI (688). Representatives of KME, EM and IMI held a EDWD dinner the night before in the Mövenpick-Airport Hotel (689). BCZ confirmed its participation (690).

(388) On 3 February 2000, IMI analysed margins of the German and United Kingdom copper plumbing tube market. It concluded:

"Statistical information would indicate that, with the new markets in the former Eastern Europe, the overall demand for copper plumbing tube is growing. Furthermore, market information is that those manufacturers producing engineering type tubes are very busy and that most factories are full. Under these circumstances the reticence of KME to apply the laws of supply and demand to pricing is difficult to understand." (691)

(389) According to an internal note of Wieland, the "EDWD-Arbeitsgruppe" met on 28 March 2000 in Brussels ("Sheraton Airport"). The meeting was organised by Wieland. Participants included at least two employees of Wieland, two of OTK, one of Buntmetall (692) and one of KME (693). The agenda of an IMI employee suggests his Participation (694). BCZ confirmed its participation (695).

(390) OTK pointed out that a meeting was held in Zurich in March or April 2000. Participants included KME, BCZ, IMI, Wieland (696) and OTK (697).

(391) On 2 May 2000 (11:30), KME met with (698) OTK (699).

(392) From 14 until 17 May 2000 an IWCC "Joint Meeting" was organised in Toronto. Besides a number of other persons, participants at the ECI Board meeting included KME, Wieland, OTK, Halcor and BCZ (700). In preparation of the IWCC meeting, Wieland wrote an internal note: "The profit situation is still unsatisfactory. In the context of the taking place of the Joint IWCC meeting in Toronto mid May, I would like to ask you to update the European situation." (701). The conclusion has to be drawn that, on the occasion of the IWCC meeting, discussions with the goal to increase prices took place among competitors.

(393) On 15 May 2000, an EDWD meeting on the operational level took place in Amsterdam Schiphol Airport ("Skyport Corporate Meeting Center"). Participants included Wieland, KME and Buntmetall (702). KME and OTK confirmed their participation (703). The agenda of an IMI employee suggests his participation (704).

(394) An internal memo "Sanitary Tubes Europe - Market Development on the Markets D, GB, F, E and NL" ("Kupfer-Installationsrohre Europa - Marktentwicklung auf den Märkten D, GB, F, E und NL") of Wieland dated 18 May 2000 summarises content and consequences of the meeting of 15 May 2000 (705).

"It remains our objective to increase our earnings level on these markets to the target level of DEM 200,- /kg. This objective has absolute priority over the attempt to bring about market share movements in our favour...

Germany: ... At the end of April we published the new price lists for SANCO(r) and WICU(r) tubes, but at the same time we noted that our competitors had increased their rebates. According to our estimates, the rebate situation is currently as follows:

SANCO(r) domestic - max. 58,5%, SANCO(r) export 61,7%, RAL (706) (DVGW) 62-63%. In addition we shall try in the next few weeks to manage the price indices in such a way that we gradually get closer to the target processing earnings of DEM 200.-- /kg...

United Kingdom: ... On the existing List 11, rebates of up to 50 % are, according to information in our possession, being granted on imported tubes. We had agreed internally on a maximum rebate of 40,5%. There is therefore a considerable discrepancy here..." (707)

(395) Attached to the memo are sales volume charts of 1999 and 2000 with target and sales volumes for KME, Wieland, OTK, IMI, and BCZ (708).

(396) BCZ indicated that an EDWD meeting occurred in Amsterdam on 26 May 2000 (709).

(397) The agenda of an IMI employee contains an entry that states "elephants" for 21 June 2000 (710).

(398) On 3 July 2000, Wieland wrote an internal memo containing 1997 market shares of KME, IMI, OTK and Wieland in France, Germany, the Netherlands, Spain and the United Kingdom (711). A separate line of the table compares the total demand figures for 1997 to those provided by the IWCC. Under "further steps", Wieland lists "quantity and price targets in the five main markets on the basis of the situation in 1997", "observation of competitor behaviour, especially those of outsiders, and determination of the way to proceed", "verification and new determination of action against a background of rapid price adaptation to the top", "development of a European price list" and "price target from January 2001: 100 Euro" (712). According to Wieland, this document constitutes an internal document that served as a basis for strategic planning and does not summarise any agreements. Whether it served as a preparation for a meeting could not be recalled (713).

(399) On 19 July 2000, an EDWD meeting took place in Zurich ("Airport Forum", Room 202 "Eduard Amstutz"). The minutes list as participants OTK, IMI, BCZ, Buntmetall, KME, Wieland, EM and KME (714). Wieland and KME confirmed the meeting (715); OTK indicated that a meeting occurred on 17 July (716). On 21 July 2000, KME sent a round-fax to OTK, IMI, Wieland and BCZ concerning the "EU Water Directive" (717).

(400) From (at least) 2 August 2000 to (at least) 9 January 2001, TMX informed HME, OTK, BCZ, Wieland, Desnoyers, Wednesbury, Buntmetall and IMI about their price increases in the French market (718).

(401) It appears that around 17 August 2000, German sales volume information was exchanged between KME, Wieland (and Buntmetall) and (possibly) IMI concerning WICU, CUBO, Eurosan, TALOS, Hetcu and Kuterlex tubes on a table. The origin (and possible destination) of the table is unknown (719).

(402) According to a confirmation circulated by BCZ on 8 August 2000, a EDWD took place on 31 August 2000 at Zurich Airport ("Unique Conference Center", Room 202 "Eduard Amstutz"). A dinner was organised by KME the day before in the Dolder Grand Hotel, Zurich (720). KME (721), IMI, Wieland, Buntmetall, IMI, BCZ and OTK attended (722). Wieland confirmed that the meeting was a sales representative meeting (723), KME that the meeting took place (724), and OTK explained that the meeting was a joint top and operative level meeting (725). According to OTK, the meeting led to a decision "to re-organize commercial co-operation and to establish the Steering Group and the Working Group" (726).

(403) Wieland possessed a table dated 31 August 2000 containing rebate structures of France, Germany, the Netherlands, Spain and the United Kingdom for SANCO, SANCO imports, Imports, OTK tubes, United Kingdom tubes and United Kingdom imports (727) with the following comments:

"Germany: Valid as from 4'9.2000; Spain: As from 5'9.2000 new price lists accept no new fixings; France: rebates are being adhered to. In order to attain objective, index is being adjusted. New list in process of being drawn up. United Kingdom: With List 5, objective not attained as rebates again extended 104 L/100m; as from 31.8.00 new List 6 reb. see left. Appl. Until 18'9.00; as from 29'9.00 new List 7; RW fixings max. 10 days; rebates w. up to e. 00 unchanged." (728)

(404) On 4 September 2000, TMX reported to KME about discussions with Mueller about the introduction of the half-hard tube into the French market (729).

(405) According to Wieland, on 29 September 2000, an operative level meeting was held in Zurich Airport ("Unique Conference Center"). OTK, Wieland, IMI, EM and KME participated. (730)KME, IMI, and EM met the night before at the Mövenpick Hotel to prepare the meeting. (731) KME and BCZ confirmed its participation (732). OTK circulated the minutes (733).

(406) The information exchange with respect to sales volume data for the five Member States France, Germany, the Netherlands, Spain and the United Kingdom is evidenced by an internal table summarising KME's sales volume: The phone number of OTK was noted on the table with the instructions to provide only the total volume figures (and not the internal breakdown by subsidiaries) (734). KME possessed a table, containing "real" market shares of 1997-2000 and "target" market shares for 1997- 2000. A separate table is labelled "Shipments of 5 into 5 markets" (735).

(407) In an internal memo dated 9 October 2000, an employee of Wieland summarised a "meeting of 02.10.00" (736).

"Germany: ... Through indexing alone the price targets set (for SANCO(r) Germany > DEM 200,-- /kg) probably cannot be achieved. A new price list will therefore follow at the end of the year.

The current rebate structure will be substantially changed by this price list. The maximum rebates will then be around 30 %.

The present rebate level (around 60 %) places SANCO(r) at a disadvantage compared with EURO-SANCO(r) and RAL/DVGW, as it leads to a price spread which the market is less and less prepared to pay for the SANCO(r) brand.

Such a price list must be prepared by the end of October...

Spain: On 5'9.2000 Outokumpu introduced a new price list on the Spanish market. We have followed this new price list and the amended rebate structure. As far as we can see, our competitors have done the same. Depending on the marketing stage, rebates now lie between 46 and 52%... Bonuses are no longer being granted...

United Kingdom: In the UK, List 7 was introduced on the market at the end of September. List 8 will likely be published in October. According to information in our possession, the competition, apart from Mueller, has followed this development. Mueller seems to be allowing itself more time." (737)

(408) Attached to the memo are detailed sales volume lists containing "real volumes" and "target volumes" for the years 1997, 1998, 1999 and 2000 (1. Quarter) for five Member States (France, Germany, the Netherlands, Spain and the United Kingdom) (738).

(409) An EDWD meeting (operative level) was held on 16 October 2000 in Zurich Airport ("Unique Conference Center"). The minutes were circulated by KME to Wieland, EM, OTK, BCZ and IMI (739). OTK, Wieland, KME and BCZ confirmed this meeting (740).

(410) On 17 October 2000, a representative from Wieland attended a "Water Directive" meeting in London (741). Agendas of KME and IMI suggest that employees of both companies participated (742). It appears that also OTK participated (743).

(411) On 19 October 2000, KME sent a letter to IMI concerning the complaint of the European Financial Management Association (EFMA) to the German standardisation committee against DIN 50930-6. DIN 50930-6 is a norm that defines the scope of use of copper tubes for drinking water. KME notices that the complaint of EFMA, in which IMI was involved, "is essentially controversial to the interests of the German copper fabricators." "With this I consider the never written but always respected agreement, to let the local market leader determine the policy on his own market, to be not existing anymore..." (744).

(412) An internal memo of Wieland dated 20 October 2000 summarises the results of one of the meetings (most likely the one of 16 October) (745).

"Spain: We have been able over the past few weeks to get much closer to our price targets in Spain. The maximum rebate is 52 %. We are still proceeding on the assumption that at the end of the year we shall achieve the reference figure of 100 euro. Obviously it is not yet clear what the market leader intends doing with regard to the introduction of half-hard products...

France: AFNOR is currently occupying itself with the half-hard grade. Talks are planned with the French producers; also with Müller, of whom we know that under no circumstances one is able and willing to offer half-hard. The reference price is currently around 90 euro (rebate max. 66 %), so here too the chances of achieving 100 euro by year's end are good. At the same time we must try - and here we must follow the market leader - to harmonise measurements (14x1/16x1).

United Kingdom: the 100 Euro mark has already been reached.

Germany: The reference price for non-SANCO tubes is DEM 153,- ; for SANCO tubes, DEM 200,- has already been reached. We shall try to raise the price level by a further DEM 50,-/100kg via the price indices as well as through a new price list. In this connection, the rebate structures must be reshaped (reduced). The rebate levels for resellers must also be redefined. The difference between SANCO and EUROSANCO should be 2%.

Netherlands: We are proceeding on the assumption that HME is introducing a new price list - probably in early December. If this proves to be the case, we shall apply this to deliveries as from 1.1.2001. The current reference price level is 80 Euro.

Other markets (Ireland, Poland, Austria, Belgium): We should determine price policy in these markets internally, especially as far as Austria and Poland are concerned.

Reports to World Bureau: We intend to continue reporting figures." (746)

(413) In a fax dated 19 October 2000, KME instructs TMX "to prepare a new price list with a new structure and considering a rebate of ~ [...] %. The price list should be introduced asap. In this context please think of how far we could integrate a harmonisation of the dimensions 14, 15 and 16 mm into this price list. At least we should point out in the price list that the dimensions 14 + 16 mm are discontinued." (747)

(414) An EDWD meeting was held on 15 November 2000 in Zurich. Participants included representatives from Wieland and KME (748), and, according to addresses of the fax with which the minutes were circulated, also OTK, BCZ, IMI and EM (749). KME, Wieland and BCZ confirmed this meeting (750).

(415) On 27 November 2000 representatives of IMI visited Wieland (751).

(416) On 28 November 2000, IMI's copper tube business received advice from IMI's Company Secretary/Solicitor:

"In the normal course of business an exchange of confidential information between two competitors is likely to be regarded as evidence of co-ordinated behaviour, such as, market sharing or price fixing. Such co-ordinated behaviour is probably an infringement of competition law and information exchange in that context would therefore be unlawful. The exchange of specific pricing and customer information is particularly difficult to justify on any legitimate basis and should normally be regarded as unlawful.. (752)

(417) OTK stated that an operative level EDWD meeting took place on 11 December 2000 in Zurich. Participants included representatives from KME, Wieland, IMI, BCZ and OTK (753). Wieland, KME and BCZ confirmed their participation (754).

(418) According to a fax sent by IMI on 4 December 2000 to IMI, BCZ, KME, EM, OTK and Wieland, a "EU Water Directive Meeting" dinner was organised on 19 December 2000 in the restaurant "Altes Zollhaus Zons", Düsseldorf (Zons) (755). The following day, 20 December, continued with a "EU Water Directive Meeting" in Düsseldorf (Arabella Sheraton Hotel) organised by IMI and/or KME. Participants at the meetings of 19 and 20 December included representatives from KME, Wieland, IMI, BCZ and OTK (756). OTK indicated that the meeting was a top level meeting, while Wieland explained that there was an operative and top level meeting (757). BCZ confirmed its participation (758).

(419) According to KME's summary "New pricelist 2001 - introduction into the market", KME planned to publish its new price list mid-December, Wieland before Christmas, and "others" beginning of January (at the latest 05.01.2001). The new price lists would be applied "all together 15.01.2001". No price line reductions would be granted. The differentiation between German SANCO and RAL (759)/EURO-SANCO would be at a maximum of 2% (= DEM 33,-/100 kg) (760). The new price list "must be applied in a way that all parts of it (del credere, bonus etc") are included. As KME, non-SANCO and EURO-SANCO producers work with a rebate structure.. Standard rebates and highest rebates were to be applied according to different quantities. Non-SANCO producers would not increase the highest rebate of 40 %. BCZ and TMX would keep to a rebate being between German SANCO tube and RAL tubes. Differences between German SANCO and non-SANCO had to be reduced to max. DEM 20,-. In Germany, rebate structures and existing customers' relationships had to be respected (761). As of 1 January 2001, rebates would be reduced from 59%-63% to 35-40 % for non- SANCO producers, from 58,5%-62,5% to 34,5%-38,5% for EURO SANCO producers, and from 57%-61% to 33-38% for German SANCO producers. (762) After these price increases, and by applying different price lines (763), the "added value" would develop between 23 August 2000 and end December 2000 (with the application of the new price list) as follows: From 186,-/100kg to 272,-/100 in the category "59%", and from 126,-/100kg to 200,-/100kg in the category "63%" (764). A table including also 24 January 2001 suggests that the "added value" was further raised in 2001 (765).

(420) An internal memo of Wieland dated 8 January 2001 summarises a meeting, likely the one of 19/20 December (766).

"Germany: Our new price lists were published before Christmas. They will be applied in January. With them we shall achieve the target of 100 for the standard measurement. The spread between SANCO and RAL tubes is still too wide, however, and as a result there continues to be a risk of price destabilisation in the market.

France: In this market the earnings target of 100 has been achieved.

United Kingdom: 100 was achieved in the market, but we are obtaining no supplement for half-hard finishing. Netherlands: - The price level is unsatisfactory. Dealers. rebate structures are uncoordinated, with the result that we are well off target in Holland at EUR 75.

Europe list: There is an urgent need to standardise structures in Europe at the time of the next price lists. This will include also uniform payment objectives and uniform metal rules..." (767)

9.2.13. 2001

(421) KME had tables containing a detailed breakdown of capacity for 2001 of each competitor in the major markets of the world ("D", "I", "F", "U.K.", "B", "NL", "A", "GR", "Spain", "SF", "S", [...]") (768). The capacity figures are broken down by capacity for plumbing and for industrial tubes.

(422) Despite Mueller's withdrawal from the EDWD meetings and its termination of its participation in the WBMS data exchange system (see recitals (223), (308), (344) and (349)) in 1999, representatives of Mueller confirmed on 25 January 2001 that certain "discussions on prices had taken place up to the current week" (769).

(423) According to an internal note of IMI and KME (770) and as indicated by OTK, an operative level (EDWD) meeting occurred on 29 January 2001 in Zurich ("Airport Forum", room 201, "Jacob Ackeret"). Participants included participants from OTK, KME, Wieland, IMI, and BCZ (771). KME and BCZ confirmed this meeting (772). An internal memorandum from OTK contains some details about planned or ongoing arrangements such as, concerning Spain, "Proposal: from everybody 30 % less volume in March-April" (773).

(424) According to an invitation and confirmed by OTK, a EDWD occurred on 27 February 2001 in Düsseldorf (Arabella Sheraton Airport Hotel, room "Montgolfier") (774). OTK, KME, IMI and BCZ participated (775). The planned subject of the meeting was a joint discussion on introducing a thinner wall half-hard copper plumbing tube (776). The initial proposal to set up a technical working group, where every company would share its technical and market knowledge, was formulated by OTK (777). KME and BCZ confirmed this meeting (778).

(425) Wieland indicated that on 5 March 2001, an operative level meeting was held in Zurich. Participants included employees of Wieland and KME (779).

(426) On 12 March 2001, shortly after the meeting, Wieland summarised in an internal memo (780) that "the main problem in the German market is now no longer the quality of earnings but the downward trend in quantities." "The new price list introduced in Spain at the beginning of September is taking hold only slowly, but with it processing earnings of EUR 100 are being attained with a 50 % rebate. In the United Kingdom, List 02/01 is, according to market observations, now being universally applied." .The aim in France is to introduce a new price list in the course of the year". "The Dutch market continues to cause problems" (781). Concerning the switch to "half-hard tubes", neither Desnoyers nor BCZ will be able to follow KME and OTK so that both companies will face problems at the beginning (782). Attached to the memo were tables containing sales volumes ("old" and "new"), market shares and a "deviation" figure of KME, Wieland, OTK, IMI and BCZ for Germany, France, Spain, the Netherlands, the United Kingdom and Poland for 1999 and from 01/2000 for each separate month until 01/2001 (783). The Commission concludes that these tables originate from the meeting of 5 March 2001.

(427) After an ECI meeting in the Kempinski Hotel Gravenbruch in Frankfurt on 14 March 2001 (784), an "elephants" (EDWD) meeting was held at Frankfurt Airport ("Airport Conference Center", Room No 2). (785) The room was reserved for 8 persons (786). OTK confirmed the participation of two KME employees, two Wieland employees, one IMI employee, one BCZ employee and two OTK employees (787). Wieland and BCZ confirmed their participation (788).

(428) The next EDWD top level ("Steering Group") meeting was planned for 27 March 2001 in Frankfurt (789). Notes found at OTK indicate that a European price list should be ready "in March". Targets concerning this price list were "payments", "tariff structure", "bonus structure" and apparently to be applied on 1 January 2002. Further targets were the "consolidation of the targets (the United Kingdom)" and "market stability" (790).

(429) On 30 March 2001, OTK Copper Tubes sent a fax to Mueller Europe SA indicating for Spain the change of the index effective as of the same day (791).

(430) According to diary entries of IMI, Elephant meetings were at least planned on 4 May 2001 in Hamburg, (792) 15 June 2001 in Düsseldorf and 16 October in Munich (793). OTK (794), KME (795) and Wieland (796) submit that these meetings did not take place, and that EDWD meetings were discontinued in March 2001, following the Commission's on-the-spot investigations (797). The Commission has no evidence that any of these meetings took place.

E. - APPLICATION OF ARTICLE 81(1) OF THE TREATY AND ARTICLE 53(1) EEA

10. ARTICLE 81(1) OF THE TREATY AND ARTICLE 53(1) OF THE EEA AGREEMENT

10.1. Applicability

(431) Article 81(1) of the Treaty prohibits as incompatible with the common market all agreements between undertakings, decisions by associations of undertakings or concerted practices which may affect trade between Member States and which have as their object or effect the prevention, restriction or distortion of competition within the common market, and in particular those which directly or indirectly fix purchase or selling prices or any other trading conditions, limit or control production and markets, or share markets or sources of supply.

(432) Article 53(1) of the EEA Agreement contains a similar prohibition. However, the reference in Article 81(1) of the Treaty to "trade between Member States" is replaced in the EEA Agreement by a reference to "trade between Contracting Parties" and the reference to competition "within the common market" is replaced by a reference to competition "within the territory covered by the - [EEA] agreement".

(433) The EEA Agreement came into force on 1 January 1994. Insofar as the arrangements prior to that date restricted competition in Austria, Finland, Iceland, Liechtenstein, Norway or Sweden (then EFTA Member States) they did not violate Article 53(1) of the EEA Agreement. For the period preceding that date, the only provision applicable to these proceedings is Article 81 of the Treaty.

(434) After the accession of Austria, Finland and Sweden to the Community on 1 January 1995, Article 81(1) of the Treaty became applicable to the arrangements insofar as they affected competition in those markets. The operation of the arrangements in Norway and Iceland remained in violation of Article 53(1) of the EEA Agreement. In practice, it follows that insofar as the cartel agreements operated in Austria, Finland, Norway, Sweden and Iceland they constituted a violation of the EEA and/or Community competition rules as from 1 January 1994.

10.2. Jurisdiction

(435) The Commission is the competent authority to apply both Article 81(1) of the Treaty and Article 53(1) of the EEA agreement on the basis of Article 56 of the EEA Agreement. In this case the turnover of the parties achieved in the territory of the EFTA states is less than 33% of their turnover in the EEA, and the primary effects of the arrangements in question are on trade between Member States and on competition in the Community. The effect on trade between Member States was shown in recital (26).

11. THE NATURE OF THE INFRINGEMENT

11.1. Agreements and concerted practices

(436) An agreement for the purposes of Article 81(1) of the Treaty can be said to exist when the parties, expressly or implicitly, jointly adopt a plan determining the lines of their respective action (or abstention) on the market (798). It does not have to be made in writing; no formalities are necessary, and no contractual sanctions or enforcement measures are required. The agreement may be express or implicit in the behaviour of the parties, since a line of conduct may be evidence of an agreement. If an undertaking is present at meetings that have a manifestly anti-competitive purpose, unless it publicly distances itself from what is agreed it will be considered to be a party even if it does not in fact abide by the outcome of the meetings (799). Furthermore, it is not necessary, in order for there to be an infringement of Article 81(1) of the Treaty, for the participants to have agreed in advance upon a comprehensive common plan. The concept of agreement in Article 81(1) of the Treaty may apply to the inchoate understandings and partial and conditional agreements in the bargaining process which lead up to the definitive agreement.

(437) An agreement for the purposes of Article 81(1) of the Treaty does not require the same certainty as would be necessary for the enforcement of a commercial contract at civil law. Moreover, in the case of a complex cartel of long duration, the term .agreement. can properly be applied not only to any overall plan or to the terms expressly agreed upon but also to the implementation of what has been agreed on the basis of the same mechanisms and in pursuance of the same common purpose. As the Court of Justice, upholding the judgement of the Court of First Instance, has pointed out in Case C-49-92P Commission v Anic Partecipazioni SpA (800) infringement of Article 81(1) of the Treaty may result not only from an isolated act but also from a series of acts or from continuous conduct. That interpretation cannot be challenged on the ground that one or several elements of that series of acts or continuous conduct could also constitute in themselves an infringement of that article.

(438) Although Article 81 of the Treaty and Article 53 of the EEA Agreement draw a distinction between the concept of "concerted practice" and that of "agreements between undertakings", the objective is to bring within the prohibition of those Articles a form of co-ordination between undertakings by which, without having reached the stage where an agreement properly so-called has been concluded, they knowingly substitute practical cooperation between them for the risks of competition (801).

(439) The criteria of coordination and cooperation laid down by the case law of the Court, far from requiring the elaboration of an actual plan, must be understood in the light of the concept inherent in the provisions of the Treaty relating to competition, according to which each economic operator must determine independently the commercial policy which it intends to adopt in the common market. Although that requirement of independence does not deprive economic operators of the right to adapt themselves intelligently to the existing or anticipated conduct of their competitors, it strictly precludes any direct or indirect contact between such operators the object or effect of which is either to influence the conduct on the market of an actual or potential competitor or to disclose to such a competitor the course of conduct which they themselves have decided to adopt or contemplate adopting on the market (802).

(440) Although in terms of Article 81(1) of the Treaty the concept of a concerted practice requires not only concertation 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 a concertation and remaining active on 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 caught by Article 81(1) of the Treaty even in the absence of anti-competitive effects on the market (803).

(441) Moreover, it is established case law that the exchange, between undertakings, in pursuance of a cartel falling under Article 81(1) of the Treaty, of information concerning their respective deliveries, which not only covers deliveries already made but is intended to facilitate constant monitoring of current deliveries in order to ensure that the cartel is sufficiently effective, constitutes a concerted practice within the meaning of that article (804).

(442) However, in the case of a complex infringement of long duration, it is not necessary for the Commission to characterise the conduct as exclusively one or other of these forms of illegal behaviour. The concepts of agreement and concerted practice are fluid and may overlap. Indeed, it may not even be possible to make such a distinction, as an infringement may present simultaneously the characteristics of each form of prohibited conduct, while when considered in isolation some of its manifestations could accurately be described as one rather than the other. It would however be artificial analytically to sub-divide what is clearly a continuing common enterprise having one and the same overall objective into several different forms of infringement.

(443) In its PVC II judgement, the Court of First Instance confirmed that "[i]n the context of a complex infringement which involves many producers seeking over a number of years to regulate the market between them, the Commission cannot be expected to classify the infringement precisely, for each undertaking and for any given moment, as in any event both those forms of infringement are covered by Article [81] of the Treaty" (805).

11.2. Single, complex and continuous infringement

(444) A complex cartel may properly be viewed as a single continuous infringement for the time frame in which it existed. The agreement may well be varied from time to time, or its mechanisms adapted or strengthened to take account of new developments. The validity of this assessment is not affected by the possibility that one or more elements of a series of actions or of a continuous course of conduct could individually and in themselves constitute a violation of Article 81(1) of the Treaty.

(445) Although a cartel is a joint enterprise, each participant in the agreement may play its own particular role. One or more may exercise a dominant role as ringleader(s). Internal conflicts and rivalries or cheating may occur, which will not, however, prevent the arrangement from constituting an agreement or a concerted practice for the purposes of Article 81(1) of the Treaty where there is a single common and continuing objective.

(446) The mere fact that each participant in a cartel may play the role which is appropriate to its own specific circumstances does not exclude its responsibility for the infringement as a whole, including acts committed by other participants but which share the same unlawful purpose and the same anti-competitive effect. An undertaking which takes part in the common unlawful enterprise by actions which contribute to the realisation of the shared objective is equally responsible, for the whole period of its adherence to the common scheme, for the acts of the other participants pursuant to the same infringement. This is certainly the case where it is established that the undertaking in question was aware of the offending conduct of other participants or could have reasonably foreseen it and was prepared to take the risk (806).

(447) In fact, as the Court of Justice stated in its judgement in Case C-49-92P Commission v Anic Partecipazioni (807), the agreements and concerted practices referred to in Article 81(1) of the Treaty necessarily result from collaboration by several undertakings, who are all co-perpetrators of the infringement but whose participation can take different forms according, in particular, to the characteristics of the market concerned and the position of each undertaking on that market, the aims pursued and the means of implementation chosen or envisaged. It follows that infringement of that article may result not only from an isolated act but also from a series of acts or from a continuous conduct. That interpretation cannot be challenged on the ground that one or several elements of that series of acts or continuous conduct could also constitute in themselves an infringement of Article 81 of the Treaty (808).

(448) The fact that each undertaking takes part in the infringement in ways particular to it does not suffice to exclude its responsibility for the entire infringement of Article 81(1) of the Treaty, including conduct put into effect by other participating undertakings but sharing the same anti-competitive object or effect. Such a circumstance may nevertheless be taken into account when assessing the seriousness of the infringement which it is found to have committed. Such a conclusion is not at odds with the principle that responsibility for such infringements is personal in nature, nor does it neglect individual analysis of the evidence adduced, in disregard of the applicable rules of evidence, or infringe the rights of defence of the undertakings involved.

11.3. Nature of the infringement in this case

11.3.1. Agreement and concerted practices (complex infringement)

(449) The facts described under Part D of this Decision demonstrate that (i) KM(E) and Boliden during the period from June 1988 (first documented joint meeting see recital (163)), and (ii) TMX and EM (from 1995: the KM(E)-group), IMI, Outokumpu and Wieland from September 1989 (first documented joint meeting of the main copper plumbing tube competitors, see recitals (196)-(198), (236)) until March 2001 (for the last documented meeting see recitals (427)-(430)) made the following arrangements with regard to the European market of plain copper plumbing tubes (as far as Wieland and KM(E) are concerned, at least since 1991, also plastic-coated copper plumbing tubes were concerned, see recitals (149), (159)-(162)):

- stabilised their market shares by allocating sales volumes by country (see, for example, recitals (137)-(138), (158), (199)-(226), (236)-(245), (265), (273)-(274), (282)-(284), (290), (298), (300)-(302), (306)-(308), (325), (326), (329), (335)-(338), (345), (350), (353), (356), (394), (412), (420) and (426));

- agreed upon price increases or co-ordinated prices (see, for example, recitals (130)-(135), (180), (181), (190), (199)- (226), (236)-(245), (249), (253), (265), (273)-(274), (277), (281), (290), (298), (300), (321), (326), (335), (336), (338), (345), (350), (366), (378), (394), (403), (407), (412), (413), (419), (420), (426)) and implemented these (see, for example, recitals (277), (412), (413));

- ensured implementation of the market allocation and price agreements/coordination by a monitoring system consisting of a market leader arrangement for various European territories (see, for example, recitals (199)-(225), (245)), as well as of the regular exchange of confidential information on commercial strategies, sales volumes and targets, and occasionally prices and rebates (see, for example, recitals (139)-(146), (158), (199)-(225), (236)-(244), (249), (265), (301), (302), (306)-(308), (326), (335), (336), (338)).

(450) The facts described in Part D of this Decision also demonstrate that (i) during the period from October 1997 until January 2001 (the month of its withdrawal from cooperation in the plumbing tubes market, see recital (422)) Mueller (Desnoyers/Mueller SA and Wednesbury/Mueller Ltd"), (ii) from August 1998 until March 2001 BMA and HME, and (iii) from August 1998 until at least beginning of September 1999 Halcor - together with Boliden, IMI, Outokumpu, Wieland and the KME group - made the following arrangements with regard to the European market of plain copper plumbing tubes:

- stabilised their market shares by allocating sales volumes by country (see, for example, recitals (199)- (226), (325), (326), (329), (335)-(338), (345));

- agreed upon price increases or co-ordinated prices (see, for example, recitals (335), (336), (338));

- ensured implementation of the market allocation and price agreements/coordination by a monitoring system consisting of a market leader arrangement for various European territories, as well as of the regular exchange of confidential information on commercial strategies, sales volumes and targets, and occasionally on prices and rebates (see, as far as relevant also for this period of infringement the references in recital (449)).

(451) The overall scheme qualifies as an "agreement" between undertakings within the meaning of Article 81 of the Treaty in the sense that in their meetings the undertakings concerned expressed their joint intention to conduct themselves on the market in a specific way, as the evidence set out in Part D demonstrates, and with the object to refrain from competition within the concerned territories (see, for example, recitals (199), (237)-(245), (335)). It is therefore an infringement by object (see recitals (498)-(510)).

(452) Furthermore, even though it is not necessary to show that the participants had agreed in advance upon a comprehensive common plan, the description of the overall scheme, to take just a few examples, in recitals (199)-(226), (236)-(245) and (335)-(340), demonstrates that the participating parties of the described meetings agreed upon such a comprehensive plan. Sales volume allocation as well as price cooperation, together with a monitoring system to ensure compliance with the common rules, were all part of this overall plan. Agreements were implemented (see, for example, recitals (277), (412), (413)) and anti-competitive price raises led, at least for certain companies and time periods, to substantial .additional profits. to the detriment of the customers (see recitals (180), (181), (213), (277), (280), (321), (335), (338), (350), (366), (378), (394), (407), (412), (419), (420), (426), and Section 20.3.1.2). The common aim was the control of the European market for (plain) copper plumbing tubes. Indeed, this was explicitly expressed on several occasions (see, for example, recitals (236), (237), (242), (336)). With respect to price agreements, KME argues that "the actual decision to [adapt prices ...] was taken independently by each producer" (809). The Commission considers that the fact that each producer independently chose to respect price agreements does not prevent the behaviour in question from being qualified as agreement.

(453) The term "agreement" applies not only to the overall plan but also to the implementation of what had been agreed in pursuance of the same common purpose of controlling the market. One of the actions taken to implement this overall plan was the appointment of market leaders who monitored national markets and informed the other cartel members of the evolution within their respective territories. Moreover, among other things, the parties created spreadsheets in order to facilitate dissemination of the market information (see, for example, recitals (199), (237)-(245)), which enabled them to review the implementation of the volume targets (see the references in recitals (449) and (450)).

(454) Some factual elements of the illicit arrangement could also aptly be characterised as a concerted practice with the object of controlling volumes and prices. Whilst there clearly existed an agreement behind the action taken to ensure implementation through the market leader arrangement and the exchange of confidential information, as described, for example, in recitals (199)-(226), (236)-(245) and (335)-(340), the operation of this arrangement through the actual regular exchange of the sales volumes and the presentation or exchange of price information between the undertakings, namely concerning rises of price lines as well as the applicable discount scheme, could also be regarded as adherence to a concerted practice to facilitate the coordination of the parties' commercial behaviour. Through this, the producers in question were able to monitor the current prices and market shares. Given the existence of this sophisticated and institutionalised system of information exchange and coordination, the Commission has to conclude that the parties could not continue to operate on the market without using the knowledge and contents of the information exchanged (with respect to the implementation see, for example, the references in recitals (449) and (450)).

(455) In view of their identical purpose, the various agreements and concerted practices formed part of a scheme of fixing market shares (in order to raise or maintain prices (810) above the competitive level) and, during certain periods, prices or target prices and monitoring compliance through competitor meetings and telephone contacts as well as through exchanges of faxes and electronic data. This scheme was part of a series of efforts made by the undertakings in question, in pursuit of a single economic aim, namely to avoid competition and, as a consequence, the normal movement of prices on the market. The Commission understands that the use of notions such as "confirmation", "revision" or "modification", in some documents (see, for example, recital (335)) demonstrates a series of contacts and refers to previously established arrangements. It would thus be artificial to split up such continuous conduct, characterised by a single purpose. The fact is that the participants took part - over a period of over eleven years (BCZ, IMI, KME (including KM(E), TMX and EM), Outokumpu and Wieland) or, respectively, over three years (Mueller, BMA, and HME) or one year (Halcor) - in an integrated scheme constituting a single infringement, which manifested itself in both unlawful agreements and unlawful concerted practices (811).

(456) With respect to the SANCO-, WICU- and Cuprotherm arrangements, the Commission notes that the "licence" cooperation as described in Section 6 included many hardcore restrictions prohibited under Article 4 of the Commission Regulation (EC) No 772-2004 on the application of Article 81(3) of the Treaty to categories of technology transfer agreements (812). Therefore, the cooperation goes clearly beyond restrictions permissible in the framework of technology transfer licence agreements. This applies all the more to the conditions for a possible exemption under the (old) Regulation Commission Regulation (EC) No 240-96 of 31 January 1996, on the application of Article 85 (3) of the Treaty to certain categories of technology transfer agreements (813). The licence cooperation failed to meet certain core conditions of the exemption under the Regulation of 1996 (in particular the number of parties to the licence agreements). In this respect, it has to be stressed that the new Regulation on the application of Article 81(3) of the Treaty to categories of technology transfer agreements is considerably less strict than the one of 1996, which was in force at the time when the infringement was committed.

(457) On the basis of the above considerations, the Commission considers that the complex of behaviours conducted by BCZ, IMI, KME (including KM(E), TMX and EM), Outokumpu, Wieland Werke, BMA, Halcor and HME in this case together with the underlying intention to restrict competition presents all the characteristics of an agreement and a concerted practice in the sense of Article 81 of the Treaty. For the reasons mentioned in recital (455), this is an infringement by object.

11.3.2. A single, complex and multiform infringement

(458) The KME-group (including, before 1995, KM(E), TMX and EM) and Wieland participated in one single (see recital (506)), complex (that is to say comprising agreements and concerted practices, see recitals (451)-(456)) and multiform infringement comprising the following three forms: (i) arrangements among SANCO producers, (ii) the WICU and Cuprotherm arrangements, and (iii) arrangements among the broader group of copper plumbing tube competitors (that is to say, including Outokumpu, IMI, Mueller, BMA, HME and Halcor).

(459) The Commission considers that, as far as KME and Wieland are concerned, the entire product group of (plain and plastic-insulated) copper plumbing tubes was the object of a single complex and multiform infringement. In this Decision, the term .multifom. is used to designate an infringement consisting of three separate but, as will be shown below, interconnected manifestations. These different manifestations concerned two different products, plain and plastic-insulated copper plumbing tubes. In this respect, it is noted that plain copper plumbing tubes and plastic-insulated copper plumbing tubes are not necessarily substitutable and might constitute distinct product markets when assessed under the Commission Notice on the definition of the relevant market for the purposes of Community competition law (814). However, since KME and Wieland coordinated their commercial behaviour for the entire product group often in the same meetings (joint SANCO-, WICU- and Cuprotherm meetings, see, for example, recital (190)) and with the same employees pursuing the sole purpose of avoiding and restricting competition, both products, that is to say plain and plastic-insulated copper plumbing tubes, are considered to be part of one infringement involving one product group. This view is supported by the fact that the arrangements concerning plain copper plumbing tubes necessarily had an effect also on plastic-insulated tubes, because insulators or plumbers that insulate tubes have to first purchase plain tubes produced by copper plumbing tube manufacturers. Neither KME nor Wieland have argued that the above mentioned arrangements should be treated as separate infringements.

(460) BCZ participated in one single (see recital (506)), complex (that is to say comprising agreements and concerted practices, see recitals (451)-(456)) and multiform infringement consisting of two separate but, as will be shown below, interconnected manifestations, that is to say the arrangements among SANCO producers on the one hand, and arrangements among the broader group of copper plumbing tube competitors (that is to say, including Outokumpu, IMI, Mueller, BMA, HME and Halcor) on the other hand. The manifestations were interconnected because they concerned the same product, plain copper plumbing tubes, and all SANCO producers participated at the same time also in the broader European arrangements. From the perspective of the SANCO producers, the cooperation in both manifestations served only one goal, to restrict and control competition with respect to plain copper plumbing tubes. It could not be proven that Boliden was aware of the WICU and Cuprotherm arrangements, or could have reasonably forseen such conduct, and was prepared to accept the risk.

11.3.3. A single and complex infringement

(461) IMI, Outokumpu and, for the period after October 1997, Mueller participated in a single and complex infringement concerning plain copper plumbing tubes (meetings and contacts of the group of five and of the group of nine). It could not be proven that IMI, Outokumpu or Mueller were aware of the SANCO, WICU or Cuprotherm arrangements, or could have reasonably forseen such conduct, and were prepared to accept the risk.

(462) With respect to BMA (before its acquisition by Wieland), HME and Halcor, it is noted that these three companies participated in a single and complex infringement consisting only of meetings and contacts of the .group of nine. concerning plain copper plumbing tubes.

11.3.4. Continuity of the infringement

(463) The Commission considers that this single, complex and, with respect to BCZ, the KME-group and Wieland, multiform infringement was continuous from 3 June 1988 as far as KM(E) and BCZ are concerned (see recital (163)), and from 29 September 1989 as far as TMX and EM (comprising the KME-group as of 1995), IMI, Outokumpu and Wieland are concerned until March 2001.

(464) The Commission has to establish continuity by producing sufficiently precise and coherent proof that the alleged infringement was continuous. For this standard of proof to be met, the Commission can deduct from fragmentary evidence. "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" (815). Also for the proof of continuity, different pieces of evidence are to be interpreted in their overall context. As will be further demonstrated below, the overall scheme of controlling and restricting competition in the copper plumbing tube market persisted and, in particular, the will to do so, as confirmed by Outokumpu. In the light of such a persisting scheme and corresponding will to infringe competition law, it is not sufficient to assume that participants interrupted their infringement if their cooperation became simply less visible or their contacts less numerous for a certain period. For terminating its participation, a party rather has to publicly distance itself from the cartel's activities and, at the same time, to entirely withdraw from cooperation with respect to all of its competitors. A participating company in an infringement has certainly not withdrawn from the infringement if it continues to have bilateral contacts with competitors participating in the cartel, where market conditions, market strategies, volumes, prices or the arrangements of the remaining participants are discussed.

(465) The question of continuity of the infringement only arises with respect to BCZ, IMI, KME (including KM, TMX and EM), Outokumpu and Wieland. As different considerations apply for BCZ, this undertaking will be addressed separately.

(466) BCZ, IMI, KME (including KM, TMX and EM), Outokumpu and Wieland continued (i) the same type of information exchange, coordination and organisation of contacts (on the top- and operational level); (ii) the focus of the infringement remained the same (volume allocation and price co-ordination through coordinating the change of price lines and agreeing on rebates); as far as SANCO producers are concerned, (iii) contacts, information exchange and coordination between the SANCO producers continued without any interruption (the same applies for the WICU- and Cuprotherm arrangements). In addition, there is no indication that any of the parties publicly announced its termination of its participation in the arrangements and/or entirely stopped its cooperation with respect to all parties and all elements of the infringement.

(467) Before the Statement of Objections, only KME contested the conclusion that there had been a continuous infringement. KME alleged a termination between "1990 and December 4, 1992" (816), and between July 1994 and July 1997 "following a failed attempt to revive the arrangements in mid-1996" (817). However, in its reply to the Statement of Objections, KME acknowledged that anti-competitive contacts continued to take place during the period 1990 to 1992.

(468) In its reply to the Statement of Objections, Wieland claims interruptions from the beginning of 1990 until December 1992 (almost three years), and from mid-1994 until spring 1996/97 mostly referring to explanations given by Outokumpu and KME (818). IMI believes it should be treated in the same way as BMA, HME, Halcor and Mueller. (819)

11.3.4.1. Period from beginning of 1990 until December 1992

(469) With respect to the alleged interruption from 1990 until December 1992, only Wieland claims an interruption referring to its internal revision, to a description of the broader European cartel arrangements provided by KME (see recital (203)) and to a quote of an employee of Outokumpu (820) who was interviewed by the Commission (821).

(470) The Commission identified a number of illegal competitor contacts during the period from 1990 until December 1992 (see recitals (241)-(264)). The probative value of this evidence as well as the finding of continuity by the Commission for the period from (beginning of) 1990 until December 1992 were not contested by any other party. In addition, Outokumpu did not identify any interruption in the broader European arrangements during this period. Nor did Outokumpu comment on the SANCO arrangements that continued during this period without interruption, a fact that is not explicitly contested by Wieland. Second, Outokumpu's employee, whom Wieland referred to, indicated in the very same paragraph quoted by Wieland that he "think[s] that somewhere at the end of 1992 the [...] system was stopped, because in 1992 [the information] was still collected [...]", thereby, and also with the remainder of his statement, precisely confirming that the illegal information exchange and cooperation continued from September 1989 until end of 1992, when the [...] system was replaced by discussing IWCC data. (822) The information exchange system that the parties established in September 1989 (the "[...] spreadsheet") was carried on throughout the period (see recital (249)). Third, KME, in its submission of 17 February 2003, had not identified any interruptions concerning the SANCO arrangements, in which also Wieland participated. Last, KME, after having reviewed the facts presented in the Statement of Objections, corrected its statement indicating that there was no interruption during this period.

(471) Therefore, the Commission considers that it has proven the continuity of the infringement in the form of the SANCO arrangements and the broader European arrangements for the period from 1990 until 1992.

11.3.4.2. Period from July 1994 until July 1997

The parties' arguments

(472) KME submits that the Commission, in its assessment of continuity concerning the period from July 1994 until July 1997, should take account of the following elements. First, Outokumpu's employees state that the meetings were discontinued in 1994 (823). Second, Outokumpu stated that in 1996 the re-initiation of meetings was discussed. KME believes that this "appears to indicate that the arrangements had been previously abandoned, which would confirm the recollection of KME's employees" (824). Third, as to the appropriate standard of proof in cases involving the question of continuity of an infringement, KME acknowledges that no document exists in the Commission's file that would manifest KME's intention to terminate the agreement as of July 1994. However, as a general matter, to be obliged to provide a document evidencing the will to terminate an illegal arrangement is essentially an impossible evidentiary burden to meet, and should thus not be the appropriate standard of proof. Where meetings are discontinued by all companies, such an evidentiary requirement would appear unreasonable. Fourth, KME's employees remember that meetings were discontinued. Last, based on the evidence cited in the Statement of Objections, no meetings appear to have taken place at the European level between 8 July 1994 and 1 February 1996 (825).

(473) Wieland alleges an interruption in the period "1994 until 1996/1997" (826). The Commission understands that Wieland means by "1996/1997" an interruption until February 1996, when, according to Wieland, an unlawful meeting occurred (see recital (286)). In support of its allegation, Wieland points to (i) the discontinuation of the [...] information exchange and start of the discussion on the basis of IWCC figures (827), (ii) explanations given by Outokumpu (recital (214)) and one of its employees (828), (iii) explanations provided by KME (829), (iv) an internal Memo dated 11 October 1994, in which Wieland noted that KME was moving away from a joint product and market policy, (830) (v) notions like "reactivation" (paragraph 393 of the Statement of Objections) used by the Commission, which, according to Wieland, imply an interruption (831), (vi) explanations like the construction boom in Eastern Germany that made cooperation unnecessary (832), (vii) Wieland's assumption that the termination of the cooperation must have been communicated among the cartel members (833), (viii) Wieland's view that producers realized in 1996/97 because of the declining demand that price erosion created financial problems and therefore created a new system of market coordination (834).

(474) In IMI's view, the absence of a formal announcement of withdrawal from the European cartel is not relevant "per se". Similarities of object, method and practice do not appear to be of themselves indicators of continuity. With respect to the period from June 1994 to April 1996, IMI could not identify any element pertaining to its participation in the wider European cartel activities or any specific evidence in the Commission's file indicating that meetings in 1995 (see recital (285)) were cartel related (835).

(475) Outokumpu confirmed its statement (836) that there was a .period of lesser intensity. between "August 1994 until around summer 1997" with an "endeavour [in spring of 1996] to reactivate the cooperation" prior to 1994 with three to four meetings (837). In reply to a question raised by the Commission, Outokumpu stressed that although it did not believe that much went on at this time, it could not 'say that nothing went on. (838).With respect to contacts on the occasion of trade association meetings, Outokumpu states that "parties met regularly in trade association meetings, but there are no indications that any unlawful discussions occurred" (839). Outokumpu has no recollection of a meeting during 1995 (840). Mueller's indications that from .at least. 1995, "periodic meetings and phone calls" occurred [information not relevant for the EU proceedings] (for example meetings in the United Kingdom and in France) (841). Outokumpu has no recollection of the meetings referred to in recitals (286)-(288) (842). Outokumpu believes that SANCO meetings appear to have continued (843).

The Commission's view

(476) IMI, KME (including KM, TMX and EM), Outokumpu and Wieland did not terminate the infringement and continued to have illegal contacts, albeit in a less intense form, from July 1994 until July 1997. The infringement was never terminated with regard to all of its elements (see also the considerations in recital (594)). KME (including KM, TMX and EM) and Wieland continued their illegal cooperation in the framework of the SANCO arrangements.

(477) As far as the broader European arrangements are concerned, from July 1994 until July 1997, contacts between the parties took place in a less organised way or in different forms, sometimes only bilateral, amongst certain competitors (for example SANCO producers) and less frequently, but, as will be explained in recital (486), followed the scheme that was developed in September 1989.

(478) First, the Commission considers that the statements of Outokumpu's employees, in particular the one to which KME refers, do not prove any termination. Outokumpu's employee did not have sufficient recollections. Asked for the first time whether there was any specific interruption in the sense that competitors agreed not to meet anymore, for instance in 1994 or 1995, the employee in question answered that "here were disruptions or at least quiet periods" specifying, when asked a second time, that "one problem in those meetings was the fact that everybody knew that the number which had been given by the companies were not the right numbers. That is why there was also quite a lot of distrust between those companies" (844). Having been put the question for the third time, Outokumpu's employee confirmed (referring to the question), he "believe[s] that is the best of my understanding" that there was a meeting where participants declared that they decided to discontinue (845). Outokumpu's employee was neither in a position to provide any specific date nor, without indicating such a date, to describe the content and details of any such meeting. The Commission's view not to consider this statement as a proof for any termination is confirmed by Outokumpu, which itself did not want to interpret the statement of its employee as meaning that all elements of the illegal cooperation were terminated.

(479) Second, with the exception of KME (see recital (482)) and possibly Wieland, none of the participants claimed, nor is there any evidence proving, that any of the participants formally ended its involvement in the cartel or publicly demonstrated its turning away from the arrangements and its willingness to terminate the infringement with regard to all of its elements while rejecting at the same time to engage in any future cartel activities. Mueller, Outokumpu and initially Wieland did not identify any discontinuity of the infringement throughout the relevant period from September 1989 until March 2001. Outokumpu described that there was a "lower intensity period in the cooperation and information exchange" "from around August 1994. until around summer 1997". Mueller, describing [...] European-wide arrangements, did not identify any period of interruption (846). Wieland originally stated that regular meetings had occurred at least since 1993. In its reply to the Statement of Objections, Wieland submits that an interruption occurred and assumes that the termination must have been communicated without substantiating this claim.

(480) Taking into account the sophisticated accomplishment and the institutionalised dimensions of this cartel from September 1989 until at least July 1994, the Commission considers that in any case, any meeting leading to an official termination of the infringement would have been equally and clearly recalled by representatives of Outokumpu and/or other participants.

(481) None of the participants were in a position to report any (specific) meeting during which members, including KME, officially declared their withdrawal from the arrangements. In fact, while Outokumpu was able to reconstruct and comment on the vast majority of the meetings throughout the whole period of the infringement, from September 1989 until 2001, it did not describe or recall any specific meeting that would have lead to any "formal" termination. BCZ confirmed that, at least until mid-1995, there had been no meeting in which any of the parties would have announced its termination and/or terminated its participation. Neither Mueller nor Outokumpu have indicated that they were in any way aware of KM(E)'s (TMX's or EM's) decision to terminate its participation for the future, although they did notice when other participants, like BCZ distanced themselves from the arrangements for a limited period of time. The corresponding statements of Outokumpu, Mueller, initially also of Wieland and Boliden prove that the will to restrict and control competition continued throughout the "lower intensity period of cooperation". The fact that there was no (public) withdrawal of any of the participants and, furthermore, the continued, albeit less frequent, contacts among the competitors confirm their willingness and expectation not to stop their co-ordination.

(482) Third, the Commission considers that the alleged recollection of KME's employees that meetings and contacts were terminated by participants in July 1994 is not credible. The only reported meetings during this period were the ones of 16 June 1994 and 8 July 1994. Accordingly, only during one of these meetings could the parties supposedly have decided to terminate their cooperation. However, even if such termination had been declared or decided, the continued co-ordination after these two dates demonstrates that the participants did not intend to terminate the infringement with respect to all its elements. This conclusion is substantiated by the following elements.

(483) Until November 1994, the parties coordinated and implemented price increases in the United Kingdom-market, which resulted, according to one participant, in the price of copper tubes rising by approximately 60 % overall (see recital (277)). Substantial price increases in the United Kingdom market until the end of 1994 are evidenced by two economic NERA reports submitted by IMI and KME, and, in addition, confirmed by a former executive of the copper plumbing tube industry (see recital (213)). Based on these reports, it appears that price increases had effects far beyond the date of November 1994. In addition, OTK had contacts with Wieland until August 1994 and with EM until December 1994. At least with EM, Outokumpu discussed confidential information. An alleged termination of an agreement at a certain point in time cannot be accepted, when, at a later stage, this very agreement is implemented and the effects remain in force, as is the case in the present case.

(484) As far as KM(E), TMX, EM and Wieland are concerned, there is evidence that the European arrangements continued in the form of the SANCO arrangements between the SANCO producers, and concerning KM(E) and Wieland, the arrangements continued with respect to WICU and Cuprotherm tubes. These producers had a steady and continuous coordination and/or information exchange with respect to volumes and/or prices and/or rebates. For the "quiet period" from July 1994 until July 1997, besides the meeting of 26 September 1995, where it is proven that KME and Wieland discussed sensitive business information (see recital (172) and below), KME lists eleven other SANCO meetings (847) where it was reasonably likely that commercially sensitive information was exchanged, a fact that is not contested by Wieland. Also BCZ, as already explained, confirming the statement of a former executive of the copper plumbing industry, remembers that illegal cooperation, at least among SANCO producers, continued until mid-1995 (when BCZ discontinued its participation in these meetings). With respect to the meeting of 26 September 1995 (the association .Arbeitskreis Kupferrohre"), KME indicates that besides Wieland and KME, two other German companies - MKM and Standardwerke Werl - also participated in the meeting. In KME's view, the meeting was thus unrelated to the European arrangements or .SANCO arrangements. subject of the Statement of Objections (848). The Commission considers that discussions described in recital (172) clearly relate to sensitive business strategies, the discussion of which violates competition rules. The fact that such discussions occurred on the occasion of a working group meeting within an association confirms the statement of the testimony (see recital (285)), that illegal discussions occurred on the occasion of such meetings, in particular during a time period for which KME and Wieland claim that illegal contacts had been terminated. The fact that discussions might have focused on Germany or that additional companies participated, does not affect the liability of KME and Wieland for infringing competition law nor does it exclude considering these discussions as part of the multiform infringement, in which KME and Wieland were involved. In fact, the Commission considers that the illicit cooperation between KME and Wieland concerning SANCO copper plumbing tubes cannot be separated from other parts of the multiform infringement by KME and Wieland.

(485) Fourth, even if almost no agreements were reached and/or implemented with regard to volumes or prices during the period from May 1994 until July 1997, this does not imply that the infringement concerning copper plumbing tubes had been terminated with regard to all of its elements.

(486) There is a clear continuity of method and practice of the cartel scheme throughout the entire period from September 1989 until March 2001. In fact, a series of meetings in 1996 (see recitals (288)-(293)) illustrates that participants seamlessly took up and re-intensified the arrangements of the period from 1989 until July 1994 that, in the end, were successfully continued in July 1997. The 1996/1997 meetings can only be understood as the continuation of the 1994 meetings. Participants did not have to set up a (totally) new form of coordination, but rather improved the system initially established. The main part of the cooperation was the exchange of sales volumes data and, on its basis, the allocation of volume quotas. Volumes were first exchanged through the [...] spreadsheets. From around 1993, IWCC data was used as a basis for agreeing on volume-targets until at least mid 1994. In 1996, certain data was exchanged directly among the parties to re-intensify cartel activities as they had occurred before (see recital (290)). Directly after the start of meetings in July 1997, participants set up the WBMS data exchange system, the use of which can be considered as a more efficient and sophisticated continuation of the system established in September 1989.

(487) When the mutual trust among the participants was finally re-intensified in July 1997 after a number of meetings between April and July 1996, the European copper plumbing tube market continued to be discussed amongst the same undertakings in a similar way by top level meetings that were followed by operational level meetings. A number of top and operational level company representatives participated throughout all three stages of the cartel (849). By 1997, the scope of data exchange was more focused and concentrated on five main countries, with occasional discussions also on other countries. However, the form of the data exchange and the basic principles with regard to fixing and monitoring volumes remained the same. Similarity of methods is a strong indicator for continuity.

(488) The major difference in the operation of the cartel was that, since the meeting in July 1997, the participants had established an efficient information exchange system that proved reliable and ensured the desired results (see recitals (223) and (301)). The reliability of the information exchange after July 1997 is evidenced by contemporaneous statements of participants (see, for example, recitals (326) and (350)). Consequently, the Commission considers that the conduct in question constituted a single continuous infringement of Article 81(1) of the Treaty in which each participant must bear its responsibility for the duration of its adherence to the common scheme.

(489) Furthermore, Mueller, OTK and KME reported national cartels in Germany, France, the United Kingdom, the Netherlands, Spain and Belgium, some of which continued, according to the parties, throughout the "quiet period" (for example, the cartels relating to France, the United Kingdom, Germany and the Netherlands) (850). In this case, a termination of the European-wide cartel activities would have required a complete and definite termination of any cartel activity with respect to copper plumbing tubes involving any of the countries and/or parties that were the subject of, or involved in, the European-wide cartel. No such withdrawal of any of the parties occured.

(490) Finally, with respect to IMI, IMI cannot be treated in the same way as HME, BMA, Halcor or Mueller. Evidence suggests that HME, BMA, Halcor and Mueller did not participate in any meetings and did not have any contacts concerning the European level for a period of more than three years. For this reason, continuity with respect to these companies could not be established. IMI, on the contrary, (i) implemented heavy price increases in the United Kingdom in reaction to cartel agreements until November 1994 (see recital (483)), (ii) participated as one of the core participants in numerous meetings in 1996 taking an active role in negotiating the terms for reintensifying the cartel successfully (see recital (288)-(293)), (iii) took an active role in enlarging the number of participants by inviting and encouraging a number of competitors that had withdrawn from the meetings (specifically Mueller and HME) to participate again in the cooperation in mid 1997 and August 1998 (see recitals (216), (299), (305) and (325)), and (iv) played a crucial role by taking over the role as a market leader for the United Kingdom (see recital (222)). In this sense, the continuous participation of IMI was crucial for the functioning and continuation of the cartel. IMI's core role in the infringement can in no way be compared to the participation of smaller competitors such as HME, BMA, Halcor and also Mueller. Last, IMI did not provide any coherent explanations in which way it supposingly withdrew from the arrangements. For all these reasons, the Commission does not see any basis for treating IMI in the same way as HME, BMA, Halcor and Mueller.

(491) In the light of the explanations given, the Commission considers that the period from July 1994 to July 1997 can be more appropriately viewed as a period of decreased pressure from the market because of the German reunification combined with a crisis resulting from mistrust between the cartel participants, rather than the termination of one infringement and starting of another, as confirmed by Outokumpu (see recitals (214) and (215)). It is natural that discussions over a period longer than eleven years may have involved organisational changes and a shifting constellation of undertakings and participation. Also, there may have been periods of tension that led to a certain increase in competition. Tension inside the cartel might have been caused by the companies submitting information that was considered unreliable by the other members of the cartel. However, in 1997, participants re-organised the information exchange system to render it reliable, this time through the WBMS, and agreed to submit accurate figures.

11.3.4.3. Boliden

(492) The Commission has established that BCZ did not terminate the infringement during the period under investigation (June 1988-March 2001) with regard to all of its elements but continued to participate in illegal arrangements, albeit for a considerable time period in a less intense form.

(493) BCZ submits that meetings and exchange of information within the SANCO Club essentially comprised two types of horizontal issues; a more or less legitimate trademark and patent licence arrangement and an unlawful price and market sharing cartel (851). BCZ's participation in the .legitimate side. of the SANCO arrangements lasted from April 1988 until March 2001 (852). With respect to the illegal activities as they are described in this Decision, BCZ participated from 1988 to 1995. After 1995, Boliden has not been able to verify the existence of, or participation in, SANCO meetings with the exception of one meeting in June 1999 (853). BCZ indicated that as a consequence of settlement negotiations concerning WICU tubes, it met with KME on 21 November 1997 and provided correct sales volumes for WICU and SANCO tubes for the period from 1992 until 1997. According to BCZ, OTK also participated in this meeting. BCZ rejected the joint offer of KME and OTK to participate in EDWD meetings (see recital (310)) (854).

(494) The Commission acknowledges that BCZ did not participate in any meeting at a European level between mid-1995 and August 1998, and from 10 December 1998 until October 1999. It is, however, established that BCZ submitted sales volumes for SANCO tubes to the SANCO secretariat during the entire period from 1988 to 2001 and received the volume sales information of its SANCO competitors. The data exchange, far from being the implementation of a "legitimate" licence agreement, served as a basis for market sharing and volume allocation among SANCO producers, as BCZ knew, and was illegal. In the absence of a termination of Boliden's participation in both parts of the multiform infringement with respect to all of its elements, Boliden's participation has to be viewed as continuous.

11.3.4.4. Halcor

(495) Halcor indicates that for a period of five months, between 29 October 1998 and 19 March 1999, it did not participate in any meetings. After the meetings of 19 March 1999 and 28 April 1999, it did not attend any further meetings (855). While it continued supplying data to the WBMS office until August 1999, Halcor submits that its employees believed that there was no anti-competitive implication in continuing to send their statistics to WBMS for a short time "just to cover their withdrawal from the scheme". Halcor believes that the continued contribution of Halcor to the WBMS was of no assistance to the remaining members. Also, it was of no use for Halcor except for the entirely legitimate purpose of being able to work out its own market share. In Halcor's view, the supply of data to WBMS cannot be considered as continued adherence to a cartel agreement. The presence of the person declaring the data at meetings was essential in order for the scheme to operate as a quota monitoring mechanism. Halcor stresses that on its own and without each contributor attending meetings to justify its sales, the WBMS scheme was entirely in conformity with competition rules (856).

(496) The Commission considers that Halcor continued the illegal cooperation until September 1999, when Halcor decided not to submit data to the WBMS scheme anymore (see recital (376)). The Commission notes that Halcor has not alleged any interruption due to the fact that it did not participate in any meeting for five months (between October 1998 and March 1999). With respect to the WBMS data exchange, it is important to note that Halcor was aware that this information system was set up for illegal purposes, specifically to monitor quota allocations. By submitting its sales volumes, Halcor contributed to the volume allocation scheme and enabled competitors to better monitor the market. Competitors, with the help of Halcor, continued to verify if volume allocations had been respected by subtracting their individual sales figures from aggregated WBMS figures. In order to withdraw from the illegal cooperation, Halcor had to stop its participation in the illegal arrangements entirely. In addition, the relevant date for determining duration of the infringement depends also on the implementation of the agreement and the use of exchanged information in determining one's commercial strategy after a meeting occurred. Halcor's continued supply of sales volumes to the WBMS, can only be understood as meaning that Halcor had not taken a final decision to completely withdraw entirely from the illegal arrangements. Halcor therefore continued to participate in the illegal cooperation, albeit in a minor form. The Commission further notes that the effects of the cartel lasted far beyond Halcor's withdrawal from the cartel at in September 1999. In favour of Halcor, the Commission therefore concludes that Halcor participated in a continuous infringement for the entire period from August 1998 until at least beginning of September 1999 by withdrawing from the arrangements by stopping to supply sales figures in September 1999.

11.3.4.5. Other parties

(497) As far as BMA, HME and Mueller are concerned, this Decision should only be addressed to these parties for their participation in the infringement during the period after October 1997 or August 1998, respectively. The continuity of their participation has not been contested.

12. RESTRICTION OF COMPETITION

12.1. Object

(498) Article 81(1) of the Treaty and Article 53 (1) of the EEA Agreement expressly include as restrictive of competition agreements and concerted practices which (857): - directly or indirectly fix selling prices or any other trading conditions, - limit or control production, markets or technical development, - share markets or sources of supply.

(499) These are the essential characteristics of the horizontal arrangements under consideration in this case. Market sharing and control of production occurred through allocation of market shares, as well as the agreements on market leaders for different European territories (see references in recitals (449) and (450)). These arrangements aimed at freezing and stabilising market shares and sales volumes in respect of these territories, implying that the producers were restricted from competing for market shares. Freezing market shares also enabled the parties to succeed in increasing the market price or preventing it from declining at the pace of the market forces (858). The anti-competitive object is explicitly stated in evidence presented in the factual description (see, for example, recitals (236), (237), (242), (335), (350)).

(500) Specifically, the fixing of a price, even one which merely constitutes a target, affects competition because it enables all the participants in a cartel to predict with a reasonable degree of certainty what the pricing policy pursued by their competitors will be (859). By expressing a common intention to apply a given price level for their products, the producers concerned cease independently determining their policy in the market and thus undermine the concept inherent in the provisions of the Treaty relating to competition (860).

(501) Agreeing upon a rebate structure, target prices and/or price lines, initiating price rises by the market leader, in particular during the periods from 1988 until end of 1994 and from July 1997 until March 2001 (see, for example, recitals (206), (207), (222), (277) and the references in footnote 592) as well as exchanging information on other commercial terms, such as delivery and payment conditions, albeit only occasionally, are examples of the fixing of sales prices and other trading conditions (see references in recitals (449) and (450)). Prices being the main instrument of competition, the various collusive arrangements adopted by the producers had the purpose of inflating prices to their benefit and above a level, which would have been determined by free competition.

(502) Price fixing and market sharing by their very nature restrict competition within the meaning of both Article 81(1) of the Treaty and Article 53 (1) of the EEA Agreement.

(503) The parties took explicit action to conceal their meetings and to avoid detection of their anti-competitive agreements and documents. To this effect, for example, after 1997, they used the "code-name" "EC Drinking Water Directive Meeting" in communications and to set up meetings, drafted misleading minutes that did not reflect the content of the meeting, and never mentioned any of their arrangements in any written communication. Last, after 1997, they established a "legal" data exchange system, the WBMS system, as a basis for their arrangements, and subsequently disclosed the individual sales figures during their meetings.

(504) Regarding the anti-competitive object of the exchange of information, the arrangements have to be seen in the light of all circumstances. They served to attain the single objective of freezing market shares and co-ordinating prices to maintain a price level that was not based on competition and enabled the undertakings to adapt their strategy to the information received from competitors.

(505) The permanent exchange of information was intended to guarantee the stability of the illegal scheme. In the event of a manifest imbalance in market shares or prices, the conflict could be settled through discussion, proposals, persuasion or even pressure (861).

(506) This complex of agreements and concerted practices, as described in Part D of this Decision, had therefore one and the same object: the restriction of competition within the meaning of Article 81(1) of the Treaty and Article 53(1) of the EEA Agreement. Thus, it must be considered as one single infringement.

(507) In this case, while Mueller, Outokumpu, KME, Wieland, Halcor, BCZ and IMI admit the existence of the anticompetitive agreements and practices described in recitals (449) and (450), KME, Outokumpu, IMI, Wieland, BCZ and Halcor have advanced defensive arguments, such as a negligible effect on prices and competition, difficult economic environment (overcapacity), lack of implementation, lack of retaliatory mechanisms, frequent deviation, the buyers' purchasing power, the threat from bigger competitors to participate in the anti-competitive arrangements and/or unawareness that certain elements of the behaviour could be considered illegal (862). The parties have argued that prices eroded and the cartel had no effect on the market. As will be further discussed under Section 20.3.1.2, evidence suggests that for certain periods the arrangements led to substantial price increases in major European markets.

(508) It is, however, settled case-law that for the purpose of application of Article 81(1) of the Treaty and Article 53(1) of the EEA Agreement 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. In accordance with settled case-law, it is not necessary to show actual anti-competitive effects where the anti-competitive object of the conduct in question is proven (863). It should be added that the defensive arguments mentioned in recital (507) do not exonerate the authors of the infringement from having committed it.

(509) In this case, it is therefore not necessary, given the manifestly anti-competitive object of the agreement, to demonstrate an adverse effect upon competition. Price fixing and market sharing by their very nature restrict competition within the meaning of Article 81(1) of the Treaty. The fixing of target prices and market shares together with the exchange of confidential information must at the very least have distorted, if not eliminated, the free play of market forces in the establishment of a competitive price level.

(510) Whilst the competition-restricting object of the arrangements is sufficient to support the conclusion that Article 81(1) of the Treaty and Article 53(1) of the EEA Agreement have been infringed, in this case it may be concluded that, on the basis of the elements which are put forward in this Decision, the Commission has also proven that the anti-competitive agreements have been implemented and that actual anti-competitive effects of the cartel arrangements have taken place.

- The implementation of the cartel decisions was ensured by a monitoring scheme through market leaders and the regular exchange of confidential information (see, for example, the references in recitals (449) and (450)). In the absence of proof to the contrary, it may be presumed that the competitors in question took into account the information exchanged in determining their own conduct on the market.

- The implementation of the cartel decisions was also ensured by the frequent contacts between competitors. The fact that they met regularly over a period of more than ten years - more than a hundred bilateral or multilateral contacts are established - to discuss volumes, target volumes, prices, rebates, occasionally customers or other commercial terms is an indication that their arrangement must have had some kind of success. The periods of crisis and deviation from the agreed principles, which appears to have occurred in 1994, 1995 and 1996 with respect to certain parties (recitals (463)-(493)), may be considered normal in the life cycle of a long-lasting cartel.

- The parties participating in the infringement from June 1988 or September 1989 until March 2001, demonstrated a constant effort to continue or implement arrangements that they allegedly considered inefficient by improving the system of data collection and dissemination and rendering the infringement more effective (for example, principal focus of discussions on five Member States).

- There is also sporadic evidence of internal instructions to implement a price increase agreed upon at a meeting with competitors (see recitals (412) and (413)) and of notes reporting the success of price cooperation (see, for example, recitals (180), (181), (213), (277), (321), (335), (338), (350), (366), (378), (394), (407), (412), (419), (420) and (426); for a detailed discussion of this evidence see Section 20.3.1.2).

- As to the stabilisation of market shares, the regular review of the market share development at meetings enabled parties to monitor possible deviations in order to re-establish the agreed percentages. The market share development of the participants shows that their shares remained relatively stable throughout the period of infringement (Annex). The parties accepted certain losses of market shares as a consequence of their arrangements in order to generate additional profits (see, for example, recitals (326) and (350)).

13. EFFECT UPON TRADE BETWEEN MEMBER STATES AND BETWEEN EEA CONTRACTING PARTIES

(511) The continuing anti-competitive arrangements between the copper plumbing tube producers had an appreciable effect on trade between Member States and between EEA Contracting Parties.

(512) Article 81(1) of the Treaty is aimed at arrangements which might harm the attainment of a single market between the Member States, whether by partitioning national markets or by affecting the structure of competition within the common market. Similarly, Article 53(1) of the EEA Agreement is directed at arrangements that undermine the achievement of a homogeneous European Economic Area. In order for an agreement, decision or concerted practice to affect trade between Member States, it must be possible to foresee with a sufficient degree of probability on the basis of a set of objective factors of law or fact that the agreement, decision or concerted practice in question may have an influence, direct or indirect, actual or potential, on the pattern of trade between Member States (864).

(513) As demonstrated in Section 1.5 of Part B, the "Inter-State Trade", the market for copper plumbing tubes is characterised by a substantial trade between Member States. There is also a considerable volume of trade between the Community and EFTA countries belonging to the EEA.

(514) The application of Articles 81(1) of the Treaty and 53(1) of the EEA Agreement to a cartel is not, however, limited to that part of the members. sales that actually involve the transfer of goods from one Member State to another. Nor is it necessary, in order for these provisions to apply, to show that the individual conduct of each participant, as opposed to the cartel as a whole, affected trade between Member States (865).

(515) In this case, at least for a certain period, the cartel arrangements covered virtually all trade throughout the Community and EEA, albeit only indirectly through the cartelization of the major European markets. The existence of a volume allocation system and the price-fixing must have resulted, or was likely to result, in the diversion of trade patterns from the course they would otherwise have followed (866).

(516) Insofar as the activities of the cartel related to sales in countries that are not members of the Community or the EEA or to sales before these countries became members of the Community or the EEA, they lie outside the scope of this Decision.

14. ADDITIONAL CONSIDERATIONS

14.1. The SANCO arrangements and KME's role

(517) The Commission has explained in recital (126), the extent to which it relies on the statements of the former executive of the copper plumbing tube industry, which are confirmed by BCZ. The Commission considers that it could not be established whether KM(E) compelled BCZ after its acquisition of UCZ in 1987/1988 to take part in the arrangements involving SANCO copper plumbing tube producers and the broader group of European competitors. Also, it could not be established whether KME functioned as a ringleader among the SANCO producers or on a broader European level.

Boliden's arguments

(518) In principle, BCZ confirmed the statements of one of its former executives ("the witness"). BCZ submits that when the negotiations on the renewal of the licence agreements with KME started in 1988, KME presented BCZ with an ultimatum. "In order for BCZ to be allowed to succeed UCZ in the WICU and SANCO agreements, BCZ would have to: (i) agree to cooperate with KME and the other SANCO producers (Wieland, Tréfimétaux and EM/LMI) in various market activities including the reporting of its sales figures to KME and participating in SANCO meetings; (ii) limit its sales into Germany to a maximum of 10,000 tonnes per year; and (iii) refrain from producing SANCO tubes at its production facility in Gusum, Sweden". Furthermore, "KME made it clear to BCZ that a close cooperation with KME and the other SANCO producers was a precondition for KME to be willing to assist BCZ in the re-registration process of BCZ's products with national certification authorities" (867).

(519) BCZ submits that in order "to avoid KME's censorship", it did not report correct sales figures to KME and other SANCO producers. KME possessed a variety of instruments enabling it to enforce its will on BCZ. Besides implicit threats of massive sales at dumped prices in BCZ's main markets, instruments have included explicit and realized threats to hinder BCZ's renewal of national authorisations for its products, influencing BCZ's licensees not to pay their licence fees to BCZ, (868) and to terminate BCZ's WICU and/or SANCO licences (869).

(520) BCZ contests KME's argumentation that KME was as dependent on BCZ as BCZ was on KME during the negotiations in 1988. BCZ indicated that KME's licence under the SANCO patent was not affected by BCZ's acquisition of UCZs assets.

KME's arguments

(521) KME contests BCZ's statements and the statements of the witness (870). KME believes that the witness is unreliable and biased, because he holds a personal grudge against KME and that his statements should therefore be disregarded in their entirety. In this respect, KME pointed to various documents of the file evidencing the difficult relationship between the witness and KME, and to the fact that the witness constantly violated BCZ's contractual obligations. KME believes that in the interviews with the Commission, he played down BCZ's role by presenting BCZ as victim of an aggressive large company (KME) that only reluctantly participated in any of the illicit arrangements. The fact that the witness' statements were not reflected in BCZ's submission proves, according to KME, that they are unfounded (871).

(522) KME submits that a number of statements of the witness are wrong. First, the SANCO arrangements were not the core arrangement around which the broader European arrangements developed. KME submits that the vast majority of the SANCO meetings had nothing to do with arrangements described in the Statement of Objections because they were attended by technical representatives and not by sales representatives. The SANCO licence cooperation was neither created nor maintained in order to serve as a platform for anti- competitive contacts. Contacts concerned certain national plain plumbing tube markets (mainly Germany and Belgium). Similar contacts occurred in several other countries such as, for example, the United Kingdom. Hence, the classification of these meetings as "national meetings" would be more appropriate. SANCO volume collection only concerned Belgium, France, Germany, Italy, the Netherlands and Switzerland (872).

(523) KME recalls that in 1988 and 1989, KME and BCZ met a number of times in order to negotiate on a possible distribution cooperation. Also rebates and volumes were discussed. After September 1989 until at most the mid-1990s, KME, Wieland and BCZ discussed in bilateral or trilateral meetings volumes and rebates in particular concerning Germany and Belgium (in the same way discussions were held amongst United Kingdom producers and French producers). At the latest from the mid-1990s, any arrangements involving KME took place mainly on a bilateral basis with Wieland. Discussions focused on Germany: Other markets may have been reviewed occasionally but, if so, mostly for informative purposes (873). KME stresses, however, that the principal purpose of the meetings were technical or advertising issues (874). According to KME, the SANCO arrangements were not a "second level" European cartel but essentially bilateral meetings among KME and Wieland mainly concerning the German market. KME believes that the testimony mixed up European and SANCO meetings. The SANCO meetings listed in the Statement of Objections for the period of 1988 until 1995 confirm that SANCO meetings were not the systematic and European-wide arrangements described by the testimony (875).

(524) KME submits that the Commission should take into account for its assessment that the volume collection under the SANCO licence agreements was considered legal. Also, KME points out that SANCO meetings were not the only or principal anti-competitive contacts "below" the European arrangements, and were not its origin or cause (876). KME, however, does not contest that the data obtained was also used to monitor certain illegal volume arrangements (877).

(525) KME contests that it had a leading role concerning pricing decisions in Europe. With respect to prices, each domestic producer played a decisive role in its home market (878).

(526) KME did not coerce BCZ to join the SANCO arrangements (879). KME submits that as far as statements of the testimony concern the period before June 1988, they should not be taken into consideration (880).

The Commission's conclusion

(527) The Commission notes that Wieland had no recollections concerning the SANCO cooperation. In particular, Wieland did not confirm that KME played any leading role in the cooperation among SANCO producers. Although it appears to be correct that KME influenced BCZ's licensees not to pay their licence fees to BCZ, it cannot be excluded that the sole reason for this was that BCZ did not respect contractual agreements. BCZ itself indicated that it considered these licence agreements legal. These kind of tensions caused by BCZ's breach of these agreements, even though they appear to have been illegal, can therefore not have been seriously understood by BCZ as a threat by KME to BCZ to participate in cartel arrangements. In the absence of additional documentary evidence with respect to KME's role, the Commission cannot conclude that KME coerced BCZ to participate in the cartel arrangements.

(528) In this respect it has to be stressed that under competition law, KME did not have any obligation to license its trademark or patents to BCZ. As far as BCZ's involvement might have been motivated by business motives to secure licence agreements with KME (for example, concerning WICU tubes), this cannot serve as an excuse for BCZ to participate in a cartel.

(529) It is also not proven, as claimed by BCZ, that the SANCO producers had jointly agreed not to purchase any of UCZ's assets in order to assure that the production capacity would disappear from the market (881). KME contested this allegation and substantiated its contestation with credible documentary evidence (882).

(530) The fact that the Commission could not establish whether BCZ was coerced by KME to participate in the arrangements, does not affect BCZ's position in the current proceedings. BCZ, had it suffered from pressure by KM(E), should have denounced such behaviour to the competition authorities that ensure that competition rules are respected.

14.2. WICU and Cuprotherm arrangements

(531) With respect to the exchange of sensitive information as regards WICU and Cuprotherm products, KME does not contest that such an exchange occurred in particular among Wieland and KME concerning Germany. However, KME stresses the small significance of WICU and Cuprotherm tubes, and correspondingly the negligible importance of these products for this investigation.

14.3. The parties. claim that certain arrangements were considered legal

(532) Wieland, KME and BCZ claimed that their employees considered legal certain elements of the SANCO, WICU and Cuprotherm arrangements such as the licence agreements as well as the exchange of volume data for the purpose of allocating advertising costs during the relevant period. Accordingly no effort was made to conceal the exchange of volume data under the licence agreements or the licence partner meetings (883).

(533) The Commission notes that Wieland's, KME's and BCZ's employees knew of the illegal nature of the volume allocation and price co-ordination, these also being elements of the SANCO, WICU and Cuprotherm arrangements. Therefore, the Commission does not have to assess possible implications of any error as to the prohibited nature of other elements of the SANCO, WICU and Cuprotherm arrangements.

14.4. Alleged coercion of Halcor

(534) Halcor was neither coerced to participate in the meetings nor presented with any ultimatum that could have had a relevant effect on Halcor's will to participate in the illegal arrangements. Even in the hypothetical case of coercion, Halcor would have had the possibility to contact competition authorities instead of participating in the illicit arrangements.

(535) Halcor claimed to have been coerced into participating in the infringement. However, after having received the Statement of Objections, Halcor corrected essential elements of its company statement (for example that Halcor did not expect to discuss market shares and sales in the 28 August 1998 meeting). Further elements were corrected in Halcor's submission of 5 December 2003 (884). Halcor now submits that it was coerced by credible threats of retaliation on its home market into participating in the arrangements, presented with an ultimatum in the group of the nine sweepers. meetings on 28 August 1998, and forced into the arrangements by the .Big Five. acting as a cohesive group (885). Halcor made clear that the basis of its allegations was in particular the note dated 29 August 1998 from a representative of Halcor, which in Halcor's view, constitutes proof for its allegation (886). Further to this note, Halcor quotes a number of documents dated before August 1998, in which competitors mention certain possible "actions" against Halcor or other competitors (887). Halcor stressed that it could not identify one or other of the producers as "the coercer".

(536) In the Commission's view, the documents to which Halcor refers, do not prove any coercion or ultimatum by the group of the five. As far as the documents concern internal notes of competitors that date from before the meeting of 28 August 1998, they are not relevant for any possible coercion during or after that meeting since during that meeting, Halcor was not aware of any possibly existing reflections of competitors with respect to coercion. In this respect, it is important to note that Halcor has not claimed that before the meeting of 28 August 1998, it had suffered from any illegal pressure from its competitors.

(537) With respect to the interpretation of the note dated 29 August 1998, first, Halcor does not claim, and it is not proven, that KME or Wieland individually (or jointly) coerced Halcor (see (538)). Second, the documents evidence that OTK was not even present in the meeting of 28 August 1998 (see recital (325)). Although an employee of KME reassured Halcor that Outokumpu would "follow the other four" (see recital (326)), there are no indications that Outokumpu exercised pressure and/or was aware or accepted that the other producers did so. Finally, there is no mention of IMI's or Mueller's conduct (888). Under such circumstances, it is not possible to conclude that OTK, IMI or Mueller participated in any kind of coercion.

(538) In fact, instead of proving coercion, the notes of Halcor's employee indicate that Halcor negotiated, as any other company, the best possible volume allocation for itself. It is not surprising that an attempt to increase one's own volume quota leads to reactions from other competitors that would have had to renounce a corresponding volume share. Halcor's representative imagined "what will happen in Europe if they activate their spare capacity" and thought that "at hard confrontation we [Halcor] would lose more". These considerations demonstrate that Halcor took a conscious decision to participate in cartel meetings to avoid the "hard confrontation" and risk of competition. This happened despite the fact that other competitors such as OTK did not wish to participate in "such meetings" or Mueller, for reasons of competition law concerns, resigned during a time, while Halcor was participating in the arrangements (see recitals (344) and (349)). Halcor's role was not exclusively passive, as Halcor tries to suggest. It is evidenced by Halcor's notes of the meeting of 29 October 1998 that, at least in that meeting, Halcor actively took part in discussions on pricing in the United Kingdom (889).

(539) KME (890), IMI (891), Mueller (892), OTK (893) and Wieland (894) contested Halcor's allegations. After having reviewed the file and after having been confronted with other parties. arguments at the Oral Hearing, Halcor did not explain or clarify its role in cartel meetings related to the United Kingdom (as mentioned in various submissions of other parties and Mueller's presentation during the Oral Hearing) or other previous meetings on the European level (and also did not contest its involvement), and why it felt threatened by competitors, with which it had illegally co-operated before. Halcor's statement of 18 April 2003 that prior to 28 August 1998 it had not had any meetings with its European competitors regarding commercial issues, is contradicted by documentary evidence and statements of competitors (895).

14.5. No proof of retaliatory measures by Wieland and KME against IMI

(540) IMI submits that it conducted an aggressive commercial policy towards its competitors and was the victim of individual retaliatory measures by KME and Wieland (896). The Commission considers that the documents, to which KME referred, as well as the remaining file, do not demonstrate any kind of individual retaliatory measures by KME and Wieland towards IMI. The first document (897) reports certain sales of KME in the United Kingdom market, which, according to Wieland, could possibly be understood as a reaction against sales of IMI in Germany. The second document (898) mentions rebates that IMI granted in reaction to competition from HME. At the time, Wieland expected reactions from KME and BCZ and believed that it had to adapt its own rebate policy accordingly. Both documents only demonstrate competitors. reactions to competitive pressure from IMI but not any retaliation.

14.6. Considerations on certain evidence

(541) A number of documents and tables, used as evidence in this decision, were found at the premises of one of the participants and might not have been subject to an exchange with other parties. Examples are the tables referred to in recital (348) (see also, for example, recital (366) and (421)). KME claims that these tables are of an internal nature and seek to give an overview about current price offers in order to better position itself in the market. According to KME, the tables are based on KME's own market intelligence stemming in particular from customers providing KME with offers from competitors (899).

(542) The Commission cannot exclude that certain information contained in such tables is based on KME's own market intelligence. However, with respect to those tables, given the level of details and the completeness of the information for the majority of the European countries contained in the tables, the overlap with information, which was subject to exchange amongst the parties, and the close temporal link to many meetings, the Commission concludes that the compilation of information in the tables, to a large extent, is the result of the illicit information exchange of sensitive business information amongst the participants of the infringement. The same applies to similar evidence presented in the factual part of this Decision.

F. - ADDRESSEES

15. LIABILITY FOR THE INFRINGEMENT

15.1. General principles

(543) The subject of Community and EEA competition rules is the "undertaking", a concept that is not identical with the notion of corporate legal personality in national commercial or fiscal law. The term "undertaking" is not defined in the Treaty. It may, however, refer to any entity engaged in economic activity. According to the circumstances, it may be possible to treat the whole group or individual subgroups or subsidiaries as the relevant "undertaking" for the purposes of Article 81 of the Treaty and Article 53 of the EEA Agreement.

(544) With regard to the liability of the parent company over its subsidiaries' conduct the Courts have consistently referred to an absence, on the part of the subsidiary, of "autonomy in determining its course of action in the market" (900). In this regard, it may be presumed that a wholly-owned subsidiary, in principle, necessarily follows the policy laid down by the parent company and thus does not enjoy such an autonomous position (901).

(545) The Court of Justice examines, among other things, whether the parent company was in a position to exert a decisive influence on its subsidiary's commercial policy or whether the subsidiary was autonomous (902). Apart from the situation of a wholly-owned subsidiary, this power is exercised where the facts of the case show that the subsidiary is not "autonomous" in its behaviour on the market, and vice versa. This may, for example, be the case where the parent company has directly instructed its subsidiary to adopt that behaviour (903), where the parent was otherwise actively implicated in the infringement, for example by representing the interests of its subsidiary in the cartel meetings (904), or where the parent company had been aware of the infringing behaviour in question but had not intervened to put an end to it (905).

15.2. The liability of the parent companies in this case

15.2.1. The liability of Boliden AB, Boliden Fabrication AB and Boliden Cuivre & Zinc SA

(546) The Commission notes that Boliden AB controlled the entire capital of Boliden Fabrication AB and, indirectly, of Boliden Cuivre & Zinc SA throughout the duration of the infringement (see recital (27)). Therefore, the Commission presumes that Boliden AB had full (100 %) effective control and decisive influence over its subsidiary's commercial policy. Boliden AB and Boliden Fabrication AB have not contested that they had full control and decisive influence over BCZ. In addition, it is not contradicted by any element in the file. The management of Boliden AB, Boliden Fabrication AB and Boliden Cuivre & Zinc SA was interconnected.

(547) Consequently, this Decision should be addressed to Boliden AB, Boliden Fabrication AB and Boliden Cuivre & Zinc SA, which should be held jointly and severally liable for the illicit activities of Boliden Cuivre & Zinc SA and its subsidiaries for the duration of the infringement from June 1988 until March 2001.

15.2.2. The liability of Buntmetall Amstetten GesmbH, Austria Buntmetall AG and Wieland Werke AG

(548) Buntmetall Amstetten GesmbH was wholly owned by Austria Buntmetall AG throughout the duration of the infringement (see recital (30)). Therefore, the Commission presumes that Austria Buntmetall AG had full (100 %) effective control and decisive influence over its subsidiary's commercial policy. This presumption is neither contradicted by any element in the file nor contested by Austria Buntmetall AG.

(549) Since 9 July 1999, with its 75,1 % share (acquired from its private owners), increased to 82,8 % on 1 October 1999 and to 83,3 % on 30 November 2000, Wieland exercised sole control over Austria Buntmetall AG (see recital (30)). This assessment is not contested by Wieland and is confirmed by the fact that after the acquisition of Buntmetall's shares former Buntmetall employees switched to Wieland (see recital (33)). In addition, Wieland negotiated with KME on the future use of a Buntmetall brand (see recital (119)). This demonstrates that Wieland exercised decisive influence on Buntmetall's commercial behaviour. Because of its own participation, Wieland was informed about the activities of Buntmetall.

(550) Consequently, this Decision should be addressed to Buntmetall AG and Buntmetall Amstetten GesmbH, which should be held jointly and severally liable for the illicit activities of Buntmetall Amstetten GesmbH from August 1998 until 8 July 1999.

(551) From 9 July 1999 until March 2001, Wieland Werke AG, Buntmetall AG and Buntmetall Amstetten GesmbH should be held jointly and severally liable for the illicit activities of Buntmetall Amstetten GesmbH and by consequence this Decision should be addressed to those entities for the period in question.

15.2.3. No liability of Viohalco SA

(552) As indicated in recital (35), through different holding companies, Viohalco SA holds 65,55% of Halcor SA's shares. The Commission notes that Halcor SA is not wholly owned by Viohalco SA and considers that there is not sufficient evidence in the file to conclude that Viohalco SA exercised control over Halcor SA.

15.2.4. The liability of IMI plc, IMI Kynoch Ltd. and IMI Yorkshire Copper Tube Ltd.

(553) IMI plc controlled the entire capital - through its fully owned subsidiary IMI Kynoch Ltd. - of Yorkshire Copper Tube Ltd. (YCT) throughout the duration of the Infringement (906). Therefore, the Commission presumes that IMI had full (100 %) effective control and decisive influence over its subsidiary's commercial policy. This presumption is not contradicted by any element in the file. In addition, the management of IMI and YCT was interconnected. IMI plc. and IMI Kynoch Ltd. have not contested the finding of control over and conclusion of liability for their subsidiary.

(554) IMI requested that the fine be imposed on IMI plc rather than on YTC, which was sold to KME. The Commission considers, that IMI plc is entitled to pay the fine for IMI Kynoch Ltd. and Yorkshire Copper Tube Ltd., for which it should be held jointly and severally liable. However, the mere sale of a company does not discharge a company from its responsibility.

(555) Consequently, this Decision should be addressed to IMI plc, IMI Kynoch Ltd. and Yorkshire Copper Tube Ltd., which should be held jointly and severally liable for the illicit activities of YCT and its subsidiaries from September 1989 until March 2001.

15.2.5. The liabilities within the SMI/KME Group

15.2.5.1. Liabilities

(556) The Statement of Objections was addressed both to SMI and its current subsidiaries KME, TMX and EM. The latter three companies were considered to be jointly and severally liable for each other's behaviour as part of the same cartel during the period 1990-2001, when they all belonged to the SMI-group (recitals (44) and (50) to (56)). Furthermore, SMI was an addressee, bearing joint and several liability for the illicit activities of EM and TMX since the beginning of the infringement in September 1989 and of KM (KME since 1995) since it joined the SMI-group in 1990, until the end of the infringement in March 2001.

- KME's arguments

(557) SMI maintains that it cannot be held liable (jointly and severally with KME, EM and TMX) for the behaviour challenged, its main argument being that as a pure financial holding company SMI was not involved in the operational business of its subsidiaries nor did it participate in the arrangements described in the Statement of Objections. This claim is further substantiated in KME's reply to the Statement of Objections in which further information is adduced to demonstrate that SMI did not exercise decisive influence over its own commercial policy or that of its subsidiaries. These documents include copies of the most relevant intragroup services agreements and related documents, as well as documents indicating that after the restructuring of the group in 1995, KME had the legal responsibility for the management of the group.

(558) With regard to the period from 1989 to 1995, KME points out that SMI's shareholding in EM (which in turn held 100 % of TMX) was only between ca. 41-52%. It contests the Commission's interpretation that SMI's important role in appointing EM's board members and the three common members (out of 11 or 12) of the management boards of SMI, TMX, and EM is evidence of decisive influence of SMI on EM from 1989 to 1995.

(559) While KME has not contested the attribution of liabilities among KME, EM, and TMX, it has argued, for the purposes of determining the relative weight of the cartel participants, that the companies composing the group did not always follow a common commercial policy. To show the existence of intra-group competition and that no overall common commercial policy existed, KME refers to various documents in the Commission's file (907). With regard to the period from 1989 until the end of 1993, it cites two documents, namely a report concerning an intra-group meeting of 5 June 1991 (908) and the TMX 1993 Annual report which refers to the intra-group competition as a reason for creating EMT, a common sales organisation of TMX and EM (909). In KME's view, the fact that the companies of the group had to take the measures referred to in these documents (that is to say, a decision to coordinate the market behaviour for each existing costumer and to create a common sales organization for EM and TMX) to stop competing against each other demonstrates that competition existed between them and that they carried out their commercial policies autonomously.

(560) With regard to the period following those measures (that is to say after November 1993), KME submits that the coordination or the elimination of intra-group competition, which continued until the restructuring of KME in 1995 and, to a lesser extent, until 1999, was only partially successful. In support of this claim, KME refers to the NERA Report that found that each subsidiary followed its own commercial policy (910).

- The Commission's view

(561) After having examined the views expressed by SMI and KME with regard to SMI's position in these proceedings, the Commission concludes that this Decision should not be addressed to SMI. The Commission points out, however, that the legal responsibility of the management of a company does not necessarily coincide with the business reality.

(562) It should be noted that unlike Outokumpu OYj, SMI did not have 100 % control over its subsidiaries. In contrast to the situation of Outokumpu OYj for which the Commission can establish direct involvement in and awareness of the cartel at a time when, according to Outokumpu, the infringement started, the Commission is unable to demonstrate that SMI was either involved in the cartel or aware of it, nor could it be established in this case that SMI managed the commercial policies of its subsidiaries or gave them instructions relating thereto.

(563) The Commission takes note that the evidence referred to in recital (557) expressly states that since the restructuring of the SMI-group in 1995, KME has been fully responsible for the business management of the group from a legal standpoint. As a matter of fact, KME has only contested the liability of the holding company SMI. On the contrary, it has not disputed, in this context, the joint and several liability of KME, EM and TMX in the period 1990-2001 during which all these companies have been part of the SMI-group, nor does it contest the joint and several liability of EM and TMX in the period preceding KME's entrance into the group in 1990, as established in the Statement of Objections.

(564) Nevertheless, based on the evidence adduced, it appears appropriate to distinguish two separate periods for the purposes of imputation of liability within the SMI-group, once SMI has been exonerated from such liability. During the first period including the years 1988 to 1995, KME's management board was different from that of its sister companies (recital (53)), and KME's operational management and reporting structures appear to have been coordinated with those of EM and TMX only after the restructuring of the group in 1995 (recitals (54) and (56)). Furthermore, the incidents of intra-group competition and other evidence reported above in recital (559) suggest that the entities of the group competing against each other on the market were mostly KME and TMX. These elements together lead the Commission to consider that KME was a separate undertaking from EM and TMX until 1995, regardless of the fact that it joined the SMI-group already in 1990. It may therefore be concluded that during the period from 1988 to 1995, KME is liable only for its own conduct.

(565) On the other hand, EM and its wholly-owned subsidiary until 1995, TMX, must be regarded as one economic unit and thus a single undertaking distinguished from KME until the restructuring of the group. Further to 100 % control of EM over TMX, a number of other elements support the presumption that the subsidiary did not follow an autonomous commercial policy. KME itself has stated that Italian managers were introduced to TMX organisation at the board level and TMX business plan and commercial strategies were aligned with those of EM since 1987 (recital (50)). The commercial policies of EM and TMX were thus intertwined and the companies were closely involved in each other's strategic and organisational management. When the parent company and its subsidiary both manufacture the same product and furthermore participate in the same cartel, as in this case, it is hardly conceivable that each of them would conduct its own autonomous policy on the market of the product in question and make independent decisions with regard to competitively sensitive issues, in particular, prices, sales and production volumes. Accordingly, in the period 1989-1995, EM bears liability for its own conduct and should be held jointly and severally liable for the illicit behaviour of its subsidiary TMX.

(566) As to the period following the restructuring of the SMI group in 1995, since which KME has controlled 100 % of the capital of both EM and TMX, the management of these entities was closely interconnected. The composition of the management boards of these companies was reorganised so that there were significant overlaps between the entities of the group (recital (54)) and their operational management was coordinated (recital (56)). Accordingly, and considering the reasoning exposed in recital (565), KME and its wholly-owned subsidiaries must be considered to have acted as a single undertaking on the market during the period 1995-2001. The presumption of control based on KME's 100 % shareholding in EM and TMX, which is further supported by significant management links and economic reality, has not been rebutted by sufficient evidence.

(567) Based on the above, KME should be held liable for its own infringement from June 1988 until 22 March 2001. In addition, while KME should bear joint and several liability for EM's and TMX's illegal conduct during the period from 1995 to 2001, EM should be held jointly and severally liable for TMX's conduct during the period from 1989 to 1995.

15.2.5.2. Succession

- Europa Metalli

(568) As regards Europa Metalli, the entity that started the infringement in 1989 was Europa Metalli-LMI SpA ("EM-LMI") (recital (236)). EM-LMI contributed its industrial operations to its newly founded subsidiary Europa Metalli SpA ("EM") in 1995 and ceased to exist as a legal entity thereafter. As EM-LMI's successor, EM should bear the liability for the infringement committed by its predecessor.

15.2.6. The liability of Mueller Industries, Inc., WTC Holding Company, Inc., Mueller Europe Ltd., DENO Holding Company, Inc. and DENO Acquisition EURL

(569) Mueller Industries Inc. controlled the entire capital of Wednesbury Tube & Fittings Company Ltd./Mueller Europe Ltd. and Desnoyers SA/Mueller SA throughout the duration of the infringement (see recitals (57)-(61))[...].

(570) Considering that Mueller SA has been declared insolvent, is in liquidation and that Mueller can be held liable for its subsidiaries, the Commission has decided not to open proceedings against Mueller SA.

(571) Mueller has confirmed its control over Mueller Europe Ltd. and Mueller S.A (911)[...]during the whole period of the infringement from October 1997. In November 1999, Wednesbury Tube & Fittings Company Ltd. changed its name to Mueller Europe Ltd. and Desnoyers SA to Mueller SA A mere change of a company's name does not affect its liability. Thus, Mueller Europe Ltd. and Mueller SA are responsible for Wednesbury Tube & Fittings Company Ltd. and Desnoyers SA, respectively.

(572) Furthermore, the Commission considers that the facts demonstrate that Mueller Europe Ltd. (formerly Wednesbury) and Mueller SA (formerly Desnoyers) participated jointly in the infringement. Often, they were represented by business leaders of either of the two companies and co-ordinated their participation. Thus, they were necessarily aware of each others' illegal behaviour throughout the entire period of the infringement (912). When the companies of the same group all manufacture the cartelized product and furthermore participate in the same cartel, it is hardly conceivable that each of them would conduct its own autonomous policy on the market of the product in question and make independent decisions with regard to competitively sensitive issues, in particular, prices, sales and production volumes. This finding is not contested by Mueller.

(573) Consequently, this Decision should be addressed to Mueller Industries, Inc., WTC Holding Company, Inc., Mueller Europe Ltd., DENO Holding Company, Inc. and Mueller SA's holding company DENO Acquisition EURL, which should be held jointly and severally liable for the illicit activities of Wednesbury Tube & Fittings Company Ltd./Mueller Europe Ltd. and Desnoyers SA/Mueller S.A for the period 21 October 1997 to January 2001.

15.2.7. The liability of Outokumpu OYj and Outokumpu Copper Products Oy

(574) In the Statement of Objections, the Commission considered that Outokumpu OYj controlled the entire capital of OCP throughout the duration of the infringement from September 1989 until March 2001. It presumed that Outokumpu OYj had full (100 %) effective control and decisive influence over its subsidiary's commercial policy, and noted that this presumption was not contradicted by the party nor by any element in the file. Accordingly, Outokumpu OYj was considered to be jointly and severally liable for the illicit activities of its wholly owned subsidiary Outokumpu Copper Products Oy (OCP).

(575) In its reply to the Statement of Objections, Outokumpu contests the liability of Outokumpu OYj. Only Outokumpu Copper Products. management was involved in the infringement, and no member of the parent company Outokumpu. Furthermore, Outokumpu Copper Products Oy was formed and fully in operation by December 1988. This is before the beginning of the period considered by the Commission in these proceedings. (913)

(576) The presumption of liability in case of a wholly-owned subsidiary can be rebutted by adducing evidence that the subsidiary determines autonomously its course of action in the market, implying that the parent does not exercise effective control over the commercial policy of the subsidiary (914). This does not necessarily imply that the subsidiary enjoys such autonomy specifically with respect to the infringement. Hence, it is not necessary to show that the parent company directly participated in the cartel meetings or other illicit competitor contacts.

(577) Although the Commission has limited its Statement of Objections to the period starting in September 1989, it notes that Outokumpu explained that the infringement, although in a less organised form, had already started in 1987. In this respect, it is important to note that Outokumpu Copper Products Oy was registered as a corporate entity in 1988 (see recital (62)). Outokumpu Oyj has held 100 % of OCP's shares since the formation of the latter as a separate legal entity. OCP's establishment occurred in 1988, one year after the starting date of the infringement indicated by Outokumpu. The parent company was therefore responsible for the copper tube activities in the initial period and consequently directly involved in the infringement. The infringing behaviour itself can be positively traced back to the parent company, even if none of Outokumpu's staff above Outokumpu Copper Products as a Division or a company was implicated in these proceedings and the highest representative who participated in any meetings in this case was at the level of the Presidents of Outokumpu Copper Products.

(578) During the infringement period following OCP's formation on 30 December 1988 (from September 1989 until 22 March 2001), the Commission presumes Outokumpu OYj's had effective control over the commercial policy of its wholly-owned subsidiary. There are no elements in the Commission's file showing real business autonomy of OCP, nor has Outokumpu been able to adduce sufficient evidence to rebut this presumption. Accordingly, Outokumpu Oyj and OCP must be regarded as a single undertaking for the purposes of this Decision. Moreover, there are letters in the Commission's file showing that the chief executive officer of Outokumpu OYj had meetings and contacts with the vice president of Europa Metalli in 1993 to discuss the market situation in copper and copper alloy semis (915). He also intervened to suggest meetings between OCP's and Europa Metalli's management (916). Hence, the Commission has valid reasons to assume that the top management of Outokumpu Oyj was involved in the commercial policy of its subsidiary OCP.

(579) Consequently, Outokumpu OYj should be held jointly and severally liable with Outokumpu Copper Products Oy (OCP) for the illicit activities of the latter in the period from September 1989 until 22 March 2001.

16. ADDRESSEES OF THIS DECISION

(580) It is established by the facts as described in this Decision, and taking into consideration the liabilities and successions defined in Section 15 above, that the following companies are considered to have directly participated in the infringement found in this Decision for the respective periods: Boliden AB, Boliden Fabrication AB and Boliden Cuivre & Zinc SA; Buntmetall Amstetten GesmbH and Austria Buntmetall AG; Halcor SA; HME Nederland BV; IMI plc, IMI Kynoch Ltd. and IMI Yorkshire Copper Tube Ltd.; KM Europa Metal AG, Tréfimétaux SA, Europa Metalli SpA; Mueller Industries, Inc., WTC Holding Company, Inc., Mueller Europe Ltd., DENO Holding Company Inc., and DENO Acquisition EURL; Outokumpu Oyj and Outokumpu Copper Products Oy; and Wieland Werke AG (for details see recital (597)).

(581) EM and TMX are considered to have formed a single undertaking during the period from 1989 to 1995 (recital (565)), bearing joint and several liability for the infringement in that period. EM is also liable for the conduct of its predecessor, EM-LMI from the beginning of the infringement on September 1989 until it succeeded the latter as a legal entity (recital (568)).

(582) The Commission considers that KME, EM and TMX have formed a single undertaking since 1995 (recital (566)), bearing joint and several liability for the infringement during the period 1995-2001 (for details see recital (597)).

G. - DURATION OF THE INFRINGEMENT

17. STARTING AND ENDING DATES RETAINED FOR THE PURPOSES OF THESE PROCEEDINGS

(583) As the exact date on which the collusion between the addressees started can no longer be established with certainty, the Commission in this case limits its assessment under competition rules and the application of any fines to the period from 3 June 1988 or 29 September 1989, respectively, these being the dates of the first documented meeting between the two SANCO producers KM(E) and BCZ on the one hand (see recitals (163), (164)) and between the European producers on the other hand (see recitals (236) to (241)). To some companies, the Decision is only addressed for their participation as of 21 October 1997 (see recital (306) and (592)) or 27/28/29 August 1998 (see recitals (325), (326) and (592)). None of the parties have contested these starting dates, albeit certain parties claimed interruptions (see recitals (463)-(497)).

(584) As to the termination date, the Commission retains 22 March 2001 as the relevant date (the date of the Commission inspections) except for Mueller, which terminated its participation in January 2001, and for Halcor, which stopped participating in September 1999.

(585) As to Europa Metalli, it participated regularly in the European cartel meetings of the copper plumbing tube producers and was part of the SANCO club, implying also its involvement in that part of the multiform infringement (recitals (236)-(245)). The Commission notes that the entity's participation in the initial meetings until its reorganisation was as the legal person "Europa Metalli-LMI SpA" which contributed its copper plumbing operations to its newly founded subsidiary Europa Metalli SpA in 1995 and ceased to exist as a legal entity thereafter. Because Europa Metalli-LMI SpA ceased to exist and Europa Metalli SpAcontinued Europa Metalli-LMI SpA's copper plumbing tubes activity, the transfer of the plumbing tube business is considered a mere corporate reorganisation and succession within the same undertaking. Such a reorganisation does not affect the duration of the infringement.

(586) With regard to TMX and to a certain extent also Europa Metalli, the fact that individually these companies occasionally did not participate in the meetings is not relevant when determining their involvement in the cartel, because EM and/or KM(E) represented TMX and/or EM and agreed on volumes that concerned the entire group. Also, TMX's and Europa Metalli's conduct continued to be affected by the cartel beyond their individual membership. Both companies also participated in the SANCO club. Their participation in the arrangements is not contested.

(587) The Commission has demonstrated that BCZ, despite its resignation from cartel meetings for certain periods, continued to participate in the illegal information exchange of the SANCO producers (see recitals (492)-(494)).

(588) With respect to Mueller and its European subsidiaries, the facts described under Section 5 demonstrate that Mueller's involvement in the cartel lasted from 21 October 1997 until January 2001 despite the fact that it discontinued its participation in the EDWD meetings as of 10 December 1998 (see recital (344)) and in the WBMS information exchange system as of 10 December 1999 (see recitals (306)-(308)). The Commission notes that Mueller continued certain arrangements until January 2001 with competitors that took part in the European arrangements. Thus, it appears that Mueller did not completely withdraw from its participation in the cartel until its cooperation with the Commission in January 2001 (see recital (422)).

(589) BMA's and HME's participation started on 27 and/or 28 and/or 29 August 1998 and ended on 22 March 2001 (recital (325) and, for the reasons concerning the end of the duration, see recital (590)). Halcor's participation started on 27 and/or 28 and/or 29 August 1998 (recital (325)) and ended in September 1999 (for the reasons concerning the end of the duration, see recital (496)).

(590) The Commission notes that even though the last known cartel meeting took place on 14 March 2001 (recital (427)), as confirmed by the parties, the relevant date for determining duration of the infringement depends on the implementation of the agreement rather than on the date of the last cartel meeting. Further meetings were scheduled for the period after the date of the Commission's inspections (see recitals (428)-(430)). This evidence supports the conclusion that the implementation of the cartel agreements continued at least until 22 March 2001, when the Commission carried out its inspections pursuant to Article 14(3) of Regulation 17.

18. APPLICABILITY OF LIMITATION PERIODS

(591) Pursuant to Article 1 of Regulation (EEC) No 2988-74 of the Council of 26 November 1974 concerning limitation periods in proceedings and the enforcement of sanctions under the rules of the European Economic Community relating to transport and competition (917), the power of the Commission to impose fines or penalties for infringements of the substantive rules relating to competition is subject to a limitation period of five years. For continuing infringements, the limitation period only begins to run on the day the infringement ceases (918). Any action taken by the Commission for the purpose of the preliminary investigation or proceedings in respect of an infringement interrupts the limitation period and each interruption starts time running afresh (919).

(592) In this case, the Commission's investigation started with the surprise inspections pursuant to Article 14(3) of Regulation No 17 on 22 March 2001. Hence, for infringements ceased prior to 22 March 1996 no fines may be imposed. The Commission considers that this limitation period applies to Buntmetall, HME, Halcor, Desnoyers and Wednesbury as far as their involvement in the European-wide infringement until 1994 is concerned. First, the Commission did not have sufficient evidence to establish continuity of their participation for the years 1995, 1996 and 1997 until October 1997 or August 1998, respectively. Second, evidence of their participation before 1994 has mainly been occasional. With respect to Boliden, IMI, the KME(-group), Outokumpu and Wieland it has been demonstrated under Section 11.3 that this infringement is a single infringement, therefore no limitation period is applicable.

19. PERIODS OF REDUCED CARTEL ACTIVITY

(593) Boliden, IMI, KME, Wieland and OTK have claimed that in this case the Commission should reduce the duration by periods during which the cartel activity was significantly reduced or suspended.

(594) As already mentioned in recital (445), the Commission highlights that although a cartel is a joint enterprise, each participant in the agreement may play its own particular role. Internal conflicts and rivalries or cheating may occur, but will not however prevent the arrangement from constituting a continuous agreement/concerted practice for the purposes of Article 81(1) of the Treaty where there is a single common and continuing objective and will to restrict competition, as in this case. An undertaking may be held responsible for a cartel as a whole, even if it has directly participated only in one or some of its constitutive elements, if it knew or should necessarily have known that the collusion in which it participated was part of a global plan, which covered all the constitutive elements of the cartel in question. Under such circumstances, the fact that the undertaking in question did not directly participate in all the constitutive elements of the global cartel does not relieve it from the responsibility for the infringement of Article 81(1) of the Treaty (920).

(595) With regard to the periods from 1990-December 1992 and July 1994 until July 1997, for which parties have requested a reduction of duration because of the reduced cartel activity, it is sufficient to refer to explanations given in recitals (463)-(491), where the Commission explained that the cartel activity was never entirely interrupted. Boliden's continuous participation is discussed in recitals (492)-(494). The appropriate amount of increase for each year of the infringement is assessed below (see Section 20.3.6).

(596) The parties' claims according to which the duration of the infringement should be reduced are therefore rejected.

(597) On the basis of the above, the Commission considers that the following companies should bear responsibility for their respective infringements and be addressees of this Decision for the following duration:

- Boliden AB, Boliden Fabrication AB and Boliden Cuivre & Zinc SA jointly and severally liable for the period from 3 June 1988 until 22 March 2001 (12 years, 9 months).

- Austria Buntmetall AG and Buntmetall Amstetten Ges.m.b.H jointly and severally liable for the period from 27 and/or 28 and/or 29 August 1998 until 8 July 1999 and jointly and severally liable with Wieland Werke AG for the period from 9 July 1999 until 22 March 2001 (total 2 years, 6 months).

- Halcor SA individually liable for the period from 27 and/or 28 and/or 29 August 1998 until at least the beginning of September 1999 (1 year).

- HME Nederland BV individually liable for the period from 27/28/29 August 1998 until 22 March 2001 (2 years, 6 months).

- IMI plc, IMI Kynoch Ltd. and IMI Yorkshire Copper Tube Ltd. jointly and severally liable for the period from 29 September 1989 until 22 March 2001 (11 years, 5 months).

- KM Europa Metal AG individually liable from 3 June 1988 until 22 March 2001 (total 12 years, 9 months), and jointly and severally liable with Tréfimétaux SA and Europa Metalli SpA for the period from 20 June 1995 to 22 March 2001.

- Europa Metalli SpA and Tréfimétaux SA jointly and severally liable for the period from 29 September 1989 to 19 June 1995 and jointly and severally liable with KM Europa Metal AG from 20 June 1995 to 2001 (11 years, 5 months). From 1989 to 1995 Europa Metalli SpA bears the liability as EM-LMI's successor.

- Mueller Industries, Inc., WTC Holding Company, Inc., Mueller Europe Ltd., DENO Holding Company, Inc. and DENO Acquisition EURL jointly and severally liable for the period from 21 October 1997 to 8 January 2001 (3 years, 2 months).

- Outokumpu Oyj and Outokumpu Copper Products Oy jointly and severally liable for the period from 29 September 1989 until 22 March 2001 (11 years, 5 months).

- Wieland Werke AG individually liable from 29 September 1989 until 22 March 2001 (11 years, 5 months), and jointly and severally liable with Austria Buntmetall AG and Buntmetall Amstetten Ges.m.b.H for the period from 9 July 1999 until 22 March 2001.

20. REMEDIES

20.1. Article 7 of Regulation No 1-2003

(598) Where the Commission finds that there is an infringement of Article 81(1) of the Treaty or Article 53(1) of the EEA Agreement it may require the undertakings concerned to bring such an infringement to an end in accordance with Article 7 of Regulation (EC) No 1-2003.

(599) While the undertakings concerned have informed the Commission of having taken the necessary steps directly after the inspections to ensure that their representatives no longer take part in anti-competitive meetings and other collusive contacts, it is necessary to ensure with absolute certainty that the infringement has ceased. It is therefore indispensable for the Commission to require the undertakings to which this Decision is addressed to bring the infringement to an end (if they have not already done so) and henceforth to refrain from any agreement, concerted practice or decision of an association which might have the same or a similar object or effect.

(600) The prohibition should apply not only to secret meetings and multilateral or bilateral contacts but also to the activities of the undertakings insofar as they involve, in particular, diffusing individualised sales statistics.

20.2. Article 23(2) of Regulation No 1-2003

(601) Under Article 23(2) of Regulation (EC) No 1-2003, the Commission may by decision impose fines on undertakings where, either intentionally or negligently, they infringe Article 81(1) of the Treaty and/or Article 53(1) of the EEA Agreement. Under Article 15(2) of Regulation No 17 (921) 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. The same limitation results from Article 23(2) of Regulation (EC) No 1-2003.

(602) Pursuant to both Article 15(2) of Regulation No 17 and Article 23(3) of Regulation (EC) No 1-2003, the Commission must, in fixing the amount of the fine, have regard to all relevant circumstances and particularly the gravity and duration of the infringement, which are the two criteria explicitly referred to in these Regulations. The fine imposed must reflect any aggravating or attenuating circumstances.

(603) In this case, the cartel constituted an intentional infringement of Articles 81(1) of the Treaty and 53(1) of the EEA Agreement. With full knowledge of the illegality of their actions, the leading producers of copper plumbing tubes combined to set up a secret and institutionalised system designed to restrict competition in a major industrial sector. The intentional nature of the infringement is shown, among others, by the the parties. expressions of their common aim to jointly control the European market for copper plumbing tubes, legal advice received from in-house lawyers (see, for example, recital (294)), and by the precautions they took to conceal the cartel (see references in recitals (449) and (450)). For a certain time, Outokumpu did not participate in top-level meetings because of concerns with the legality of the meetings under competition law (see recitals (322) and (329)).

20.3. The basic amount of the fines

(604) The basic amount is determined according to the gravity and duration of the infringement.

20.3.1. Gravity

(605) In assessing the gravity of the infringement, the Commission takes account of its nature, its actual impact on the market, where this can be measured, and the size of the relevant geographic market.

20.3.1.1. Nature of the infringement

(606) The infringement in this case consisted mainly of market-sharing and price-fixing practices, which are by their nature the most serious restrictions of competition. The Commission, therefore, considers that this infringement constituted by its nature a very serious infringement of Article 81(1) of the Treaty and Article 53(1) of the EEA Agreement.

(607) Wieland submits that secrecy is the very nature of cartels and "damage of the public" its consequence; therefore it is a motive of the legislator in creating Article 81 of the Treaty and already taken into account by the legal framework for the determination of fines. Accordingly, it cannot be used a second time to justify a particular seriousness of an infringement (922).

(608) The Commission points out that the existence of the infringement as defined by Article 81(1) of the Treaty does not imply a specific nature or gravity. Not every infringement of that Article is a cartel, and not every infringement qualifies as a very serious infringement. The nature of an infringement is relevant specifically in the determination of the fines (see recital (605)). In the fines, the Commission is also required to assess the impact of an infringement. The Commission considers secret hardcore cartels, by their nature, as one of the most serious violations of the competition rules. In this Decision, the fact that the parties engaged in a cartel implies that they committed an infringement of Article 81(1) of the Treaty. The secrecy of the cartel and the damage of the customers are used to assess the seriousness of the infringement (its nature and its impact on the market).

20.3.1.2. The actual impact of the infringement

(609) In their earlier submissions and respective replies to the Statement of Objections, Boliden, Halcor, IMI, Outokumpu, KME and Wieland (and BMA) have put forward several arguments to show that the cartel had no or only a limited impact on the market, due to various factors, such as overcapacity, buyer power, difficult economic conditions and loose implementation of the agreements.

The parties' arguments (923)

(610) KME believes that harm to the consumers was limited. In fact, it believes that, at least for plain plumbing tubes, "the result overall was substantially as if no cartel had existed" (924).

(611) KME claims that its gains from the arrangement, if any, were small (925). It argues that Wieland's memo of 10 June 1998 does not state what the reasons for the additional profits were, that is to say, an improved market situation and/or the DWD meetings. The employee who drafted the memo generally referred to an improved profit situation in the market. HME, although at the time not a participant in the DWD meetings, also made additional profits of around DM 11 million (926). KME contests the testimony's statements with respect to additional profits (927). KME notes that in June 1988, the German conversion margin amounted to EUR [...]/kg. In June 1995, the conversion margin had increased by [...] % to EUR [...]/kg. At the same time, costs of living in Germany had increased by more than 20 % (for example, due to inflation) (928). KME submits that it is very difficult to objectively estimate the irregular gains, since increases may be influenced by a number of factors such as inflation, increases in raw material costs, increases in labour costs. Any estimate would therefore require an in-depth economic analysis (929).

(612) With regard to the lack of impact on prices, KME submitted a report by a group of expert economists of NERA Economic Consulting entitled "An Analysis of the Impact of Industry Information Exchanges on Copper Tube Prices in Europe, Plumbing Tubes" (the "KME-NERA Report"). That report, made upon KME's request, analyses whether and to what extent the prices charged by KME and its subsidiaries increased as a result of the discussions during the 1990s. The analysis is based on a data set constructed on the basis of all of KME's available invoices and customer information that contains data concerning orders and quantities delivered to every customer between 1990 and 2002. The principal findings of the NERA Report are that: "The contacts among copper tube producers had no impact on the final full prices charged by Europa Metalli's and Tréfimétaux to their customers for any of their products. The contacts held by KME AG regarding plain plumbing tubes did not affect the prices charged to customers of plain plumbing tubes. The prices charged by KME AG for WICU and Cuprotherm tubes increased as a result of their contacts with Competitors". In KME's view, the report shows that prices were only affected to a limited extent and only for WICU and Cuprotherm tubes. Since the "actual impact" is one of the three key elements in assessing gravity, the lack of economic effect needs to be taken into account (930). KME stresses that the participants. submissions, in particular those of KME, Outokumpu and Halcor, confirm the lack of impact of the arrangements (931). Finally, a comparison of US and European conversion margins shows that prices in Europe are considerably lower (932).

(613) Outokumpu submitted a Memorandum entitled "Economic context of the European copper tubes industry" to demonstrate the limited impact of the cartel on the market (933). Outokumpu explained that the arrangements led to very limited impact on the market. The cooperation did not concern all producers and was not constant, quite loose and only periodical, and planned objectives had in practice not been attained (934). Cooperation concerned the conversion margin only (935).

(614) Outokumpu explained that since the early 1990s, Outokumpu's conversion prices have developed unsteadily and in a comparable way to the general economic development (936). Outokumpu concludes that cooperation was not distorting prices but that many factors kept price development independent (937). The long-standing overcapacity situation and competition only occasionally, if ever, made cost based pricing possible. Mostly, pricing was driven by competition. Purchasers exercise buyer power (938). According to Outokumpu, the prices increased due to the German reunification (1990-1993). From 1993-1997, the situation stabilised. Good price levels attracted more capacity, which led to oversupply. As a result, prices decreased. From 1999-2000, extremely high demand in industrial tubes led to a shift of capacity from sanitary tubes. The shortage of supply drove prices up.

(615) Wieland, in turn, does not deny that the cartel had certain effects but it submits that these effects were relatively minor and fluctuating. It argues that for certain periods of time, prices even reached competitive levels due to the suspension of the agreements.

(616) In support of this allegation, Wieland submitted figures on the development of its conversion margin (non-deflated) and indicated that on a deflated basis, in 2000 the conversion margin represented 70 % of the value of 1989. Wieland concludes that there was no relation between the infringement and the rise in prices (939).

(617) Wieland further submits that the target prices were set at the highest, slightly above hypothetical competitive prices. Wieland claims, in line with KME, that for the purposes of determining gravity the Commission has to take into account that the participants did not derive economic advantage from the cartel.

(618) Wieland submits that the following factors lead to low influence of the infringement on the copper plumbing tube market: buyer power (see recital 171), the limitation of agreements to certain customers, over-capacity of producers, a permanently difficult market situation, and the fact that the arrangements concerned only part of the market (940). Further reasons are set out in the following recitals:

(619) First, exchanged information was publicly available. Thus, the information exchange could not cause any distortion of competition (941).

(620) Second, the cartel had a defensive nature. The primary target of the arrangements had not been price increases, but the avoidance of price erosion (942). Occasional discussions on target price would not justify a different assessment. Prices were always adapted to market conditions by granting additional rebates (943). Conversations only concerned harmonization of prices and not the increase of prices (944). The market leader adapted prices to market conditions by either increasing or lowering them (945). Despite the arrangements, the industry has not been profitable, which is supported by bankruptcies and takeovers from 1988 until 2002 (946).

(621) Third, price competition from non-European copper plumbing tube producers, noncartel members and among the cartel members, led to natural price control (947). The market share of the cartel members was too small to control the market. Wieland considers that according to a memo dated 11 December 1998, substantial competitive pressure was exercised by Halcor, HME, Boliden and BMA that had a combined share of ~ 22,2% (and not of 20,8% as originally estimated by Wieland) (948). The parties therefore lost market shares (949). They had to compete with other materials (950).

(622) Fourth, Wieland submits that the arrangements were not implemented. Information provided was incorrect; arrangements were ineffective and not implemented (951). Heavy price fluctuations and declines of up to 30 % contradict a functioning cartel (952). An effective system of control did not exist (953). Control was only exercised on the basis of the figures submitted by the parties (954). However, for certain durations and countries, control was facilitated by export statistics (955). Agreed rebates were not respected (956). Competitors cheated with different methods, which lead to the erosion of prices (957).

(623) Halcor submits that its participation did not in any way impact upon its behaviour in the market (958). Halcor asks the Commission to prove that given certain circumstances its conduct as proved by the evidence amounted to an infringement of Article 81(1) of the Treaty. Such circumstances include the coercion of Halcor to attend, its trivial role in the scheme of collusion, its non-implementation of any agreements, the one-way nature of the volume allocation scheme, which worked only to the benefit of the group of five, and the absence of any demonstration of an actual effect upon the market.

(624) IMI provided the Commission with a study on the .Conditions in the Copper plumbing tube industry 1987-2001.. The study concludes that the sector has many characteristics of a mature commodity product market and that there has been a continuing and significant overcapacity throughout the period 1990-2001. This situation led to a decline in gross margins (and conversion margins) in the major markets of United Kingdom and Germany over that period. Even if, in the period 1992-1996, YCT margins improved, it was due to exceptional circumstances (demand boom in the German market, exchange rate particularly favourable for United Kingdom producers, non plumbing copper tubes market expansion which took away capacity from plumbing tubes). The study also concludes that the national producers. share of their domestic markets reduced to the benefit of the other European producers. Nevertheless the national producers remained in general market leaders in their domestic market. The study refers to reports of independent market observers published between 1993 and 2000 alleging that fierce price competition took place. Finally, the study indicates that distributors and plumbers. merchants had considerable market power and alleges that they used it.

(625) The impact of the discussion on prices in the United Kingdom market is measured by a further study provided by NERA Economic Consulting, London, on behalf of IMI plc, entitled "An Analysis of the Impact of the Alleged Cartel in the United Kingdom Market for Copper Plumbing Tubes" (the "IMI-NERA Report"). It is based on IMI's transaction data and covers the period from March 1988 to August 2003. The transaction data are aggregated per month resulting in a time series of [...] monthly observations for the United Kingdom market. As in the KME-NERA report a dummy variable is implemented to measure the price impact of discussions. The approach followed by the IMI-NERA Report is to compare on the one hand the price level in periods without discussions to those with discussion (comparison over time). Several variables are included to control for shifts in demand and cost factors as well as a time trend. Several model alternatives are measured. Furthermore, a second approach is employed by measuring a long-term relationship between the copper tubes price and other control variables during the cartel period. Based on these estimates the price in the post-cartel period is estimated under the assumption of on-going discussions and compared with the factual price after the cartel investigation in April 2001 to December 2002. The principal findings of the IMI-NERA Report are that: "the highest estimated increase in transaction prices due to the cartel ranges from [...] % to [...] %, depending on the test adopted. All these estimated increases however are statistically insignificant; that is, there is a high probability that there is no price effect of the cartel,...".

(626) Finally, IMI provided the Commission with another report on the "Financial performance of IMI plc's Copper plumbing tube Business 1989-2001". The report claims that although in the period 1989-2001 the reported trading profit of YCT fluctuated from [...] in 1992 to [...] in 1996, this is not representative of IMI's return from its copper tubing business. [Summary of financial information provided by IMI] Similar arguments explaining the low profitability of Outokumpu's copper plumbing tube business were submitted by Outokumpu.

The Commission's view

General considerations

(627) The Commission emphasises that, contrary to KME's interpretation of the case law, the actual impact of an infringement on the market is not the main element of the assessment of gravity, it is just one of the three: the nature, the actual impact and the geographic market (see (605)). Under each of these elements, the infringement is assessed for its gravity. Moreover, irrespective of the Commission's finding that the infringement had a restrictive effect, the fact that it had a restrictive object which was intrinsically very serious must, in any event, be a more significant factor in the Commission's categorisation of the infringement as "very serious" than factors relating to its effects. The effect which an agreement or concerted practice may have had on normal competition is not a conclusive criterion in assessing the proper amount of the fine. As confirmed by case law, factors relating to the intentional aspect, and thus to the object of a course of conduct, may be more significant than those relating to its effects, "particularly where they relate to infringements which are intrinsically serious, such as price-fixing and market-sharing" (959).

(628) The Court of First Instance has indicated in the ADM Case that "it follows, first, that in the case of price agreements, there must be a finding by the Commission that such agreements have in fact enabled the undertakings concerned to achieve a higher level of transaction price than that which would have prevailed had there been no cartel" (960). There is no need for the Commission to quantify in detail the extent to which prices differed from those which might have been applied in the absence of these arrangements. This cannot always be measured in a reliable manner, since a number of external factors may simultaneously have affected the price development of the products, thereby making it extremely difficult to draw conclusions on the relative importance of all possible causal effects. Indeed, this difficulty is apparent in the parties' arguments relating to the various factors that have affected the price level and it is aggravated by the length of the infringement.

(629) Although there are certain elements in the file that allow a careful estimation, for a limited time period, of the impact of the cartel on prices, it is impossible for the Commission to determine precisely what the evolution of prices during the entire period of infringement would have been in the absence of the cartel. As stated in the ADM Case (961), the fact that the participants informed each other about their sales volumes and price levels was likely to influence their conduct within the cartel and in the market.

(630) The Commission further notes that the arrangements were not only aimed at raising prices but, in particular, at preventing prices from declining at the pace determined by market forces. Indeed, influencing prices by volume restrictions and setting target prices was at the heart of the whole discipline imposed by the cartel rules. Contemporaneous estimates of business people involved in the arrangements and certain other elements show that prices were higher than under normal conditions of competition. This as well as other explanations given in recitals (452)-(454) and (510), allow the conclusion that arrangements were implemented.

(631) In order to contest the existence of any significant impact on the market, the parties have presented two types of arguments, one according to which the arrangements had no effects on prices (or did not lead to price increases) and another according to which the arrangements were not profitable for the participating undertakings.

(632) Certain parties supported the first type of arguments with the economic/econometrical studies mentioned in recitals (612) and (625). As a preliminary remark, it should be noted that, from a conceptual standpoint according to case law, the impact of a cartel does not have to be assessed at the level of one undertaking or even a group but at the level of the global cartel. The Court of Justice has indeed ruled that "lastly, when considering how the effects of the infringement had been taken into account, the Court of First Instance did not have to examine the individual conduct of the undertakings when, as it rightly pointed out at paragraph 280, the effects to be taken into account in setting the general level of fines are not those resulting from the actual conduct which an undertaking claims to have adopted, but those resulting from the whole of the infringement in which it had participated" (962). It should therefore be concluded that a report which examines the impact of the cartel on a single or few undertakings does not match the requirements set up by case law and cannot be conclusive in this respect. This applies also to other arguments provided by the other parties on the impact they claim they experienced individually.

The impact on prices

(633) Despite the above reasoning, the Commission carefully investigated lines of information to estimate the price effects of the cartel agreement on the markets. First, statements of managers on price effects of the infringement at specific points in time have been matched with the actual price developments as provided by the parties. Second, the Commission carefully investigated the econometric evidence (the KME and the IMI-NERA Report) provided by the parties.

(634) Based on this assessment the Commission comes to the conclusion that as concerns prices, at least for a certain time period, arrangements were effective. While significant price effects can be detected in most markets, the data suggests the most severe impacts of the infringement being effective in the German market for the product classes Cuprotherm and WICU.

(635) In order to evaluate the impact, it should also be considered that the copper plumbing tube producers participating in the arrangement at their peak represented 84,6% of the total EEA market. Given that the Commission, in principle, did not include national arrangements in its analysis, this figure appears to be understated. In fact, according to the parties, various "competitors" have participated in national cartels often also including distributors (for details see the references in recitals (106) and (107)).

(636) Therefore, the Commission concludes that prices of copper plumbing tubes have evolved in the period 1989-2001 differently from a purely competitive scenario (that is to say in the absence of the cartel).

(637) It should be recalled that Wieland has admitted that target prices were set at the highest, slightly above hypothetical competitive prices (see recital (617)). Wieland also admitted that, for certain periods of time, prices even reached competitive levels due to the suspension of the agreements (see recital (615)). KME and Outokumpu admit the existence of an impact, although limited (see recitals (610) and (613)). The KME/NERA report concluded that the prices charged by KME AG for WICU and Cuprotherm tubes increased as a result of their contacts with competitors (see recital (612)).

(638) Indeed, numerous elements in the file show that, at least for a certain time period, arrangements were effective.

(639) From 1988 until mid-1995, the cartel arrangements were described as highly efficient by one top-level participant (see recital (213)). According to his explanations, during the indicated period the arrangements succeeded in raising the conversion margin in Belgium over 30 %. In Germany, additional earnings generated as a consequence of the arrangements were estimated to amount to approximately 1 EUR/kilo for local SANCO-producers by the end of 1995. In the United Kingdom, the conversion margin doubled throughout the indicated period (963). According to Mueller, the parties. price-fixing resulted in an overall rise of the price of copper tubes by approximately 60 % during the period from April until November 1994 (see recital (277)) (964). With respect to the United Kingdom, this is in line with the price increase empirically observed. For instance, the IMI-NERA Report states a price increase from [...] GBP/kg in April 1994 to [...] GBP/kg in November 1994 in the United Kingdom for IMI's transaction. This translates to a percentage price increase of [...] % from April to November 1994. If one considers the slightly wider period from November 1993 to April 1995 the prices rose from [...] GBP/kg to [...] GBP/kg, implying a percentage increase of [...] %. The Commission notes that an overall rise of the price of copper tubes by approximately 60 %, as confirmed by Mueller, corresponds roughly to a doubling of the conversion margin, as indicated by the former participant of the meetings. Given that the former top-level participant based his statement on recollections in response to questions raised in an oral interview, the testimony's statement cannot be considered untrustworthy just because the testimony did not recall the precise price increase or the exact period for which the price-fixing was successful. The parties did not substantially contest the price rise in the United Kingdom as described in the Statement of Objections, with the exception of IMI and KME. IMI's contestation appears to be surprising given that IMI proposed a specific pricing action, which was also successfully implemented (see recital (277)). KME's study doesn.t address and analyse the specific price development in the United Kingdom. In addition, Boliden stated at the oral hearing that it confirmed the statement of its former participant of the cartel meetings. The Commission considers therefore that it is proven that the prices in United Kingdom rose overall by [...] % in the year 1994 due to the discussions. The Commission notes that although meetings amongst a large group of producers on the European level might have not occurred after the last meeting described in Part D in spring 1994, prices continued to increase in the United Kingdom market at least until April 1995.

(640) A contemporaneous analysis of business people dated 10 June 1998 (recitals (180), (181) and (321)) found that in the first five months of 1998, the "additional profit" in the German market went up on average by DEM 30,- % kg (965). Based on this price rise, during that time period, the KME group was estimated to have generated additional profits of DEM 3,9 million in Germany. In the Commission's view, the expression "additional profit". has to be understood as gains from increased income, and not as financial profits.

(641) Since July 1997, the "additional profit" in Germany increased by around DEM 70,- % kg so that "the additional profit is even substantially higher" (966). Calculated on the basis of "only D[E]M 30,- % kg", the additional profit of Wieland amounted to DEM 1,54 million. The profit situation, that is to say, the "additional profit", concerning France and the United Kingdom was estimated to be more or less the same as in Germany, while it was slightly lower in the Netherlands and much lower in Spain (where Wieland indicated a need to catch up) (967). On the basis of improved prices, HME was estimated to generate additional profits on the basis of 15.000 tons/year of DEM 11 million in 1998 compared to 1997 (968).

(642) An internal memo of Wieland dated 10 June 1998 states: "Price situation: Since our internal findings in June 97, processing prices have risen by DEM 70.- on average in the five target markets. The differences between Spain and the UK are too large. They are approx. DEM 100". KME contests that this increase is related to the cartel (see recital (611)). In support, KME points out that during that period, HME did not participate in the DWD meetings. The Commission notes, however, that the Wieland memo is dated 10 June 1998, the day after the meeting between Wieland and KME and, according to Wieland, is a neutrally drafted report of the cartel meeting (see recital (180) and (226)). Furthermore, according to KME, HME was involved in a national cartel related to the Netherlands (see recital (489)). This might explain, why Wieland included an analysis of the cartel benefits of HME into the memo. Another reason might have been that HME benefited from the cartel arrangements of others that covered or affected Benelux countries.

(643) Contemporaneous notes of a Mueller executive (see recital (335)) confirm such price Increases (969). Conversion margins (970) rose from July 1997 until October 1998 in the following way:

- Germany: from 145 to 200 (German SANCO), from 75 to 170 (Euro SANCO), 60 to 140 (Importers); in addition, a strategy was agreed to further increase prices between 30 and 45;

- France: from 103 or 92 to 177-187 (TMX), from 75 to 173-187 (Euro SANCO), from 67 to 141-164 (Non SANCO); in addition, a strategy was agreed to further increase prices between 3 and 23;

- Netherlands: from 150 to 196 (KME), from 145 to 170-186 (other); in addition, a strategy was agreed to further increase prices between 24 and 29;

- Spain: from 141 to 134-155 (KME), from 81 to 128-155 (OKC), from 65 to 120 (others), in addition, a strategy was agreed to further increase prices between 15 and 45;

- United Kingdom: from 175 to 230-247 (United Kingdom producers), from 160 to 201-225 (other); in addition, a strategy was agreed to further increase prices between 10 and 24;

(644) Notes of Halcor (see recital (338)) confirm planned price increases, for example, concerning Germany "e.g. for KME - from 200 to 230." (971)

(645) Contemporaneous calculations of additional profits for the year 1998 confirm substantial additional profits (for further explanations see recital (642)). With respect to Germany, "KME - has the biggest market share - [and therefore] earnings-wise, profits the most in absolute terms from the good price level that prevails". On the basis of an earnings improvement of "just DEM 50/kg", as was "certainly" achieved in "'98 compared with '97" based on "actual quantities" in "'98[,] KME has achieved an earnings increase of approximately DEM 30 million while we have achieved just under DEM 9 million". It is further clear that the parties took a conscious decision to raise prices with the risk of losing certain market shares: "Our proposal as far as the new participants are concerned is therefore that we should continue [...]. This is definitely better than jeopardising everything we have achieved so far [...]; this should be clear to every participant given the millions earned, as mentioned above, especially to the KME group, which was and still is the one that stands to gain the most in absolute terms from the whole exercise". (see recital (350)).

(646) According to the market analysis of Wieland, the price level that was achieved, and based on which every participant was in a position to earn "additional millions", was as of 15 January 1999 as follows (see recital (350)): Germany: 209 (German SANCO), 164 (DVGW); France: 177; Netherlands: 204; Spain: 160; United Kingdom: 193.

(647) KME's compilation of sales data of 1998 (972) tends to confirm Wieland's analysis (see recital (366)). First, KME's "Industrial Result" analysis shows that in 1998, the KME group profited in total figures by far the most from the price situation in the market. For IMI, Mueller, Wieland and Halcor, the price situation appears to have also led to profits (973). In the core markets of the arrangements, the conversion margin situation for plain copper plumbing tubes, expressed in "Euro%kg" and "calculated without any commissions and margins" was as follows (974):

- Germany: [...] (KME-group), [...](Wieland), [...] (BCZ), [...] (OTK), [...] (IMI), [...] (Mueller), [...] (BMA), [...] (HME), [...] (Halcor);

- France: [...] (KME-group), [...] (Wieland), [...] (BCZ), [...] (OTK), [...] (IMI),[...] (Mueller), [...] (BMA), [...] (HME), [...] (Halcor);

- Netherlands: [...] (KME-group), [...] (Wieland), [...] (BCZ), [...] (OTK), [...] (IMI), [...] (Mueller), [...] (BMA), [...] (HME);

- Spain: [...] (KME-group), [...] (others);

- United Kingdom: [...] (KME-group), [...] (IMI), [...] (Mueller), [...] (others);

(648) For the years 1999 and 2000, according to market intelligence of KME, the conversion margin situation developed as follows (in "D[E]M 0/0 kg") (975):

< emplacement tableau >

(649) At the beginning of October 2000 (see recital (407)), however, Wieland explained (978) that "through indexing alone the price targets set (for SANCO(r) Germany > DEM 200,-- /kg) probably cannot be achieved. A new price list will therefore follow at the end of the year." At the same time, price increases in Spain and in the United Kingdom were successful: "On 5.9.2000 Outokumpu introduced a new price list on the Spanish market. We have followed this new price list and the amended rebate structure. As far as we can see, our competitors have done the same. Depending on the marketing stage, rebates now lie between 46 and 52 %... Bonuses are no longer being granted. United Kingdom: In the UK, List 7 was introduced on the market at the end of September. List 8 will likely be published in October. According to information in our possession, the competition, apart from Mueller, has followed this development. Mueller seems to be allowing itself more time." (979)

(650) Already in mid-October (980) successes were reported from Spain, France, the United Kingdom and Germany (recital (412)): "Spain: We have been able over the past few weeks to get much closer to our price targets in Spain. The maximum rebate is 52%. We are still proceeding on the assumption that at the end of the year we shall achieve the reference figure of 100 euro. France: AFNOR is currently occupying itself with the half-hard grade. Talks are planned with the French producers; also with Mueller, of whom we know that one is able and willing to offer half-hard under no circumstances. The reference price is currently around 90 euro (rebate max. 66 %), so here too the chances of achieving 100 Euro by year's end are good.United Kingdom: the 100 Euro mark has already been reached. Germany: The reference price for non-SANCO tubes is DEM 153,- ; for SANCO tubes, DEM 200,- has already been reached. We shall try to raise the price level by a further DEM 50,-/100kg via the price indices as well as through a new price list". (981)

(651) KME's summary "New pricelist 2001 - introduction into the market" summarised the strategy of the parties (see recital (419)) and stated the consequences for price developments. The "added value" was analysed to develop between 23 August 2000 and the end of December 2000 (with the application of the new price list) as follows: From [...]kg to [...] in the category "[...] %", and from [...]kg to [...]kg in the category "[...] %". (982) On 8 December 2000, the parties achieved EUR [...]kg (German SANCO) and EUR [...] kg (non-SANCO) (983). A table including also 24 January 2001 suggests that the "added value" was further substantially raised in 2001 (984).

(652) An internal memo of Wieland dated 8 January 2001 analyses the success of the collusion (see recital (420)) (985): "Germany: Our new price lists were published before Christmas. They will be applied in January. With them we shall achieve the target of 100 for the standard measurement. - France: In this market the earnings target of 100 has been achieved. United Kingdom: 100 was achieved in the market, . Netherlands: - The price level is unsatisfactory. at EUR 75". (986)

(653) On 12 March 2001, shortly after the meeting, Wieland further commented on the success of the price collusion (see recital (426)) (987). "The main problem in the German market is now no longer the quality of earnings but the downward trend in quantities." "The new price list introduced in Spain at the beginning of September is taking hold only slowly, but with it processing earnings of EUR 100 are being attained with a 50 % rebate. In the United Kingdom, List 02/01 is, according to market observations, now being universally applied." "The Dutch market continues to cause problems". (988)

(654) In conclusion, there is clear evidence for very serious price effects of the cartel in the years 1994/95 and 1998 in all of the core markets (Germany, France, United Kingdom, Spain and Netherlands). Evidence for serious price effects can also be found for later years, in particular for 2000 and 2001.

Econometric evidence

(655) Two econometric studies were submitted by the parties (KME and IMI) to prove statistically, as well as economically, insignificant price effects of the infringement. The Commission received the raw data and estimation procedures employed by the parties. After a careful investigation of the two Reports, the Commission concludes that the Reports do not disprove serious price effects as shown in the in recitals (633)-(654). However, the Reports indicate that the impact on prices was most severe for the markets served by the KME subsidiary, especially for the product classes Cuprotherm and WICU. The arguments on which these conclusions rest will be discussed in the following recitals.

The KME-NERA Report

(656) One general weakness of the KME-NERA Report is the limited set of so called control variables - these are variables which control for other factors that influence the prices. The authors fail in finding control variables - beside the copper price, GDP and buyer power - representing shifts in demand, costs or in the market structure relevant for the industry concerned. Instead a broad set of dummies (for product family, brand, region and country) and time variables (a linear time trend, a country specific time trend and yearly fixed effects) is included. However, the authors do not provide any robustness checks allowing to assess whether these variables are correlated with the discussion variables (so called collinearity problem). In the case of collinearity, statistically insignificant estimates of the impacts of the alleged infringements just reflect the limits of the methodology applied to distinguish the effects of the infringement from other factors but not the insignificance of the infringement on prices itself.

(657) Furthermore, from the three substantiated control variables a significant and intuitive (negative) effect could only be estimated for the buyer bower variable. GDP is insignificant in 8 out of 9 model alternatives presented on page 34 of the report and the cost of copper has a counterintuitive negative effect on the full price of copper tubes for one subsidiary (Europa Metalli) (989). Given these results it is unclear why the consultants claim that "it is important to note that the parameter coefficients of the control variables included in these regression models have the expected signs" (page 33). For the subsidiary Europa Metalli at least the presented results cannot be considered as being based within an economically reasonable framework, given the negative effect of the copper price on the full price and no other substantiated control variable having a significant impact on prices beside buyer power (in one of the two presented model alternatives).

(658) Finally, estimations conducted by the Commission have shown that a slight change in the functional form - measuring the effects in linear form and not in a log-linear functional form - changes significantly the effects measured for the KME subsidiary:

In this case an average price increase of [...] % is estimated and for individual discussions price effects between [...] % to [...] % (all statistically significant at least at [...] % level) can be observed (990). The parties do not provide any economic argument why the log-linear functional form was applied or why a different representation - like the linear one - is not equally reasonable from an economic point of view.

(659) In sum, the KME-NERA Report does not in the Commission's view rebut the empirical evidence provided in the previous paragraphs based on the documentary evidence. It fails to find substantiated control variables representing shifts in demand or cost factors like labour, energy or capital cost. Given the broad set of dummy and time variables it is unclear whether price-effects of the alleged infringements are captured by these variables too, hence underestimating the actual price effects. For the KME subsidiary the data support price increases between [...] % to [...] % of the real full price. It has to be kept in mind that this is an average price increase in all periods and countries considered to be affected by discussions by the consultants (see Appendix D of the Report for a full list). That is, the results imply a permanent price increase for all European transactions listed in Appendix D of the KME subsidiary during 1991-2001 between [...] %-[...] % - a result perfectly in line with very serious price effects in some periods and some countries. However, the Report indicates that the impact on prices was most severe for the markets served by the KME subsidiary and here especially for the product classes Cuprotherm and WICU.

IMI-NERA Report

(660) The second econometric evidence put forward, the IMI-NERA Report, concentrates on the United Kingdom market only. It provides two different types of tests. Both tests undertaken compare the price before the Commission's cartel inspection (period from October 1989 to April 2001) with the price after the inspection (period April 2001 to December 2002) after controlling for several control variables.

(661) Both tests rely on the assumption that after the Commission's inspections prices converged back to the competitive price level in the short term. If the cartel agreements had had a lasting effect after the Commission's investigation, both statistical tests provided in the report would have significantly underestimated the price effects of the agreements. The relatively short time period available for comparison (April 2001 to December 2002) limits the power of the tests. Furthermore, the price predictions provided by the consultants show an increasing divergence between the actual price and the predicted price from beginning 2002 on - indicating significant price divergence in the mid- to long term (figure 4.1 at page 26 of the report).

(662) Furthermore, the report states that - with respect to its first set of models - "throughout test 1, the only explanatory variable that has a robust sign and is always statistically significant is the LME copper price" (page 19 of the IMI-NERA report). Test 2 finds one additional significant variable - the gas price index. Whether its influence remains robust in the various model alternatives estimated is, however, not further explored by the authors. Hence, the Report fails to explain effects of other variables on the copper tubes price other than the copper price. Taking this result seriously, it means that any changes of the conversion margins (which is the copper tubes price minus the copper price) cannot be explained by cost or demand changes - in support of the evidence provided in previous paragraphs.

(663) In addition, the lack of substantiated control variables reflecting shifts in demand or costs limits the accuracy of prognosis. For instance, the industry data provided by the KME-NERA report presents a significant structural break in the year 2001 in the United Kingdom market (Figure 2.4 on page 8 of the KME-NERA report). Based on these figures "apparent consumption of copper water tubes" in the United Kingdom dropped by 40-50 % in comparison to year 2000, "domestic production and exports to Member States of copper water tubes" in the United Kingdom declined by roughly 80 % in comparison to year 2000 (991). The effect of this structural break - given the lack of substantiated control variables - on the two tests are not addressed by the authors and limits their explanatory power.

(664) Finally, the IMI-NERA Report assumes a constant effect of the cartel discussions on prices. The Commission however considers - beside the legal fact of one single infringement - the intensity of discussions to vary significantly, resulting in very serious price effects in some periods. Such a more complex price effect of the cartel can only be poorly measured by a variable assuming a constant permanent price effect. For instance, the Commission conducted its own analysis by allowing a linear price increase over the period April until November 1994 assuming a lasting effect thereafter. Such an approach is consistent with the factual evidence cited before (see the discussion in recital (639)). Taking the model as provided by the parties in Table 3.2 of the IMI-NERA Report but replacing the "cartel dummy" variable with the newly created "cartel dummy" variable shows a statistically significant (at 5% level), and positive, price effect of the discussions during these periods. The estimations predict a price increase by 0,0565 GBP/kg per month over this period, resulting in a price increase of 0,452 GBP/kg from April until November 1994. This result is consistent with the results observed earlier through a simple price comparison during this period. Furthermore, evidence presented in the facts (see recital (639)) supports the presumption that these price increases are being induced by the alleged infringements.

(665) In sum, the econometric evidence provided by the IMI-NERA Report is, first, based on the weak assumption that the cartel discussions had non-lasting effects after the Commission's investigations; second, it fails to provide demand or cost factors influencing the copper tubes price other than the copper price; and third, it assumes a constant price effect of the discussions. Hence, it cannot overrule the clear factual evidence provided, for example, in the recitals (277) and (639), and even supports this evidence with regard to the United Kingdom market, as it shows that changes in the conversion rate cannot be explained by changes in demand or cost parameters.

(666) With respect to the exchange of sensitive information as regards WICU and Cuprotherm products, KME confirms that such an exchange occurred in particular among Wieland and KME concerning Germany. However, KME stresses the small significance of WICU and Cuprotherm tubes, and correspondingly the negligible importance of these products for this investigation. During the years 1989 until 2001, the share of KME's WICU and Cuprotherm tubes in the EEA copper plumbing tube market (including Switzerland) was on average between [...] % (992). However, the Commission notes that the combined share of Wieland and KME on the EEA market for plastic-coated copper plumbing tubes amounted to [...] % in value or [...] % in volume. Both parties generated a large extent of their sales in Germany, the market on which their cooperation focused. According to the NERA report submitted by KME, prices charged by KME for WICU and Cuprotherm tubes increased as a result of its contacts with competitors concerning plastic coated copper plumbing tubes in the entire EEA.

The impact on profitability

(667) The Commission notes that general arguments of the parties are not sufficient to rebut the detailed and contemporaneous evidence presented by the Commission in the Statement of Objections, which is summarised in recitals (633)-(654).

(668) The studies and the arguments provided by the parties indicating that the copper plumbing tube had a low average profitability over the infringement period 1989-2001 fail to prove the absence of an impact of the cartel on the market. The documentary evidence described in Part D and recalled in recitals (639) to (653) shows that at these precise points in time certain price increases decided by the cartel were successfully implemented and led to substantial additional conversion margins, incomes and/or profits. These occurrences are sufficient to demonstrate that the evolution of prices and profits would have been different in the absence of the cartel. It should also be added that the low average profitability is clearly compatible with the concept of a defensive cartel as submitted by Wieland and with an industry with high exit costs in which it is preferable to suffer (some) losses than to close the business. It should be recalled that even a negative profitability (in economic terms: losses) is not a proof of absence of impact if the cartel made it possible to reduce or minimize such losses. Now clearly, the occurrences listed by the Commission show that the cartel obtained revenues significantly different from a competitive situation.

(669) The Commission notices that with respect to the comparison of US and European conversion margins, the data submitted by KME is ambiguous. First, according to the table (993) submitted by KME, until mid 1997 the conversion in Germany (SANCO) was higher that in the US. Second, with respect to the conversion margin from mid 1997 until the end of the cartel in 2001, the Commission notes that the German conversion margin (SANCO) indicated in the table does not correspond to internal data found at the premises of KME and quoted in the Statement of Objections (see recital (648)). Based on the assumption, however, that the US conversion margin figures were accurate, for a limited period of time, from mid 1997 until March 2001, the US conversion margin level was higher than in Germany. This, however, does not demonstrate that the cartel did not have effects on prices in the European market. In this context, it is important to note that according to the table, right after the cartel terminated, the German conversion margin (SANCO) steadily decreased between mid 2001 and mid 2003 by approximately 50 %.

Conclusion

(670) The Commission concludes that prices would have developed in a different way, either eroded more or increased less, as the case may be, in the absence of the anti-competitive agreements.

(671) Even the reported failures to achieve the target prices are far from rebutting in any convincing manner the Commission's demonstration and from proving that the implementation of the cartel agreement could not have played any role in the setting and fluctuation of prices in the copper plumbing tubes market. The fact that in spite of the cartel's efforts the results sought by the participants were not entirely achieved may illustrate the difficulties encountered by the parties in increasing prices in a specific market situation, but it does not prove in any way that the cartel had no effect on the market, or that prices were not kept at an artificial level. It should also be borne in mind that "not entirely achieved" involves a certain degree of success and that the subsequent initiatives were designed to complete the efficiency of what had partially succeeded. The Commission also considers that the impact of a cartel is not limited to prices, especially where the object of the anti-competitive behaviour also concerns market allocation. In this case, one of the objectives was to stabilise market shares, as Outokumpu, KME, Mueller and Wieland Werke have explained (see recitals (199)-(226)). None of the participants have contested the Commission's finding in the Statement of Objections that the market shares remained relatively stable throughout the period of the infringement, although parties stressed that customers fluctuated between the participants (994) (see table in Annex).

(672) Furthermore, it must be pointed out that the effects of the cartel cannot only be assessed given the actual market structure and capacities employed, but that keeping firms and capacities in the market and preventing markets from opening towards foreign competition are also negative impacts on the market to be taken into account. Indeed, part of the arrangements concerned the organisation of a mechanism of market segregation: national markets were given a market leader who would decide the price variations; price levels were decided for other national producers and the importers of the various types of products, provoking de facto an "organised competition". Such a segregation made it possible to maintain different price levels in the different domestic markets that would have not existed (or not to that extent) in the absence of the cartel. In a competitive market environment, such a segregation would have been limited by the the established attitude of the various producers to attempt to gain volumes in the other countries in order to compensate the losses of volumes in their domestic markets.

(673) In the light of the foregoing and the efforts put by each participant into the organisation of the cartel, there is sufficient proof that, regardless of the fact that implementation was from time to time disturbed by the parties' deviation from the agreed principles, the anti-competitive scheme has overall had an impact on the market, although it is not possible to quantify it precisely. The Commission considers that the parties concerned by this Decision have not been able to rebut its finding as to the actual impact of the infringement on the copper plumbing tubes market in the EEA.

20.3.2. The size of the relevant geographic market

(674) Outokumpu argues that since Norway, Sweden, Finland and Austria only joined the EEA in 1994, any unlawful activity relating to these countries between September 1989 and that date falls beyond Community/EEA jurisdiction. Accordingly, Outokumpu's market share should be determined taking this fact into account when assessing the impact on the market. (995) However, with respect to the calculation of the basic amount of the fine, in the light of the Commission's jurisdiction, the EEA should be the appropriate geographic scope for any fine assessment. (996) Halcor underlines that it only participated in arrangements covering five Member States and that it realized a large amount of its sales outside those five Member States. (997)

(675) It is important to note that the cartel covered the whole of the Community and, following its creation, the EEA. Since 1996, the cartel focused on discussing five Member States accounting for roughly 72% of consumption in the EEA. At the same time, the parties had a common understanding not to compete on each others domestic markets. Discussions therefore focused on those five Member States, where cooperation was most necessary to avoid competition.

(676) In this context, the Commission takes into account the territory affected by the cartel, that is to say the geographic extent of the copper plumbing tubes business, as a whole and not the territorial scope of the activity of each individual undertaking. Accordingly, OTK's argument relating to its activities in Norway, Sweden, Finland and Austria prior to their entry into the EEA must be rejected. With respect to Halcor's arguments, it is considered that a cartel controlling the Community/EEA-wide copper plumbing tubes market, and in particular national areas in which prices are above average, necessarily affects competition in the entire geographic market concerned, which is at least the EEA. This general consideration is all the more true in this case, where Halcor also entered into the cartel to protect its domestic market (see especially the considerations of Halcor's representative, recital (326)), just like Buntmetall, Boliden and, for example, Outokumpu, KME and Wieland entered into the cartel to protect their domestic markets from competition (see recital (290)). The note of Halcor's representative further shows that as part of its bargain on market shares, also the Greek market was discussed. Halcor therefore benefited from the cartel as other participants.

(677) As far as the WICU/Cuprotherm arrangements are concerned, it is important to note that, according to the KME-NERA Report, prices charged to customers for these products, no matter in which geographic area within the Community/EEA, were affected by these arrangements.

20.3.3. The Commission's conclusion on gravity

(678) The Commission also takes into account that plain and plastic coated copper plumbing tubes are a very important industrial sector, with an estimated market value in the EEA of EUR 1151 million, based on the total price of copper plumbing tubes charged to customers in 2000, which was the last full year of the infringement. The estimated market value is calculated by adding to the accumulated turnover of the parties the turnover of main competitors. In 2000, the estimated EEA market value of plain copper plumbing tubes amounted to approximately EUR 970 million. Copper plumbing tubes are widely used. The cartel may potentially have affected a large proportion of households in the EEA.

(679) The arguments of the parties according to which turnover is not a relevant measure of the importance of the market, or that the infringement took place only on the conversion margin, suggesting therefore that the price of copper should not be considered, are not acceptable. The Court of First Instance ruled that: "as is confirmed by the case-law on the application of Article 85(1)(a) of the EC Treaty, the prohibition of agreements and concerted practices which directly or indirectly fix prices also extends to agreements relating to the fixing of a part of the final price (see, in particular, Case T-29-92 SPO and Others v Commission [1995] ECR II-289, paragraph 146). It follows, in particular, that AST's argument that most of the final price of stainless steel was not the subject of an agreement is irrelevant" (998). Hence, in this case an agreement on part of the price of tubes is an agreement on the entire price of tubes. When evaluating the economic importance of the industrial sector affected by an infringement, it would not be justified to subtract the price of raw materials, irrespective of how the price of such raw materials is formed. It is also important in this respect that the coordination of price lines, which include the copper price within a predefined range, served as an instrument to increase the conversion margin for the tubes (see, for example, recital (206)).

(680) Taking all the foregoing factors into account, it can be concluded that the undertakings concerned by this Decision have committed a very serious infringement of Article 81(1) of the Treaty and Article 53(1) of the EEA Agreement.

20.3.4. Differential treatment

(681) Within the category of very serious infringements, the scale of likely fines makes it possible to apply differential treatment to undertakings in order to take account of the effective economic capacity of the offenders to cause significant damage to competition, as well as to set the fine at a level which ensures that it has sufficient deterrent effect. The Commission notes that this exercise seems particularly necessary where there is considerable disparity in the size of the undertakings participating in the infringement. For this purpose, the undertakings concerned can be divided into different categories according to their relative importance in the market concerned, subject to adjustment where appropriate to take account of other factors and especially the need to ensure effective deterrence.

(682) In the circumstances of this case, which involves several undertakings, it will be necessary, in setting the basic amount of the fines, to take account of the specific weight and therefore the real impact of the offending conduct of each undertaking on competition. In this context, the specific weight is distinguishable from the importance of the undertaking in question in terms of its size or economic power. The proportion of turnover derived from the goods in respect of which the infringement was committed is likely to give a fair indication of the scale of the infringement on the relevant market (999). Whilst an undertaking's market shares (based on turnover or sales volume) cannot be a decisive factor in concluding that an undertaking belongs to a powerful economic entity, they are nevertheless relevant in determining the influence which it may exert on the market affected by the infringement (1000). Moreover, the market share of any given party to the cartel also gives an indication of its contribution to the effectiveness of the cartel as a whole or, conversely, of the instability which would have affected the cartel had it not participated.

(683) As the basis for determining the relative importance of the undertakings in this infringement, the Commission considers it appropriate to take into account the respective turnover of each undertaking with the concerned product according whether each party participated in the infringement with respect to both plastic and plain copper plumbing tubes or only plain copper plumbing tubes. Accordingly, only with respect to the KME group and Wieland, the combined turnover with plain and plasticcoated copper plumbing tubes in the EEA is used. This approach is justified by the fact that the cartel was primarily EEA-wide, and that its object was inter alia to fix prices and allocate markets in the most important markets of the EEA, that is to say France, Germany, the Netherlands, Spain and the United Kingdom. The comparison is made on the basis of the product market share in the EEA in the last full year of the infringement (2000):

<emplacement tableau>

- The parties' arguments

(684) KME argues that KME's importance in the copper plumbing tube market is less significant than stated in the Statement of Objections. Since it is the largest producer in two large European markets, France and Germany, it has necessarily also a large market share in the EEA. Furthermore, the individual market shares of KME, TMX and EM are not higher than the market shares of other producers (see already recitals (559) and (560)). KME argues that in particular until the restructuring of KME in 1995, KME, TMX and EM competed against each other, running their own sales network in various domestic markets until 1999. The KME-NERA Report confirms that KME, TMX and EM did not follow a single commercial policy. To assess the group's importance on the basis of the combined market shares of KME, TMX and EM would significantly overstate their "real impact" on competition in the copper plumbing tube market. (1001)

(685) Wieland believes that Wieland's market share is relatively small (1002). Outokumpu submits that the Commission should consider the average market shares of the companies concerned over the period of the infringement in order to assess the specific weight of each company (1003). While Outokumpu in 2000 held an EEA market share of some [...] %, the KME group reached [...] % and Wieland [...] %. In past years, the difference has been much greater (1004). Furthermore, SANCO producers jointly held some [...] % (1005). In Outokumpu's view, the distinction between the SANCO arrangements and the broader arrangements should be reflected in the Decision and the possible fines (1006). Last, not all copper producers were in the cooperation for the whole period. Only after 1998 did nine companies participate (1007). IMI argues it should be treated in the same way as BMA, HME and Halcor, and, at the very least should not be placed in the same category as the SANCO club members (1008).

(686) Wieland argues that the combined share of the cartel members amounted only to around 68% of the plain copper plumbing tube market (1009).

(687) Halcor submits that in the five markets, which were the subject of the arrangements, Halcor's position was marginal or non-existent. Halcor did not have effective economic capacity to cause significant economic damage to other operators. In the meetings it played a minimal role and kept a low profile. According to Halcor, it had no intention of respecting "any of the dictates" of the major producer. Rather it attended some of the meetings "in order to avoid a hard confrontation and to have a better idea of what these large companies were plotting to do". In Halcor's view its "input" was strictly a one-way process while the arrangement operated entirely for the benefit of the "old members", and Halcor derived none (1010). Halcor submits that it had no commercial or economic motivation for its participation other than to avoid the execution of the threats that had been made against it (1011). Accordingly, it should be put in the lowest group (together with BMA). In Halcor's view, simply using market shares as the determinant of the respective starting points for the groups would be too rough; the scale of any infringement allegedly attributed to Halcor would therefore require any potential fine imposed upon it to be based on a higher differential, in comparison with the starting point for the largest undertaking, than the simple arithmetical ratio between their respective market shares. Given the tighter cooperation between the group of the five, they had a bigger weight on the market.

The Commission's view

(688) The Commission considers that for the purposes of evaluating the relative weight of the participants within the affected geographic area, it is generally appropriate to take into account their market shares of the product in question in the last full year of the infringement, that is to say the year 2000 in this case (see table in recital (683)).

(689) The Commission points out that although SANCO producers had separate arrangements, it could not be proven that their cooperation was substantially tighter during the years 1988-1995 than that one on the broader European level (see in this respect recital (126)), in particular since the description of KME contained in recitals (204)-(207), concerning price cooperation, was not contested by any of the parties. In the years after 1995 until 2001, the cooperation only involved the KME group and Wieland and mainly focused on the German market. Accordingly, the cooperation between SANCO producers does not justify a different treatment.

(690) The fact that a more intense cooperation existed among the group of the five than among the group of nine does not justify any different treatment of the four producers that did not participate in all meetings. The four producers, which did not belong to the group of five because of their size, profited proportionally in the same or a similar way as the other five competitors from the increased price level during the years 1998 until 2001 (see above 20.3.1.2). They were not, as Halcor suggests, victims of the arrangements but contributed through their participation to the success of the cartel as a whole. Their participation can be considered as important for the functioning of the cartel, since four additional competing producers could have rendered the cartel substantially less sufficient. The limited impact of their participation due to their size is therefore appropriately taken into consideration by groupings that are based on market shares. Halcor's claim to have been a victim of the group of five has already been rejected (see above Section 14.4).

(691) Wieland's argument that the combined share of the cartel members amounted only to around 68% of the plain copper plumbing tube market, has to be rejected. It is not contested by any party that also during the period from 1989 until 1994, at least Halcor, Mueller's subsidiaries and HME participated for a considerable time period in the infringement. [Information not relevant to the EU proceedings]]Therefore, it has to be concluded that the cartelized share of the plain copper plumbing tube market was for a considerable time period close to the share of the nine companies during 1998.

(692) The Commission notes that the product-group affected by the single infringement of this case was plain and plastic-coated copper plumbing tubes. Consequently the combined market size should be used as a basis for calculating the groupings. The Commission notes that this approach is more advantageous for all parties, and in particular for those that did not participate in arrangements concerning plastic-coated copper plumbing tubes. The plain copper plumbing tubes sales should be used for the purposes of grouping the companies with respect to those companies that participated only in arrangements concerning plain copper plumbing tubes. With respect to the KME group and Wieland also sales of plastic-coated copper plumbing tubes have to be taken into consideration. With a total share of around [...] % of the plain and plastic-coated copper plumbing tubes market in 2000 based on value, the KME-group is the largest player on the EEA market for copper plumbing tubes and will therefore be placed in the first category. IMI, Mueller, the Wieland-group and Outokumpu, having relevant market shares (based on value) between [...] % and [...] %, are placed in a second category, consisting of companies that can be considered as medium-size operators within the EEA market of copper plumbing tubes. Boliden with a market share of [...] % forms the third group. HME (year 2000) and Halcor (year 1999), with market shares of around [...] % (based on value), are placed in a fourth category, consisting of small operators within the EEA market of copper plumbing tubes.

(693) On the basis of the foregoing, the appropriate starting point for a fine resulting from the criterion of relative importance in the plain and plastic-coated copper plumbing tube market concerned is for each category as follows:

- KME-group: EUR 70 million

- Wieland Werke: EUR 23,8 million

- IMI-group: EUR 23,8 million

- Mueller-group: EUR 23,8 million

- Outokumpu-group: EUR 23,8 million

- Boliden-group: EUR 16,1 million

- Halcor: EUR 9,8 million

- HME: EUR 9,8 million

(694) As EM and TMX formed a single undertaking in the period 1989-1995 (recital (565)), they are jointly and severally responsible for the respective part of the infringement. Similarly, KME AG, EM and TMX formed a single undertaking (the .KME-group") in the period 1995-2001 (recital (566)), and they are jointly and severally responsible for that part of the infringement. The Commission cannot therefore take possible intragroup competition into consideration when assessing the relative weight of the participants in the cartel in the period following the restructuring. Rather, the basic amount of the fine should be divided in two parts, one for the period 1988-1995 and one for the period 1995-2001. The first part (EUR 35 million) should be divided into two equal parts between KME AG, on the one hand, and EM and TMX (jointly and severally), on the other hand. The second part (EUR 35 million) should be attributable jointly and severally to KME AG, EM and TMX.

(695) After July 1999, Buntmetall and Wieland Werke AG formed a single undertaking (Wieland-group). For this period they should be held jointly and severally liable. Before July 1999, Buntmetall and Wieland Werke AG should be held separately liable for their respective involvement in the infringement. Accordingly, the starting amount should be divided in two parts, each of which reflects the duration of involvement, one for the period September 1989 until July 1999 (EUR 19,52 million) and one for the period July 1999 until March 2001 (EUR 4,28 million). The part for the period September 1989 until July 1999 is further divided in two parts, one for Buntmetall and one for Wieland Werke AG, taking again into consideration the respective duration of involvement.

(696) This division should be as follows:

- EUR 35 million for KME-group (jointly and severally among KME AG, TMX and EM);

- EUR 17,5 million for KME AG,

- EUR 17,5 million for EM and TMX (jointly and severally)

- EUR 3,25 million for Wieland-group (jointly and severally among Wieland and Buntmetall)

- EUR 19,52 million for Wieland Werke AG

- EUR 1,03 million for Buntmetall

(697) On this basis, the appropriate starting point for a fine should be for each undertaking as follows:

- Halcor: EUR 9,8 million

- HME: EUR 9,8 million

- Boliden-group: EUR 16,1 million

- IMI-group: EUR 23,8 million

- Mueller-group: EUR 23,8 million

- Outokumpu-group: EUR 23,8 million

- Wieland-group: EUR 3,25 million

- Wieland Werke AG: EUR 19,52 million

- Buntmetall-group: EUR 1,03 million

- KME AG: EUR 17,5 million

- KME-group EUR 35 million

- EM and TMX EUR 17,5 million

20.3.5. Deterrence

The parties' arguments

(698) Wieland and Outokumpu submit that the principle of proportionality has to ensure that the calculation of the fine takes into consideration the share of the copper plumbing tube business within the company, and is not based on other unrelated activities (1012). Wieland claims that its company is considerably smaller (EUR 1,225 billion; the turnover further decreased) than both KME (EUR 2,054 billion) and Outokumpu (EUR 5,558 billion) (1013). The share of conversion margin turnover realised with Wieland's copper plumbing tube business in the EEA amounts to [...] % of the total turnover of Wieland (1014). Outokumpu's copper business generated less than [...] % of its total turnover with the sale of copper plumbing tubes in the EEA (1015).

(699) In addition, Outokumpu questions the fairness of the Commission's general approach in fining .for deterrence. by looking at the turnover of a wider business unit, covering other businesses than that specifically involved in an infringement. This would unfairly penalise larger companies involved in many markets entirely unrelated to the infringement. In Outokumpu's view, to use the turnover of Outokumpu Oyj (the group) as a basis for the calculation of the fine would be patently unfair and disproportionate. In Outokumpu's view, its involvement is much less than others, and Outokumpu appears to have gained much less than others (1016). Fines should only be based on involvement and impact, not on whether companies are conglomerates or not (1017).

(700) KME submits that it is a medium-sized undertaking with limited legal and economic resources.

The Commission's view

(701) In order to ensure that the fines imposed have a sufficient deterrent effect and take into consideration the fact that large undertakings have a legal and economic knowledge and infrastructures which enable them more easily to recognise the illegal nature of their conduct, the Commission may adjust the starting amount of fine. For this purpose, total turnover is the figure, which gives an indication of the size of the undertaking and of its economic power, which must be known in order to assess whether a fine will deter it (1018).

(702) The Court of First Instance has approved the Commission's approach consisting in applying a multiplying factor. In a recent judgement, it stated that insofar as the amount of fine "was further multiplied by 2.5 in order to take into account the applicant's position as a European group, that weighting was not applied on the basis of the applicant's total turnover" and that "the multiplier of 2.5 has no proportional link with the difference between the applicant's and the other undertakings' total turnover" (1019).

(703) In this case, the Commission considers it appropriate to apply a further upward adjustment in Outokumpu's case to take account of its size and overall resources. In this assessment, it is appropriate to take into account the overall world-wide turnover of the group (over EUR 5 billion), since, according to Outokumpu, the parent company (Outokumpu OYj) itself initiated the infringement in 1987/1988 and has controlled 100 % of the capital of OCP throughout its involvement in the cartel. Therefore, the starting amount of its fine determined in recital (697) should be multiplied by 1.5 to EUR 35,7 million.

(704) The fact that Outokumpu was given a deterrence factor (of the same magnitude) in the Industrial tubes Decision does not constitute a violation of the ne bis in idem principle because the multiplying factor applied to the starting amount of the fine itself is linked to the starting amount determined for a specific product and the share of the undertaking in the product subject of the Decision. Should the Industrial Tubes Case (1020) and this one have been decided in one single decision, one fine would have been imposed on Outokumpu for each cartel. Each of the fines for these cartels would have been adjusted on its own merits, including deterrence.

(705) Concerning the argument that undertakings which have recently grown in turnover would be unfairly penalised in terms of deterrence, the Commission considers that, as deterrence aims at preventing future infringements, it appears appropriate to look at the size of the undertaking when the fine is imposed and not when the infringement was committed. In a recent judgement, the Court of First Instance has established that: "In the light of the case-law referred to at paragraph 239 above, the Commission was therefore entitled to take the view that, owing to its enormous worldwide turnover by comparison with the turnovers of the other members of the cartel, SDK could more readily raise the necessary funds to pay its fine, which, if the fine was to have a sufficiently deterrent effect, justified the application of a multiplier. None of SDK's arguments to the contrary can be upheld" (1021). Given the size of the total fine in percentage of the turnover, the 1,5 multiplier does not appear to be disproportionate.

20.3.6. Duration of the infringement

(706) As discussed in recitals (463)-(497) and (583)-(597) the infringement involving the addressees of this Decision, that is to say Boliden group, Buntmetall group, Halcor SA, HME, IMI group, Mueller-group, Outokumpu group, Wieland Werke, KME AG, TMX and Europa Metalli, with different companies involved for different periods, started at the latest on 3 June 1988 and continued at least until 22 March 2001. The following companies committed a continuous infringement for the respective duration indicated:

KME-group: 5 years, 7 months

in addition:

- KME: 7 years, 2 months;

- EM/TMX: 5 years, 10 months

- Wieland-group: 1 year, 8 months

in addition:

- Wieland Werke AG: 9 years 9 months

- Buntmetall-group: 10 months

- IMI-group: 11 years, 5 months

- Mueller-group: 3 years, 2 months

- Outokumpu-group: 11 years, 5 months

- Boliden-group: 12 years, 9 months

- Halcor: 12 months

- HME: 2 years, 6 months

(707) KME, Outokumpu and Wieland argue that with respect to the determination of the fine, quiet periods should either have as an effect that the previous years are timebarred (Article 1(b) of Regulation (ECC) 2988-74) or, at least, that the quiet years are taken into consideration as a suspension (1022). KME believes that in view of the less organised and partly discontinuous nature of the contacts, the period between 1988 and 1994 should only partially be considered. Wieland, KME and Outokumpu point to low intensity periods concerning in particular the period from 1990 until 1992 and from 1994 until 1997. In particular during these periods, normal, or almost normal, levels of competition existed (1023). KME believes that the Commission's practice should be in accordance with the Case Luxemburg Breweries (1024), where the fine for a cartel lasting 14 years was increased by only 100 % (1025).

(708) Boliden, as explained above, stressed that after 1995 it did not participate in SANCO meetings or EDWD meetings until 27 or 28 August 1998, when its participation started, lasting until its formal withdrawal on 10 December 1998 (see recital (344)). In October 1999 (see recital (382)), it re-entered the EDWD meetings until March 2001.

(709) Outokumpu submits that, with respect to the fines, it should also be taken into account that it sought to distance itself from the cooperation in 1998. (1026)

(710) It must be noted that the Commission's practice in previous decisions does not itself serve as a legal framework for the fines imposed in competition matters, since that framework is defined solely in Regulation No 17 (1027) (and now in Regulation No 1-2003). Hence, KME's argument in recital (707) based on a previous Commission Decision must be rejected. The current policy of the Commission for cartel cases is to increase the fines by 10 % per year for infringements exceeding five years. This has resulted in increases in the amount of the fine for duration of more than 100 % in several recent cases (1028).

(711) With respect to OTK's alleged attempt to distance itself from meetings in 1998, it is noted that OTK continued to follow the arrangements without attending meetings (see recitals (326) and (329)). This fact is not contested by OTK. In view of OTK's commitment to continue its cooperation, OTK is considered to have fully participated in the arrangements in 1998.

(712) Outokumpu argues that it qualifies for a reduction on the basis of cooperation outside the scope of the 1996 Leniency Notice, referring to the Industrial Tubes Case where the Commission followed the line adopted in Commission notice on immunity from fines and reduction of fines in cartel cases (1029) ("2002 Leniency Notice") and did not increase the fine for duration or gravity, as far as one of the parties provided evidence relating to facts previously unknown to the Commission (1030). Accordingly, the period from September 1989 to the end of 1997 (at least until 16 September 1997) should be subtracted from the duration in its case, since it was the first to admit that the arrangements took place during those periods and to submit a list of meetings that closed the gap in the Commission's file concerning the first period. These arguments are addressed in recitals (758) and (759).

(713) As indicated in recitals (564) and (565), KME AG was a separate undertaking from that formed by EM and TMX during the period from 8 June 1988 to 19 June 1995. The increase for duration should therefore be calculated separately for those two undertakings for that period of seven years. For the rest of the duration from 20 June 1995 to 22 March 2001, in other words five years and eight months, the increase should be common for the whole KME-group. In the same way, the duration was calculated for the Wieland-group on the one hand and Wieland-Werke AG and Buntmetall on the other (see recitals (695)).

(714) The starting amount of the fines determined for gravity should therefore be increased by following percentages:

- KME-group: 55%

in addition:

- KME: 70 %

- EM/TMX: 55%

- Wieland-group: 15%

in addition:

- Wieland Werke AG: 95%

- Buntmetall-group: 0 %

- IMI-group: 110 %

- Mueller-group: 30 %

- Outokumpu-group: 110 %

- Boliden-group: 125%

- Buntmetall-group: 10 %

- Halcor: 10 %

- HME: 25%

20.3.7. Potential fines in parallel proceedings

(715) Wieland, KME and Outokumpu have requested that the Commission should consider the fact that they have been the subject of an additional fine in the parallel proceeding concerning industrial tubes (Case COMP/E-1-38.240). In support, KME refers to the Commission decision in Specialty Graphites (1031) in which the fine imposed on one of the companies was reduced by 33% to take account of its delicate financial position and the fact that it recently had received a large fine.

(716) Wieland submits that the separation of the two Cases, Industrial and Copper Plumbing Tubes, was artificial and should in any case not lead to any increased fine compared to the one that would have been imposed had both cartels been addressed in a single decision. Deterrence is already ensured by the fine in the Industrial Tubes Case (1032). Equally Outokumpu claims that it should not face a double penalty because of its involvement in the two Copper Tubes Cases. Both the Industrial Tubes and the Copper Plumbing Tubes cartel had their reason in the poor economics of the copper tubes sector. Second, it would be unfair to consider the turnover of the larger part of Outokumpu's business for the basic amount twice, that is to say in both Cases. Such an approach would lead to a double penalty for deterrence (1033). Coming at the same time, neither fine will deter the other activity. There should be a reduced assessment for deterrence to avoid a double penalty (1034).

(717) The Commission considers that the fact that the Commission is conducting investigations on several cartel Cases in which the same undertakings are involved (although the Cases were initiated as one and later separated) does not prevent it from imposing, if appropriate, the maximum amount in each infringement. The splitting was decided when it became apparent that the infringements were different, performed by different players and concerned different products. In any event, unless sufficient deterrence is already achieved, an obligation on the Commission to take account, when determining the fine, of an undertaking's participation in multiple cartels would be tantamount to conferring an unjustified advantage on undertakings who commit multiple parallel infringements. Each separate infringement merits a separate fine. If not, an undertaking involved in one or more cartels would have nothing to lose by entering into further cartels. It could then derive unjustified profits from additional cartels without any risk of a fine for that behaviour. Imposing a fine for each separate infringement serves to deter such behaviour. Outokumpu's arguments were sufficiently addressed under (704).

(718) It should be noted that in its Specialty Graphites Decision, the Commission reduced the amount of the fine imposed on an undertaking because the undertaking in question was in a serious adverse financial situation and a significant fine had relatively recently been imposed on it by the Commission. The Commission considered that, in those particular circumstances, imposing the full amount of the fine did not appear necessary in order to ensure effective deterrence. That conclusion took, in particular, account of the fact that the aggravating circumstance of recidivism did not apply to the undertaking in question. Apart from the Case of KME, which will be discussed below (see Section 20.8), none of these conditions apply in this Case.

20.3.8. Conclusion on the basic amounts

(719) The basic amounts of the fines should therefore be as follows:

- KME-group at: EUR 54,25 million

in addition:

- KME at: EUR 29,75 million

- EM/TMX: EUR 27,13 million

- Wieland-group: EUR 3,74 million

in addition:

- Wieland Werke AG: EUR 38,06 million

- Buntmetall-group: EUR 1,03 million

- IMI-group at EUR 49,98 million

- Mueller-group at EUR 30,94 million

- Outokumpu-group at EUR 74,97 million

- Halcor at EUR 10,78 million

- HME at EUR 12,25 million

- Boliden-group at EUR 36,225 million

20.4. Aggravating circumstances

(720) The repeated infringement by Outokumpu constitutes an aggravating circumstance. Outokumpu was addressee of Commission Decision 90-417-ECSC Cold-rolled Stainless Steel Flat Products (1035) (hereinafter "Stainless Steel Case").

(721) Outokumpu contests, however, the Commission's finding of recidivism in its regard as addressee of the Decision in the Stainless Steel Case. According to Outokumpu, that case involved a very different situation, since it was in a quasi.public context where Outokumpu was acting under government influence and in the belief that the arrangements were publicly endorsed. It also argues that pressure was initiated by the institutions of the Community itself. In such circumstances, the Commission accepted that this was not a straightforward infringement and imposed no fine (1036). Outokumpu submits that the conditions set out in the Thyssen Case are not met; in that case the Court of First Instance stated that .recidivism, as understood in a number of national legal systems, implies that a person has been committing fresh infringements after having been penalised for similar infringements. (1037).

(722) As further argument, Outokumpu stresses that different businesses were concerned, involving different units and employees in different locations, as well as a different treaty provision (Article 65 of the ECSC Treaty). In OTK's view, therefore, to link the two infringements (one in copper plumbing tubes and the other in stainless steel) would be to penalise a company with many different businesses in comparison to smaller companies with only interests in copper tubes. OTK also notes that the Commission's Decision in the Stainless Steel Case was adopted in July 1990, some two years after the starting date of the infringement in industrial tubes sector (September 1989), and therefore cannot form the basis for recidivism for the earlier period (1038).

(723) In conclusion, OTK maintains that fining for deterrence and recidivism would be unfair and disproportionate, since the infringements are not the same nor have they been committed in the same business. It therefore considers that if the Commission increases fines for deterrence twice (in the Industrial Tubes and Sanitary Tubes Cases) and again for recidivism (by making a link to a completely unrelated business), it would mean that Outokumpu could end up with a triple penalty, just for being a large company with many operations in many different sectors (1039).

(724) The Commission considers that a repeated infringement occurs when an undertaking, which has been addressee of a Commission Decision in the past as party to an infringement, is later found responsible for another infringement of the same type. In addition to ordering the undertaking to end the infringement, the function of such a Decision is to warn and deter the undertaking in question from committing similar infringements in future, even if for some reason no fine is imposed. The Commission also considers that as the Decision in the Stainless Steel Case was adopted after the infringement in the copper plumbing tube sector had already been initiated, Outokumpu's executives involved in the latter should have taken measures to end the infringement. Continuing it after being warned by a Decision in a different product sector amounts to recidivism.

(725) This Decision concerns the same type of infringement as the Stainless Steel Case which concerned fixing of quotas and prices to control production and share markets (1040). As for Outokumpu's argument relating to a different treaty provision, it is sufficient to recall that according to case law Article 65 of ECSC Treaty is equivalent to Article 81(1) of the Treaty (1041).

(726) That Outokumpu continued its infringement in the copper plumbing tubes sector after being ordered to end its infringement in the stainless steel sector by a Commission Decision clearly shows that the previous Decision did not have a sufficiently deterrent effect on Outokumpu's market behaviour. Hence, future deterrence, with respect to Outokumpu but also with respect to other potential companies in a similar situation, has to be ensured by increasing the amount of fine in this case. Consequently, the gravity of the infringement is aggravated in Outokumpu's case by the fact that it has been subject to a previous Decision finding a similar infringement. This aggravating circumstance justifies an increase of 50 % in the basic amount of the fine to be imposed to Outokumpu. A 50 % rate is not the maximum rate applied by the Commission but the normal rate employed in cases involving recidivism under the current policy.

20.5. Attenuating circumstances

(727) The Commission considers that among the attenuating circumstances invoked by the parties, the following should be analysed.

20.5.1. Non-implementation in practice of the arrangements

(728) Almost all parties have requested that the Commission take into account, as an attenuating factor, that the cartel was not fully implemented. Because non-compliance was not punished and difficult to detect, deviation was frequent. In this respect, the parties cite a number of incidents to demonstrate that the cartel was not fully implemented (1042). KME and IMI further refer to the NERA Reports demonstrating that their pricing policy adopted in the market was "competitive".

(729) Unlike the impact of a cartel on the market, which must be assessed for the cartel as a whole, implementation of the agreements is to be analysed separately for each participant. In order to determine whether the agreements were implemented in practice, it is necessary to ascertain whether the circumstances, which the cartel members plead, are such to conclude that during the period in which they were party to the infringing agreements they actually avoided applying them by adopting competitive conduct in the market (1043). The fact that an undertaking which participated in collusion on prices with its competitors did not behave at all times on the market in the manner agreed with its competitors, is not necessarily a matter which must be taken into account as an attenuating circumstance when determining the amount of the fine to be imposed. An undertaking which, despite colluding with its competitors, follows a more or less independent policy on the market may simply be trying to exploit the cartel for its own benefit (1044).

(730) It would therefore be necessary for each individual undertaking to show that it systematically and clearly refrained from applying the restrictive agreements. The simple fact of cheating on the other cartel members cannot thus be held as an attenuating factor. Even if certain decisions were not fully implemented, this did not prejudice the implementation of the cartel as a whole. In this case, it is apparent that none of the participants systematically refrained from implementing their agreements.

(731) It is also established case law that the implementation of agreements on target prices and other commercial terms does not necessarily require that these exact prices and conditions be applied. In line with the Court of First Instance's judgement in the ADM Case (1045), when there is an agreement relating to price objectives rather than to fixed prices, "it is clear that implementation of that agreement simply meant that the parties would endeavour to achieve those objectives." The failure to apply the agreed price targets does not necessarily constitute an attenuating circumstance. The agreements can therefore be held to be implemented when the parties fix their prices in order to move them in the direction of the target agreed upon.

(732) The Commission concludes that in this case, the implementation of the cartel decisions was ensured through the monitoring scheme consisting of the market leaders and the regular exchange of confidential information. In such circumstances, it may be presumed that the competitors in question took into account the information exchanged in determining their own conduct on the market. It has been established that the participants regularly exchanged their sales figures, first in form of the "[...] Spreadsheet", then during meetings based on IWCC data, finally through the WBMS data exchange. Subsequently, the parties compared their sales figures to target volumes in their meetings. Thus, it is proven that the agreement to exchange confidential information was implemented in practice by each of them. It is therefore sufficient to observe, in line with the ADM Case (1046), that by informing each other about their sales volumes the participants implemented the agreement in question, irrespective of whether the information supplied was correct.

(733) The implementation of the cartel decisions was also ensured by the frequent contacts between competitors. The periods of tension and deviation from the agreed principles, which occurred in particular during the period from mid/end 1994 until mid-1997, may be considered normal in the life cycle of a long-lasting cartel.

(734) With regard to the implementation of the price agreements, the Commission has (sporadic) evidence of internal instructions to implement price increases agreed upon at a meeting with competitors (see the references in recitals (510)). Furthermore, as repeatedly described, statements of Mueller and Boliden as well as notes of Wieland confirm the success of price implementations (for details see above Section 20.3.1.2).

(735) Some elements of the agreements might have been more effectively implemented than others.

(736) The Commission notes that in Part D, elements of proof were adduced that the parties implemented several of the cartel agreements in practice. The Commission has also shown implementation in recital (510). This attenuating circumstance is therefore not applicable to any of the addressees of this Decision.

20.5.2. Limited benefit derived from the infringement

(737) Most parties maintain that the fines should be reduced due to the limited profits or absence of economic advantage deriving from the infringement. Moreover, Wieland disagrees with general statements of the Commission that the hypothetical market price cannot be determined and claims that the Commission is obliged to estimate the hypothetical price. If it is no longer possible to adduce proof of additional profits derived from the infringement, it should not lead to any disadvantage of the parties but the burden of proof should be on the Commission.

(738) Arguably, some of the participants made unsatisfactory operating profits or losses on their European activities. But unlike the parties suggest, the Commission does not consider that in general non-benefit from a cartel, or a lack of economic advantage resulting from participation in such an infringement, could be either an attenuating factor or reduce the gravity of the infringement. The Court of First Instance did not reject the Commissions view that the Commission is under no obligation to take into account the profits derived from the infringement. (1047) It is generally difficult to determine what profits each undertaking has derived from its participation in the infringement, although a number of important indications for considerable profits are contained in the file. Where there has been a serious and deliberate infringement of Article 81 of the Treaty and Article 53 of the EEA Agreement, that infringement may be considered to be sufficiently important so that the Commission does not have to attach particular importance to the actual profits.

(739) It should be also recalled that according to Wieland the objective of the cartel was to stop the price erosion (recital (620)). The fact that in certain years the participants allegedly did not earn such high economic or financial benefits from the infringement is compatible with this objective, especially in a sector in which there are high exit costs. The economic or financial benefits derived by the offenders cannot be restricted to super-profits. A loss which is smaller than it would have been in the absence of the cartel also constitutes an economic or financial benefit.

(740) Finally, as far as the Commission demonstrates the existence of such advantages and the fact that the fine does not exceed the advantage, in line with the purpose of fines, the Commission has to increase the amount of the fine by an aggravating circumstance. The fact that the Commission cannot evaluate such benefit for a considerable period of the infringement does not transform this element into an attenuating circumstance. In other words, the absence of an aggravating circumstance does not constitute an attenuating circumstance.

20.5.3. Economic difficulties in the copper plumbing tube sector

(741) All the parties have submitted that the copper plumbing tube market has suffered from over-capacity since the late 1980s. In particular, market entry or expansion by incumbents led to price erosion and low profitability.

(742) In recent case law, the Court of First Instance has confirmed that the Commission is not required to regard as an attenuating circumstance the poor financial state of the sector. The Court of First Instance has also stated that just because the Commission has taken account in earlier cases of the economic situation of a sector as an attenuating circumstance it does not necessarily have to continue to observe that practice. As a general rule, cartels come into being when a sector encounters problems. If the parties. reasoning were to be followed, the fine would have to be reduced as a matter of course in virtually all cases (1048).

(743) The Commission maintains that the situation in the copper plumbing tubes sector cannot be compared with the situation described in the Commission Decisions in Cases Alloy Surcharge (1049) and Seamless Steel Tubes (1050). In the Seamless Steel Tubes Case, the Commission concluded that "Since the 1970s, the Community steel market has been affected by a long, serious crisis, the most notable features of which have been the continuous fall in demand and the collapse of prices. These market conditions have brought with them serious problems of overcapacity, low plant-utilisation rates and prices failing to cover total production costs and ensure the profitability of firms. The crisis in the steel market has not just hit ECSC steel but has also affected the non- ECSC sectors, which include the pipes and tubes covered by this decision" (recital 25 of the Decision). In addition, "With regard in particular to the pipe and tube industry in the Community, since 1980 Community production has been severely restructured in order to adapt capacity to changing market conditions. By the end of 1990, seamless pipe and tube production capacity had been reduced by about 20 %. Between 1988 and 1991, more than 20000 jobs were lost. Since early 1991, the worsening situation of Community production, combined with the growing influx of imports, has resulted in draconian decisions having to be taken concerning the continued reduction of capacity to core levels and in the closure of several production mills in Germany, Italy and the United Kingdom" (recital 26 of the Decision).

(744) In the Alloy Surcharge Case, the Commission found that: "On the other hand, the economic situation in the sector at the end of 1993 was particularly critical. The price of nickel was rising rapidly, while the price of stainless steel was very low. It should be noted that this particular situation applies only to the very beginning of the concerted action" (recital 83 of the Decision).

(745) In this case, according to the NERA Report prepared at the request of KME, "with the exception of Germany, the consumption of copper plumbing tubes presented an increasing trend in all major European markets over the 1990s (...). Consumption in Germany grew extraordinarily fast until the mid-nineties but started to decrease at the end of the demand boom brought about by the reunification process. In spite of the positive trends observed in the market over the last decade, the 21st century began with important decreases of consumption in almost all European countries". (emphasis added by the Commission) The Commission concludes from this that for the largest part of the duration of the infringement, copper plumbing tubes consumption has been increasing. Although a consolidation process leading to the reduction of the number of players in the copper tube market was implemented in the last 20 years, the situation appears to be substantially different from the difficulties suffered by the steel sector.

(746) The Commission must therefore conclude that this sector was not in a crisis similar to the Cases Alloy Surcharge and Seamless Steel Tubes during the infringement period and that a reduction of fine is thus not justified.

20.5.4. Gradual drifting to illegality

(747) Wieland invokes, as an attenuating circumstance, the fact that the cartel behaviour was initiated and intensified gradually over the years.

(748) The Commission rejects this argument. It has been established that by September 1989 participants in the infringement, in particular the SANCO producers, had already started price cooperation (see under Sections 6.4.1.1 and 8.2). The participants' attempts to conceal the discussions in their meetings at the beginning of the infringement period also show that they knew about the illegal nature of these discussions from the beginning, as is apparent in recital (236). Descriptions concerning the SANCO producers as well as KME's descriptions of the arrangements, both of which are not contested by Wieland, do not suggest that the cartel developed gradually.

20.5.5. Termination of the infringement

(749) A number of parties maintain that the Commission should take into account the fact that they ceased to participate in the cartel immediately following the dawn raids and prior to the Commission's letter pursuant to Article 11 of Regulation 17.

(750) The Commission considers that the immediate cessation of the illegal behaviour cannot in general be regarded as an attenuating circumstance in cartel cases involving deliberate infringements. According to the Court of First Instance, "[a]n undertaking's reaction to the opening of an investigation into its activities can be assessed only by taking account of the particular context of the case" and "the Commission cannot therefore be required, as a general rule, either to regard a continuation of the infringement as an aggravating circumstance or to regard the termination of an infringement as a mitigating circumstance..." (1051).

(751) Claims to obtain a reduction of fine based on the termination of the infringement should therefore be rejected.

(752) Halcor withdrew from the arrangements in beginning of September 1999, more than a year before the Commission's inspections were carried out at the premises of other parties. Mueller withdrew in 2001, before the inspections and when it started cooperating with the Commission. In the case of Halcor, Halcor does not merit any reduction of the fine because Section B(3) third indent of the Guidelines on the method of setting fines imposed pursuant to Article 15(2) of Regulation No 17 and Article 65(5) of the ECSC Treaty (1052) ("the fining guidelines") does not provide for an attenuating circumstance if a party terminates its participation before the Commission intervenes. In the case of Mueller, the termination of its involvement in the illegal activity no later than the time at which it disclosed the cartel is a requisite for total immunity from fines (Section B, point (c) of the 1996 Leniency Notice).

20.5.6. Compliance programme

(753) Most parties have requested that the Commission take into account that they have adopted antitrust compliance programmes. Furthermore, Wieland submits that, as a consequence of the proceedings in the Industrial Tubes Case (38.240), it has already run a compliance program and has obliged all employees to respect competition rules. It confessed, regrets and has accepted its guilt. Regret and confession are mitigating factors in the Member States' legal order. There is no risk of repetition. In Wieland's view, there is therefore no need for deterrence with respect to Wieland (1053).

(754) The Commission welcomes any initiatives to set up antitrust compliance programmes. Nevertheless, it should be borne in mind that, whilst it is important that an undertaking should take steps to prevent fresh infringements of Community competition law from being committed in the future by members of its staff, that does not alter the fact that an infringement has been committed. Thus, the mere fact that in certain of its previous Decisions the Commission took the implementation of a compliance programme into consideration as an attenuating factor does not mean that it is obliged to act in the same manner in any given case (1054), especially where the infringement in question is, as in this case, a clear infringement of Article 81(1)(a) and (b) of the Treaty and Article 53(1)(a) and (b) of the EEA Agreement (1055).

(755) The Commission therefore does not accept any claims that adoption of a compliance programme should be taken into account as an attenuating factor.

20.5.7. Alleged coercion of Boliden by KME

(756) It could not be established whether a number of statements made by Boliden with respect to having been coerced by KME are true. The Commission considers that Boliden's position is not affected by this. In particular, no mitigating circumstance would have applied to Boliden, had coercion been proven. A company that is coerced by other participants to participate in a competition law infringement should inform public authorities. In the light of such an option, which respects the law, a participation in illegal cartel activities cannot be justified.

20.5.8. Cooperation outside the scope of the 1996 Leniency Notice

(757) The Commission notes that unlike point 23 of the 2002 Leniency Notice, the 1996 Leniency Notice does not provide for any specific reward to a leniency applicant that discloses facts previously unknown to the Commission and affecting the gravity or duration of the cartel. It is therefore appropriate to consider any such cooperation under the attenuating factors.

(758) The Commission considers that Outokumpu's cooperation qualifies for an attenuating factor in this regard. Outokumpu was the first to disclose the whole duration of the European cartel in the copper plumbing tubes sector, and, in particular, was the first to provide decisive evidence and explanations to prove continuity of the infringement during the period from July 1994 until July 1997 (and the period from 1990 until end of 1992). Outokumpu's contribution was necessary to establish a continuous cartel starting from September 1989 until March 2001. Prior to Outokumpu's leniency application, only based on the evidence obtained from Mueller and from the inspections (see also recital (777)), the Commission could not have established the duration and continuity of the infringement from September 1989 until July 1997.

(759) The Commission considers that Outokumpu should not be penalised for its cooperation by imposing on it a higher fine than the one that it would have had to pay without its cooperation. Therefore the basic amount of Outokumpu's fine should be reduced to the hypothetical amount of the fine that would have been imposed on Outokumpu for a three and a half-year infringement. In the light of the above, the basic amount of the fine to be imposed on Outokumpu should be reduced by the lump sum of EUR 40,17 million for effective cooperation outside the scope of the 1996 Leniency Notice.

(760) The Commission also considers that KME's cooperation qualifies for an attenuating factor in this regard. Although the Commission had certain isolated indications that the illegal behaviour also concerned plastic-coated tubes and more solid evidence with respect to information exchange concerning plastic-coated tubes in the Statement of Objections, it was only with KME's contribution that it could establish the existence of a single continuous and complex infringement with respect to WICU/Cuprotherm tubes starting from at least beginning of 1991. Wieland did not contest KME's contribution, which was also presented in the Oral Hearing.

(761) The Commission considers that the KME group should not be penalised for its cooperation. Therefore the basic amount of the fine to be imposed on the KME group should be reduced. The appropriate point of reference for the reduction is the relative importance of the plastic coated tubes sector compared to the plain copper plumbing tubes. Based on this criterion, although without making a precise mathematical calculation, the basic amount for the fine should be reduced by a lump sum of EUR 7,93 million.

20.5'9. No reduction because of administrative procedure

(762) KME submits that the Commission should take into consideration that certain factors hindered KME's efforts to prepare a full response to the Statement of Objections. In particular, KME considers the deadline for the reply to the Statement of Objections of eight weeks as too short (1056). In the Commission's view, such a circumstance cannot be considered as an attenuating circumstance for the calculation of the fine but relates to the issue of the effective exercise of the rights of defence. In this respect, the Commission notes that the parties had more than nine weeks to submit their written replies to the Statement of Objections and that, in particular, KME was able to submit a very detailed reply of almost 130 pages. The Commission also notes that KME could also further explain its view during the Oral Hearing held on 28 November 2003 and was allowed, as were the other parties, to comment in writing afterwards on certain issues. It therefore appears that KME's rights of defence have been respected.

20.6. Application of the 10 % turnover limit

(763) Most parties maintain that the relevant figure for the purposes of determining the 10 % ceiling for fines is the worldwide 2002 added value (that is to say the conversion value) turnover instead of its consolidated turnover based on full price (that is to say the added value and metal price). Addressees of the Industrial Tubes Decision further maintain that the 10 % turnover limit must be applied prior to any reduction for leniency and to the combined amount of the two fines of the two cases, the Industrial Tubes Case (COMP/E-1-38.240) and this case.

(764) Outokumpu maintains that it would be unfair and disproportionate for the Commission to take as a basis a turnover greater than that of Outokumpu Copper Products Oy for Europe for the purpose of any fine.

The Commission's view

(765) With regard to the parties' argument pertaining to the conversion value turnover, reference is made to the discussion in recital (679). Turnover reflects what is charged to customers and therefore is the relevant figure.

(766) The amount of the fine calculated by taking account of any attenuating or aggravating circumstances may not exceed 10 % of the worldwide turnover of the undertaking concerned. According to established case law, the Commission does not have to limit the maximum amount of the fine to 10 % of the turnover in the relevant product and geographical market, but turnover is to be understood as meaning the total turnover of the undertaking concerned (1057).

(767) The Commission refers to its reasoning in recital (717) to rebut the argument that the 10 % turnover limit should be applied to the combined amount of fines in the Industrial and Plumbing Tubes Cases. The fact that the Commission is conducting an investigation on several cases (although they were initiated as one and later separated) does not prevent it from imposing, if appropriate, the maximum amount in each infringement.

20.7. Application of the 1996 Leniency Notice

(768) The addressees of this Decision have co-operated with the Commission, at different stages of the investigation into the infringements for the purpose of receiving the favourable treatment set out in the 1996 Leniency Notice. The Commission therefore examines in the following Section, in chronological order, whether the parties concerned satisfied the conditions set out in the notice.

20.7.1. Mueller group

(769) Mueller, the current parent company of WTC Holding Company, Inc., Mueller Europe Ltd., DENO Holding Company, Inc. and DENO Acquisition EURL, was the first undertaking to inform the Commission about the existence of a cartel in the copper plumbing tubes sector affecting the EEA market in the 1990s. The [...] evidence [...] Mueller provided [...], prior to the Commission's investigation, enabled the Commission to establish the existence, content and the participants of a number of cartel meetings held in particular in 1989, 1994 and from 1997 until 1998 and various cartel related contacts until 2001, as well as to undertake inspections on 22 March 2001 and thereafter.

(770) Mueller immediately put an end to its involvement in the infringement before starting its cooperation with the Commission (see recital (597)). It continuously provided the Commission with all relevant information, documents and evidence available, and maintained full cooperation throughout the investigation by numerous submissions. Mueller neither compelled other participants nor did it act as an instigator nor did it play a determining role.

(771) Mueller should therefore benefit from total exemption from any fine.

20.7.2. Outokumpu

(772) Outokumpu informed the Commission about its willingness to cooperate with the Commission on 9 April 2001 (when a second inspection at its premises was undertaken). It provided documentary evidence in its possession on 30 May 2001, shortly after the Commission's inspections on 22 and 23 March 2001 and on 9 April 2001.

(773) The documentary evidence, corporate statements and witness testimonies provided by Outokumpu cover a period extending from 1987 to 2001. In its Memorandum dated 30 May 2001, Outokumpu provided a description of the cartel including a non-exhaustive list of the multilateral meetings of copper plumbing tube producers (with indication of the dates, locations and participants), as well as a number of additional documents it had found in its internal audit. It also described the context of a number of handwritten notes and other documents found during the inspections at its employees' offices, which made it possible to connect these documents to specific cartel events. This submission was completed by oral explanations given by Outokumpu's employees at interviews conducted at the Commission's invitation in Brussels on 4 and 11 February 2003, as well as by a response of 21 February 2003 to the Commission's request for information sent in preparation of the interviews. It should be noted that the interviews of 4 and 11 February 2003 were initially planned for June 2002, and the employees in question had agreed to submit to the interviews at that time, but the Commission postponed the questioning of these employees for its internal reasons.

(774) The Commission notes that the list of meetings attached to Outokumpu's first submission of May 2001 contained gaps concerning certain periods of the infringement (notably the years 1990 until 1995 and beginning of 1999 and February 2000), although Outokumpu provided general descriptions of the meetings for this period that proved to be accurate. These gaps were, however, subsequently filled to a satisfactory extent in particular by Outokumpu's submission of 21 February 2003, in which Outokumpu replied to certain questions raised during the interviews with its representatives, and also by statements of its representatives confirming the cartel activities. The fact that Outokumpu no longer had specific recollections concerning all cartel meetings throughout the whole period of the infringement does not alter the Commission's conclusion that Outokumpu's cooperation was complete.

(775) Outokumpu's cooperation in this matter began nearly a year and a half before that of the other participants. The Commission therefore accepts that Outokumpu's early assistance allowed the Commission to better understand the infringement and interpret the documents obtained in the inspections. The information submitted by Outokumpu in the form of important documentary evidence, corporate statements and executive interviews was detailed and therefore extensively used by the Commission in the pursuance of its investigation. That information was also used to draft requests for information that contributed to trigger the admission by Halcor of its participation in the cartel. Outokumpu thus assisted the Commission significantly in establishing the facts on which this Decision is based.

(776) Outokumpu does not qualify for a non-imposition of a fine or a very substantial reduction of at least 75% in its amount under Section B of the 1996 Leniency Notice. More specifically, it does not meet the condition set forth in point (a) of Section B, since it did not inform the Commission about the cartel before the Commission undertook an investigation, ordered by Decision, in this case.

(777) Furthermore, Outokumpu does not qualify for a substantial reduction from 50 % to 75% under Section C of the 1996 Leniency Notice, as the Commission investigations ordered by Decision provided sufficient grounds for initiating the procedure leading to a Decision in this case. The inspections produced direct evidence on the existence of the cartel primarily in the period of 1989, 1994 and from July 1997 to March 2001. While the evidence and indices before and after that period, including documents concerning the first known cartel meeting on a broad European level in September 1989 and further meetings in 1994, were only sporadic, the Commission considers that it could have opened proceedings in this case and established a continuous infringement from at least July 1997 to March 2001 without Outokumpu's cooperation. Nevertheless, since Outokumpu was the first to disclose the whole duration and, supported by documents, continuity of the infringement, for its cooperation outside the scope of the 1996 Leniency Notice was granted an attenuating factor (recitals (757) to (759)).

(778) Under Section D of the 1996 Leniency Notice, an undertaking which does not comply with all the conditions set out in Sections B or C of that Notice can still benefit from a significant reduction of 10 % to 50 % of the fine that would otherwise have been imposed. The Commission notes that before the Statement of Objections was sent, Outokumpu materially contributed to establishing the existence of the infringement, and after having received the Statement of Objections, it has informed the Commission that it does not substantially contest the facts on which the Commission has based its allegations. Outokumpu therefore fulfils the conditions set out in Section D of the 1996 Leniency Notice, qualifying for a significant reduction in a fine (10 %-50 %).

(779) In accordance with Section D of the 1996 Leniency Notice and in view of Outokumpu's early and extensive cooperation, Outokumpu should, accordingly, be granted a 50 % reduction of the fine that would otherwise have been imposed if it had not co-operated with the Commission.

(780) The total fine imposed on Outokumpu should therefore be EUR 36,14 million.

20.7.3. KME group

(781) KME claims that its cooperation qualifies for a 50 % reduction under Section D of the 1996 Leniency Notice (1058). KME claims that it provided the Commission with important information and maintained continuous cooperation throughout the investigation (1059). KME submits that its cooperation was fully voluntarily (1060). KME believes that it materially contributed to establishing the existence of the infringement and co-operated fully voluntarily (1061).

(782) More specifically, KME argues that besides Outokumpu, it was the only company to have provided an overall description of the arrangements for the period 1988 to 2001. It provided a detailed description of the subject matters discussed during the meetings such as, for example, rebates and pricelines. KME's submission contained a general description of the so-called EDWD meetings as of 1997, including duration, participants and subject matter. KME also explained the market leader system and provided details on national contacts. KME provided the Commission with a list of 41 meetings at the European level between 1992 and 2001 (after Wieland had provided its list), and a list of meetings concerning the national level. Furthermore, KME provided information on SANCO, WICU and Cuprotherm products (1062).

(783) KME did not start cooperating with the Commission until it responded to a letter addressed to it pursuant to Article 11 of Regulation No 17 in July 2002 in the Industrial Tubes Case (38-240), a case that was only separated from the Copper Plumbing Tubes Case after the inspections. At the same time, the Commission acknowledges that the cooperation was not initiated by a letter sent to KME pursuant to Article 11 of Regulation No 17 in the Copper Plumbing Tubes proceedings. The on-site inspections ordered by a Commission Decision were carried out as early as March 2001. Being aware of the Commission's investigation for more than 18 months, KME did not offer its cooperation before it had been approached again by a formal Commission intervention in the Industrial Tubes Case. The importance of the Commission's intervention also for these proceedings from KME's point of view is demonstrated by the fact that KME believes the Commission should apply certain rules like the 10 % turnover threshold to the sum of fines imposed in both proceedings. KME's application for leniency in the Copper Plumbing Tubes Case was introduced only on 15 October 2003, nearly a year and a half after Outokumpu's.

(784) The Commission considers that KME contributed materially to establishing the existence of the infringement for its full duration before the Statement of Objections was sent, for which adequate recognition should be accorded.

(785) With regard to the complete duration of the infringement, the Commission notes that KME was only the second to provide a description, after the Commission had already received decisive evidence from Outokumpu. As far as a list of meetings is concerned, to a certain extent Outokumpu and especially Wieland provided the Commission with evidence before KME. With respect to the establishment of continuity, KME's cooperation was of minor importance.

(786) The Commission recognises, however, that as opposed to Wieland, KME's description of the cartel cooperation extended also to these early years of the infringement 1988-1993. However, KME claimed an interruption for 1990 until end of 1992 and from mid 1994 until mid 1997. For the quiet periods, and in particular for the period from 1990 until end of 1992, KME was not able to add any value. However, KME did explain content and duration of national arrangements. As far as the period from 1988-1995 is concerned, the Commission had already sufficient evidence before KME's leniency application. Compared to Wieland, KME did not provide any contemporaneous documents that prove the content of the infringement. The travel expenses that were compiled by KME, to a large extent, had already been collected during the inspections and although they suggested the presence of certain individuals at certain meetings, they did not allow the Commission to establish the specific anticompetitive contents of such meetings. KME did not comment on any contemporaneous cartel documents collected at its premises.

(787) As far as descriptions of the content of the meetings are concerned, that is to say, rebates and pricelines, KME, to a certain extent, confirmed the detailed description of the former executive of the copper plumbing tube industry.

(788) It must be pointed out, however, that the Commission had previously obtained documentary evidence concerning a number of meetings and exchanges of confidential information during this period as a result of the inspections and from other sources, as set forth, for example, in recitals (769), (770), (773)-(775) and as is apparent from the description of facts in Section D.

(789) The Commission accepts KME's claim that it has both provided new evidence and corroborated existing evidence for the entire period of the infringement from 1988 to 2001. In contrast, it does not agree with KME's assertion that the Commission had been provided with only very limited descriptions of the anticompetitive arrangements prior to KME's reply to the Commission's request of information, as already pointed out above in recital (787).

(790) With regard to the specific meetings first disclosed by KME, the Commission notes that the establishment of a cartel does not require an exhaustive list and description of all meetings and competitor contacts that occurred.

(791) KME's leniency contributions should be assessed in a way that avoids any hypothetical disadvantages that could appear to arise from information obtained through interviews of employees of Outokumpu and Outokumpu's subsequent submission.

(792) KME was not the first undertaking to provide the Commission with decisive evidence on the copper plumbing tubes cartel, as required under point (b) of Section B of the 1996 Leniency Notice, and therefore it does not qualify under Section C which refers to the conditions set out in Section B, points (b) to (e). Nevertheless, under Section D of the said Notice, an undertaking which does not comply with all the conditions set out in Sections B or C can still benefit from a significant reduction of 10 % to 50 % of the fine that would otherwise have been imposed.

(793) After having received the Statement of Objections, KME has informed the Commission that it does not substantially contest the facts on which the Commission has based its allegations.

(794) After due consideration of all these circumstances, the Commission considers that KME fulfils the conditions set out in Section D(2) first and second indent of the 1996 Leniency Notice and grants it a 35 % reduction of the fine that would have been imposed if it had not cooperated with the Commission.

(795) The total fine imposed on the companies of the KME-group should therefore be EUR 67,08 million (of which EUR 32,75 million on KME group; EUR 17,96 million on KME AG; and EUR 16,37 million on the undertaking formed by EM and TMX).

20.7.4. Wieland group

(796) Wieland Werke (including Buntmetall) did not start cooperating with the Commission until it responded to a letter addressed to it pursuant to Article 11 of Regulation No 17 in July 2002 in the Industrial Tubes Case (38.240), a case that was separated from the Copper Plumbing Tubes Case after the inspections. At the same time, the Commission acknowledges that the cooperation was not initiated by a letter sent to Wieland pursuant to Article 11 of Regulation No 17 in the Copper Plumbing Tubes proceedings. The on-site inspections ordered by a Commission Decision were carried out as early as in March 2001. Being aware of the Commission's investigation for more than 20 months, Wieland only offered its cooperation in this case on 23 January 2003, after it had been approached again by a formal Commission intervention in the Industrial Tubes Case. The importance of the Commission's intervention also for these proceedings from Wieland's point of view is demonstrated by the fact that Wieland recognises that the two proceedings are interconnected to an extent that Wieland considers the separation of the two cases as artificial.

(797) Wieland (including Buntmetall) provided a detailed description of the functioning of the cartel since 1993 (claiming that there was an interruption between .from mid-1994 until spring 1996/97"), and provided important explanations for the understanding of contemporaneous memos of its employees, meaning that the memos were .neutrally. drafted summaries of cartel meetings, covering the period from mid-1997 until 2001 mostly found during the Commission's inspections. Wieland was the first to provide a list of meetings for the period from 1996 until 2001. In particular for the year 1996, Wieland disclosed a number of meetings that helped reconstruct the extent of cartel activity during the alleged "quiet period". This information complemented information provided by Outokumpu. Consequently, the Commission considers that Wieland Werke contributed materially to establishing the existence of the infringement before the Statement of Objections was sent, for which adequate recognition should be accorded. The Commission notes, however, that in its application for leniency Wieland had not recollections concerning the period before 1993. Wieland's leniency contributions should be assessed in a way that avoids any hypothetical disadvantages that could appear to arise from information obtained through interviews of employees of Outokumpu and Outokumpu's subsequent submission. In its reply to the Statement of Objections, Wieland has not contested the starting date of the infringement but submitted that there were interruptions. It did not substantially contest the facts.

(798) Wieland Werke (including Buntmetall) was not the first undertaking to provide the Commission with decisive evidence on the copper plumbing tubes cartel, as required under point (b) of Section B of the 1996 Leniency Notice, and therefore it does not qualify under Section C of the Notice, which refers to the conditions set out in Section B, points (b) to (e) thereof. Nevertheless, under Section D of the Notice, an undertaking which does not comply with all the conditions set out in Sections B or C can still benefit from a significant reduction of 10 % to 50 % of the fine that would otherwise have been imposed.

(799) After due consideration of all these circumstances, it can be concluded that Wieland Werke and Buntmetall fulfil the conditions set out in Section D(2) first and second indent of the 1996 Leniency Notice. The Commission grants Wieland and Buntmetall a 35 % reduction of the fine that would have been imposed if they had not cooperated with the Commission.

(800) The total fine imposed on Wieland Werke should therefore be EUR 27,8411 million (of which EUR 2,43 million on Wieland group; EUR 24,7416 million on Wieland Werke; and EUR 0,6695 million on Buntmetall).

20.7.5. Halcor

(801) Halcor applied for leniency and started cooperating with the Commission immediately after having received a letter addressed to it pursuant to Article 11 of Regulation No 17 in March 2003. No inspections at the premises of Halcor were carried out beforehand. The letter sent pursuant to Article 11 of Regulation No 17 in March 2003 was Halcor's first contact with the Commission in the Copper Plumbing Tube Investigation, although the fact that inspections had taken place was public information in the copper plumbing tube industry, since certain parties had published press releases about the inspections.

(802) Halcor submits that the reduction on the grounds of leniency should be no lower merely because a request for information was sent. The information request was very broad. Halcor voluntarily provided information without any legal obligation to respond to the information request. The information given should not be excluded from eligibility for leniency.

(803) A considerable part of the information provided was in reply to the letter addressed to Halcor pursuant to Article 11 of Regulation No 17 and therefore falls, as such, within the ambit of the undertaking's duty to fully reply to these requests as set out in Article 11. The Commission acknowledges, nevertheless, that Halcor's reply to the Commission's Article 11 letter, in which it also applied for leniency, exceeded its obligation to reply. Before sending its reply, it provided as part of its leniency application a number of documents contemporaneous to the infringement. In its reply, it described the functioning of the cartel. The Commission points out that as far as Halcor provided self-incriminating information, this information should be rewarded in the context of Halcor's leniency application.

(804) Halcor provided detailed descriptions of the functioning of the cartel and its participation in the meetings, although its participation had already been established before Halcor's cooperation. In this respect it is noted that the only potentially new element and source of a material contribution to the establishment of certain elements of the infringement would have been Halcor's allegation that it was coerced or presented with an ultimatum by the group of the five into participating in the infringement, should this have been proven. However, the Commission considers that Halcor made a self-serving statement without sufficient grounds. It is important in this respect that after having reviewed the file and after having been confronted with other parties. arguments at the Oral Hearing, Halcor did not add further value to the establishment of the facts by explaining and clarifying its role in previous cartel meetings on a European level or other meetings involving participants of the European-wide cartel (which were mentioned in various submissions of other parties and Mueller's presentation during the Oral Hearing), and why it felt threatened by competitors, with which it had illegally cooperated before. Thus, up to that point, Halcor's cooperation was limited. The specific value of Halcor's leniency application lies in contemporaneous notes of cartel meetings proving price coordination and volume allocation. Consequently, the Commission considers that Halcor provided the Commission with evidence, which materially contributed to establishing the existence of the infringement before the Statement of Objections was sent. This evidence, however, as already explained had only limited value because the Commission was already in the possession of evidence to prove the infringement for the relevant period. For this cooperation, adequate recognition should be accorded. After having received the Statement of Objections, Halcor informed the Commission that it did not substantially contest the facts on which the Commission has based its allegations.

(805) Halcor was not the first undertaking to provide the Commission with decisive evidence on the copper plumbing tubes cartel, as required under point (b) of Section B of the 1996 Leniency Notice, and therefore it does not qualify under Section C which refers to the conditions set out in Section B, points (b) to (e) of the said Notice. Nevertheless, under Section D of the Notice, an undertaking which does not comply with all the conditions set out in Sections B or C can still benefit from a significant reduction of 10 % to 50 % of the fine that would otherwise have been imposed.

(806) The Commission takes into consideration that no inspections had been carried out at the premises of Halcor. In favour of Halcor, the Commission assumes that Halcor started cooperating at the same time as Wieland and KME, since both of these companies applied for leniency after having received a letter pursuant to Article 11 of Regulation No 17 an Information Request in the related Industrial Tubes Case.

(807) After due consideration of all these circumstances, it can be concluded that Halcor fulfils the conditions set out in Section D(2), first and second indent of the 1996 Leniency Notice. It should be granted a 15 % reduction of the fine that would have been imposed if it had not co-operated with the Commission.

(808) The total fine imposed on Halcor should therefore be EUR 9,16 million.

20.7.6. Boliden group

(809) After having received the Statement of Objections, the Boliden group applied for leniency. Boliden admitted the infringement and did not contest the facts. In addition, Boliden clarified certain factual details. Under Section D of the 1996 Leniency Notice, an undertaking which does not comply with the conditions set out in Sections B or C can still benefit from a significant reduction of 10 % to 50 % of the fine that would otherwise have been imposed.

(810) Given the cooperation of Mueller, Outokumpu, the KME-group, Wieland and Halcor, as well as the inspections, the infringement had already been established in its entirety. The non-contestation of the infringement by Boliden was therefore of minor importance for establishing the infringement. In accordance with Section D of the 1996 Leniency Notice the Commission grants Boliden a 10 % reduction of the fine that would otherwise have been imposed if it had not cooperated with the Commission.

(811) The total fine imposed on Boliden should therefore be EUR 32,6 million.

20.7.7. IMI group

(812) After having received the Statement of Objections, the IMI group applied for leniency. IMI admitted the infringement and did not contest the facts. Under Section D of the 1996 Leniency Notice, an undertaking which does not comply with the conditions set out in Sections B or C can still benefit from a significant reduction of 10 % to 50 % of the fine that would otherwise have been imposed.

(813) Given the cooperation of Mueller, Outokumpu, the KME-group, Wieland and Halcor, as well as the inspections, the infringement had already been established in its entirety. The non-contestation of the infringement by IMI was therefore of minor importance for establishing the infringement. In accordance with Section D of the 1996 Leniency Notice the Commission, accordingly, grants IMI a 10 % reduction of the fine that would otherwise have been imposed if it had not co-operated with the Commission.

(814) The total fine imposed on IMI should therefore be EUR 44,98 million.

20.7.8. Conclusion on the application of the 1996 Leniency Notice

(815) In conclusion, with regard to the nature of their cooperation and in the light of the conditions as set out in the 1996 Leniency Notice, to the addressees of this Decision should be granted the following reductions of their respective fines:

(a) Mueller group: immunity from fines

(b) Boliden-group: a reduction of 10 %

(c) Halcor: a reduction of 15 %

(d) HME: no reduction

(e) IMI-group: a reduction of 10 %

(f) KME-group: a reduction of 35 %

(g) Outokumpu-group: a reduction of 50 %

(h) Wieland-group: a reduction of 35 %

20.8. Ability to pay and other factors

20.8.1. Ability to pay

(816) In its reply to the Statement of Objections, the KME group asked the Commission to take into consideration its potential inability to pay a high fine and the fine imposed on it in the parallel case (Case COMP/E-1-38.240 - Industrial tubes) (1063).

(817) As an introductory remark, KME considered that the Commission has in a recent discussion mis-stated its own fining guidelines, arguing that the test of the inability to pay should be based on the proof of (i) a risk of an immediate bankruptcy resulting from the payment of the fine and (ii) that such bankruptcy should be regarded as unacceptable given .the specific social context. (1064). According to KME the risk of immediate bankruptcy is stricter than the real ability to pay criterion in the fining guidelines and is therefore contrary to the principle of legitimate expectations.

(818) [Arguments presented by KME]

(819) [Arguments presented by KME]

(820) [Arguments presented by KME] (1065)

(821) As a preliminary remark, the Commission notes that KME does not substantiate its claim that the "risk of immediate bankruptcy" is a stricter test than the "real ability to pay". Secondly, the Commission disagrees with such interpretation. In order to be eligible for any adaptation of the fine according to the requirements set out in the fining guidelines (that is to say the real ability to pay or, in practical terms, inability to pay), the undertaking must demonstrate that it could not meet its contractual obligations (debts, including the payment of the fine) and therefore risk an immediate bankruptcy. Indeed, the risk of immediate bankruptcy is the only reliable test that an undertaking or its shareholders do not have the necessary resources to pay the fine and it is not that they just do not wish to pay the fine.

(822) Even if the first requirement was satisfied, a reduction of the fine could not be granted unless related to "a specific social context".

(823) The annual reports of SMI and KME show that their financial position has deteriorated in 2003 as compared with 2002. SMI's loss in 2003 was due to a series of elements: the operating losses (due to the negative economic conditions, the structural overcapacity of the sector and the reduction of sales prices), the costs of restructuring the group (see recital (824)), the depreciation of KME's goodwill and the imputation of the fines in the two Commission cartel proceedings (EUR 39,81 million for the Decision of 16 December 2003 and an estimated EUR 80 million (1066) in view of this Decision; this estimate was necessary both to evaluate the amount of the recapitalisation to be made by the shareholders and to assess the relative weight of GIM SpA and SMI SpA in the context of the projected merger). These elements and taxes explain the difference between SMI's operating result (risultato ordinario) and the net result: EUR -236,1 million. As far as the parent company is concerned (SMI SpA), the loss was covered by the use of the available SMI reserves and the reduction of the equity.

(824) [Commission's summary of information provided by KME]

(825) The Commission also notes that, according to the consolidated accounts of the SMI group (SMI mother company (1067) and KME group) submitted by KME, the group reached a net profit of EUR 5,1 million in the tax year/accounting year 1996-1997, EUR 19,1 million in 1997-1998, EUR 33,5 million in 1998-1999, EUR 46,9 million in 1999-2000, EUR 38,6 million in 2000-2001 and EUR 6,8 million in the six-month period of July-December 2001. In 2002, for the first time the group suffered a loss of EUR 19,3 million (although it achieved an operating profit). It should be noted that in 2002, the KME continued its expansion taking over a competitor (Yorkshire Copper Tube Ltd) and selling a company not belonging to its core activity. The SMI group had investments in industrial activities totalling EUR 62 million (the SMI group was able to invest about EUR 600 million between 1996 and 2002) (1068). Also in 2002, KME provided a dividend to its shareholders.

(826) In 2003, the loss reached EUR 236,1 million. Although this loss appears huge, it is mainly composed of extraordinary items like the restructuring costs and the setting aside of provisions for the fines, which will not appear in future fiscal years. In particular, as EUR 80 million have been entered as a provision to pay the fine imposed by this Decision [no further expense will fall on the 2004 fiscal year. In 2003 the SMI group was able to invest EUR 55,8 million in industrial activities. The increase in debts in 2003 was almost entirely due to the acquisition of a stake of a large Italian company, for EUR 19,1 million. It should also be noted that at the end of 2003 KME still had EUR 115,25 million of retained earnings.

(827) In the first quarter of 2004, the KME group has made a small positive net income, which has to be compared with a loss in the same period of 2003. Such improvement of the situation also confirms that the result of 2003 was heavily conditioned by exceptional items.

(828) [Commission's summary of information provided by KME]

(829) The argument that the KME group is facing the present negative market conditions and a difficult situation as concerns the price of copper could apply to any other company that incurs a fine from the Commission in this case. Indeed, if the Commission accepted this argument for the KME group but still imposed fines on any other company in this proceeding, it would rightly stand accused of discrimination.

(830) [Commission's summary of information provided by KME]

(831) [Commission's summary of information provided by KME]

(832) [Commission's summary of information provided by KME]According to case-law: "In any event, recognition of an obligation requiring the Commission to take account, when determining the fine, of an undertaking's loss-making financial situation would be tantamount to conferring an unjustified competitive advantage on undertakings least well adapted to the conditions of the market. (1069). A reduction of the fine for a group mainly confronted with current general market conditions and whose losses mainly depend on the concentration of exceptional financial costs in one year would confer on it a greater competitive advantage with regard to the other producers.

(833) In addition, SMI/KME has not presented sufficient arguments to claim that its alleged inability to pay must be seen in a specific social context. [Commission's summary of information provided by KME].

(834) A fortiori, the Commission rejects the claims of other undertakings (Outokumpu and Wieland (1070)) which have made unsupported general statements on their ability to pay.

20.8.2. Other factors

(835) KME requested the same benefit granted to SGL in the Specialty Graphite Case where a reduction of 33 % of the fine was granted to it because it was both in a serious adverse financial situation and a significant fine had been imposed on it by the Commission relatively recently.

(836) In fact, on 18 July 2001, the Commission imposed a fine on SGL of EUR 80,2 million for its participation in the graphite electrodes cartel. On 17 December 2002, the Commission imposed a further fine on SGL of EUR 27,75 million for its participation in the isostatic specialty graphite cartel and the extruded specialty graphite cartel (covered by the Speciality Graphite Decision (1071)). All of these different cartel activities occurred simultaneously.

(837) On 3 December 2003, the Commission imposed another fine of EUR 23,64 million on SGL, including a 33% discount for the same reasons as in the Electrical and Mechanical Carbon and Graphite Products Case (hereinafter, Electrical and Mechanical). However, in the same Decision, the Commission decided not to grant that reduction to another company, Carbone Lorraine, which also claimed to be in difficulty and had also been fined in the isostatic specialty graphite cartel.

(838) In line with the Electrical and Mechanical Decision, the Commission observes that the (first) fine imposed on the KME group in the industrial tubes cartel corresponded to [...] % of its worldwide turnover in 2002. In comparison, the first fine imposed on SGL corresponded to 6,35% of its global turnover in 2000 and the first fine imposed on Carbone Lorraine corresponded to less than 1 %. The sum of the first and the second fine imposed on SGL corresponded to 8,75 % of its turnover in 2001. The sum of the first and the second fine imposed on Carbone Lorraine corresponded to more than 6,93 % of its turnover in 2002. In this case, the sum of the two fines imposed on the KME group represents 5,6% of its worldwide turnover.

(839) The Commission concludes that as a proportion of the annual turnover, the first fine imposed on SGL is more than three times higher than that imposed on the KME group and that the sum of both fines imposed on the KME group is far below that imposed on SGL.

(840) [Commission's summary of KME's financial situation] The Commission notes, however, that contrary to SGL which has been in a serious adverse financial situation for several years (at least since 2000), the KME group was a relatively healthy company until 2002 (included). In addition, the deterioration of the situation in 2003 is partly due to the general economic situation (which started recovering in 2004) and mainly to exceptional circumstances that are not going to appear again in 2004. The submitted figures for the first quarter of 2004 do not yet include the recapitalisation decided in May.

(841) The KME group is therefore not entitled to any reduction of its fine on the grounds of "other factors".

20'9. The amount of the fines imposed in these proceedings

(842) In conclusion, the Commission sets the fines to be imposed pursuant to Article 15(2) Regulation No17 and Article 23(2) of Regulation No 1-2003 as follows:

- Boliden-group EUR 32,6 million

- Buntmetall-group EUR 0,6695 million

- Halcor EUR 9,16 million

- HME EUR 4,49 million

- IMI-group (1072) EUR 44,98 million

- KME-group EUR 32,75 million

- KM Europa Metal AG: EUR 17,96 million

- Europa Metalli SpA and Tréfimétaux SA: EUR 16,37 million

- Mueller-group: EUR 0

- Outokumpu-group: EUR 36,14 million

- Wieland-group: EUR 2,43 million

- Wieland Werke AG: EUR 24,7416 million

HAS ADOPTED THIS DECISION:

Article 1

The following undertakings infringed Article 81(1) of the Treaty and - from 1 January 1994 - Article 53(1) of the EEA Agreement by participating, for the periods indicated, in a complex of agreements and concerted practices consisting of price fixing and market sharing in the copper plumbing tubes sector:

(a) Boliden AB, together with Boliden Fabrication AB and Boliden Cuivre & Zinc SA, from 3 June 1988 until 22 March 2001;

(b) Boliden Fabrication AB, together with Boliden AB and Boliden Cuivre & Zinc SA, from 3 June 1988 until 22 March 2001;

(c) Boliden Cuivre & Zinc SA, together with Boliden AB and Boliden Fabrication AB, from 3 June 1988 until 22 March 2001;

(d) Austria Buntmetall AG:

(i) together with Buntmetall Amstetten GesmbH, from 29 August 1998 at the latest until 8 July 1999, and

(ii) together with Wieland Werke AG and Buntmetall Amstetten Ges.m.b.H, from 9 July 1999 until 22 March 2001;

(e) Buntmetall Amstetten GesmbH:

(i) together with Austria Buntmetall AG, from 29 August 1998 at the latest, until 8 July 1999, and

(ii) together with Wieland Werke AG and Austria Buntmetall AG, from 9 July 1999 until 22 March 2001;

(f) Halcor SA from 29 August 1998 at the latest, until at least beginning of September 1999;

(g) HME Nederland BV from 29 August 1998 at the latest, until 22 March 2001;

(h) IMI plc together with IMI Kynoch Ltd. and Yorkshire Copper Tube Ltd. (formerly: IMI Yorkshire Copper Tube Ltd"), from 29 September 1989 until 22 March 2001;

(i) IMI Kynoch Ltd. together with IMI plc and Yorkshire Copper Tube Ltd. (formerly: IMI Yorkshire Copper Tube Ltd"), from 29 September 1989 until 22 March 2001;

(j) Yorkshire Copper Tube Ltd. (formerly: IMI Yorkshire Copper Tube Ltd") together with IMI plc and IMI Kynoch Ltd., from 29 September 1989 until 22 March 2001;

(k) KM Europa Metal AG:

(i) individually, from 3 June 1988 until 19 June 1995, and

(ii) together with Tréfimétaux SA and Europa Metalli SpA, from 20 June 1995 to 22 March 2001;

(l) Europa Metalli SpA.:

(i) together with TMX, from 29 September 1989 to 19 June 1995, and

(ii) together with KM Europa Metal AG and Tréfimétaux SA, from 20 June 1995 to 22 March 2001.

(m) Tréfimétaux SA:

(i) together with Europa Metalli SpA, from 29 September 1989 to 19 June 1995, and

(ii) together with KM Europa Metal AG and Europa Metalli SpA, from 20 June 1995 to 22 March 2001.

(n) Mueller Industries, Inc., together with WTC Holding Company, Inc., Mueller Europe Ltd., DENO Holding Company, Inc. and DENO Acquisition EURL, from 21 October 1997 until 8 January 2001;

(o) WTC Holding Company, Inc., together with Mueller Industries, Inc., Mueller Europe Ltd., DENO Holding Company, Inc. and DENO Acquisition EURL, from 21 October 1997 until 8 January 2001;

(p) Mueller Europe Ltd., together WTC Holding Company, Inc., Mueller Industries, Inc., DENO Holding Company, Inc. and DENO Acquisition EURL, from 21 October 1997 until 8 January 2001;

(q) DENO Holding Company, Inc., together with WTC Holding Company, Inc., Mueller Europe Ltd., Mueller Industries, Inc. and DENO Acquisition EURL, from 21 October 1997 until 8 January 2001;

(r) DENO Acquisition EURL, together with WTC Holding Company, Inc., Mueller Europe Ltd., DENO Holding Company, Inc. and Mueller Industries, Inc., from 21 October 1997 until 8 January 2001;

(s) Outokumpu Oyj together with Outokumpu Copper Products Oy, from 29 September 1989 until 22 March 2001;

(t) Outokumpu Copper Products Oy, together with Outokumpu Oyj, from 29 September 1989 until 22 March 2001;

(u) Wieland Werke AG:

(i) individually from 29 September 1989 until 8 July 1999, and

(ii) together with Austria Buntmetall AG and Buntmetall Amstetten GesmbH, from 9 July 1999 until 22 March 2001.

Article 2

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

(a) Boliden AB, Boliden Fabrication AB and Boliden Cuivre & Zinc SA jointly and severally EUR 32,6 million

(b) Austria Buntmetall AG and Buntmetall Amstetten GesmbH jointly and severally EUR 0,6695 million

(c) Austria Buntmetall AG, Buntmetall Amstetten GesmbH and Wieland Werke AG jointly and severally EUR 2,43 million

(d) Halcor SA individually EUR 9,16 million

(e) HME Nederland BV individually EUR 4,49 million

(f) IMI plc, IyMnoIc Kh Ltd. and Yorkshire Copper Tube Ltd. (formerly: IMI Yorkshire Copper Tube Ltd") jointly and severally EUR 44,98 million

(g) KM Europa Metal AG individually EUR 17,96 million

(h) KM Europa Metal AG, Tréfimétaux SA and Europa Metalli SpA jointly and severally EUR 32,75 million

(i) Europa Metalli SpA. and Tréfimétaux SA jointly and severally EUR 16,37 million

(j) Outokumpu Oyj and Outokumpu Copper Products Oy jointly and severally EUR 36,14 million

(k) Wieland Werke AG individually EUR 24,7416 million

The fines shall be paid, within three months of the date of the notification of this Decision to the following account:

Account N°

001-3953713-69 of the European Commission with :

FORTIS Bank, Rue Montagne du Parc 3, 1000 Brussels

(Code SWIFT GEBABEBB - Code IBAN BE71 0013 9537 1369)

After expiry of that period, interest shall automatically be payable at the interest rate applied by the European Central Bank to its main refinancing operations on the first day of the month in which this Decision is adopted, plus 3,50 percentage points, namely 5,52 %.

Article 3

The undertakings listed in Article 1 shall immediately bring to an end the infringements referred to in that Article, insofar as they have not already done so. They shall refrain from repeating any act or conduct described in Article 1, and from any act or conduct having the same or similar object or effect.

Article 4

This Decision is addressed to: 1. Wieland Werke AG, Graf-Arco-Strasse 36 89079 Ulm Germany

2. Buntmetall Amstetten GesmbH Fabrikstrasse 4 3300 Amstetten Austria

3. Austria Buntmetall AG Fabrikstrasse 2 A - 2551 Enzesfeld Austria

4. Mueller Industries Inc 8285 Tournament Drive Suite 150 Memphis, Tennessee 38125 USA

5. Mueller Europe Ltd. Oxford Street, Bilston West Midlands WV14 7DS United Kingdom

6. DENO Holding Company Inc. c/o The Corporation Company 30600 Telegraph Road Bingham Farms MI 48025 USA

7. DENO Aquisition EURL 17, Rue de la Baume 75008 Paris France

8. Outokumpu Oyj Riihitontuntie 7 D 02201 Espoo Finland

9. Outokumpu Copper Products Oy Riihitontuntie 7 A 02201 Espoo Finland

10. KM Europa Metal AG Klosterstrasse 29 49074 Osnabrück Germany

11. Tréfimétaux SA 11, bis rue de l'hôtel de ville 92411 Courbevoie France

12. Europa Metalli SpA Via dei Barucci, 2 50127 Firenze Italy

13. Halcor SA 252, Piraeus Street 17778 Athens (Tavros) Greece

14. Boliden Cuivre & Zinc SA Rue du Forneau 43 4030 Liège Belgium

15. Boliden AB Box 5001 194 05 Upplands Väsby Sweden

16. Boliden Fabrication AB Box 5001 194 05 Upplands Väsby Sweden

17. Yorkshire Copper Tube Ltd. (formerly: IMI Yorkshire Copper Tube Ltd") East Lancashire Road Kirkby Liverpool, L33 7TU United Kingdom

18. IMI Kynoch Ltd. Lakeside, Solihull Parkway Birmingham Business Park Solihull B37 7XZ United Kingdom

19. IMI Plc Lakeside, Solihull Parkway Birmingham Business Park Solihull B37 7XZ United Kingdom

20. HME Nederland B.V. Veerweg 14 5145 NS Waalwijk The Netherlands

21. WTC Holding Company, Inc. c/o The Corporation Company 30600 Telegraph Road Bingham Farms MI 48025 USA

This Decision shall be enforceable pursuant to Article 256 of the Treaty and Article 110 of the EEA Agreement.

ANNEX <emplacement tableau>

Notes 1 OJ L 1, 4.1.2003, p. 1. Regulation as amended by Regulation (EC) No 411-2004 (L 68, 6.3.2004, p. 1)

2 OJ 13, 21.2.1962, p. 204/62. Regulation as last amended by Regulation (EC) No 1216-1999 (OJ L 148, 15.6.1999, p. 5)

3 OJ L 354, 30.12.1998, p. 18

4 OJ [...], [...], p. [...]

5 See 32123. Unless otherwise states, numbers refer to the page numbers in the Commission's file. According to a study of Boliden, 45% are used for water tubes/plumbing, 52% for heating systems and 3% for gas pipes.

6 See 22851, 22859.

7 See Case No COMP/M.3284 - Boliden/Outokumpu, recitals 35, 63.

8 See 32539.

9 See 11415, 11416, 29662.

10 See 32540, 22575.

11 OJ L 330, 5.12.1998, p.32. Directive as amended by Regulation (EC) No 1882-2003 of the European Parliament and of the Council (OJ L 284 , 31.10.2003, p. 1). Implementing national laws were adopted between 1999 and 2002. The Directive introduced new environmental quality parameters for drinking water, one of which was the reduction of copper in drinking water from 3 mg/l to 2 mg/l.

12 See Case No COMP/M.3284 - Boliden/Outokumpu, recitals 33-35.

13 Wieland on the one hand, p. 22886-22899, and Outokumpu on the other hand, reply to the Statement of Objections of 10 and 17 November 2003, Annex 3, submitted slightly contradictory but similar data from the same source, www.kwd-globalpipe.com (KWD). According to the data submitted by Wieland, the .water distribution/heating tubes/total. share in Europe (with respect to the geographic reference area see footnote 17) developed as follows (from 1992 to 2001): Copper from 60,6 % to 47,9 %, steel from 15,2 % to 7,5 %, plastic/multi-layer from 24,5 % to 44,6 %. Also KME, in its reply to the Statement of Objections of 7 November 2003, p. 106, submitted data from KWD (the precise geographic reference area of which, however, is unclear ("Europe")), which slightly contradicts the data submitted both by Outokumpu and Wieland. According to estimates submitted by Outokumpu, p. 23751, on the other hand, the shares of competing materials developed as follows: .Water distribution/heating tubes/total. (apparently: Europe): from 1993 to 2001: Copper from 61% to 63 % (2000: 64%), steel from 18 % to 8 %, plastic/multi-layer from 21 % to 29 %. IWCC data, submitted by Outokumpu, concerning the years between 1993 and 2001 (EEA) shows similar trends as the KWD data: concerning copper a decrease of the share from 64 % to 59 % (water distribution), an increase from 67 % to 69 % (heating tubes), concerning steel a decrease from 18 % to 6 % (water distribution), a decrease from 18 % to 9 % (heating tubes), and concerning plastics/multi-layer tubes an increase from 18 % to 22 % (water distribution) and from 15 % to 21 % (heating tubes).

14 See 32130.

15 The ECPPC, London, is an action program of leading copper tube and copper fittings producers to promote the benefits of copper plumbing and heating systems to the construction industry. For further details see recital (72).

16 See European Copper Plumbing Promotion Campaign (ECPPC), http://www.copperplumbing.org/uk/statistics.htm. It is not known whether the trend of steady increase in the whole of Europe continued after 1999 or not. See also recital (388). Outokumpu confirmed in 2000 a .good demand. in sanitary tubes, 23720.

17 See 22887 (Annex 4.2 of Wieland's submission of 23 January 2003; Western Europe includes: Austria, Belgium, Switzerland, Germany, Denmark, Spain, France, Finland, United Kingdom, Greece, Italy, Ireland, Norway, the Netherlands, Portugal, Sweden). KME, in its reply to the Statement of Objections of 7 November 2003, p. 106 submitted different and contradicting figures stemming from the same source as Wieland's figures (KWD, see footnote 13).

18 According to data submitted by Wieland, demand in Western Europe amounted to 751,0 million metres in 1997 and to 751,4 million metres in 1999.

19 See 23719.

20 See 22850.

21 See 32122.

22 See 32122, 32539.

23 .Plastic-coated. copper plumbing tubes are used as a synonym for .plastic-insulated. copper plumbing tubes.

24 Brands of the companies involved in the proceedings or competitors mentioned in this Decision are the following: plain plumbing tubes: 'SANCO. (KME Group, Wieland, Boliden); .OK-ESSEM., .Tube. and .OCSA. (Outokumpu); .YORKEX., .BRITISH T. and .YORKSHIRE. (IMI); .TALOS. (Halcor); .BLACK LABEL. (Wednesbury); 'sTREAMLINE. (Mueller); .NOCARBON. (Foam); .HETCU. (MKM); 'sUPERSAN. (Buntmetall); 'sECURUS. (HME); .CARBONFREE. (Feinrohren); plastic-coated plumbing tubes: "WICU" (KME Group, Wieland; from 1988-1998 also Boliden); .FINCUPLAST., .FINCU. and .PRISOL. (Outokumpu); .CUBO. (Boliden); .KUTERLEX. (IMI); .BLACK LABEL. (Wednesbury); .HETCU-PLUS. (MKM); .EUROSAN. (Buntmetall); plumbing tubes for heating applications: .Cuprotherm. (KME Group, Wieland, Boliden); .YORKTHERM. (IMI); .HETCU-THERM. (MKM).

25 See, for example, 32541; see also 22575.

26 See 23724.

27 In Case No COMP/M.3284 - Boliden/Outokumpu, recitals 36-39, the Commission defined the geographic market at least regional for continental Europe and wider than regional for the Nordic countries.

28 See, for example, 29445.

29 Based on estimates of different parties. See also Case No COMP/M.3284 - Boliden/Outokumpu, recital 36.

30 See 29445, 22575, 24640, 24641, 24642.

31 In the United Kingdom, price lists are also issued without price lines. By consequence, price lists are issued more often, see 26407.

32 See 32541. For details see also 29442.

33 See 32542, 29434.

34 See 32541, 32542. The price line changes frequently, in many cases daily. 29442.

35 According to IMI, the price of copper can fluctuate by as much as GBP 30 per tonne on a daily basis (currently around 3% of the value), see 26396.

36 See 26406, 26407.

37 See Annex.

38 This estimate is based on turnover information provided by companies active in the plastic-coated copper plumbing tube industry.

39 Buntmetall's market share of around 2 % is already taken into account in the share of the Wieland group.

40 See 11417-11424.

41 See 32293, 32294.

42 See 32293.

43 BMSL is not directly owned by Boliden, but by Boliden Holding UK, a subsidiary of Boliden Fabrication AB in Stockholm, Sweden. However, within the Boliden Group, BMSL acts operationally as a subsidiary of BCZ.

44 See 32294, 32295.

45 See 22856-22858.

46 See 23134, 23135, 32485-32487.

47 The IMI group has gone through substantial restructuring since 1992, see 26395, 26396.

48 From 1987 until 1998, IMI Yorkshire Copper Tube Ltd. ("YTC") was part of IMI's Building Product division, which was renamed in 1998 to Hydronic Controls.

49 See 24625.

50 See 24626, 25104, 25105.

51 KME explained that the name of "EM" changed several times during the relevant period due to restructurings and acquisitions. In the following, the term refers to La Metalli Industriale SpA. (until 1986), LMI-La Metalli Industriale (until 1987), LMI-La Metalli Industriale SpA. (until 1995) and Europa Metalli SpA. (since 1995). See 32538, footnote 1.

52 See 24626.

53 See 24626.

54 See 24626.

55 See 24628.

56 See 24628.

57 See 33438.

58 See 24629.

59 See 33442, 33443.

60 See 33442-33456.

61 See 33442-33456.

62 See 33442-33456.

63 See 29528.

64 See 29529.

65 [...]

66 [...]

67 [...]

68 [...]

69 [...]

70 [...]

71 [...]

72 [...]

73 [...]

74 See 29555, 29556.

75 See 24157, 24158.

76 See 24157, 24158.

77 See 32541.

78 See http://www.coppercouncil.org/

79 See 23594.

80 See http://www.eurocopper.org/

81 See 1862. Organs and working groups are: Board of Directors, ECI Communications Committee, ECPPC (Tubes), ERCC (Roofing), Program Review Committee, ECEC (European Copper Environmental Com"), 11326, 11327.

82 For details concerning Outokumpu's involvement see 11325, 11326.

83 See http://www.copperplumbing.org/, and 8611-8620.

84 See http://www.kupfer-institut.de/front_frame/index.php

85 See http://www.wv-m.de/welcome.asp?page_id=16

86 See http://www.dvgw.de/

87 See http://www.afnor.fr/portail.asp

88 See http://www.uni.com

89 OJ C 207, 18.7.1996, p. 4.

90 [...]

91 According to Article 34 (2) of Regulation (EC) No 1-2003, "procedural steps taken under Regulation No 17 and Regulations (EEC) No 1017-68, (EEC) No 4056-86 and (EEC) No 3975-87 shall continue to have effect for the purposes of applying this Regulation".

92 See 29654-29694.

93 See 11709, 11710.

94 See 11717-11732 (Fax of May 30, 2001; Letter of June 1, 2001) and annexes. On 19 November 2001, Outokumpu Oyj submitted a supplementary Memorandum entitled "Economic context of the European copper tubes industry"; see 23714-23754.

95 KME authorised the Commission to use any information submitted in the Industrial Tubes Case COMP/38240 in the present proceedings that it believes to be relevant, 25523.

96 See 22023-22058.

97 See 33419-33433.

98 See the note of KME dated 15 June 1989 that was written in the context of bilateral meetings that appear to have served as a preparation of the European copper plumbing tube meetings: "maintain domestic discussions".

99 [Information not relevant to the EU proceedings]

100 [...]

101 [...]

102 With respect to national arrangements see 32043, 0015-0017, 0020-0023, 0023-0025, 0036-0038, 0085-0090, 0354-0357 and the corresponding Annex 2 and 3 (0619-0848), 0357, 0358 and corresponding Annexes 5a-5c (0393-0398), 0358-0362 and corresponding Annexes 6a-6c (0399-0404), 0363, 0364 and corresponding Annex 8 (0409-0411), 0365 and corresponding Annexes 9a-9c (0412-0417), 0365-0367 and corresponding Annex 10 (0418-0420), 0367-0369, 0370, 0371 and corresponding Annexes 14a-14f (0440-0477), 0371-0373 and corresponding Annexes 15a-15k (0478-0505), 0373-0374 and corresponding Annex 16 (0506-0509, under .Trend of Trade"), 0384, 0385 and corresponding Annexes 24a-24c (0613-0618), 0385, 0386, 0447, Geneva meetings for the French market took place on or around these dates: December 16, 1996; January 29, March 10, July 23, November 21, 1997; January 20, March 4, June 12, September 8, November 13, 1998; February 11, 1999; see also 32543, 0985-0987 (faxes were sent: for example 4 June 1997, 12 August 1998, 18 March 1999, June 1999, 24 July 2000), 0988, 990, 991, 25491, 25492, 1283-1292, 31127, 7055-7057, 11230, 32616, 32622, 32630, 32631, 29677, 12078, 15892, 25494-25496, 15895-15902, 15947-15957, 32720, 32721, 32543-32544, 32054-32059, 22522-22674.

103 See, for example, recital (267), 0363, 0406-0408; 0365-0367, 0419, 0420; 0384, 0385, 0613-0618, 15952. Further references are included in footnote 102.

104 See 23550, 32700, 32703; regarding statistical information of the IWCC see 16858-16965.

105 See 32541, 32722.

106 See 32722.

107 See 0357, 0358, 0394-0400, 0367-0369, 15951, 15952, 32700, 32701, 32703, 32706; see also 32541, 22875.

108 See 23550, also 29656; below recital (256).

109 See 29656, 32052. According to KME, the SANCO? trademark was established in 1981.

110 See KME's explanations on page 32052, also Outokumpu's statement 23551.

111 See 29673. Before its bankruptcy, the secretariat was held by UCZ.

112 See 32299.

113 See 24631.

114 See 32298.

115 See 32300 and 24631. KME explained that it also holds the SANCO(r) International trademark

(since July 24, 1981) and the SANCO(r) Europe trademark (since 1 April 1996), 24631; see also the overview on 24661. According to the licence contract, KME is "the owner of the trademark SANCO(r) which has been registered and protected in the countries of the EC and various other countries", see 32379 and 24376. See also the trademark licence contract between KME and Wieland that indicates KME as the holder of the trademark rights for example in the following geographic regions and European countries: Germany, Europe, Denmark, Finland, United Kingdom, Ireland, Norway, Poland and Sweden.

116 See 24631, the expiration is between 2008 and 2014.

117 See 29656, 32052, 32299, 32300.

118 See 32299-32303.

119 See 29656, 32052. For the licence contracts see 24633.

120 See, for instance, the provision Art. 3 paragraph 1.2 of the "Patent License and Know-how Agreement (SANCO?)" of 1990 attached to the letter of 18 December 1990 from KME to Outokumpu, in which KME proposed to Outokumpu to join the SANCO club, 24019, 24030, 24031. In 1992, the fees amounted to an initial payment of ECU 100.000,-, a minimum royalty p.a. of ECU 81.000,-, and ECU 22,70/t up to 5.000 t/y, ECU 14,60/t up to 7.000 t/y, and ECU 9,70/t for more than 7.000 t/y, see 24054.

121 See below recital (141).

122 See 24633.

123 [...]

124 [...]

125 [...]

126 [...]

127 See 24633 and 24634.

128 See 24634.

129 See 25388, 29656, 29657.

130 See 29676, 29677, 33419; 33431, 33432.

131 See also the meeting described in recital (310), when Boliden met with KME to provide correct volume figures.

132 See 9706 (internal Wieland Memos, in which Wieland discusses the "equal product" (or one brand) strategy for SANCO and Cuprotherm tubes. An exception was made for the Austrian market"), also 9708. Furthermore see the explanations of KME, recitals (151)-(153).

133 See 29664, 29665

134 See 33418-33433, 33423.

135 See 33427, 33428, but also 29662, paragraph 31-43 of KME's reply of 7 November 2003.

136 See 29661.

137 See 29659, 29669, 29671, also 33421.

138 See 29671.

139 See 29671.

140 See 29658, 29659.

141 See 33422.

142 See 29657-29659, 29664, 29690, 33422.

143 See 29657, 29664.

144 See 29664, 29669.

145 See 29658, 29659, 29689 and 29690.

146 See 29665.

147 See 29659, 29665, 33422.

148 See 33423.

149 See 29659, 33420.

150 See, for example, recital (169).

151 See 29690, 33420-33422.

152 See, for example, 29671, 29672 and 29675.

153 See 29657-29659, 29672, 29692, 33420. The allocation of customers involved also agreements on the rebates to be granted.

154 See 29660.

155 See 29660.

156 See 29667-29669, 29676, 33424. Switzerland as a location was chosen in order to conceal the meetings and reduce the risk of detection.

157 See 29673 and 29674.

158 See 29669, 29673 and 29674. For examples of export/import statistics and data that was submitted

In such a context by Wieland, see 7586-7590.

159 See 32307, 32429-32431.

160 Royalties include a yearly minimum licence fee. See footnote 119, 29656, 24634-24248.

161 See 29656, 29663, 29664, 29689, 32301, 32304.

162 See 24637-24639.

163 For exceptions see recitals (119), (158) and (360).

164 See 29670.

165 See 29669.

166 See 29669, 29674.

167 See 29671, 29672.

168 See 23994.

169 See 23994, 24019, 24023, 24044-24055.

170 See 29680.

171 See paragraph 100 of Wieland's reply of 7 November 2003, which, according to oral explanations

given by Wieland, is not to be understood as any contestation of facts. Wieland explained that it does not have any recollection with respect to any arrangements concerning SANCO, WICU and Cuprotherm tubes.

172 According to KME, meetings took place on the following dates: 1992: 08.01'92/09.01'92, 23.06'92, 07.08'92, 23.11'92/26.11'92; 1993: 04.06'93, 01.09'93, 29.11'93/30.11'93, 08.12'93, 21.12'93; 1994: 12.04'94/15.04'94, 21.09'94; 1995: 05.04.1995, 27.04'95, 29.11.1995; 1996: 24.01'96, 08.03'96, 03.04'96, 03.06'96, 16.10'96; 1997: 24.06'97, 27.06'97; 1999: 16.06'99, 22.09'99, 21.10'99, 18.11'99/19.11'99; 2000: 09.02.00/10.02.00. See 22522-22526 (summary) and 22527-22568 (travel expenses and supporting evidence). Boliden recalled meetings concerning either WICU or SANCO in 1996: 16.12'96; 1997: 11.12'97; 1998: 29.7'98, 19.02'98, 24.02'98, 30.07'98; 1999: 9.3'99, 28.5'99, 21.6'99; 2000: 12.12.00; 32304, 32305. Travel expenses of Wieland show participation of its employees at the following dates: 1994: 25.5'94 (9460); 1996: 8.3'96, 26.3'96, 17.4'96, 30.4'96, 5.6'96, 21.6'96, 21.8'96, 10'9'96, 26'9'96, 8.11'96, 19.12'96 (9293, 9294, 9454, 9453, 9524); 1997: 14.2'97, 17.2'97, 27.6'97 (9441, 9447); 1998: 19.2'98, 9.6'98 (9281, 9430, 9512); 1999: 15.3'99, 21.6'99 (9268, 9361).

173 See 32055.

174 See 32055, 22522-22526 (summary) and 22527-22568 (travel expenses and supporting evidence).

175 See 32054, 32055.

176 See 32055.

177 See 32055. With respect to price line increases of KME between 13.11.2000 and 9.4.2001 see 31320- 31337.

178 The products were broken down by "Stangen"/"Ringe" (bars/rings). Boliden confirmed the exchange of such information, see 32340-32347, 32302-32304.

179 See 8467.

180 See 8467: ". falls von KME an unserer Absatzentwicklung herumgemäkelt wird!"

181 See 8467.

182 See 8478.

183 See 8477 (including the years 1985-1987, and figures from January-December 1988). Boliden was copied on the circulation of this table.

184 See 8476 (including the years 1986-1988, and figures from January-December 1989). Boliden was copied on the circulation of this table.

185 See 8485 (including the years 1987-1989, and figures from January-December 1990). Boliden was copied on the circulation of this table.

186 See 8474 (including the years 1987-1990, and figures from January-December 1991). Boliden was copied on the circulation of this table.

187 See 8473 (including the years 1989-1992, and monthly figures from January-December 1993). Boliden was copied on the circulation of this table.

188 See 8472 (including the years 1989-1993, and monthly figures from January-December 1994). Boliden was copied on the circulation of this table.

189 See 8471 (including the years 1990-1995, and monthly figures from January-December 1996).

190 See 8470 (including the years 1990-1996, and monthly figures from January-December 1997).

191 See 8469 (including the years 1990-1997, and monthly figures from January-December 1998).

192 See 8468 (including the years 1992-1999, and monthly figures from January-March 2000).

193 Not broken down by "Stangen"/"Ringe" (bars/rings).

194 See 8479-8486 (including the years 1987-1999, and monthly figures from January-December 1990, 1991, 1992, 1993, 1994, 1996, 1998, and from January-March for 2000). Boliden was copied on the circulation of these tables.

195 See 8487-8510 (including the years 1984-1999, and monthly figures from January-December 1987, 1988, 1989, 1990, 1991 (only non-SANCO), 1992, 1993, 1994, (only January-August) 1996, 1997, 1998, and from January-March for 2000). Boliden was copied on the circulation of the tables of 1989, 1990, 1991, 1992, 25.01.1994, 24.01.1995, 21.09.1995, 21.01.1997, 26.01'98, 18.01'99, 20.04.2000. Boliden was only copied on tables that contained non-SANCO figures from 1989-1992.

196 See 8511-8521, also broken down by "Stangen" and "Ringe" (including the years 1984-1999, and monthly figures from January-December 1987, 1988, 1989, 1991, 1992, 1993, 1994, 1996, 1997, 1998, and from January-March for 2000). Boliden was copied on the circulation of the tables of 1988, 1989, 1990, 1991, 1992, 25.01.1994, 24.01.1995, 21.09.1995, 21.01.1997, 26.01'98, 18.01'99, 20.04.2000.

197 See 8522-8533, (including the years 1987-1999, and monthly figures from January-December 1987, 1988, 1989, 1991, 1992, 1993, 1994, 1996, 1997, 1998, and from January-March for 2000). Boliden was copied on the circulation of the tables of 1989, 1990, 1991, 1992, 25.01.1994, 24.01.1995, 21.09.1995, 21.01.1997, 26.01'98, 18.01'99, 20.04.2000.

198 See 7572; the period covers 1992-01/2001. In addition, concerning all indicated countries, see 7567-7572 (SANCO).

199 Including the months of 1986, 1987, 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1996, 1997, 1998, 2000, and January-March 2000.

200 See 8534-8548.

201 Precisely for the years 1986, 1988, 1990, 1991, 1992, 1994 (26.01") and 1995 (24.01").

202 See 8549-8555; also 7573-7581 (WICU).

203 Including the latest monthly figures (starting from 1996); see 8556-8565.

204 See 8564.

205 See 31059.

206 MHH, Foil Co, Schorendorf, Babcock, Metallhandel Neuss and HSM.

207 See 31055 "im gemeinsamen Preisinteresse".

208 See 31055.

209 See 31056.

210 See 31057.

211 It seems that the abbreviation "de. M." stands for an executive of Boliden.

212 See 31053, 31054. See also the note summarising a preparatory telephone conversation between Boliden and KME 31060.

213 See 31053, 31054.

214 See 31155.

215 See 31155.

216 See 9844-9846.

217 See 9845.

218 See 9846.

219 See Wieland memo of 14 March 1997, 8038, 8040.

220 See 7770 (agenda).

221 See 7770 (agenda).

222 See also the rest of the document 31140.

223 See 8534.

224 See 8602-8609.

225 See 8590-8600.

226 See 8571-8589.

227 See 8567, 8568.

228 See 8568-8570.

229 See 8566-8609; KME was agreed to be the market leader for SANCO and WICU tubes, Wieland for Cuprotherm (CT-FBH), see 8567.

230 See 8602.

231 See 8602.

232 See 8603.

233 See 8604.

234 See 6905.

235 See 6906-6913.

236 See 6909, 6910.

237 See 31142, 31143-31154.

238 See 31143.

239 See 6587 (travel expenses), 7751 (agenda).

240 See 6597 (travel expenses).

241 See 7884.

242 See 6823.

243 The information was supposed to flow from Buntmetall, Mueller and Halcor to Wieland, and from Outokumpu, "MKM" and BCZ to KME.

244 See 6824.

245 See 7747 (agenda).

246 See 32567-32576, 32580.

247 See for details 32567-32576.

248 See 32569.

249 See 32571, 32572, 32573.

250 See 32580 (E-mail).

251 See 32578.

252 See 23550, 23551, 23994.

253 See below recital (238).

254 See 32700.

255 See 11763, 32700 and 23551, 23552.

256 See 32700.

257 See 32700, 31815-31817.

258 See 32700.

259 See 32700.

260 See 32700, see also 29433. With respect to the exchange of price lists see for example 0479-0505 (Spanish price lists of Outokumpu from 18/09/97-1/2000).

261 See 32700, 32701. Operational level meetings were sometimes held in connection with the IWCC copper water tube committee.

262 See 32701.

263 See 32540, 32541. Depending on the occasion, participants included Wieland, Outokumpu, Boliden, HME, Desnoyers, IMI, Wednesbury, KME, Tréfimétaux and EM.

264 See 32540, and the correction of the statement in 32053.

265 See 32541.

266 See 32053, 32054; 32540-32542.

267 See 32541-32543.

268 See 32541, 32542.

269 See 32542 (including footnote 6).

270 See 32542.

271 See 22867.

272 See 22867, 22868, 22875.

273 See 22870.

274 See 22872.

275 Market leaders were KME (Germany), KME/Tréfimétaux (France), Outokumpu (Spain), IMI (the United Kingdom), HME (the Netherlands).

276 See 22872, 22873.

277 See 22873, 22874. The agreements did not include certain payment conditions and bonus payments.

278 See 22874.

279 See 22874.

280 See 22875. According to an internal table of Wieland (see 8633-8635), Wieland employees attended, for example, the following meetings in 1992: 31.03/01.04/02.04 IWCC-Meeting in Paris, Copper Tube ECPPC Steering Committee Meeting, and conversations between the meetings (Z/H); 24.04 ECPPC Meeting in Frankfurt (Z/H), 22.05 ECPPC Meeting in London (Z/H); 12.08. Copper Tube Meeting with Wieland in Düsseldorf (Z/H); 23.09. Copper Plumbing Tube Meeting in Frankfurt (Z/H, ZR/Mo), 29.10. IWCC Copper Water Tube Meeting in Zurich (Z/H), 03./04.12 Copper Plumbing Tube Meeting in Regensd. (Z/H).

281 See 22875.

282 See 22876.

283 See 22877.

284 See 22877, 22878.

285 See 33419, 33420, 33423 with respect to the absence of any interruption see 33424.

286 See 33432, 33433.

287 See 32701.

288 See 23552, 23553.

289 See 23552, 23553.

290 See 23553, 32701, 23997.

291 See 23552.

292 See 32042. It is indicated that meetings lasted at least until first quarter of 2000. With respect to the relevant dates, see above recitals (199)-(202).

293 See 0374-0377 and corresponding Annexes 17a-17g.

294 For example of the DKI, WV Metalle, ECPPC, ECI, and ICA.

295 See 22877.

296 See 0028, 32042. This happened with the purpose to conceal the meetings, see 32706.

297 See 32042.

298 See 22877.

299 See 22877. In addition see recital (216).

300 See 32706.

301 See 32043; see also 23553, 23554, 32702-32704; 32544, 32545; 22877.

302 See 32043.

303 See 0025, 0026. For further details see 32707, 22875-22878.

304 See 16972, 16973.

305 See 0026-0028. For further details see 32707, 22868-22874.

306 See 32043.

307 See 0027, 0028; see also 32705, 32706, 32707.

308 See 100-104, 2069-2072, and the invoice for Jan-Dec 1998, 2058, 2059. IMI wrote that "it makes sense for you to "do the honours" and invoice the other participating companies". Also see the agreement sent to IMI on 4 February 1998, 2061, 2062: Participants were IMI, KME, Wednesbury, Outokumpu, and Wieland. Also other employees of the companies were involved in setting up the scheme and the respective correspondence. Also see 2060-2078 (2069), 11541-11548, 32667, 23997, 32703, 32544, 32545, 22871, 22872. See also recital (306).

309 See the form to be filled out by each participant, 2064, and the scheme according to which the data was submitted 2065-2067. See also 30002-30011.

310 See 32043, 0228, 0374-0377 and corresponding Annexes 17a-17g; 2066, point 6 of the scheme. It has to be noted that the scheme was reviewed by lawyers. Provision N° 6 was identified by lawyers as potentially raising competition law concerns, 2073: "If these guidelines or the arrangement as such fell under the scrutiny of a competition official[...] this no doubt would raise unnecessary questions and doubts about the legality of the arrangement. Therefore, it would [be] advisable to modify these guidelines so that they would require each producer supplying any particular market to provide WBMS with an estimate of the total consumption of that market based on which WBMS then would determine an average of all such estimates".

311 See 0374-0377 and corresponding Annexes 17a-17g, 0377-0378 and corresponding Annexes 18 and 19, 0988, 0989; also 25492, 1283-1287.

312 See 32043.

313 With respect to participants of KME, Wieland, IMI, Boliden and Outokumpu see 32704.

314 With respect to participants of KME, Wieland, IMI, Boliden and Outokumpu see 32704. With respect to HME's and Wednesbury's involvement see 32706.

315 See 23553, 23554; see also 32702-32704, 32706.

316 See 32706.

317 See 23553, 23554, see with respect to the functions of the market leader 32705. For details concerning the function of the market leader see 32707.

318 See 32704, 32705.

319 See 32544. The setting of price guidelines by the market leader involved the proposition of pricelines that the others would follow (see recital (206)).

320 See 32544. According to KME, "non-core members were [] mainly involved in promotional and marketing activities." (footnote 9, 32544).

321 See 1358-1360.

322 See 22877. This explanation applies to the memos of 10.06.1998 (7555-7557), 21.10.1998 (7518), 18.08.1999 (9723), 20.10.2000 (7871-7873), 08.01.2001 (7862-78639), (22956-22965) and of 17.09.1997, 10.09.1997, 06.10.1997, 02.02.1998, 30.10.1998, 11.12.1998, 23.02.1999, 18.03.1999, 29.04.1999, 20.05.1999, 28.06.1999, 23.07.1999, 08.09.1999, 18.05.2000, 09.10.2000 and of 12.03.2001 (22253-22264). With respect to the participation in meetings of Wieland's employees see the travel expenses 9268-9328, 9329-9470, 9471-9501 and 9502-9537.

323 For details see 22870, 22871.

324 The Commission understands that the notions "Steering Group" or "Vorstandsrunde" refer to top level meetings.

325 The Commission understands that the notions "Working Group" or "Sales Representatives" refer to operational level meetings.

326 Other terminology for EDWD is "Drinking Water Directive" meetings ("DWD. meetings).

327 See 23550. See also 7187, 30981.

328 See, for example, the handwritten notes: 11087, 11088 (undated), 11088-11091 (9.1.1989), 11091- 11094 (16.2.1989), 11094-11095 (7.3. "Madrid"), 11096-11098 (10.11.1988), which evidence discussions and/or the exchange of sensitive information and contain information on the nature of the cartel.

329 See 11087, 11088.

330 See 11088-11091.

331 See 11091-11095.

332 See 11094, 11095.

333 See 30979.

334 See 7184, 7185. The participation of KMEis confirmed by notes of Outokumpu, see 11440.

335 See the initials on 30967.

336 See 30968-30977.

337 "Stangen/Profile aus Cu, Bänder aus Cu, Bleche/Scheiben aus CU", "Rohre aus CU, gerade, Wanddicke > 0,6 mm", "</= 0,6 mm", "Rohre aus CU, in Ringen, Wanddicke > 0,6 mm", "</= 0,6 mm", and further tubes of "CuZn".

338 See 30968-30977.

339 See 11440 and the notes of this meeting 11444.

340 See 11440-11443.

341 See 11438, where the possibility of negotiation with IMI and Wednesbury is discussed.

342 See 11436-11439. See also the notes of an employee of Outokumpu 11431-11433 and 11434-11435 related to two other meetings.

343 See 7183.

344 A number of undated "posters" stemming from IMI show notes apparently stemming from competitor meetings (subjects were, for example, market statistics, monitoring of shares; price/customers structures; "rules of the game); see 28812-28859.

345 See 11456-11461, 11763, 32708.

346 See 32708.

347 See 11460, 11461, 32657, 32658.

348 See 11460, 11461, 32657, 32658.

349 See 11462.

350 See 23551, 11763, 32708.

351 See 32708, 31816-31818.

352 See 31818.

353 See 23551, 23552, 32708.

354 See recital (249). See also notes of an employee of Outokumpu that contain (Outokumpu's) volume information of Germany; Italy, France, the United Kingdom, Ireland, Benelux, Spain, Portugal, Greece, Austria, Switzerland, and Scandinavia (Sweden, Finland, and Norway), 11463; 23551, 31816.

355 See 32708.

356 See 23552.

357 See 23551, 23552, 32708, 32709 ("a few times"). See, for example, notes of an employee of Outokumpu of a Frankfurt meeting (11099-11102), which report on volumes and the intention of freezing market shares and arrange an information exchange.

358 See 11099-11102.

359 See 23551.

360 See 11763.

361 See 11763, 32708.

362 See 32709.

363 See 32709.

364 See 32708, 32709.

365 See 32709.

366 See 23551.

367 See 23551.

368 See 11460, 11461, 32657, 32658, 11763.

369 See 32709, 31819, 31820.

370 See 0018, 15949.

371 See 0018, 0019.

372 See 0019.

373 See 0019, 0986, 15949.

374 See 32709.

375 See 31135 "dürfte. realistisch sein".

376 See 31135.

377 See 32626.

378 See 23635-23636.

379 See 32626.

380 See 5308-5312.

381 See 15949, 15950.

382 See 32627.

383 See the handwritten tables 32632-32639, and 32630, 32631.

384 See 32630.

385 See the handwritten tables 32632-32639.

386 See the invitations/propositions for a meeting on 27 February 1992 (11273) and 18 June 1992 (11272).

387 Tables contain sales volume information that dates back as far as 1987. Tables date from 9 March 1992 to 18 December 2000 (5462-5508). See also tables dated 27.3.1998, 5408-5410; handwritten notes 5411-5414; tables of 18.02.1998, 5415-5420, of 13.02.1998, 5421-5427; handwritten tables "basis Januar 1998", 5428-5431; handwritten notes 5434-5435.

388 See 10826 (letter describing the general content of the meeting), 10827, 10816-10818 (memo). Concerning a meeting between EM and Boliden see 10847-10851, a contact between EM and Outokumpu 10830, 31514.

389 See 10816-10818.

390 See 10809, 10811.

391 See 31495.

392 See 10819; see also correspondence between IMI and EM 11023, 11024.

393 See 10819, 10820, 10853, 10982, 11019-11021, 11023, 11024.

394 See 10820, 10853, 10982 (invitations), 10985 (confirmation Wieland).

395 See 22789-22814.

396 See 31506, 10854. EM was excused.

397 See 23552.

398 See 32709, 22815 (invitation).

399 See 22815.

400 See 31536, 22815.

401 See 32709.

402 See 32709.

403 See 22367, 22369-22372 (travel expenses, and .Bewirtungskosten - Abrechnung. for the participants of the working group).

404 See the general description of KME of the Zurich-meetings in 32540-32542, 32052, 32053.

405 See 32053.

406 See 32507, 31540, 31448: the meeting was originally planned for 21 February 1993 or 18 or 23 March 1993, 31450, 32710.

407 See 32507, 31540, 31448

408 See 0357-0358, 0393-0398.

409 For details see 0358-0362 and 0400-0404, 15951.

410 See 22367, 22373-22376 (travel expenses, and "Bewirtungskosten-Abrechnung" for a SANCO meeting).

411 See recital (264) and 32540-32542, 32052, 32053.

412 See 23616, 23617, 23597.

413 See 32710.

414 See 32710.

415 See 22367, 22377, 22378.

416 See 32540-32542, 32052, 32053.

417 See 32600-32607.

418 See 11032-11033.

419 See 22367, 22379-22381 (travel expenses).

420 See recital (264) and 32540-32542, 32052, 32053.

421 See 0019-0020, 15951.

422 See 15951.

423 See 15951 with further explanations concerning information exchange and monitoring.

424 See 0367-0369.

425 See the notes on page 0425 and explanations on 0367-0369.

426 See 32710.

427 See 3450-3453.

428 See 10832-10841.

429 See the memo that was most likely prepared for a meeting with Wieland, 11034-11035, the knowledge of Outokumpu of Wieland's capacity, 11038.

430 See 22367, 22382 (travel expenses).

431 See 22738.

432 See 22784.

433 See 22785, 22786.

434 See 22787-22788.

435 See 32540, 32541, 32052, 32053. See 22367, 22383 (see, however, the contradicting dates on the travel expense report).

436 See 22367, 22383 (travel expenses) on the one hand, and 32710 on the other.

437 See 32710, 22738-22743.

438 See 22738, 22739.

439 For details see 22740, 22741.

440 See 22742, 22743, 32710.

441 See 0367 and corresponding Annex 11 (0422, 0423).

442 See 0367.

443 For example, from June 1994 until December 1996, KME's employees participated at least in the following events concerning copper plumbing tubes: 21/22 June 1994 an ECPPC Working group in Brussels; between 10-14 July 1994 an ECPPC meeting in Switzerland; 19/20 July 1994 an ECPPC meeting in Frankfurt; 18-20 October 1994 an ECPPC Steering Committee in Brussels; 21-23 November 1994 an ECPPC/Promotion Committee in Brussels; 12/13 December 1994 an ECPPC meeting in Frankfurt; 21-23 February 1995 a Promotion Committee in London; 07/08 March 1995 a meeting in Zurich; 24 March 1995 a meeting concerning the promotion campaign in London; 17/28 March 1995 a meeting of the promotion committee in Paris; 29-31 May 1995 an ECPPC Promotion campaign in Barcelona; 29 August 1995 an ECPPC-Meeting for Curpotherm-Werbung in Frankfurt of Switzerland; 11/12 September 1995 an ECPPC Promotion Campaign in Kopenhagen; 18 September 1995 an ECPPC Steering Committee in Brussels; 12 December 1995 an ECPPC Steering Committee in Brussels; 31 January-1 February 1996 an ECPPC meeting in Brussels; 18-19 March 1996 an IWCC-Meeting and ECPPC-Meeting Committee in Barcelona; 30 May 1996 an ECPPC Promotion Committee in Frankfurt; 12 June 1996 an ECPPC meeting in Brussels or Budapest; 9-10 December 1996 in Brussels a meeting of the Promotion Committee. See also 31205-31224 (travel expenses) and 31271-31274 (minutes).

444 See 33420.

445 See KME's reply to the Statement of Objections of 7 November 2003, p. 35.

446 See KME's reply to the Statement of Objections of 7 November 2003, p. 35, confirmed in the Oral Hearing of 28 November 2003, where a representative explained that .KME could not rule out. that such contacts occurred.

447 See slide 12, 13 of "Copper Plumbing Tubes - Oral Hearing - Otuokumpu Speaking Notes" of 28 November 2003.

448 See 22875.

449 See 32711.

450 See KME's Reply to the Statement of Objections of 7 November 2003, p. 35, 36.

451 See 22868.

452 See 22868; see also the table dated 23.04.1996, 31137.

453 See 32711.

454 See 22875.

455 See 32711 - Internal documents of Outokumpu of 10 April 1996 state that "if the situation in Spain does not improve, we will hit strongly back - and then a war with KMO is unavoidable!" (11040). Concerning Spain: "Kabelmetal has been disturbing competition and it does not seem to respect the market structure and price. It would be a shame if the results of the finally promising AENOR campaign would be voided because of KME's actions; the consequences would be like the Lappish war"; concerning Poland: "IMI and BMA have destroyed the market price", 11041.

456 See 32711.

457 See 11764, 32711, also 32543.

458 See 32711.

459 See 32711.

460 See 22367, 22385-22390 (travel expenses).

461 See 32543, but also recital (264) and 32540-32542, 32052, 32053.

462 See 22744-22750.

463 See 22876. According to Wieland, this meeting concerned high executives.

464 See 32711.

465 See 22868.

466 See 32711. Also see the notes of Outokumpu of the previous meeting 22746, 22749.

467 See 11764.

468 See 32711.

469 See 32711.

470 See 32711, 32712.

471 See 11381-11383. See also 16973: According to Mueller, it was reported that Outokumpu, at a later stage, did not attend a (number of) meeting(s) because of Outokumpu's concern with the European Commission.

472 See 7765, 22876. Wieland indicated that participants included usually: KME, IMI, Outokumpu, Wieland. On only one or occasional meetings participated: Boliden, HME and Buntmetall.

473 See 32712.

474 See 7770.

475 See 22868. According to Wieland, typically, the following companies were represented: KME, Tréfimétaux, Outokumpu, IMI, Boliden, Mueller, Halcor, HME, Wieland and Buntmetall. Certain companies and/or employees might not have been present at each of the meetings.

476 See 32712.

477 See 32722, 22751.

478 See 22751.

479 See 0345.

480 For further background see 16969, 16970.

481 See 11764, 32712, 22776, 22777 (invitation); 9441, 9520 (travel expenses). Minutes and an expense report confirm this, see 22408, 22409.

482 See 32712.

483 See 11764, 32544, 22868, 22392, 22404-22410 (invitation, minutes, travel expenses). KME indicated that also Outokumpu participated. Its participation is evidenced by the minutes. According to Wieland, this meeting concerned the sales representative of participating companies.

484 See 32544, 22404, 22405.

485 See 32544.

486 See 7771 (agenda), 11764, 32712, 32713, 22768-22770 (invitation), 22771-22775 (notes).

487 See 32712, 32713.

488 See 22771-22775.

489 See 7771 (agenda), 22876, 22392, 22411, 22412. According to Wieland, this meeting concerned high executives. An executive of Wieland summarized the meeting, see 22155, 22156.

490 See 11764, 32713.

491 See 32713.

492 See the complete memo (including tables) under 22969-22978.

493 See 22969-22975. See also the memo dated 6 October 1997, 22157-22159.

494 See 0025, 0026.

495 See 0026; the offer was made on a "Novotel Note"; 16970, 16971.

496 See 0026.

497 See 0374-0377. See also the tables dated 19.5.1997 and an internal note of IMI of 13 October 1997, 3425-3432, 15896-15901, 15953-15955. See also recital (223).

498 See 0376.

499 See 0376, 0377.

500 See 17709-17716.

501 See 0374-0376, 0510-0542 (Annexes 17 a-g). Concerning Mueller's correspondence with the WBMS, see also 0377-0378, 0849-0984 (from 4/2/98-10/12/99).

502 [...]

503 See, for example, 0377-0378, 0849-0860 (0860) and 0860-0984. With respect to communication of Desnoyers concerning WBMS see also 1283-1287.

504 See 31182 (travel expenses), 11764, 32713, 22762-22767.

505 See 11764, 22762-22767.

506 See 22868, 22392 and 22413-22420 (minutes, travel expenses).

507 See 32713.

508 See 32713.

509 See 11764, 32713.

510 See 22868 ("sales representative meeting"), 22393, 22421-22428 (minutes, travel expenses).

511 See 31182, 11764, 22762-22767.

512 See 31116 (travel expenses), 11764, 32713, 32714.

513 See 32713, 32714.

514 See 22868. According to Wieland, this meeting concerned the sales representative of participating companies. See 22393, 22429-22432 (minutes, travel expenses). See also the (undated) table containing 1997 volumes and market share of competitors for "F, UK, G, NL, Sp" found at KME, 6972.

515 For the complete memo (including the table) see 22160-22168.

516 See 22160, 22161.

517 See 22162-22167.

518 See 22168.

519 See 31185 (travel expenses), 9423 (travel expenses), 22868.

520 See 11765 and the correction in 32714, 22393, 22433-22439 (minutes and travel expenses).

521 See 22434, 32714.

522 See 7763, 22876.

523 See 32714.

524 See 0028, 0029, 17717, 17718 (minutes), see also 15897, 15953.

525 See 11765, 32641, 32642 (minutes), 32714, 22868, 22394, 22440-22444 (minutes, travel expenses).

526 See 0028, 0029, 0988, 31187 (travel expenses), 15953, 11765, 22641, 22440, 22441.

527 The strategy is not a finalised one and still had to be internally reviewed, see 5406.

528 See 9406 (travel expenses), 22869.

529 See 22876.

530 See 22394, 22445, 22446.

531 See 32715.

532 See 1377, 32105, 1825, 2056-2057, 2226, 2227, 7736-7738 (minutes), 7740 (booking confirmation), 7761 (agenda), 16971, 22876, 22943-22947, 32715.

533 See 7555-7557.

534 See 0030, 0378, 0988, 1284, 15898, 15953, 25501-25504, 15959, 15963-15965, 17719 (minutes).

535 See 22869, 22394, 22447 (minutes).

536 See 0030, 22447, 31111, 31112 (travel expenses), 31189 (travel expenses), 22869.

537 See 0030.

538 See 32715-32720.

539 See 0378, 0545.

540 See 1284.

541 See 0030, 22870-22873.

542 See 22394, 22448-22450 (travel expenses), see also 32715.

543 See 22450 (travel expenses).

544 See 2057, 7735, 16971, see also 32715.

545 See 32715.

546 See 0030-0032, 0378, 0379, 0545, 0989, 1284, 31108-31110 (travel expenses), 31190 (travel expenses), 9406 (travel expenses); 15899 (including Annex 11), 15954, 25505, 15967, 25506, 17720 (minutes).

547 Boliden confirmed the participation of HME, see 32522. With respect to the exchange of HME's price lists to competitors see 0371, 0372, 0441-0477 (from 02/1996-07/2000). A handwritten remark on one of the price lists confirms as of which date the price list will be effective (1/10/00), 32409, 32413.

548 See 0030-0031.

549 See 0379, see for example Mueller's invitation to Halcor 0547.

550 See 0030-0032, 0989, 1284, 31108-31110 (travel expenses), 31190 (travel expenses), 11765 (without specific recollection, 32715), 22869, 22395, 22451-22454 (minutes and travel expenses). According to KME, the meeting of 28 August took place in the Zurich Hilton Hotel, the meeting of 31 August at the Airport Forum. Apparently in preparation of the meeting, KME asked Tréfimétaux for a "market update", see 5452-5464. 23428-23432 (minutes, notes), see also 23433, 25321-25325, 32409, 32413.

551 See 22869.

552 See 22395, 22455. Participants of this meeting included KME, Wieland, IMI, HME, EM, Buntmetall, Halcor and Boliden.

553 See 0032, 1284. However, sales data was exchanged also between participants of the group of the five and the group of the nine, see 15954-15957.

554 This appears to refer to the information exchange system described under recital (223).

555 See 23385-23392, 23429-23432; 25335-25338.

556 See recital (326) and 23433.

557 See 25323.

558 See 22501.

559 See 0032-0033, 0379-0380, 0549 (invitation), 0551 (confirmation), 0989, 1285, 15898 (including Annex 13), 15955, 17721, 9406 (travel expenses).

560 See 11765, 32640 (invitation), 32715, 32716, 22869, 22396, 22456 (minutes), 22457 (travel expenses), 23435 (invitation), 25325.

561 Boliden confirmed the participation of HME, see 32522.

562 See 32640, 22396, 22456, 31191 (travel expenses).

563 See 0379-0380, 0989, 1285, 32640-32642, 22869. According to Wieland, this meeting concerned the sales representative of participating companies.

564 See 23412.

565 See tables 23412, 23413.

566 See 23413.

567 See 23413-23415.

568 See recital (306).

569 See 0380 and corresponding Annexes 22a-22c (0552-0562), 989, 32591-32599; 23492-23512, 32306, 32423-32427.

570 See 0092, 0124-129, 0260 (Outline Programme), 1377, 32106, 1826 (diary), 6565 (travel expenses), 7760 (agenda), the agenda of a Mueller employee had a reference to an "elephants"-meeting on September 17, 1998, 16972.

571 See 0125-129, 1826 (diary), 6565 (travel expenses), 7760 (agenda).

572 See 32716.

573 See 23384, 23437-23440, 23393, 23416, 25325, 25326.

574 See 23463, 23464 (invitation from HME) and business cards 23465, 25326.

575 See 23384, 23442-23463.

576 See 0033, 0034, 0097, 0098, 0130, 0131; 1377, 32107, 1827, 2079, 2080 (minutes), 6566, 30926 (travel expenses), 7519-7522 (minutes), 7731-7734 (minutes), 7759 (agenda), 11210-11211 (minutes), 11214- 11217 (travel expenses), 16972, 16973, 17722, 17723 (minutes), 22876.

577 See 0033, 0034.

578 See 22876, 22948-22949.

579 See 32716.

580 See 16972.

581 For further details see 7517, 22959.

582 See 0034-0035, 0225, details about the content 0227. See also 0380-0384 and corresponding Annexes 23a-23v (0563-0567, 25487, 25488, 0570-0612), 990; 1285. Furthermore 31192 (travel expenses); 9396 (travel expenses); 15898, 15955, 15956, 25507, 15970, 25508, 25509, 15973, 25510, 15975, 25511, 17724. Boliden confirmed the participation of HME, see 32523.

583 LLW 22869, 22396, 22458-22460. Outokumpu indicated that it might not have participated (see 32716), 23384, 23467, 23473 (invitation), 23467-23472 (correspondence with HME), 23394-23401, 23421-23424 (notes), 25326. Also see 11403, 11404, 11405 (confirmation by Wieland, invitations); 32409, 32414, and 32438, 32439 (travel expenses).

584 See 0232-0248, 0226-0228, 22460 and 31192.

585 For details, reference is made to the notes on 0232-0248 and the summary submitted by Mueller in 0226-0230.

586 See the summary submitted by Mueller 0227.

587 See notes 0233. "Margin" means "the net selling price before settlement discount minus the raw material cost (i.e. the price of the copper) and a small premium to cover the financing and purchasing cost for the purchasing of the copper. The margins shown are expressed in Deutsche Mark per 100 kg of the most popular copper tube (15 x 1 in Germany, Spain, the Netherlands; 14 x 1 in France and 15 x 0.7 in the UK). They were set as the appropriate margin to be obtained by the participants after applying all discounts and rebates, except the payment discount, and based on a copper price of LME plus 100 USD (17 DEM) per 100 kg. These margins were meant to be the lowest available on the market (i.e., a "floor" on the margins that the participants should receive)." See 0227.

588 See the summary submitted by Mueller 0227.

589 See 0233-0235 and the tables that are part of the report 0236-0248. The meeting was also summarized by Wieland, see 22169-22176.

590 See 0227, 0228.

591 See the tables with the proposed volume allocation 0238-0241 and 0244-0247 as well as Mueller's explanations 0229, 0230.

592 See 23394-23401, 23412-23426 (notes).

593 See 23418.

594 See 22173.

595 See 22176.

596 See the columns "G5 Performance to end of July (%)", "Mueller Allocation" and the subcolumns "% of G5", "tonnes p.m.", "% of total market", "Actual (Jan-Aug)" and "%" in the table of 0249 and the explanations given by Mueller on 0231.

597 See 32409, with respect to the content see 32306-32308.

598 See 22873, 22874, 22933-22941 containing lists for the following dates: 30.10'98, 11.12'98, 22.2'99, 29.4'99, 22.7'99, 8'9'99, 11.10'99, 26.11'99, 18.12.00.

599 DVGW stands for "Deutscher Verein des Gas- und Wasserfaches e.V." (German Association of Gas and Water Specialists), see recital (75).

600 See 22933.

601 See 32734.

602 See 0035-0036, 990, 15956, 25112, 15979, 15980, 25513, 17725, 17726 (minutes).

603 Boliden confirmed the participation of HME, see 32523.

604 See 0035-0036, 990, 31118 (travel expenses), 31194 (travel expenses), 9384 (travel expenses), 22869; Boliden confirmed its brief participation, 32409.

605 See 22869, 22397, 22461-22465 (minutes, travel expenses), 23384, 23475, 23476 (tables from the meeting). See on the other hand 25326, where Halcor explains that it is not able to confirm its participation in the meeting.

606 See travel expenses, 22465.

607 For the complete memo including the attached tables see 22177-22183.

608 For details see 22179, 22180.

609 See 22180: "Hier sieht es so aus, als ob Outokumpu nunmehr wirklich seine Marktführerrolle übernimmt. Deshalb sollten wir das Preis- bzw. Rabattniveau unbedingt halten, um so zur Stabilisierung beizutragen."

610 See 22181, 22182.

611 See 11765.

612 See 11766.

613 See 7030-7047.

614 See 0036, 0037; 990.

615 See 10492-10495 and also the information on D, E, F and NL (10494, 10495), 22334-22337. See also the undated preparatory memo of Wieland 22348, 22349.

616 See minutes of the Copper Industry Meeting including discussions on the EU Drinking Water Directive, 7729, 7730, 7758 (agenda), 22876. With respect to internal co-ordination of KME see 6825.

617 See 1377, 32108, 1824, 2222, 2223, 2373, see also the preparatory letter of 3 February 1999 2081, 7110 (Travel expenses: "Info": Boliden, Müller, Wieland), minutes of the ECI board meeting 7114, 7115, 22876.

618 See 32716, 22876, 22950, 22951.

619 Boliden was not able to confirm the participation of HME, see 32523.

620 See 31101, 31102 (travel expenses), 31195 (travel expenses), 9384 (travel expenses), 22869.

621 See 22869, 22397, 22466-22468.

622 For the complete memo see 22184-22190.

623 See 22187-22189.

624 See 1377, also 32108 (1.4'99 .IWCC Steering Committee minutes - corrected"), 6568 (travel expenses), 7757 (agenda); 32716.

625 See 32716.

626 See 22876.

627 See 31196 (travel expenses), 9384 (travel expenses), 22869, see also travel expenses of Tréfimétaux, 4877.

628 See 23384, 23478-23489 (correspondence with KME/EM), 23402-23406, 23421-23424 (notes), 25326.

629 See 23402-23406, 23421-23424 (notes).

630 For the complete memo see 22191-22198.

631 See 22191-22195.

632 See 22197, 22198.

633 See 32148, 32149 (SANCO price list in Dutch and French); see also 4433-4438 (Wieland's price lists for Austria).

634 See 31197 (travel expenses).

635 See 31197 (travel expenses), 22869, 22397, 22469, 23384, 23491 (confirmation of meeting), 23407- 23410, 23425, 23426 (notes), 25327.

636 For the complete memo see 22199-22206.

637 See 22202-22204. See also the notes of Halcor with respect to the content of the meeting, 23407-23410, 23425, 23426 (notes).

638 See 22205, 22206. The rebate table was updated with respect to the United Kingdom by the memo dated 20 May 1999 (see 10496, 10497, 22207, 22208).

639 See 11765.

640 See recital (347) and 11766.

641 See 1377, 32109, 1828, 1863-1866 ("Meeting with the Copper Industry Executives in the Taurus Conference Room"), 28329, 22876.

642 See 1377, 32109, 1828, 1863-1866 ("Meeting with the Copper Industry Executives in the Taurus Conference Room"), 28329, 6569 (travel expenses), 7756 (agenda), 9268 (travel expenses).

643 See 22876, 32717.

644 See 7755 (agenda).

645 See 1377, 32110, 1829, flight list of IMI 1996, travel expenses 6570, 7755 (agenda), 22876.

646 See 22876, 32717.

647 See 7374-7385.

648 See 7374-7382.

649 See 7383-7385.

650 For the other countries and other competitors see 7374-7376.

651 See 22398, 31104, 31105, 31106 (travel expenses). Although the meeting was named "Lizenzpartner- Meeting" by KME on document 31105, the list of participants suggests that it was a EU Water Directive Meeting (see also KME's document 31106 that talks of "Drinking Water Directive").

652 See 22398, 22470-22472.

653 See 10490, 10491.

654 See 10490, 10491.

655 See 9361 (travel expenses), 22869. According to Wieland, this meeting concerned the sales representative level.

656 See 22398, 22473, 32717, also 11766.

657 For the complete memo see 10480-10485, 22209-22214.

658 See 22213.

659 See 9361 (travel expenses), 22869.

660 See 22398, 22474.

661 See 22215-22221.

662 See 22215, 22217.

663 See 22219, 22220. Some figures of Boliden just represent estimates; Halcor figures are just available for March.

664 See 1377, 32110, 1830, 6571 (travel expenses), 7754 (agenda).

665 See 32717.

666 See 22876. According to Wieland, this meeting concerned high executives.

667 See 9723-9725, 22960-22962.

668 See paragraph 3.67 of Halcor's reply of 7 November 2003.

669 See 9361 (travel expenses), 22869.

670 See 22398, 22475.

671 See 32717.

672 For the complete memo see 22222-22228.

673 See 22222-22226.

674 See 22228. In an internal memo it is stated that Wieland knows more or less the break even point of competitors in the copper plumbing tubes business, 10603.

675 See 32738, 32717. See also the discussion in the letter of Wieland to Outokumpu and KME of 26 August 1999, 11279 concerning the ECI "It is certainly correct that forming a new structure of the European industry is necessary. But I am of the opinion that this should be discussed in circle in which the most important producers and fabricators participate."

676 See 1377, 32111, 6572 (travel expenses to Birmingham), 7753 (agenda).

677 See 32717, 22876.

678 See 9354 (travel expenses), 22869.

679 See 32718.

680 See 22398, 22476.

681 See 32409.

682 See 22876.

683 See 32718.

684 See 31103 (travel expenses), 9336 (travel expenses), 22869, 22398, 22477, 32718, 32409.

685 See 1377, 32112, 6573 (travel expenses), 7752 (agenda), 22876, see also Wieland's memo dated 30.11.1999,, 22338, 22339.

686 See 32718.

687 See 0037.

688 See 31099, 31100 (travel expenses), 9336 (travel expenses), 22869, 22398, 22478, 22479.

689 See 31100 (travel expenses). Also see the internal IMI memo with respect to the introduction of a new price list and a reduction of rebates, 32133.

690 See 32409, 32433 (travel expenses).

691 See 32146.

692 See 7727 (internal Wieland note), 22869, 22340, 22341. According to Wieland, this meeting concerned the sales representative of participating companies.

693 See 11231, 22398, 22480.

694 See 4351.

695 See 32409.

696 Notes and tables found at Outokumpu (see 11593-11598) evidence its involvement in the arrangements with competitors. Various tables indicate "target" and "real" volumes/market shares for five main countries for competitors. Undated notes state, for instance, concerning KME: "1997 figures, shares agreed to be the same". An agenda of an IMI employee points out that an ECPPC meeting was planned for 21-23 March 2000.

697 See 11766, 32718.

698 Various handwritten notes, dated and undated, were found at Outokumpu (see 11239-11259). They occasionally contain remarks related to the contested behaviour such as "E" meeting, "sanitary tubes target 2DEM/kg to tier 1.-SANCO 15 mm", "new list/panic" (see 11253, also 11257).

699 See 6633; see also 11077-11085, which contain undated handwritten notes that report of discussions of various sensitive issues and that seem to stem from the time after 1999.

700 See 6638, 32558. Minutes 31017, 7749 (agenda); Boliden confirmed the participation in the IWCC meeting, but indicated that its employees did not attend any other meeting, 32414.

701 See 7885. "Die Erlössituation ist weiterhin unbefriedigend. Im Zusammenhang mit dem IWCC-Joint- Meeting in Toronto Mitte Mai bitte ich Sie um eine Aktualisierung der europäischen Lage."

702 See 31199 (travel expenses); 9500 (travel expenses), 11766, 22869.

703 See 22399, 22481, 32718. Outokumpu further explained that also issues relating to the EU Drinking Water Directive might have been subject of the meeting. According to Outokumpu, other meetings might have also included discussions on the Drinking Water Directive.

704 See 4350.

705 See 22229.

706 See footnote 759.

707 See 22229, 22230.

708 See 22235.

709 See 32409.

710 See 4349.

711 See 7879-7881, 22342-22348.

712 See 7880, 7881. "Mengen- und Preisziele in den 5 Hauptmärkten auf der Basis der Situation 1997", "Beobachtung Wettbewerbsverhalten, insbesondere jenes von Outsidern, und Festlegung der Vorgehensweise", "Überprüfung und Neufestlegungen von Aktionierung auf dem Hintergrund schneller Preisanpassungen nach oben", "Entwicklung einer europäischen Preisliste", and "Preisziel ab Januar 2001: 100 Euro". With respect to price increases also see the internal memo of IMI, 32136, 32138.

713 See 25388.

714 See 2184, 7720-7722 (minutes of the meeting written by Wieland) and 7706-7707 (travel expenses), 7746 (agenda), invitation 7723, 7724, 7726, apologies 7725, 9305 (travel expenses); see also 6666 (travel expenses), 31098 (travel expenses), 11264-11266 (minutes).

715 See 22876, 22077-22080, 22399, 22482-22485.

716 See 11766, also 32719.

717 See 2200.

718 See 5532-5788.

719 See 31070.

720 See 2097, 32442, 32443 (invitation, agenda), 32434 (travel expenses).

721 According to an agenda, IMI participated: See 32742.

722 See 2095-2097 (invitation and agenda sent by Boliden), 3356 (minutes referring to the meeting), 7717, 7718 (invitation and agenda), 9478, 9533 (travel expenses); 11766, 11767. Concerning 31 August 2000, the letter states: "The intention is to join the expert round during the next morning from 8.00 to 10.00/11.00 am". See also 32409.

723 See 22869.

724 See 22400, 22486-22489.

725 See 11766, 32719.

726 See 32719.

727 See 9173.

728 See 9173.

729 See 30884.

730 See 3360 (minutes of the meeting of 16.10.2000 referring to the meeting), 31095, 31096, 6938, 9478 (travel expenses), 22869. According to Wieland, this meeting concerned the sales representative of participating companies.

731 See 31095.

732 See 22400, 22490-22496, 32409.

733 See 2197, 3355-3358.

734 See 31088.

735 See 6856-6858, 31090.

736 See 22236-22252 (including attached tables).

737 See 22237, 22238.

738 See the charts 22243-22252.

739 See 3359-3367 (fax, minutes, attachments), 31163-31165, 9478 (travel expenses), 11767. 22869. According to Wieland, this meeting concerned the sales representative of participating companies.

740 See 11767, 22869, 22401, 22497-22508 (invitation, minutes, additional information, travel expenses), 32409.

741 See 7692-7694 (travel expenses), 7744 (agenda).

742 See 6705. According to the agenda, the meeting took place at 56 Prince's Court, London; 32742.

743 See 7744 (agenda).

744 See 30986, 30987.

745 See 22963-22966.

746 See 22963, 22964.

747 See 5531.

748 See 31094 (travel expenses), 31163-31165 (minutes of the previous meeting), 22869. According to Wieland, this meeting concerned the sales representative of participating companies.

749 See 4595 (minutes), 4596, 4597 (invitation, agenda), 4633-4635, 22401, 22509, 22510. See also 3360.

750 See 22869, 22401, 22509-22511, 32410, 32435 (travel expenses).

751 See 7687.

752 See the Memorandum of IMI of 28 November 2000, 1406. One of IMI's employees was copied on the communication.

753 See 31093 (travel expenses), 11767, 32719, 32704.

754 See 3350-3354 (minutes and fax), 31093, 22869, 22401, 22512-22517 (circulation fax of minutes, travel expenses), 32410.

755 See 7714, 11271, 11266-11270, 32415, 22952.

756 See 32743, 27245, 27249, 2093, 2198, 32156 (fax), 6750 (travel expenses), 7675-7677 (travel expenses), 7741 (agenda), 9472 (travel expenses), 10628 (internal note and receipt Wieland), 11275 (itinerary). 11767, 32719. IMI and Wieland met on 27 November 2000 (10627, expenses Wieland) and, according to an agenda, on 18 December 2000 (10628). Apparently, the 20 December, joint-venture discussions took place between IMI, Wieland and Boliden. These discussions are not subject of the current investigation. 22869, 22876.

757 See 11767, 32719, 22869, 22876, 3350-3354 (minutes and fax), 31093, 22869, 22401, 22512-22517 (circulation fax of minutes, travel expenses).

758 See 32415, 32436.

759 "RAL" is a simplified quality mark of the Gütegemeinschaft Kupferrohr e.V. that also WICU and SANCO tubes contain on the German market.

760 See 7068.

761 See 7070, 7071.

762 See 7070-7074 and 6930-6934.

763 See both 7076 and 6933 (price line 46 (23.08.2000) and priceline 58 (end December 2000).

764 See 6933. SANCO tubes 15 x 1,0 mm.

765 See 7077: From 161,- to 250,- (category "60,5%" - between 07.09.2000 and 24.01.2001) and from 122,- to 208,- (category "63%, same period). SANCO tubes 15 x 1,0 mm. With respect to the target price of 200,- kg see recital (210).

766 For the complete memo see 7862, 7863, and 22967, 22965.

767 See 22967, 22965.

768 See, also for further countries, 31338-31340.

769 See 32041.

770 See 1833. See also 3349 (invitation) and 3354.

771 See 3084 (agenda), 11767, 32719, 32704. IMI indicated in an internal analysis of the "European Plumbing Tube Market" that "it is therefore clear that industry returns will not see sustained improvement until significant consolidation has been achieved - only then will participants recognise that capacity needs to be brought into line with demand." See 32140-32143, 32142, and also the margin comparison with the US 32155.

772 See 22402, 22518, 32410.

773 See 32675, 32676.

774 See 3347 (invitation, agenda), 31077, 11767.

775 See 3085 (agenda), 4507 (fax), 11767, 32720, 32704.

776 See 31077. The alternative proposal to setting up a working group was discussed internally at KME. The idea was to gather information first internally and then share it with the competitors involved (KME, IMI, Wieland, Boliden, Outokumpu).

777 See 31073-31076.

778 See 22402, 22519, 22520, 32410.

779 See 31201 (travel expenses); 22869.

780 According to the memo, the summary concerned a meeting of 8 March 2001. It appears, however that it summarises the meeting of 5 March.

781 See 22253, 22254.

782 See 22253, 22255.

783 For details see 22260-22263.

784 For the agenda of the ECI Board meeting, see 31003, also 31001, 31002.

785 See 3371 and 3368-3370, 7716 (announcement of the meeting by fax from Wieland to KME, Boliden, Outokumpu and IMI), 9748.

786 See 7474.

787 See 11767, 32720 and also 1377, 32117, 1833, 27240, 32128, 2214 ("elephants"), see also 2162 and 2181, confirming that IMI attended the meetings, 3368-3371, 6787, 30927, 31077; 7474 (Agenda), 22876. With respect to the participation of Boliden see 7715. An ECI board meeting was held at the same time. According to the minutes, IMI, KME, Wieland and Outokumpu were present at the ECI meeting, see 31382, 31383, and also 31376, 31377.

788 See 22876, 22952, 22954, 32410, 32415, 32444-32448.

789 See 3086 (agenda), 11106-11110, 32668-32672 (agenda, tables).

790 See 11104, 11105.

791 See 1292, 1345.

792 See 32118, 27243, 6791, 31277.

793 See 32118, 32119, 1833, 2214 ("elephants").

794 See 23554.

795 See 32544.

796 See 22876.

797 See 32544.

798 The case-law of the Court of Justice and the Court of First Instance in relation to the interpretation of Article 81 of the Treaty applies equally to Article 53 of the EEA Agreement. See recitals (4) and (15) as well as Article 6 of the EEA Agreement, and Article 3(2) of the Agreement on the establishment of a Surveillance Authority and a Court of Justice .

799 Case T-7-89 Hercules Chemicals v Commission [1991] ECR II-1711, recital 232.

800 See [1999] ECR I - 4125, at paragraph 81.

801 Case 48-69 Imperial Chemical Industries v Commission [1972] ECR 619 at paragraph 64.

802 Joined Cases 40-48-73, etc. Suiker Unie and others v Commission [1975] ECR 1663.

803 See also the judgement of the Court of Justice in Case C-199-92 P Hüls v Commission, [1999] ECR I- 4287, at recitals 161-162.

804 See, in this sense, Cases T-147-89, T-148-89 and T-151-89, Société Métallurgique de Normandie v Commission, Trefilunion v Commission and Société des treillis et panneaux soudés v Commission, [1995] ECR, p. II-1057, II-1063 and II-1191, at paragraph 72.

805 Limburgse Vinyl Maatschappij NV et al. v. Commission, joint cases T-305-94, T-306-94, T-307-94, T-313-94 to T-316-94, T-318-94, T-325-94, T-328-94, T-329-94 and T-335-94, ECR [1999], p. II- 00931, paragraph 696.

806 See Anic Partecipazioni, paragraphs 78-81, 83-85 and 203.

807 Anic Partecipazioni, cit.

808 See Anic Partecipazioni, cit., paragraphs 78-81, 83-85 and 203.

809 See 32541, 32542.

810 See recital (220).

811 Imperial Chemical Industries ,cit., paragraphs 259-260.

812 OJ L 123, 27.4.2004, p. 11.

813 OJ L 31, 9.2.1996, p. 2. Regulation repealed by Regulation (EC) No 772-2004 of 27 April 2004, OJ L 123, 27.4.2004, p. 11.

814 OJ C 372, 9.12.1997, p. 5.

815 Case C-204-00 P, C-205-00 P, C-211-00 P, C-213-00 P, C-217-00 P and C-219 P, Aalborg Portland A/S et al. v. Commission, paragraph 56 and 57 (not yet reported); see also Case T-368-00, General Motors Nederland BV and Opel Nederland BV v Commission, paragraph 88 (not yet reported).

816 See 32540, and the correction of the statement in 32053. See also recital (203).

817 See 32543, 32544. See also recitals (203) and (225).

818 See paragraph 91-97 of Wieland's reply of 7 November 2003.

819 See Annex 1 of IMI's reply of 7 November 2003.

820 See p. 31823: "There was no idea to continue. The figures were not reliable. The market shares didn.t reflect the original ideas. So I think that somewhere at the end 1992 the system was stopped [...]".

821 See paragraph 19, 20 of Wieland's reply of 7 November 2003.

822 See especially 31814-31818 with a detailed description of the meetings, although without a specification of dates, that occurred until end of 1992.

823 See paragraph 158-163 of KME's reply of 7 November 2003.

824 See paragraph 164-168 of KME's reply of 7 November 2003.

825 See paragraph 169-174 of KME's reply of 7 November 2003.

826 See paragraph 21-24 of Wieland's reply of 7 November 2003.

827 See paragraph 21 of Wieland's reply of 7 November 2003.

828 See paragraph 21 of Wieland's reply of 7 November 2003.

829 See paragraph 22 of Wieland's reply of 7 November 2003.

830 See paragraph 21 of Wieland's reply of 7 November 2003.

831 See paragraph 22 of Wieland's reply of 7 November 2003.

832 See paragraph 23 of Wieland's reply of 7 November 2003.

833 See paragraph 23 of Wieland's reply of 7 November 2003.

834 See paragraph 23 of Wieland's reply of 7 November 2003.

835 See Annex 1 of IMI's reply of 7 November 2003.

836 See 32701, 32702. See also (199), (214) and (215).

837 See paragraph 18 of Outokumpu's reply of 10 November 2003.

838 See slide 12 and 13 of Outokumpu's reply of 5 December 2003.

839 See paragraph 19 of Outokumpu's reply of 10 November 2003.

840 See paragraph 20 of Outokumpu's reply of 10 November 2003.

841 See paragraph 21 of Outokumpu's reply of 10 November 2003.

842 See paragraph 22-23 of Outokumpu's reply of 10 November 2003.

843 See paragraph 24 of Outokumpu's reply of 10 November 2003.

844 See 31945, 31946.

845 See 31947, 31948.

846 IMI argues that Mueller's recollections should not be considered by the Commission (see Annex 1 of IMI's reply of 7 November 2003). The fact that the Commission only addressed the Statement of Objections to Mueller for its participation in the infringement as of October 1997 does not exclude that Mueller has relevant recollections for the period before.

847 For details see footnote 172.

848 See paragraph 179 of KME's reply of 7 November 2003.

849 See recitals (193)-(225).

850 For details, in particular with respect to alleged quiet periods in these cartels, see the parties. submissions and recitals (106) and (107).

851 See paragraph 33-41 of BCZ's reply of 7 November 2003.

852 See paragraph 34, 35 of BCZ's reply of 7 November 2003.

853 See paragraph 36-41 of BCZ's reply of 7 November 2003.

854 See paragraph 42-52 of BCZ's reply of 7 November 2003.

855 See paragraphs 3.16-3.66 of Halcor's reply of 7 November 2003.

856 See paragraphs 3.67-4.3, 6.4 of Halcor's reply of 7 November 2003.

857 The list is not exhaustive.

858 See the explanation of KME in recital (220).

859 Case 8-72 Vereeniging van Cementhandelaren v Commission [1972] ECR 977, paragraph 21.

860 Case T-311-94 BPB de Eendracht v Commission [1998] ECR II-1129, paragraph 192.

861 See also judgement of Court of Justice in Case C-7-95P John Deere v Commission [1998] ECR I-3111, paragraph 67, upholding the judgement of Court of First Instance, see in particular paragraph 96.

862 See 23717-23725; 32539, 32540, 32301-32311, 25325, 25327, and the parties' replies to the Statement of Objections, which will be further discussed below.

863 Case T-62-98 Volkswagen AG v Commission [2000] ECR II-2707, paragraph 178.

864 See Joined Cases 209 to 215 and 218-78 Van Landewyck and others v Commission [1980] ECR 3125, at paragraph 170.

865 See Case T-13-89 Imperial Chemical Industries v Commission [1992] ECR II-1021, at paragraph 304.

866 See the judgement of the Court of Justice in Van Landewyck and others v Commission, at paragraph 170.

867 See paragraph 21, 22 of BCZ's reply of 7 November 2003.

868 This was proven by a document dated 30 October 1989 contained in Annex 5 of BCZ's reply of 7 November 2003, which was made accessible to all parties.

869 See paragraph 25-33 of BCZ's reply of 7 November 2003.

870 See KME's submission of 5 December 2003.

871 See paragraphs 9-30 of KME's reply of 7 November 2003.

872 See paragraph 31-43 of KME's reply of 7 November 2003.

873 See paragraph 44-49 of KME's reply of 7 November 2003.

874 See paragraph 50-52 of KME's reply of 7 November 2003.

875 See paragraph 53-68 of KME's reply of 7 November 2003.

876 See paragraph 69-81 of KME's reply of 7 November 2003.

877 See paragraph 419 of KME's reply of 7 November 2003.

878 See paragraph 82-106 of KME's reply of 7 November 2003.

879 See paragraph 112-140 of KME's reply of 7 November 2003.

880 See paragraph 141 of KME's reply of 7 November 2003.

881 See 33418.

882 See paragraphs 107-111 of KME's reply of 7 November 2003.

883 See paragraph 542-545 of KME's reply of 7 November 2003.

884 See for example Halcor's statement that it "was coerced into participating in the G9 sweepers. meeting on 28 August 1998 by overt threats from the major producers" (p. 1 of Halcor's reply of 7 November 2003) and Halcor's claim that it "did NOT say: - that it was coerced into going to the meeting of 28 August" (p. 1 of Halcor's submission of 5 December 2003).

885 See p 1 and paragraphs 3.3-3.8 of Halcor's reply of 7 November 2003, paragraphs 1-4.7 of Halcor's submission of 5 December 2003.

886 See paragraphs 3.10-3.12 of Halcor's reply of 7 November 2003.

887 See paragraphs 3.8 (e)-(f) of Halcor's reply of 7 November 2003.

888 See also 547, where Mueller invites Halcor to participate in the meeting of 28 August 1998.

889 See 23420.

890 See paragraph 437-457 of KME's reply of 7 November 2003.

891 See Annex 1 of IMI's reply of 7 November 2003.

892 See p. 4-16 of Mueller's reply of 7 November 2003 and Mueller's submission of 5 December 2003.

893 See p. 1, 2 of Outokumpu's submission of 5 December 2003.

894 See paragraph 28 of Wieland's reply of 7 November 2003.

895 See 25297.

896 See p. 5, 13, 14 of IMI's reply of 7 November 2003 referring to 22174 and 22217 of the file.

897 22174.

898 22217.

899 See24647, 24648 and also 24649.

900 Cf. ICI v Commission, paragraph 134; Case 107-82 AEG v Commission [1983] ECR 3151; see also Case T-354-94, Stora v Commission [1998] ECR II-2111.

901 AEG v Commission, paragraph 50.

902 Case C-286-98 P Stora v Commission [2000] ECR I-9925, paragraph 28.

903 ICI v Commission, loc.cit, paragraph 137, 138 ; Case 6-72 Europemballage and Continental Can v Commission [1973] ECR 215, paragraph 16 ; AEG v Commission, loc. cit., paragraph 51.

904 Case T-309-94, Koninklijke KNP v Commission [1998] ECR II-1007, paragraph 48 ; confirmed in Case C-248-98 P [2000] ECR I-9641, paragraph 73.

905 Cases T-308-94 Cascades v Commission [1998] ECR II-925, paragraph 158 ; T-347-94, Mayr-Melnhof v Commission [1998] ECR II-1751, paragraphs 397, 398 ; T-354-94 Stora v Commission, paragraph 83. See also Cases C-279-98 P Cascades v Commission [2000] ECR I-9693 paragraph 77; C-286-98 P Stora v Commission , paragraphs 31 and 32 ; cf., in the same sense the opinion of Advocate General Mischo in the latter Case [2000] ECR I-9925, paragraphs 40 and 51.

906 See 26270-26301 and recital (41).

907 See references in paragraphs 276-283 of KME's reply of 7 November 2003.

908 See 5279.

909 See reference in paragraph 279 of KME's reply of 7 November 2003.

910 See paragraph 281 of KME's reply of 7 November 2003.

911 See 29554, 29556.

912 For example see the correspondence between Desnoyers and Wednesbury with respect to the WBMS in 0849-0984.

913 See paragraph 51, 52 and 75-79 of Outokumpu's reply of 10 November 2003.

914 ICI v Commission, paragraph 134.

915 File p. 10831.

916 File p. 10830.

917 OJ L 319, 29.11.1974, p.1. Regulation as amended by Regulation (EC) No 1-2003.

918 Article 1(2) of Regulation (ECC) No 2988-74.

919 Article 2(1) and 2(3) of Regulation (ECC) No 2988-74.

920 Case T-334-94 Sarrió v Commission [1998] ECR II-01439, paragraph 168-169.

921 Under Article 5 of Council Regulation (EC) No 2894-94 of 28 November 1994 concerning arrangements of implementing the Agreement on the European Economic Area "the Community rules giving effect to the principles set out in Articles 85 and 86 [now Articles 81 and 82] of the EC Treaty [...] shall apply mutatis mutandis". (OJ L 305, 30.11.1994, p.6)

922 See paragraph 91-97 of Wieland's reply of 7 November 2003.

923 Certain parties presented the following arguments as an issue going to mitigation, but for analytical purposes is treated here.

924 See paragraph 190-289 of KME's reply of 7 November 2003.

925 See paragraph 458-461 of KME's reply of 7 November 2003.

926 See paragraph 462-465 of KME's reply of 7 November 2003.

927 See paragraph 466-471 of KME's reply of 7 November 2003.

928 See paragraph 471 of KME's reply of 7 November 2003.

929 See paragraph 472-474 of KME's reply of 7 November 2003.

930 See paragraph 190-201 of KME's reply of 7 November 2003.

931 See paragraph 202-204 of KME's reply of 7 November 2003.

932 See paragraph 204-207 of KME's reply of 7 November 2003.

933 File p. 23716-23754.

934 See 17624.

935 See 17625.

936 See 17625.

937 See 17625.

938 See 17625.

939 See paragraph 45, 46 of Wieland's reply of 7 November 2003.

940 See paragraph 47 of Wieland's reply of 7 November 2003.

941 See paragraph 48, 49 of Wieland's reply of 7 November 2003.

942 See paragraph 50 of Wieland's reply of 7 November 2003.

943 See paragraph 51 of Wieland's reply of 7 November 2003.

944 See paragraph 52 of Wieland's reply of 7 November 2003.

945 See paragraph 52 of Wieland's reply of 7 November 2003.

946 See paragraph 53 of Wieland's reply of 7 November 2003.

947 See paragraph 54-60, 62-66 of Wieland's reply of 7 November 2003.

948 See paragraph 63 of Wieland's reply of 7 November 2003. See 23002.

949 See paragraph 67 of Wieland's reply of 7 November 2003.

950 See paragraph 68, 69 of Wieland's reply of 7 November 2003.

951 See paragraph 70 of Wieland's reply of 7 November 2003.

952 See paragraph 71 of Wieland's reply of 7 November 2003.

953 See paragraph 72 of Wieland's reply of 7 November 2003.

954 See paragraph 73 of Wieland's reply of 7 November 2003.

955 See paragraph 74-76 of Wieland's reply of 7 November 2003.

956 See paragraph 77, 78 of Wieland's reply of 7 November 2003.

957 See paragraph 79-84 of Wieland's reply of 7 November 2003.

958 See paragraphs 5.1-5.7 of Halcor's reply of 7 November 2003.

959 Case T-141-94 Thyssen Stahl v Commission [1999] ECR II-347, paragraphs 635-636.

960 Case T-224-00, Archer Daniels Midland Company and others v Commission, judgment of 9 July 2003, not yet reported, paragraph 151.

961 Archer Daniels Midland Company and others v Commission , paragraph 279.

962 Commission v Anic Partecipazioni SpA, paragraph 152.

963 See 33432, 33433.

964 See 15951.

965 See 8602.

966 See 8602.

967 See 8603.

968 See 8604.

969 For details, reference is made to the notes on 0232-0248 and the summary submitted by Mueller in 0226-0230.

970 As explained above in footnote 587: "Margin" means "the net selling price before settlement discount minus the raw material cost (i.e. the price of the copper) and a small premium to cover the financing and purchasing cost for the purchasing of the copper. The margins shown are expressed in Deutsche Mark per 100 kg of the most popular copper tube (15 x 1 in Germany, Spain, the Netherlands; 14 x 1 in France and 15 x 0.7 in the UK). They were set as the appropriate margin to be obtained by the participants after applying all discounts and rebates, except the payment discount, and based on a copper price of LME plus 100 USD (17 DEM) per 100 kg. These margins were meant to be the lowest available on the market (i.e., a "floor" on the margins that the participants should receive)." See 0227.

971 See 23418.

972 See 7374-7385.

973 See 7384.

974 See 7377-7379.

975 See 7030-7047.

976 For explanations and origin of this figure see recital (651).

977 For explanations and origin of this figure see recital (651).

978 See 22236-22252 (including attached tables).

979 See 22237, 22238.

980 See 22963-22966.

981 See 22963, 22964.

982 See 6933. SANCO tubes 15 x 1,0 mm.

983 See 7076.

984 See 7077: From 161,- to 250,- (category "60,5%" - between 07.09.2000 and 24.01.2001) and from 122,- to 208,- (category "63%, same period). SANCO tubes 15 x 1,0 mm. With respect to the target price of 200,- kg see recital (210).

985 For the complete memo see 7862, 7863, and 22967, 22965.

986 See 22967, 22965.

987 According to the memo, the summary concerned a meeting of 8 March 2001. It appears, however that it summarises the meeting of 5 March.

988 See 22253, 22254.

989 That is, an increase of the copper price results in a decrease of the total copper tubes price.

990 The average price increase was estimated to be EUR 0.816. Given an average price of EUR 4.64 for tubes sold by KME, this translates into a percentage price increase of 18%. The other percentage values have been calculated accordingly.

991 These percentage numbers are based on absolute values provided in the figures. The first graph of figure 2.4 indicates an "apparent consumption" of roughly 60.000 t in 2000 in the United Kingdom and of roughly 30.000 in 2001. The second graph indicates "domestic production and exports" of the size of 50.000 t in 2000 in the United Kingdom and of roughly 10.000 t in 2001. For 2002 no numbers are given (KME-NERA report page 8).

992 See paragraph 149-154 of KME's reply of 7 November 2003.

993 The table was prepared by the Commodities Research Unit; see paragraph 205-207 of KME's reply of 7 November 2003.

994 See, for example, paragraph 33 of Outokumpu's reply of 10 November 2003.

995 See paragraph 61, 62 of Outokumpu's reply of 10 November 2003.

996 See paragraph 79 of Outokumpu's reply of 10 November 2003.

997 Amongst others in Halcor's submission of 25 June 2004.

998 Joined cases T-45-98 and T-47-98 Krupp Thyssen Stainless GmbH and Acciai Speciali Terni SpA v Commission, [2001] ECR II- 3757, paragraph 157.

999 Case T-220-00, Cheil Jedang Corp. v Commission, judgment of 9 July 2003, not yet reported, paragraph 91.

1000 Case C-185-95 P Baustahlgewebe v Commission [1998] ECR I-8417, paragraph 139.

1001 See paragraph 268-283 of KME's reply of 7 November 2003.

1002 See paragraph 91-97 of Wieland's reply of 7 November 2003.

1003 See paragraph 80-83 of Outokumpu's reply of 10 November 2003.

1004 See paragraph 80-83 of Outokumpu's reply of 10 November 2003.

1005 See paragraph 84 of Outokumpu's reply of 10 November 2003.

1006 See paragraph 48-50 of Outokumpu's reply of 10 November 2003.

1007 See paragraph 85 of Outokumpu's reply of 10 November 2003.

1008 See Annex 1 of IMI's reply of 7 November 2003.

1009 See paragraph 91-97 of Wieland's reply of 7 November 2003.

1010 See paragraphs 3.13-3.34 of Halcor's reply of 7 November 2003.

1011 See paragraphs 6.1 of Halcor's reply of 5 December 2003.

1012 See paragraph 4-15 of Wieland's reply of 7 November 2003, paragraph 70-74 of Outokumpu's reply of 10 November 2003.

1013 See paragraph 8 of Wieland's reply of 7 November 2003.

1014 See paragraph 15 of Wieland's reply of 7 November 2003.

1015 See paragraph 72 of Outokumpu's reply of 10 November 2003.

1016 See paragraph 94 of Outokumpu's reply of 10 November 2003.

1017 See paragraph 95 of Outokumpu's reply of 10 November 2003.

1018 Cheil Jedang Corp., v Commission, paragraphs 83 and 96.

1019 Case T-31-99 ABB Asea Brown Boveri Ltd. v Commission, [2002] ECR II-1881, paragraph 155.

1020 Case COMP/E-1-38.240 Industrial Tubes, decision of 16 December 2003, not yet published.

1021 Case T-236-01 Tokai Carbon (Graphite Electrodes), judgment of 29 April 2004, not yet reported, paragraph 241.

1022 See paragraph 91-97 of Wieland's reply of 7 November 2003; paragraph 60 of Outokumpu's reply of 10 November 2003; paragraphs 307-317 of KME's reply of 7 November 2003. IMI'claim to be treated in the same way as BMA, HME, Halcor and Mueller was rejected above (see recitals (468) and (490)).

1023 See, for example, paragraphs 307-317 of KME's reply of 7 November 2003.

1024 Commission Decision 2002-759-EC in case COMP-37.800 Luxembourg Brewers OJ L 253, 21'9.2002, p. 21, paragraphs 86 and 97.

1025 See paragraphs 307-317 of KME's reply of 7 November 2003.

1026 See paragraph 60 of Outokumpu's reply of 10 November 2003.

1027 T-23-99, LR AF v Commission, [2002] ECR II-1705, paragraph 234.

1028 See, for example, Case COMP/E-1-37.519 Methionine, decision of 2 July 2002, not yet published; COMP/E-1-37.370 Sorbates, decision of 1 October 2003, not yet published; and Case COMP/E-1-37'956, Reinforcing bars, decision of 17 December 2002, not yet published.

1029 OJ C 45, 19.2.2002, p. 3.

1030 Paragraph 23.

1031 Case 37.667, decision of 17 December 2002 (not yet published), paragraph 558.

1032 See paragraph 91-97 of Wieland's reply of 7 November 2003.

1033 See paragraph 90 of Outokumpu's reply of 10 November 2003.

1034 See paragraph 91, 92 of Outokumpu's reply of 10 November 2003.

1035 OJ L 220, 15.8.1990, p. 28.

1036 See paragraph 96-98 of Outokumpu's reply of 10 November 2003.

1037 Thyssen Stahl AG v. Commission, at paragraphs 617-625 (emphasis added).

1038 See paragraph 102-104 of Outokumpu's reply of 10 November 2003.

1039 Case T-77-92 Parker Pen v. Commission [1994] ECR II-549 at paragraphs 94-95; paragraph 105, 106 of Outokumpu's reply of 10 November 2003.

1040 For the notion of the 'same type' of infringement, see Case T-203-01, Michelin v Commission, judgment of 30 September 2003, not yet reported, paragraphs 284 et seq.

1041 See Thyssen Stahl v Commission, paragraphs 258 et seq.

1042 See paragraph 208-236 of KME's reply of 7 November 2003. The Commission notes that all quotations of evidence that KME submitted in support of its argument that parties supplied wrong volume figures date from before 1997, with the exception of two quotes that concern HME and BCZ.

1043 Joined Cases T-25-95, T-26-95, T-30-95 to T-32-95, T-34-95 to T-39-95, T-42-95 to T-46-95, T-48-95, T-50-95 to T-65-95, T-68-95 to T-71-95, T-87-95, T-88-95, T-103-95 and T-104-95 Cimenteries CBR and Others v Commission [2000] ECR II-491, paragraphs 4872 to 4874.

1044 Cascades SA v Commission, paragraph 230.

1045 Archer Daniels Midland Company and others v Commission, paragraphs 160 and 271.

1046 Archer Daniels Midland Company and others v Commission, paragraph 279.

1047 LR AF v Commission [2000] ECR II-1705, paragraphs 268 and 307. See all the more Europa Carton v Commission [1998] ECR II-869, paragraph 141.

1048 Cases T-236, 239, 244-246, 251, 252-01, Tokai Carbon v Commission of the European Communities, judgment of 29 April 2004, not yet reported, paragraph 345.

1049 Commission Decision 98-247-ECSC in Case IV-35.814 - Alloy surcharge (OJ L 100, 1.4.1998, p. 55), paragraphs 83 to 84 (the reduction for Acerinox also took into account other mitigating factors in addition to the economic crisis).

1050 Commission Decision 2003-382-EC in Case IV-E-1-35.860-B seamless steel tubes (OJ L 140, 6.6.2003, p. 1, paragraphs 168 and 169.

1051 Case T-31-99 Asea Brown Boveri v Commission, paragraph 213.

1052 OJ C 9, 14.1.1998, p .3.

1053 See paragraph 86-90 of Wieland's reply of 7 November 2003.

1054 Case T-7-89 Hercules Chemicals v Commission, paragraph 357, confirmed on appeal in Case C-51-92 P Hercules Chemicals v Commission [1999] ECR I-4235.

1055 Case T-224-00 Archer Daniels Midland Company and others v Commission, paragraph 280.

1056 See paragraphs 614-626 of KME's reply of 7 November 2003.

1057 Case T-220-00 Cheil Jedang Corp. v Commission, paragraph 60; joined cases 100 to 103-80 Musique diffusion française and Others v Commission,[1983] ECR 1825, paragraph 119, Case T-43-92 Dunlop Slazenger v Commission [1994] ECR II-441, paragraph 160, and Case T-144-89 Cockerill Sambre v Commission [1995] ECR II-947, paragraph 98.

1058 See paragraph 547-573 of KME's reply of 7 November 2003.

1059 See paragraph 548, 549 of KME's reply of 7 November 2003.

1060 See paragraph 550 of KME's reply of 7 November 2003.

1061 See paragraph 547-569 of KME's reply of 7 November 2003.

1062 See paragraph 547-569 of KME's reply of 7 November 2003.

1063 In December 2003, after KME's reply to the Statement of objections in the present case, the Commission imposed a total fine of EUR 39,81 million to the companies of the KME group for their participation in the Industrial tubes cartel. See Commission Decision of 16 December 2003 in Case COMP/E-1-38.240 - Industrial Tubes, not yet published.

1064 Joined Cases T-236, 239, 244, 246, 251 and 252-01 Carbide/Graphite v Commission, paragraph 371.

1065 [...]

1066 Estimation of KME.

1067 SMI has no production activity and acts as holding. According to KME reply to the Statement of Objections, KME is essentially the only asset of SMI.

1068 SMI's annual report for 2002.

1069 Joined Cases 96-82 to 102-82, 104-82, 105-82, 108-82 and 110-82 IAZ and Others v Commission [1983] ECR 3369, paragraph 55; Case T-319-94 Fiskeby Board v Commission [1998] ECR II-1331, paragraph 76.

1070 See paragraph 86-90 of Wieland's reply of 7 November 2003.

1071 Commission Decision of 17 December 2003 in Case COMP/37.667 - Speciality Graphites/PO.

1072 It should be pointed out that in 2002, after the termination of the infringement, SMI, KME's parent company, purchased 100 % of YCT Yorkshire Copper Tube Ltd from IMI. Given that neither SMI nor the KME group can be held responsible for (at the time) IMI Yorkshire Copper Tube Ltd's actions before the acquisition, they are not held jointly and severally liable with YCT Yorkshire Copper Tube Ltd. IMI requested the Commission to be the only addressee of the decision.

1073 The total market volume was calculated by the Commission based on turnover information and market size estimates provided by parties.