EC, November 27, 2002, No 2005-471

THE COMMISSION OF THE EUROPEAN COMMUNITIES,

Having regard to the Treaty establishing the European Community, Having regard to Council Regulation No 17 of 6 February 1962, the first Regulation implementing Articles 85 and 86 of the Treaty(1), as last amended by Regulation (EC) No 1216-1999 (2), and in particular Articles 3 and 15 thereof, Having regard to the Commission decision of 18 April 2001 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 in accordance with Article 19(1) of Regulation No 17 and with 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, Whereas :

1. PART I - FACTS

1.1. Summary of the infringement

(1) The following undertakings are the addressees of this Decision relating to the infringement of Article 81 of the Treaty : - BPB PLC (hereinafter: "BPB") - Gebrüder Knauf Westdeutsche Gipswerke KG (hereinafter: "Knauf Westdeutsche Gipswerke") - Société Lafarge SA (hereinafter: "Lafarge") - Gyproc Benelux NV (hereinafter: "Gyproc").

(2) BPB, Knauf (4), Lafarge and Gyproc entered into and participated without interruption in a complex and continuing agreement contrary to Article 81 (1) of the Treaty which was manifested in the following conduct constituting agreements or concerted practices: - the representatives of BPB and Knauf met in London in 1992 and expressed the common desire to stabilise the markets in Germany (hereinafter: "German market"), the United Kingdom (hereinafter: "UK market"), France (hereinafter: "French market") and the Netherlands, Belgium and Luxembourg (hereinafter: "Benelux market"); - the representatives of BPB and Knauf established, as from 1992, information exchange arrangements, to which Lafarge and subsequently Gyproc acceded, relating to their sales volumes on the German, French, UK and Benelux plasterboard markets; - the representatives of BPB, Knauf and Lafarge exchanged information, on various occasions, prior to price increases on the UK market; - in view of particular developments on the German market, the representatives of BPB, Knauf, Lafarge and Gyproc met at Versailles in 1996, Brussels in 1997 and The Hague in 1998 with a view to sharing out or at least stabilising the German market; - the representatives of BPB, Knauf, Lafarge and Gyproc exchanged information on various occasions and concerted their action on the application of price increases on the German market between 1996 and 1998.

(3) The undertakings concerned participated in the infringements for the following periods: - BPB: from 31 March 1992, at the latest, to 25 November 1998 - Knauf: from 31 March 1992, at the latest, to 25 November 1998 - Lafarge: from 31 August 1992, at the latest, to 25 November 1998 - Gyproc: from 6 June 1996, at the latest, to 25 November 1998.

1.2. The proceeding

1.2.1. Investigations pursuant to Article 14(3) of Regulation No 17

(4) On 25 November 1998, in response to information received (5), acting on the basis of Commission decisions pursuant to Article 14(3) of Regulation No 17, Commission officials, accompanied by representatives of the competent authorities of the Member States concerned, carried out simultaneous, unannounced investigations on the premises of the following undertakings: - BPB (United Kingdom), - British Gypsum Ltd (United Kingdom), - BPB Placo SA (France), - Rigips GmbH (Germany), - BPB Italia SA (Italy), - Knauf (Germany), - Lafarge Plâtres International SA (France), - Gyproc (Belgium).

(5) On 1 July 1999, the Commission carried out further investigations pursuant to Article 14(3) on the premises of Etex SA (hereinafter: "Etex") in Belgium and on the premises of Lafarge Gips GmbH in Germany.

1.2.2. Requests for information pursuant to Article 11 of Regulation No 17

(6) On 27 January 1999, the Commission sent BPB a request for information pursuant to Article 11 of Regulation No 17 regarding references contained in documents obtained during the investigation carried out on its premises in November 1998. BPB replied on 17 March 1999 (6), attaching a memo addressed to the Commission (7), in which it acknowledged some infringements of Article 81, involving only the exchange of business information, but arguing that the infringements had had little or no effect on competition and denying any knowledge of any agreement or understanding whatsoever on the sharing of the German market. On 28 May 1999, BPB also submitted a statement (8) in which it denied any knowledge of or participation in any agreement on the allocation of market shares in Germany. On 21 September 1999, the Commission sent a second request for information to BPB. BPB replied on 28 October 1999 (9). On 30 March 2000, the Commission sent BPB a third request for information. BPB replied on 18 May 2000 (10).

(7) The Commission sent requests for information pursuant to Article 11 to Knauf on 8 July 1999 and to Lafarge and Etex on 21 September 1999, requesting information on documents which it had obtained on the premises of those undertakings during the investigations carried out in November 1998 and July 1999. Knauf replied on 14 September 1999, Lafarge on 29 October 1999 and Gyproc on 2 November 1999. On 1 September 1999, following a meeting with Commission staff, Gyproc sent the Commission a statement on certain contacts between competitors (11).

(8) On 2 August 2002, the Commission sent a request for information pursuant to Article 11 to BPB, Gyproc, Knauf, Lafarge and Etex concerning various financial and structural data concerning those companies, to which they replied (12).

1.2.3. Administrative procedure

(9) On 18 April 2001, the Commission initiated proceedings in this case and adopted a statement of objections against BPB, Knauf, Lafarge, Etex and Gyproc.

(10) All the parties to which the statement of objections had been addressed submitted written comments in response to the objections raised by the Commission.

(11) The undertakings had access to the Commission's investigation file in the form of a copy on CD-ROM, which was sent to them on 17 May 2001. With the CD-ROM, the undertakings received a list specifying all the documents contained in the investigation file (with consecutive page numbering) and indicating the degree of accessibility of each document. In addition, the undertakings were informed that the CD-ROM gave the parties full access to all the documents obtained by the Commission during the investigation, except for business secrets and other confidential information.

(12) A hearing on the case was held on 17 July 2001. All the undertakings to which the statement of objections had been addressed took part.

1.2.4. Previous Commission measures

(13) On 13 July 1994, the Commission adopted Decision 94-601-EC (13) relating to a proceeding under Article 85 (now 81) of the EC Treaty and imposed a fine of ECU 1 750 000 on BPB's subsidiary, BPB De Eendracht NV, for having taken part in an illegal agreement in the cartonboard sector. On 14 May 1998, in Case T-311-94, the Court of First Instance of the European Communities reduced the fine to ECU 750 000 (14).

(14) On 30 November 1994, the Commission adopted Decision 94-815-CE (15) relating to a proceeding under Article 85 (now 81) of the EC Treaty and imposed on Lafarge SA (then known as Lafarge Coppée SA) a fine of ECU 22 872 000 for having taken part in an illegal agreement in the cement sector. On 15 March 2000, in Case T-43-95, the Court of First Instance reduced the fine to EUR 14 248 000 (16).

1.3. The plasterboard industry

1.3.1. The product

(15) Plasterboard is a manufactured product used as a prefabricated construction material and consisting of a sheet of gypsum plaster sandwiched between two sheets of paper or some other material.

(16) Plasterboard is cut in various sizes and thicknesses. The standard size in Europe is board 12.5 mm thick and usually 1200 mm in width, although in Germany the standard width is 1250 mm. In addition to the standard ("primary") board, there is a growing market for specialist boards which are laminated with other materials to improve fire insulation properties and soundproofing. Boards can also be coated with plastic to form prefabricated units.

(17) Plasterboard is typically used as an internal lining for walls, to form internal partitions inside buildings, as a roof lining and as a ceiling material for residential, commercial and administrative premises. The properties which make plasterboard an attractive product for the building industry are its stability, durability, easy application, fire resistance and low cost.

(18) Plasterboard manufacture is a relatively simple process involving the production of a plaster slurry which is sandwiched between two sheets of liner paper, cut to length and dried. The plaster comes from either natural gypsum (produced by quarrying) or from the gypsum produced as a by-product of other processes, particularly flue-cleaning in coal-fired power stations, which produces flue gas desulphurisation gypsum (DSG). Natural gypsum is delivered in the form of crushed rock which then has to be reduced to powder (DSG is already a powder). In both cases, however, the gypsum has to be dried before further working. The next stage, calcination, is a continuous process which feeds the powder directly to a mixer at the beginning of the production line, where it is mixed with water and additives to make a slurry. This slurry is fed onto the line where the first layer of paper is already in position. The second backing layer of liner is fed in at a set of rollers which spreads the slurry evenly across the width of the board to the required thickness. The board travels down forming lines of some 200-400 metres in length as the plaster gradually sets. When the plaster has set (but not fully dried), the board passes through a cutter which cuts the material to length before the boards are turned over (to protect the top surface) and transferred to a drying line and hardened in large ovens. Once dried, the boards are trimmed to the exact length.

(19) Output and sales are usually measured in surface area (in m²) of board produced, irrespective of thickness. Another measuring method is to convert the surface of the various types of plasterboard to the notional equivalent in standard 12.5 mm board.

(20) Although there may be alternative products for some uses (for example wet plaster for wall lining), there are no close substitutes in the construction industry for plasterboard given its cost and handling advantages and the ease and speed of installation.

1.3.2. The plasterboard market

1.3.2.1. Supply

(21) In 1998, the plasterboard markets in the European Economic Area (EEA) were worth around ECU 1 500 million, with production amounting to some 870 million m². Within the EEA, about 80% of all sales were carried out on the German, French, UK and Benelux markets covered by this proceeding. The total annual value of these markets in 1997 and 1998 was approximately ECU 1 210 million, based on a volume of some 692 million m² in 1997 and of some 710 million m² in 1998.

(22) During the period from 1992 to 1998 covered by this proceeding, and following the restructuring described in section 1.3.4, six producers were operating on the EEA plasterboard markets. The three largest were BPB, Knauf and Lafarge, which supplied throughout Europe. Gyproc sold in the Benelux countries, Germany and France. Gyproc AB, 25% of whose capital was held by BPB prior to its full takeover in October 1998, sold only in the Nordic countries. Norgips A/S (hereinafter: "Norgips"), a Norwegian company, left the German market in September 1993 to the benefit of Lafarge and focused on the Nordic markets until September 1998, when it returned to the German market, supplied from its Opole plant in Poland (see recitals (246) et seq.).

(23) Looking at the German, UK, French and Benelux markets and at the overall total for those markets, sales in 1997 and 1998 were approximately as shown in Table 1 (17). The plasterboard producers' sales during the period covered by this proceeding are broken down by year in the Annex to this Decision.

Table 1: Volume of sales in millions of square meters on the German, UK, French and Benelux markets in 1997 and 1998 (18)

<emplacement tableau>

(24) Table 2 shows, on the basis of the figures in Table 1, the plasterboard producers' market shares on the German, UK, French and Benelux markets and the total for those markets in 1997 and 1998.

Table 2: Plasterboard producers' market shares in % on the German, UK, French and Benelux markets in 1997 and 1998

<emplacement tableau>

(25) Table 3 shows the overall size, in 2001, of each of the parties to which this Decision is addressed and their relative position on the plasterboard markets in 1997 and 1998 (19).

Table 3: Size of the parties to which this Decision is addressed and relative position on the relevant plasterboard markets in 1997 and 1998 (20)

<emplacement tableau>

(26) Since the late 1980s, the market has developed from a collection of "national" markets, each dominated by the local producer, into a more "European" market in the sense that the major suppliers are now present in all the national markets (see section 1.3.4 below). As a result of amalgamations, acquisitions and extensions, the three major producers each have production operations in several Member States. BPB operates plants in the United Kingdom, Germany, France, Ireland, Spain and Scandinavia. Knauf's main production is in Germany, but it also has board factories in the United Kingdom, France, Belgium, Austria, Spain, Greece and Scandinavia. Lafarge's production facilities are located in France, the United Kingdom, Germany, the Netherlands and Spain.

(27) The producers each operate their respective production facilities and the supply-demand balance on a coordinated European level, so that a particular national market may be supplied in part from plant located in another Member State. For example, BPB supplies the German market on a regular basis not only from its factories in that country, but also in substantial quantities from Austria, France and Spain. It will be noted that, with the exception of Norgips' sales from its Polish plant as from September 1998, there were no significant imports from outside Europe during the period covered by this proceeding.

1.3.2.2. Demand

(28) Demand for plasterboard depends largely on the level of domestic and industrial and commercial construction, maintenance and refitting. Building activity is the major determinant of demand. Since there is no real substitute in the construction industry for plasterboard given its cost and handling advantages and the ease and speed of installation, demand is generally price inelastic, and, in any case, the plasterboard manufacturers are themselves the leading suppliers of alternative products such as gypsum plaster. Plasterboard is increasingly being marketed not so much as a basic product, but as part of an integrated building "system" along with accessories and fixings. Product development and new applications are an important element in marketing and strategies.

(29) In 1998, the largest national market for plasterboard in the EEA was Germany at around [240-250] million m², followed by France at almost [200-210] million m² and the United Kingdom, where consumption was in the region of [190-200] million m². The Benelux market amounted to around [60-65] million m². During the last ten years, annual demand for plasterboard in western Europe has risen from around 550 million m² to over 870 million m². The widespread use of plasterboard in the building industry means that the number of final customers is very large, ranging from large multinational construction concerns to individuals doing DIY (do-it-yourself) work. Most final customers meet their needs through builders' merchants, who purchase from the plasterboard producers; the producers supply plasterboard direct only to the largest accounts. Many of the large builders' merchants are multi-outlet chains operating nationwide or in several Member States. Distributors (wholesalers) are also used by the producers to supply a range of products to smaller builders' merchants and large end-users. Other direct customers of the board producers are DIY home improvement groups, large building contractors and original equipment manufacturers (OEMs) such as laminators who make secondary board products.

1.3.2.3. Pricing

(30) Pricing of plasterboard is complex, with a variety of discounts and rebates, so that nominal price lists may be substantially above real net market prices. Lists are published by the national producers in each national market; the 12.5 mm board price is generally used as the reference price. Prices are usually quoted on a delivered basis with an allowance for collection ex-works. The producers issue price lists to their trade customers which are subject to standard discounts for the different product groups: incentive discounts for delivery into merchant stock, load service charges, credit for pallet returns, etc. Contractors purchasing plasterboard direct from the manufacturers for specific projects also receive special prices or discounts.

(31) The pricing structure and presentation of price lists vary considerably from market to market. In the United Kingdom, for instance, the pricing system was altered in or about 1992 to reduce the wide differential between "list" prices and actual net trade prices which had encouraged sellers to adopt the "list" price (notionally the purchase price) as their own resale price. The price lists in Germany, by contrast, typically start with a published "gross price" (Bruttopreis) subject to a standard [...]% dealer rebate, a "load" rebate (Ladungsrabatt) of DEM [...] to DEM [...] per m² and special rebates (Sonderrabatte) which mean that "net" invoiced prices are [...] % or more below the published rate. The so-called "net/net" price, after special payment terms, cash discounts and year-end bonuses, is at least [...] % lower again.

(32) When they are planning a general price increase, the producers send their customers letters informing them several months in advance of the change. Whatever the apparent complexity of the pricing structure, the price changes are expressed and understood in terms of a percentage increase. However, a "headline" increase of a certain percentage does not mean that same percentage increase in "net/net" prices. Price lists incorporating the new prices are sent out nearer the effective date of the announced increase.

(33) Notwithstanding the "Europeanisation" of the plasterboard market, there can be ex-works price differences of as much as 30% between the different European countries, including the five principal national markets (Germany, France, the United Kingdom, the Netherlands, Belgium and Luxembourg), prices being significantly lower in Germany and Italy.

1.3.2.4. Cross-border trade

(34) Although the major manufacturers are now present in all national markets, and indeed may supply one market from facilities located in another Member State, cross-border trade at wholesale or distributor level is limited, despite the substantial price differences. Among the reasons cited by the producers for the relatively low level of cross-border trade are differences in standards, differences in standard board widths between countries, the non-availability in destination countries of the required accessories (joint fillers, profiles, etc.) and differences in technical language within the building trade.

1.3.2.5. The relevant market for the purposes of this proceeding

(35) Although a precise definition of the product market is not necessary, the relevant market could be defined as the market for plasterboard, a manufactured product used as a prefabricated construction material and consisting of a sheet of gypsum plaster sandwiched between two sheets of paper or some other material. As will be seen, the evidence available points to the conclusion that European plasterboard producers deemed it necessary to adopt anticompetitive behaviour, together, on the German, UK, French and Benelux markets. Although there is some evidence suggesting an agreement at European level covering not only these Member States, but also the Community or even the EEA, the Commission will, in the light of the evidence in its possession, restrict its assessment to the above-mentioned markets.

1.3.3. The producers

1.3.3.1. BPB

(36) BPB is at present one of the world's largest gypsum product groups. It operates 90 manufacturing plants and sells in 45 countries. It is the leading supplier of plasterboard and plasters in Europe, Canada and South Africa. The group manufactures a wide range of building materials, plasterboard, plasters, paperboard, paper and packaging products. It supplies more than 10% of world plasterboard demand and, both in the EEA and on the German, UK, French and Benelux markets, holds a market share of around [40-45] %. As the world's largest producer of gypsum plaster, it sold three million tonnes of plaster in the financial year 1998-99.

(37) Up until the mid-1980s, BPB was a mainly United Kingdom/Ireland gypsum and plasterboard producer (with its United Kingdom subsidiary British Gypsum having a dominant position in those markets), though it was already well established in Canada, France, Belgium and Sweden. As from the mid-1980s, through various acquisitions (21), BPB became a Europe-wide operation with an integrated production of natural gypsum and plasterboard liner. In 1987, it acquired Rigips, the second largest plasterboard manufacturer in Germany, although it is Swiss-owned. The acquisition gave BPB a significant presence in the German, Austrian, Dutch, Swiss and Italian markets, as well as a 29% share in TecnoKarton, a plasterboard liner mill in Germany whose other owners were the cartonboard producer Weig (45%) and Knauf (26%). BPB owns 46% of Gyproc Benelux NV, a company set up in 1957 by BPB and Eternit (now Etex) and holding a licence for the "Gyproc" trade mark in the Benelux countries and Germany. In France, BPB owns the Placoplâtre plasterboard manufacturer (the largest producer with over [40-45] % of the national market), as well as gypsum mines and plaster facilities. In October 1998, BPB acquired from Scancem AB the entire share capital of the Scandinavian plasterboard and gypsum producer Gyproc Group AB in which it already held 25%. BPB operates some 20 plasterboard factories in western Europe in the following countries: the United Kingdom, France, Ireland, Greece, Germany, Austria, Italy and Spain.

1.3.3.2. Knauf

(38) Founded in 1932 and having its head office and a large industrial plant at Iphofen, Bavaria (Germany), Knauf now comprises a number of private companies which are still owned by [...] or so shareholders belonging to the Knauf family (22). The company likes to present itself as the family-owned company, Knauf Westdeutsche Gipswerke: "The firm Gebr. Knauf Westdeutsche Gipswerke, Iphofen, founded in 1932, is at present not only one of the leading producers of building materials in Europe, but a group which operates worldwide and whose activities are not restricted to the production of gypsum-based materials. Despite its growth, Knauf remains a family-owned business, being owned by the Alfons and Karl Knauf families [...]." (23).

(39) Knauf Westdeutsche Gipswerke is in fact the oldest of the companies in the Knauf group and employs a large number (more than 1 000) of the group's employees; it is at present a limited partnership whose [high level representatives] are [Messrs B and C] (**). The company operates on the same premises and with the same staff (24) as another company, Gebrüder Knauf Verwaltungsgesellschaft KG, which is also a limited partnership whose [high level representatives] are also [Mr B] and [Mr C] and whose function is to administer other companies in the Knauf group. It should also be noted that, in addition to having the same management, the two limited partnerships have exactly the same shareholding structure (the same individuals holding exactly the same share in the company capital (25). Gebrüder Knauf Verwaltungsgesellschaft KG has a very small number of employees, also on the Iphofen site.

(40) Knauf is now a worldwide building materials producer operating over 60 manufacturing plants in Europe. As a private company, Knauf does not publish its accounts. Some [20-30] % of its turnover is in plaster and plasterboard (26).

(41) Knauf is the second-largest European plasterboard producer and, both in the EEA and on the German, UK, French and Benelux markets, has a market share of around [25-30] %. It also produces expanded-polystyrene-based insulation materials. Knauf operates plasterboard production facilities in Germany, Spain, France, Austria, Denmark, Greece and the United Kingdom. It entered the UK market in 1988 and built a board plant at Sittingbourne (Kent) in 1989 and another at Immingham the following year. It acquired the small Belgian producer Isolava and the Danish producer Danogips in the early 1990s. It sources its gypsum requirements from DSG in Germany and its own mineral gypsum mines in Spain.

1.3.3.3. Lafarge

(42) Lafarge is one of the world's largest manufacturers of construction materials, with factories in 65 countries. Lafarge is organised in five divisions ("branches"): cement, aggregates and concrete, roofing, plaster and speciality products. Cement and concrete together account for two thirds of Lafarge's sales. The Plaster Division, which operates 46 facilities in Europe, of which ten produce plasterboard, accounts for about 6% of the group's sales. During the period covered by the proceeding, in the EEA and on the German, UK, French and Benelux markets, Lafarge had a market share of some [20-25] %, its principal national markets for plasterboard being France, Germany and the United Kingdom.

(43) The Plasterboard Division of Lafarge is incorporated under its 99.99%-owned subsidiary Lafarge Plâtres International, which became Lafarge Gypsum International SA in 2000 (27). Lafarge Gypsum International SA owns several subsidiaries, including the French subsidiary Lafarge Plâtres (99.92% owned) and Lafarge Beteiligung GmbH (wholly owned), which itself owns (among others) 100% of the German subsidiary Lafarge Gips and the British subsidiary Lafarge Plasterboard. In 1997, Lafarge acquired through a takeover bid the British construction materials group Redland PLC, which had opened a plasterboard factory in the United Kingdom in 1988 and sold it to Lafarge in 1990 (Redland Plasterboard Ltd). Following the full takeover in 1997 and subsequent integration of Redland PLC, 1998 was the first year in which Redland's results were consolidated with those of Lafarge. In 1993, Lafarge acquired the Dutch plasterboard factory of the Norwegian producer Norgips, together with its German distribution company Norgips GmbH: this gave Lafarge a significant presence in the German market.

1.3.3.4. Gyproc

(44) During the period covered by this proceeding, Gyproc was the fourth-largest European plasterboard producer with a market share in the EEA of around [5-7]% and, on the markets covered by this proceeding, a share of more than [7-10]%, even taking account of the fact that Gyproc was not operating on the UK market. In 1972, Gyproc entered the German market and set up its own distribution network. In 1974, Gyproc set up with Knauf a joint-venture production company, Gyproc GmbH, at Harteshofen, with Knauf holding a 30% share.

(45) Etex, the Belgian building materials group, is the major shareholder in Gyproc, with a 54% holding; BPB owns the remaining 46% of the share capital. Gyproc was established in 1957, the principal shareholders at that time being BPB and Etex (then Eternit), which each held around 45% of the shares (the rest of the shares were in private ownership). Gyproc holds the licence from BPB for the Gyproc trade mark and brand name in the Benelux countries and Germany. In 1987, Etex acquired the 10% or so shares which were in the hands of private investors and thus obtained control of the company. This development followed a deterioration in the relationship between Gyproc and BPB. According to Etex, BPB's then [high level representative], having taken it upon himself to act as an "intermediary" to help Gyproc acquire Rigips in Germany, had then used his position to secure the German company for BPB instead. Gyproc then entered the UK and French markets in retaliation against BPB (but withdrew from the United Kingdom a few years later). BPB remains a significant minority shareholder in Gyproc and appoints three directors to its board (one was [Mr ...], [high level representative] of BPB from [...] to [...]). Etex appoints the other four directors.

1.3.3.5. Other producers

(46) During the period covered by this proceeding, Gyproc AB was the largest plasterboard supplier in the Nordic countries with over [50-55]% of the Nordic market and a dominant position in Sweden, Denmark and Finland. BPB owned a 25% shareholding, the rest being held by Scancem AB of Sweden. There was no connection between Gyproc AB and Gyproc (apart from the name and the shareholding of BPB). In October 1998, BPB acquired full control of Gyproc AB for GBP 95 million (ECU 140 million). Gyproc AB's turnover in 1997 was SEK 870 million (ECU 100 million).

(47) Norgips A/S is a family-owned company with its head office in Norway and supplying mainly in Norway, as well as in Sweden and the Baltic States. It sold its Netherlands plasterboard factory, which supplies mainly in northern Germany, and its German distribution company to Lafarge in September 1993, with a clause prohibiting it from competing on its previous continental European markets for five years, i.e. until September 1998 (28), with Lafarge terminating at the end of the agreement its shareholding in Norgips in the Scandinavian countries (29). Norgips owns a production facility in Opole, Poland, which started in January 1998 and sells to Germany, where it is said to have managed to acquire a market share of nearly [5-7]% in 2001 (30).

1.3.4. Industry restructuring

(48) Up to the mid-1980s, the European market in plasterboard was characterised by a large number of producers each supplying on a regional basis. The producers and their markets were as follows: - BPB, mainly in the United Kingdom, Ireland and France, but also on several other European markets through subsidiary and associate companies; - Knauf, in Germany, Austria, the Benelux countries, Switzerland, Italy and Scandinavia; - Rigips, in Germany, Austria, the Benelux countries, Switzerland, Italy and Greece; - Lafarge, in France, Italy, the Benelux countries and the United Kingdom; - Gyproc, in the Benelux countries, Germany, Austria and Switzerland; - Gyproc AB, in Scandinavia; - Norgips, in Scandinavia, the Benelux countries, Germany and France; - Danogips, in Scandinavia and Germany; - Escogips, in France, the Benelux countries and Germany; - Epysa (40% owned by Lafarge), in Spain, France, the United Kingdom, Ireland and the Benelux countries; - Arklow Gypsum, in Ireland and the United Kingdom; - Isolava, in Belgium.

(49) The consolidation of the industry began in 1987 when BPB acquired Rigips, the second-largest German producer, from its Swiss owners as part of its European expansion strategy. This acquisition had an effect which BPB, for example, describes as "dramatic" (31) in that it triggered a series of amalgamations, expansions and strategic decisions by the various producers. Knauf began its own programme of European expansion, setting up two production plants in the United Kingdom where BPB had enjoyed a virtual monopoly. The German firm also acquired Danogips and Isolava and built plants in Spain and France. Lafarge set up a new plant in Alsace and announced its intention of entering the German market. Redland PLC, the UK construction materials group, set up a board plant near Bristol in partnership with CSR, an Australian mining company. Redland also purchased Escogips in France and the Norgips factory in the Netherlands (Redland sold its plasterboard operations to Lafarge in 1990 and was itself taken over by Lafarge in 1997). BPB acquired other companies in France (Plâtres Lambert, Henry) and, on the failure of Arklow in Ireland, purchased the equipment and reassembled it in a new plant in Italy. Gyproc bought a 40% share in Nida Gips, a Polish producer formerly wholly-owned by the State and sold to Lafarge in 2001. Gyproc AB was taken over by BPB in 1998.

(50) The process of concentration in the industry during the 1980s and 1990s has left BPB, Knauf and Lafarge as the only major players in the European market, with Gyproc as a relatively minor fourth player, followed since the end of 1998 by Norgips.

(51) The main producers have an interconnected network of minority shareholdings, joint subsidiaries and joint ventures in the manufacture of products upstream, the supply of gypsum and, in some cases, the marketing of plasterboard: - BPB holds 46% of Gyproc and appoints three of its directors; - Knauf has a significant minority holding (30%) in two Gyproc subsidiaries in Germany (Gyproc GmbH and Gyproc GmbH Baustoff-produktion & Co KG, both of which are located at Harteshofen); - BPB and Knauf each have a substantial holding (29% and 26% respectively) in Technokarton, a German plasterboard liner producer; - BPB and Knauf have a joint subsidiary in Spain (32); - shared gypsum supplies may exist between competitors (33).

1.4. Details of the infringement

1.4.1. The meeting between BPB and Knauf held in London in 1992

(52) BPB states that, in early 1992, Mr [A], [high level representative] of BPB, met [Mr B] and [Mr C], his opposite numbers in Knauf, in London: "the exercise in question [the exercise of exchanging data amongst the [high level representatives] ] began some time in 1992, following a meeting between Mr [A] and [Mr B and Mr C] in London in early 1992, which according to Mr [A] was convened at [Mr B] and [Mr C]'s initiative" (34).

(53) The holding of the meeting in London "in early 1992" is confirmed by Knauf, which states, however, that its [high level representatives] do not recall who took the initiative, an aspect to which it does not attach any importance. (35)

(54) There is no documented record of this ad hoc meeting between the two competitors, which did not take place within the framework of any broader official or sectoral meetings.

(55) On the subject of the exchange of information between [high level representatives] which the Commission's investigations uncovered, BPB stated that, at that meeting, the representatives of BPB and Knauf "reached an understanding that it was in the interests of BPB, Knauf and the industry as a whole (including, ultimately, the interests of consumers) for the ruinous price war to end and for producers to attempt to compete at more sustainable economic levels" (36).

(56) BPB subsequently argued that the term "understanding" used by it should be interpreted in its most general sense as meaning a "consensus of views": having discussed the state of the industry, in particular, low prices and low or negative margins, and the resulting bleak outlook, "Mr [A, BPB] and [Mr B and Mr C, Knauf], expressed similar opinions regarding the need for producers to adjust their conflicting strategic goals in relation to market share [...] Mr [A] states that he did not agree with [Mr B and Mr C], at that meeting, either expressly or impliedly, that BPB and Knauf would collectively actually attempt to end the price war, increase prices or fix market shares" (37).

(57) BPB states that any effects of the London discussions between the [representatives] of BPB and Knauf were "short-lived": "despite the discussion, once prices edged upward [...] competition remained intense across the various European markets. Once the corner had been turned, there was vigorous competition on a more rational and sustainable basis" (38). BPB goes on to state that "there was no agreement to fix prices, or to fix market shares. Mr [A, BPB] noted the [Mr B and Mr C, Knauf]'s views and observed events; he was sceptical but eventually he tested the water, and prices were edged marginally upward. Once that turning point had been effected, competition continued between the parties and, in fact, prices continued to fall in the UK though stabilised (at least temporarily) elsewhere" (39). BPB also acknowledges that the discussion "may have been a contributory factor in accelerating the end of the price war" (40), although many factors were at work leading to the ending of the price war, which would have come to an end one way or another (41).

(58) BPB stated in its description of the London meeting in early 1992, based on a statement by Mr [A, BPB], that: "at that meeting [Mr A] and [Mr B and Mr C, Knauf] agreed to exchange sales volume information for 1991, to give themselves a reliable basis going forward to monitor whether this understanding was effective (i.e. simply to give each other a more accurate picture of the overall size of the market and thus their own market share). This was necessary because there were no reliable industry statistics" (42).

(59) Subsequently, BPB was less definite about the date on which the exchange of information between competitors was decided on. When interviewed again by BPB on the matter, Mr [A, BPB] said that he was less sure that he had agreed this at the initial London meeting and that it was more likely that the agreement was reached "at some point later in 1992, probably around the second quarter of 1992" (43).

(60) The Commission therefore finds, on the basis of BPB's statements, that there was a consensus of views expressed between BPB and Knauf at the London meeting on the need to adapt their behaviour on the plasterboard market in line with their common objective of ending the aggressive and destructive competition. The Commission also notes that a turning point was evident in prices on the various markets following the London meeting, which shows that BPB and Knauf did in fact change their behaviour following that meeting. The Commission notes in addition that, according to BPB's statements, BPB engaged in an exchange of information with Knauf at, or immediately after, the London meeting, with a view to ending the fierce competition that was prevalent in the plasterboard industry in the early 1990s.

(61) Knauf, for its part, gives the following description of the meeting (44): according to Knauf, its purpose was to discuss the joint purchase by BPB and Knauf of gypsum deposits in France and Italy. However, Knauf also states that, during the discussion, the representatives of the two undertakings touched on other subjects, which Knauf considers normal practice. Knauf states: "contrary to the Commission's supposition, there was no agreement on the ending of the price war. What may have happened is perhaps that mutual calls to commercial reason may have been exchanged to put an end to the ruinous pricing. That competition between the parties could not continue in the long term under these ruinous conditions, that consequently the price war had to end, was at that time so clear to everyone that a stated common understanding on this point ("reached an understanding") is simply soapbox oratory" (45). According to Knauf, an agreement presupposes a commitment subjectively and objectively, and this was not the case. There is no reciprocal explanation of the conduct which everyone was to adopt on the market. Knauf goes on to state that "what [Mr B and Mr C, Knauf] [...] are certain of is that no agreement or concertation to end the price war (or anything similar) was reached" (46), either on limiting destructive competition or on quotas.

(62) According to Knauf (47), the situation should be seen from the point of view of [Mr B and Mr C, Knauf]. Their undertaking had at that time not been present for long on the United Kingdom market, and [Mr B and Mr C] were not satisfied with the position achieved on it. If they had reached an agreement with Mr [A, BPB] on ending the price war, they would have also had to be satisfied with the position attained on the United Kingdom market, and that was impossible for them, an agreement would have been contrary to their interests.

(63) Lastly, according to Knauf (48), what happened on the market does not in any way prove that an agreement was reached in 1992. The "fierce" competition that was taking place on all the major markets between the operators was ruinous, since competition could take place only on prices, given the characteristics of the plasterboard market. The price war had to end some day or other, and in 1992 it should already have ended long since. In fact, according to Knauf, there was not any calming of the situation in subsequent years either in the United Kingdom or in Germany. Knauf argues that the Commission itself had noted in the statement of objections (49) that the price war had continued in the United Kingdom. According to Knauf, if an agreement had been reached, it should have been reflected at least on the main European market, namely the German market, which, according to Knauf, was not the case.

(64) Knauf states that "on the basis of the talks with [Mr A, BPB], [Mr B and Mr C, Knauf] had said they were willing to provide sales figures" (50). Knauf acknowledges that an "agreement" (51) had been reached to exchange sales information on the United Kingdom, French and German markets, and also on the Belgian market, given BPB's holding in Gyproc.

(65) The Commission notes, firstly, that Knauf does not dispute the material nature of the facts described by BPB as regards the London meeting: it not only took place, it also included a discussion on the competition situation on the market. Knauf acknowledges the possibility that the two undertakings exchanged "mutual calls to commercial reason" and expressed a "common understanding". However, Knauf disputes the meeting's coverage as described by the Commission and considers, as far as the meaning of the phrase can be understood, that there was no more mutual commitment contained in the ideas exchanged than in "soapbox oratory". However, the Commission considers that the terms used by Knauf do at the very least confirm that there was a desire to influence each other.

(66) Secondly, the Commission notes that the argument that an agreement could not have taken place, since it would have been contrary to Knauf's interests, because Knauf was not satisfied with the position it had attained on the UK market, does not appear pertinent in determining whether an agreement was or was not reached at the London meeting. It is not for the Commission to say whether such a discussion was strategically appropriate or rational, but simply to establish whether the discussion did or did not take place and what it covered. The Commission also notes that Knauf regards an agreement to end the price war in 1992 in the United Kingdom as commercially impossible. However, the evidence available, and in particular BPB's statements, point to the conclusion that the scope of the agreement covered the main European markets, but that the agreement took longer to achieve its objectives in the United Kingdom market, producing its effects only after the restructuring of the price system announced by BPB to its clients at the end of July 1992 and implemented on 31 August 1992 (52). Consequently, Knauf's interpretation does not in substance contradict the other evidence available.

(67) Thirdly, the Commission notes that, even if it were true (and there is no evidence that it is), the argument that "fierce" competition or a "price war" was bound anyway to end for economic reasons is not relevant in determining the reasons for which, and the conditions under which, the "price war" actually ended, and in particular whether anticompetitive conduct by operators was the reason for this development. The Commission also notes that Knauf's assessment that it could not be established that "fierce" competition had actually ended in 1992, and possibly a little later on the UK market, is contradicted both by Knauf itself, when it suggests that the ending of the price war was inevitable, and by BPB and Gyproc, which acknowledge that 1992 was a turning point for prices on the plasterboard market. Furthermore, it must be noted that Knauf misrepresents the statement of objections, since the Commission did not state that the price war continued in the United Kingdom, but that the process of a change in price trends took somewhat longer on that market than on the other major European markets and occurred only after a restructuring of BPB's pricing system after the summer of 1992.

(68) Comparison of the statements made by BPB and Knauf prompts the Commission to conclude firstly that - a matter which is not disputed by the two undertakings - at the meeting held in London in early 1992 between BPB and Knauf, the competitive situation on the plasterboard market in Europe was discussed by the participants, even if other subjects may have been discussed as well.

(69) The Commission notes that, although the precise nature and scope of the discussion conducted at the meeting held in London in early 1992 between BPB and Knauf are not described in identical terms by the two undertakings, their statements nevertheless are similar in substance. Thus, what Knauf describes as "mutual calls to commercial reason" is described as an "understanding" by BPB, the two companies acknowledging moreover that the purpose of this exchange of identical opinions was the desire to put an end to the "price war". The Commission considers moreover that it can rely on BPB's more precise statements, since, firstly, BPB has no reason to invent matters the revelation of which is against its interests, in that it could result in penalties for the undertaking (see section 2.1.1) and since, secondly, BPB's statements are based on precise statements given by a direct participant at the meeting, Mr [A, BPB], whereas Knauf's statements are not conclusive (53); furthermore, thirdly, BPB's statements are consistent with the other facts covered by this proceeding.

(70) The Commission accordingly concludes that, at the London meeting in early 1992, BPB and Knauf reached the joint conclusion that the fierce competition for market shares which had been prevalent between them, and more generally on the plasterboard market as a whole, for a number of years and which had resulted in price cuts and far-reaching structural changes on the market should end and that, in future, the parties should discourage any over-aggressive competition and that the market should be stabilised at the status quo.

(71) The Commission further notes that the statements made by BPB and Knauf point to the conclusion that the situation on the plasterboard market changed significantly in 1992. BPB, which stated that from 1988 to 1992 prices fell by 50-75% in real terms and that all the companies were facing heavy losses in 1992 (54), explicitly acknowledges that there was a "turning point" in that year. Knauf, even though it claims that the price war continued, argues that it was evident to all the producers that it had to come to an end. Furthermore, Knauf describes the plasterboard market as a normally operating oligopolistic market (55); Professor [...]'s study on the functioning of the German market from 1992 to 1998, annexed to Knauf's reply, confirms this conclusion (56). However, Knauf elsewhere describes the operation of the market prior to 1992 as "economic nonsense" (57), or as ruinous and unacceptable to any reasonable businessman (58). It must therefore be concluded that a change took place in 1992 that allowed a transition to be made from a type of functioning considered by Knauf to be absurd to a type of functioning considered by Knauf to be more normal on the plasterboard market.

(72) Furthermore, the Commission notes that changes in market shares after 1992 and throughout the period covered by the facts which are the subject of this proceeding were more limited than those which had occurred in the equivalent period prior to the London meeting (see the Annex to this Decision). Lastly, in contrast to the period before 1992, the competitors apparently did not attempt to establish themselves on markets on which they had not been present or to gain market shares substantially on them through the creation or development of production capacities over and above the growth in demand. German reunification and the opening-up of the economies of the central and eastern European countries created a special situation in this regard (see section 1.4.4).

(73) As far as the exchange of information is concerned, the Commission concludes from the statements of the two undertakings that BPB and Knauf decided by common agreement to carry out an exchange of internal company information, namely the volume of sales on the main European markets.

1.4.2. The exchange of information on sales volumes on the four major European markets from 1992 to 1998

1.4.2.1. Modalities

(74) In the course of the investigations carried out at BPB and Gyproc in November 1998, the Commission discovered widespread direct evidence of the exchange of information between competitors that had been carried out shortly beforehand.

(75) It should be noted that, at the beginning of 1998, i.e. before the Commission's investigations, BPB carried out an internal investigation to check the accuracy of anonymous allegations received by its main shareholders to the effect that Mr [D, BPB high level representative] had coordinated collusive arrangements in France with Knauf and Lafarge and had organised a European plasterboard cartel (59). BPB stated to the Commission that this internal investigation had not thrown up any documents showing infringements of Community competition law (60). The Commission is not in a position to check the accuracy of this statement by BPB, but notes that the possibility that the facts at issue could be the subject of an inquiry was discussed within the firm several months before the Commission's investigations took place, which means that the fact that the investigations did not take the firm by surprise must be taken into account in analysing the results of the investigations.

(76) In this regard, a number of documents were discovered dating from autumn 1998 which reveal that exchanges took place between Mr [Mr D, BPB] and the [high level representatives] of Knauf, Lafarge and Gyproc.

(77) A handwritten document entitled "End of October figures" (61) mentions figures in connection with the initials "K", "F", "GB" and "B".

(78) BPB stated (62) that this document indicated Knauf's sales volumes for plasterboard between January and September 1998 in Germany ("K"), France ("F"), the United Kingdom ("GB") and the Benelux countries ("B"), details of which Mr [C, Knauf] had passed on to Mr [D, BPB] by telephone.

(79) Another document, entitled "from Belgique"(63), sets out details of Gyproc's sales over the same period, in square metres, as noted by Mr [D, BPB], who had been provided with these figures by Mr [E] (Gyproc). The relevant markets are Belgium and Luxembourg, the Netherlands, Germany, Austria, Switzerland and France. On the same sheet of paper, Mr [D] also noted by hand the rounded sales figures for Lafarge during the same nine-month period.

(80) According to BPB (64), these figures were provided to Mr [D, BPB] by Mr [F], Mr [G]'s successor as [...] of Lafarge's plaster division as from [...].

(81) A third document entitled "End of September 98" (65) presents, for the United Kingdom, Benelux, France and Germany, a column of unheaded, typed figures and two handwritten columns entitled "BPB" and "K" respectively.

(82) According to BPB (66), the typed column corresponds to BPB's sales volume data for the period from January to September 1998, and the columns handwritten by Mr [D, BPB] 24 correspond to the figures exchanged by telephone between Mr [D] and Mr [B, Knauf]. It should be noted that the figures featured in the handwritten column entitled "BPB" are rounded to the nearest 100 000 square metres, but without any apparent distortion vis-à-vis the figures featured in the typed column, which are exact.

(83) Parallel evidence of these exchanges was discovered by the Commission during its investigations at Gyproc in the diary in which Mr [E, Gyproc] jotted down various items of information.

(84) Thus, the page of the diary for 3 March 1998 (67) gives the market shares of Gyproc, Knauf, BPB and Lafarge in Belgium, the Netherlands, Germany and France. Another page (68) features figures for each of the four producers in Germany, France and the Benelux for the first six and the first nine months of 1998.

(85) Another page of Mr [E, Gyproc]'s diary (69) shows the sales volumes of the four producers on the German market for the first nine months of 1997.

(86) It will be noted that Gyproc had not been trading in the United Kingdom for more than seven years when this information was exchanged in 1998. In addition, the figures for the latter period tally with those indicated by Mr [D, BPB] for the different producers in the documents entitled "End of October figures", "from Belgique" and "End of September 98". In Mr [E, Gyproc]' diary, the sales volumes data were also converted into market shares. Lastly, it should be noted that the figures for the first nine months of 1997 tally with those that Mr [D] indicated for Germany in his tables entitled "9 months 97" (70).

(87) Gyproc has confirmed that this exchange took place in 1998 between Mr [E, Gyproc] and Mr [D, BPB]. In his statement, the [high level representative] of Gyproc indicates that he provided "for nine months in 1998 [...] detailed figures to Mr [D] by telephone. In return, Mr [D] provided information on competitors on the German, Belgian, French and Dutch markets" (71). Gyproc stated that Mr [E] sent a fax detailing Gyproc's sales volumes to Mr [D]'s private address in London (72). Subsequently, Gyproc indicated that Mr [E] had provided Mr [D] with information on three occasions: "in April 98 (figures for the first quarter), in July 98 (figures for the second quarter) and in October 98 (figures for the third quarter)" (73).

(88) In addition, two series of tables setting out details of sales by the four major European producers were discovered in the course of the investigations carried out by the Commission on BPB's premises.

(89) The first series of tables entitled "market shares" (parts de marché) (74) features absolute values and percentages for "France", "Benelux", "Germany", "Great Britain" and the "total" for "I", "II", "III" and "IV", and for the years 1985, then 1991 and 1992 ([year]), 1993 to 1995 ("H1", "H2", [year]) and 1996 to 1998 ("Q1", "Q2", "H1", "Q3", "9 months", "H2", [year]).

(90) The second series of tables (75) features two tables per year, the first entitled "Real market share Europe [year]" (Part de marché Europe [année] réelle), and the second "Estimated market share Europe [year]" (Part de marché Europe [année] estimée) for the years 1995 to 1997 for the following countries or regions: "France", "FRG", "Switzerland", "Austria", "Belgium", "Holland", "Benelux", "Italy", "Spain", "Portugal", "Iberica", "UK", 25 "Eire", "B.I.", "Denmark", "Sweden", "Norway", "Skanska", "Finland", "Nordic", "W. Europe", "Hungary", "Czech", "Slovakia", "Poland", "Romania", "EEC(S)", "EEC(E)" and "Total". Figures are expressed as absolute values and percentages for the following four undertakings or groups of undertakings: BPB, Gyproc, AB Gyproc, the three combined forming "BPB"; Lafarge, Norgips, Epysa, the three combined forming "Lafarge"; Knauf, Danogips, Isolava, the three combined forming "Knauf"; "others".

(91) During the investigation, Mr [D, BPB] was asked about the second series of tables ("Part de marché réelle" table and "Part de marché estimée" table). According to the report drawn up on the spot by the Commission inspectors, he stated that the figures were expressed in millions of square metres, that the sources of those figures "consisted of a mixture of own estimates and talks to various people, essentially Gyproc and Knauf, but that these figures given by the latter two were only limited and in a restricted fashion and only referred to Germany, France, the United Kingdom and Benelux. On the question of the significance of "réelles" and "estimés", Mr [D] replied that the table contained a few hard figures and the second table was his own estimates. If the figures in both tables were the same, it meant that these were both estimates. On the question why he made these tables, Mr [D] replied that this was an intellectual exercise to mix hard information and estimates. On the question how he obtained the information, Mr [D] replied that he had probably received it in informal telephone conversation and not during formal meetings, that this exercise was done once a year by himself and that the information was never received in hard form and never passed on to others and was used for his own guidance only" (76).

(92) It should be noted that, when Mr [D] uses the term "hard figures", it must be understood as referring to the figures provided by his counterparts in the competing firms, i.e. to real figures - although it is possible that the competitors may occasionally have provided deliberately inaccurate figures.

(93) BPB subsequently stated (77) that, in the first series of tables (tables "I", "II", "III" and "IV") concerning the producers' respective sales on the main markets, "I" referred to BPB, "II" to Gyproc, "III" to Lafarge and "IV" to Knauf. BPB also stated that Mr [D], using information obtained from the other producers, kept computer tables showing the sales of the four producers on each of the national markets in Europe, with a quarterly update.

(94) With regard to the second series of tables ("Part de marché réelle" and "Part de marché estimée"), BPB stated (78) that this series had been kept by Mr [D, BPB] in order to compare the difference between the data provided by the other [high level representatives] ("réelle") and his personal estimates ("estimée") concerning the market situation. BPB added that Mr [D] sometimes had doubts as to the accuracy of the figures provided to him by his counterparts and regarded his own estimates as more accurate. BPB also stated that the "real" series consisted of "the figures which he [Mr D] received from the other [high level representatives]" (79).

(95) With regard to the practical arrangements for exchanging information between competitors, according to BPB (80) figures were exchanged in the course of telephone conversations. BPB also stated (81) that, on a limited number of occasions, Mr [D] posted BPB sales information to the home addresses of Messrs [G, Lafarge] and [B, Knauf] and received documents via the same channel. Mr [D] tended to receive information from Mr [E, Gyproc] in document form: he generally received faxes, sometimes handwritten and sometimes typed, 26 at his London home. Gyproc has confirmed this (82), pointing out that Mr [D] used the telephone to pass on figures to Mr [E].

(96) It should be noted that Knauf (83) and Lafarge (84), which do not dispute the substance of the Commission's description in the statement of objections of the information exchanges between competitors, have failed to provide any alternative explanation of the way in which it was possible for their data to be transmitted to Mr [D, BPB] and thus end up in BPB documents.

(97) With regard to the geographical scope of the information exchanges, the Commission notes firstly that most documents concern the four major national markets in Europe, i.e. Germany, France, the United Kingdom and the Benelux countries, and that Mr [D, BPB] stated during the investigation that the information exchanges concerned those markets only. BPB subsequently confirmed this (85).

(98) However, it is evident firstly that the document entitled "from Belgique" (86), which, according to BPB, is the result of exchanges with Mr [E] of Gyproc, mentions figures for Austria and Switzerland too. This might be due to the fact that Gyproc agreed to provide documents for all its sales, as Mr [E] stated.

(99) In addition, the "Real market share" (Part de marché réelle) and "Estimated market share" (Part de marché estimée) tables do not concern just the four major European markets, but all European markets, including Norway, Switzerland, Hungary, the Czech Republic, Slovakia, Poland and Romania. As for determining which data in those tables had been provided by the competitors, and for which markets, Mr [D, BPB]'s remark concerning those tables that "if the figures in both tables were the same, it meant that these were both estimates" does not allow clear conclusions to be drawn. This statement by Mr [D] can be taken to imply that discrepancies in the figures in the two tables for the same market and the same period mean that Mr [D] received direct information from a competitor.

(100) A detailed comparison of the two series shows that, as far as Gyproc is concerned, the "real" data do not tally with the "estimated" data for 1996 in Germany, and for all the markets in which Gyproc was active in 1997, whereas they tallied perfectly in 1995. With regard to Lafarge (87), the figures which differ in 1995 are those for Italy; the corresponding figures for 1996 concern France, Italy, Spain and Portugal; the 1997 figures concern France, Germany, Belgium, the Netherlands, Italy, Spain, Portugal and the United Kingdom. As regards Knauf88, the figures which differ in 1995 concern Germany, Belgium, Hungary, the Czech Republic, Slovakia, Poland and Romania; in 1996, Germany, Spain, Portugal and Poland; in 1997, France, Germany, Belgium, the Netherlands, Spain, Portugal and the United Kingdom (89). It must be concluded from Mr [Mr D]'s statement regarding his tables that contacts existed with Gyproc, Lafarge and Knauf, BPB's competitors, in respect of the results in 1995, 1996 and 1997, not relating exclusively to the four major European markets (France, Germany, the Benelux countries and the United Kingdom).

(101) However, in the absence of evidence to confirm the existence of exchanges on markets other than the four major European markets (France, Germany, Benelux and United Kingdom), the Commission concludes that the exchanges between competitors did certainly relate to France, Germany, the Benelux countries and the United Kingdom. This 27 conclusion is, moreover, borne out by the fact that there is direct evidence of contacts between competitors, which is confirmed by the statements of the companies which took part, and that Mr [D, BPB]'s other series of tables (tables "I", "II", "III" and "IV") concerning 1991 to 1998 has been described by BPB as being based on information obtained from competitors on the four major markets.

(102) As regards the frequency of information exchanges between competitors, BPB stated (90) that there were no arrangements between the company executives concerned to liaise at specific and regular intervals. In support of this statement, BPB emphasises that they sometimes conducted exchanges as much as two months or so after the end of the period to which the exchanged data related. BPB stated (91) that no particular arrangements had been made to exchange sales volume data on a specific date or on occasions agreed beforehand. Data were exchanged in the course of telephone conversations concerning other business matters; Mr [D, BPB] is said to have called Messrs [E] (Gyproc), [G] (Lafarge) and [B (Knauf)] on certain occasions via the companies' switchboards or on their direct lines, but discussions on sales volumes tended to be initiated by the other side. BPB pointed out (92), on the basis of Mr [D]'s statement (93), that an interval of several months would go by, interrupted by a call from Knauf or Lafarge concerning other matters, and in the course of that call the parties would realise that time had elapsed since their previous contact and they then exchanged their data for the period which had just ended. Exchanges were not subject to a strict routine, and Mr [D] received data at varying intervals.

(103) The Commission finds, however, that the "irregular" character of exchanges is not confirmed by BPB's statements to the effect that exchanges were held with increasing frequency over time, first on an annual basis, then every six months and finally every quarter. Similarly, Mr [D, BPB]'s tables point to information collected systematically and regularly. In addition, the fact, as BPB claims, that the parties sometimes exchanged information only two months after the end of the period to which the exchanged data related, even assuming it is proved to be true, is not in itself proof of the "irregular" nature of the exchange. Lastly, the claim that the exchange was "irregular", even assuming it is proved to be true, does not detract from the continuous nature of the exchange carried out throughout the period covered by this proceeding.

1.4.2.2. Objective of the parties participating in the exchange of information

(104) When asked by the Commission about the exchange of information between competitors revealed by the documents as a whole and acknowledged by Mr [D, BPB] during the investigations, BPB initially gave the following description of the information exchange engaged in until the Commission's investigations in November 1998: "The figures exchanged under these arrangements were annual figures for the major markets, namely the UK, France, Germany and Benelux. At some point Mr [D] became responsible for exchanging data in relation to France and Germany. When Mr [H] retired from BPB in [1993], Mr [D] then took responsibility for exchanging data in relation to the UK and the Benelux countries as well. At some stage (Mr [D] cannot recall precisely when but his tables suggest 1993) the parties decided to exchange information on a more frequent basis (i.e. around every six months) and then later (probably in 1995) agreed to exchange on a roughly quarterly basis" (94).

(105) BPB stated (95) that Mr [A, BPB] had ceased to be [high level representative] in [1992], but had continued [to be member] of the board until he retired in 1997. [A high level 28 representative] of BPB's plaster division, Mr [H], retired in [1993]. Mr [D], who until then had been [high level representative] of BPB in [one European country] and the group's [high level representative] for all activities in continental Europe, replaced him as [a high level representative of BPB]. Accordingly, until 1997, Mr [A] was aware that information was being exchanged. According to his statement forwarded by BPB, Mr [D] informed Mr [I], who succeeded Mr [A] as [high level representative] in 1997, only "in general terms" that he was seeking "to estimate the overall size of the plasterboard market in conjunction with the other producers" (96).

(106) BPB subsequently specified that the objective of the agreement to exchange information with Knauf was to provide Mr [A, BPB] with "a basis to assess whether there was a new mood in the industry", i.e. that "the information exchange, at a high level, would provide the degree of mutual assurance that the price war was ending" (97). BPB has moreover explicitly acknowledged that the purpose of the information exchanges effected by Mr [A] was to put an end to the fierce competition prevalent in the plasterboard industry in the early 1990s: "the subsequent two exchanges of historical data effected by Mr [A] may have served, and have been intended to serve, to assist the ending of the price war" (98).

(107) BPB thus draws a distinction, in its reply to the statement of objections (99), between, on the one hand, the first two exchanges of information effected by Mr [A, BPB] in 1992 and 1993 and, on the other, the subsequent exchanges conducted by Mr [D, BPB]. Hence, the "[Mr A] exchanges" are assumed to be linked to the London talks at the beginning of 1992 and the end of the "price war" (100), whereas the "[Mr D] exchanges" are seen as different in character and serving a different purpose, namely a simple information exchange for no other purpose than to provide information on the market to the participants; indeed, BPB accepts that such exchanges constituted an infringement of competition law. In a similar vein, BPB claims that Mr [A] was not aware that the frequency of exchanges was increased in 1993 and again in 1995 under Mr [D]'s leadership (101).

(108) However, in the annex to BPB's reply to the statement of objections, Mr [D, BPB] states that "Mr [A, BPB] collected the data in 1992 and then delegated this task to [Mr H, BPB] and myself in 1993. After Mr [H]' retirement in [1993], I remained the only person at BPB who had any involvement in the exchanges, with the full knowledge of Mr [A] who was, however, cautioning about too frequent exchanges" (102). In the reply to the statement of objections, BPB also stated that Mr [H] initially assumed responsibility for the actual information exchange with Lafarge (103).

(109) BPB also claimed that the information exchange promoted competition by increasing market transparency (104). BPB claims in this way to have minimised the scope of the information exchange by limiting the objective to better market knowledge (105).

(110) BPB also states that the tables drawn up by Mr [D, BPB] for his own company management objectives were never distributed, either within or outside BPB, but that BPB kept them for him (106).

(111) However, the Commission notes a contradiction between the statement that Mr [A, BPB] warned against too frequent exchanges and the statement that he was not aware that the frequency of exchanges had been stepped up by Mr [D, BPB].

(112) Similarly, the information on the respective roles of Mr [A, BPB], Mr [H, BPB] and Mr [D, BPB] tends to support the notion that Mr [A]'s role from the outset was to launch the exchange, the implementation of which was delegated to other BPB employees, with the direct involvement of Mr [D] under Mr [A]'s leadership as from the 1993 exchanges of the 1992 data, and that Mr [A] was kept informed until he left the company in 1997 - which suggests that these information exchanges within BPB continued throughout the period covered by this Decision. In addition, it should be noted that the same arrangements for exchanging information were used from 1992 to 1998, albeit at increasingly closer intervals. Furthermore, BPB's statements about their being two different types of exchange have not been confirmed by Knauf or Lafarge. Lastly, BPB has not provided any evidence to show that it informed the other participants of a change in the objective of the information exchange and had distanced itself from the purpose of the exchanges as previously agreed between the competitors.

(113) The Commission therefore concludes that BPB's reply to the statement of objections represents more than anything else a reinterpretation of the facts for the purpose of creating an artificial distinction between the "[Mr A, BPB] exchanges" and the "[Mr D, BPB] exchanges", without any new evidence being adduced to support this reinterpretation. Rather, BPB's initial reply and Mr [D]'s statements point to the existence of a single mechanism for exchanging information.

(114) The Commission also notes that BPB's statements allow the conclusion to be drawn that there was a link between the information exchange, the end of the price war and market stabilisation. Consequently, if - as BPB itself admits - the initial exchanges were designed to support market stabilisation, it must be concluded that the subsequent exchanges were intended to serve the same purpose.

(115) The argument that the information was not used, even if it were proved to be true, is irrelevant, since the limited use of the information by the company's [high level representative] was admitted by BPB and since, at all events, the very fact of exchanging information between competitors is liable to pose problems under the competition rules (see also recital (163)).

(116) Lastly, it should be noted that any claim by BPB that the exchanges conducted by Mr [D, BPB] were carried out in all innocence, or were at any rate harmless, is invalidated by the instruction given by Mr [D] himself at the end of March 1998 prohibiting his employees from discussing prices or market shares with competitors (see recital (275)): BPB, and Mr [D] in particular, were aware of the unlawful nature of the contacts in which he was actively participating (see also recitals (167) to (170)).

(117) Knauf has stated that its participation in the agreement to exchange information was based on the wish by [Mr B and Mr C, Knauf] to have access to market data for their own business purposes, but that such data were not available (107). It was purely for that reason that [Mr B and Mr C] agreed to make their sales figures available on an occasional basis.

(118) According to Knauf, "this was done initially on a very irregular basis and, moreover, at intervals of half a year or more. Under such arrangements, any misuse of such data for the purposes of checking any supposed agreement is simply impossible. [...] The sales figures are moreover too general for any checks to be actually possible using them. They were moreover provided ex post, so that any direct steering of the market was not at all possible" (108). Knauf also states that the exchange was carried out by informing Mr [D, BPB] in person of past sales and by receiving data in return (109). Nobody at Knauf or BPB, apart from the respective [high level representatives], was informed of the data received.

(119) As far as the purpose of the instrument established is concerned, Knauf states, after having disputed that it was used for the purpose of checking competitors, that the purpose of the information exchanges was "rather the establishment of an internal information instrument in order to be able to assess the size of the market in a more or less reliable manner. The value of such information within the firm lies amongst other things in the fact that internal planning in connection with storage costs and securing raw materials can be carried out more reliably. It is a well-known fact that plasterboard is not only a raw- material dependent product, but particularly unfavourable, as regards the relationship between the value of the goods and storage and transport costs" (110).

(120) Knauf states that a direct exchange was chosen because the previous system of data exchange via a neutral third party had not yielded satisfactory results in terms of the degree of accuracy obtained (111). But Knauf claims (112), on the basis of an internal BPB memo (113) dated 6 January 1993 on the lawfulness of an exchange of information between competitors, that there were plans to organise the information exchange "in an irreproachable form in competition-law terms".

(121) The Commission notes, firstly, that Knauf can only provide an interpretation of the facts as far as Knauf is concerned; consequently, it cannot affirm that the information collected was not used at BPB. Furthermore, with regard to BPB's internal memo, even assuming that it means what Knauf claims, and that is not how the Commission interprets it, Knauf does not provide any information or document proving that it shared what it claims were BPB's intentions.

(122) Secondly, the Commission notes that, while Knauf acknowledges the link between the London meeting and the exchange of information between its managers and those of BPB, it does not provide any details as to when or how the other producers participated in such exchange and does not say when the exchange began.

(123) Thirdly, the reference to Mr [D, BPB] as the sole contact for the information exchange contradicts the more precise data provided by BPB, which refer to Mr [A, BPB] as participating in the exchange until "at some point" (114) Mr [D] took over responsibility for the exchange as regards France and Germany, before also taking over responsibility for the exchange as regards the United Kingdom and the Benelux countries following Mr [H, BPB]'s departure in [1993].

(124) Fourthly, the Commission notes that Knauf claims that the exchange system was not appropriate to allow effective checks on the parties' conduct. However, this does not mean that there was not a desire to carry out such checks, it merely asserts that the means adopted were not effective for such purpose.

(125) Lastly, the ineffectiveness of the previous system of data exchange via a neutral third party, as alleged by Knauf, clearly did not obviate the need for the company to examine the lawfulness of a direct exchange of information with its competitors. Furthermore, the Commission notes that the reference to the "previous system of data exchange via a neutral third party" probably refers to the system that existed on the German market in the 1980s (see recital (213)), which underpinned a clearly unlawful agreement aimed at dividing the market up between competitors. Lastly, as regards the accuracy of this system, there is no reason to suppose that a direct exchange would be any more reliable than an exchange via a third party. The essential difference, and the benefit in terms of accuracy, between exchange via a third party and a direct exchange is precisely that the latter allows the market share of each participant in the exchange to be identified.

(126) The Commission notes in conclusion that Knauf's statements broadly confirm the presentation of the facts stemming from BPB's statements as regards the exchange of information and its purpose, namely the mutual monitoring of changes in each party's market shares and, by that means, ensuring that no producer disrupted established positions and market stability.

(127) As far as Lafarge's participation in the exchange of information is concerned, BPB stated during the investigation that, after the London meeting between BPB and Knauf in 1992, the information exchange process was extended to Lafarge (115).

(128) Furthermore, Lafarge's market share on the main European markets is indicated in Mr [D, BPB]'s tables in absolute terms and percentages from 1991 onwards (see recital (88)) and the Commission concludes that Knauf and BPB were aware of Lafarge's figures from 1992 onwards, which shows that Lafarge took part in the information exchange from 1992 by providing its own figures to its competitors.

(129) Lafarge acknowledges that it did "indeed gradually participate in a limited system of exchange of overall and historical sales volumes directed by Mr [D] of BPB", but states that "its participation stopped there and is in no way linked to any concerted action on stabilising market shares in Europe" (116). However, while it emphasises the inaccuracy of the initial indications provided by BPB regarding the date on which it started participating in the mechanism set up by Knauf and BPB, Lafarge does not provide either the date on which it joined the business information exchange mechanism or details of the circumstances which led it to take part in that exchange and, in particular, of the contacts which must have taken place before the exchange was begun with a view to determining the basic conditions thereof, namely its purpose, the parties involved and the frequency of exchanges.

(130) Lafarge states that "the reality of the exchange of information is very far removed from the description given by the Commission. Lafarge's participation in such a system, and obviously its establishment by BPB, stem from the need of each of the producers to have better market information. We have already emphasised the complexity of these markets and the difficulty of predicting developments on them, notably in the absence of statistics. [...] This reality was particularly marked in the early 1990s, when the producers were emerging from their traditional markets and becoming European operators. Lafarge was at that time mainly operating only in France. It had only just established itself in the United Kingdom and carried out limited exports to Germany and adjacent countries. At that time, the main European markets were also following totally different trends: the British market was declining, the French market was stable, and the German market was expanding, particularly in the East. It should also be borne in mind that, in the United Kingdom, the market was completely opaque because BPB, which published its production and sales statistics when it was alone on the market, stopped notifying them to the government as soon as other operators arrived on the market. This reality is the only reason for Lafarge's participation in any such information exchange. In this respect, it should be pointed out that the fact that Mr [G, Lafarge] did not refer to it is in no way proof that the exchange involved an infringement. Lafarge, which did not control this exchange, took part in it only in order to obtain through it some information to make up for the information which was too often missing on these few markets which were important to it, such as the United Kingdom and Germany" (117).

(131) Lafarge also argues that the information exchange did not constitute a means of detection intended to defuse conflicts, since it was "at any rate too rudimentary to serve as a means of detection", being distorted "since some of the firms had no hesitation in providing inaccurate figures, a practice which everyone was aware of" and that the system was never used as such (118).

(132) In addition, following the hearing, Lafarge provided the Commission with a statement by Mr [G], [a high level representative] of Lafarge's Plaster Division until [...], to the effect that, in view of the "serious lack of data on the relevant markets in Europe" Lafarge's competitors had suggested to him "to exchange some very general information on sales volumes. Not having much to lose, Mr [G] seized the opportunity, in return for some summary information, to gain more detailed knowledge than that available to [Mr G] on the exact size of markets, mainly in Great Britain and Germany." Mr [G] also states that he had no great confidence in the figures provided to him, which he feared were "biased" or were intended only to "mislead" Lafarge, which is why Mr [G] never communicated them to the various executives in Lafarge's Plaster Division under his authority and used them only as cross-checks. Mr [G] states lastly that he suggested the establishment at European level of a statistical information exchange system "in conformity with the rules", but came up against "some reluctance on the part of the market leaders, a reluctance which was difficult to understand, unless one tries to guess who could benefit from the resulting obscurity of the situation" (119).

(133) In its reply to the statement of objections, BPB qualified its initial statements in the following terms: "having agreed to exchange data with Knauf, Mr [A, BPB] (and presumably Knauf as well) then arranged to exchange data with Lafarge as well. Mr [A] thinks that he approached [Mr J] of Lafarge directly, though he says that it is also possible that it was Mr [H, BPB] who made the approach , as he knew Mr [J] better than Mr [A] did. (in any case, it seems that Mr [H] initially took responsibility for actually exchanging the data with Lafarge.) Mr [A] cannot recall now precisely when he (or Mr [H]) initially approached Lafarge, but he thinks that it was in the middle of 1992" (120).

(134) Asked to comment on this statement by BPB, Lafarge stated that it was "extremely hesitant", because it could not identity "the data involved, on which markets, whether it involved a one-off or regular exchange and who carried out this approach. As to the date on which this approach took place, it would probably be in the middle of 1992, but BPB has no certainty of this. Mr [J, Lafarge], for his part, has no recollection of any such meeting and did not give any instruction concerning information exchanges to the executives of Lafarge's Plaster Division, at any time whatsoever" (121).

(135) The Commission finds that the comments by BPB in its reply to the statement of objections clearly complement information already provided by that company rather than supersede it. It is therefore artificial to claim, as Lafarge does in the statement reproduced in recital (134), that the purpose or regularity of the exchange had not been specified. In addition, BPB does not hesitate in its statement, but on the contrary specifies that the contract was concluded with Mr [J, Lafarge] either by Mr [A, BPB] or by Mr [H, BPB], whereas, according to Lafarge, Mr [J] "has no recollection of any such meeting". The Commission also notes that BPB did not refer to a "meeting", as Lafarge states, but to an "approach", which can mean any form of contact - by telephone, for example - just as much as any face-to-face encounter between these two managers of competing companies.

(136) Since BPB's statements were more precise and detailed than Lafarge's, the Commission concludes that Lafarge's reply does not invalidate BPB's statements. Accordingly, the Commission concludes from the presence of 1991 figures for Lafarge in [Mr D, BPB]'s tables, from BPB's and [Mr G, Lafarge]' statements and from Lafarge's failure to dispute its participation in the information exchange that BPB at least informed Lafarge in the middle of 1992, or by the end of August 1992 at the latest, of the agreement between it and Knauf on data exchanges and that, on that occasion, Lafarge acceded to the agreement and implemented it.

(137) The Commission notes the position held by Mr [J, Lafarge], whom BPB states that it approached: Lafarge stated during the investigation in 1998 that Mr [J] was "[high level representative] of the Lafarge group" (122), which is confirmed by his biographical notice available on the group's internet site [...] (123).

(138) At all events, Lafarge's assertion that the purpose of the exchange was lawful cannot get round the fact that Lafarge clearly participated in meetings which were explicitly collusory, as will be seen later in the description of the facts relating to the Versailles meeting in June 1996 and its follow-up (see recital (220)). That meeting should at least have enlightened Lafarge, if it was unaware of it, which the Commission cannot accept, as to the fact that the information exchange had a different scope than the mere reciprocal and allegedly innocent provision of information amongst competitors, but had an anticompetitive purpose of stabilising markets in respect of which Lafarge was participating in an information exchange, i.e. - in addition to Germany - France, the United Kingdom and the Benelux countries.

(139) Furthermore, the Commission concludes from Mr [G, Lafarge]' statements that Lafarge knew, or could reasonably have known, that the information exchange constituted objectively a mechanism for monitoring the market, whatever Lafarge's intentions may have been in participating in it. Mr [G] himself wondered at the lawfulness of the information exchange, but nevertheless involved his company in it. Furthermore, Lafarge states that it was its competitors, i.e. BPB and Knauf, which contacted it, which should not have left any doubt at all in Lafarge's mind as to the anticompetitive nature of the arrangement. In addition, the fact that there was cheating or that Mr [G] suspected or feared as much indicates precisely such a monitoring objective; a simple exchange of information tends to be reliable, and there should not be any reason for cheating in such an exchange, whereas in a monitoring mechanism there will be attempts to cheat, so as not to divulge growth on any given market, which is what the mechanism is precisely there to control.

(140) With regard to Mr [G, Lafarge]' statement that he had no confidence in the information received, which consequently was not worth disseminating within the firm, the Commission notes that this makes it unclear why Lafarge would have participated in an information exchange exercise for more than six years, if the information were useless or completely unreliable.

(141) With regard to Gyproc's participation in the information exchanges on the major European markets initiated in 1992 by BPB, Knauf and Lafarge, the Commission notes firstly that information on Gyproc's market share on the various European markets is set out in the tables drawn up by Mr [D] of BPB from 1991 onwards, and that it is not disputed by any party that the agreement concerning information exchanges between BPB, Knauf and Lafarge also included Gyproc's data right from the outset.

(142) BPB stated immediately after describing the launch of information exchanges in 1992 that (124) "Mr [A, BPB] also obtained sales figures from [Mr K, Gyproc] (125) [...] and exchanged these figures along with BPB's figures, with Knauf and Lafarge". BPB subsequently reworded its initial statement as follows: "At around the same time [mid-1992], Mr [A] also obtained sales volume data for Gyproc for 1991 which he and Mr [H, BPB] exchanged with Knauf and Lafarge, but he did not tell anyone at Gyproc that he had done so. At some later point, Mr [A] cannot recall precisely when, he told Mr [E] of Gyproc that he had exchanged BPB and Gyproc data with Knauf and Lafarge" (126). BPB also stated that "Gyproc only participated directly in the post-price war exchanges from 1994 onwards" (127). BPB also stated that Mr [D, BPB] had taken part in the meetings of Gyproc's Board of Managing Directors and had received on those occasions information about the activities of the joint subsidiary of BPB and Etex (128).

(143) The Commission notes that BPB's statements can be interpreted as meaning that Gyproc did not receive information from its competitors, or even purely from BPB, but that, from a certain point in time that BPB is unable to specify, Mr [K, Gyproc] and Mr [E, Gyproc] knew that their data had been passed on by BPB to their competitors. However, the fact that Mr [A, BPB] informed Mr [K] and Mr [E] would suggest that BPB may have provided Gyproc with that information in 1992 or 1993, given that Mr [D], who had been appointed [high level representative] of BPB, subsequently took over Mr [A]'s active role in the information exchange with competitors (see recital (105)). As for the way in which BPB came into possession of Gyproc's data, it cannot be concluded from BPB's statements with any degree of certainty that Gyproc actively provided those data to BPB or that BPB used data received within the framework of relations between a parent company and its subsidiary.

(144) Mr [E], Gyproc's [high level representative], indicated in a statement to the Commission "that since 1995 he had regularly received details from Mr [D, BPB] of the competitors' German market shares. He did not receive anything else. At Gyproc's board meetings, Mr [D] received details of Gyproc's volumes in m² on all relevant markets. Mr [E] does not know whether these data were passed to the competitors by Mr [D]. In March 1998, Mr [D] warned Gyproc that BPB was being blackmailed in that an anonymous third person intended to lodge a file incriminating BPB with the European Commission. Gyproc's board responded by deciding to cease providing a country-by-country breakdown of sales volumes; Mr [D] would merely receive an overall figure. As a shareholder, Mr [D] insisted to Mr [E] that he should continue to provide the data in question. Mr [E] agreed and, for nine months in 1998, he provided detailed figures to Mr [D] by telephone. In return, Mr [D] provided information on competitors on the German, Belgian, French and Dutch markets. This information exchange stopped at the end of 1998" (129).

(145) Gyproc stated that from 1998 onwards, "following the meeting of the Board and outside that framework, Mr [D, BPB] insisted that he should continue to be provided with Gyproc data. In return, he offered to provide Mr [E, Gyproc] with his assessments of markets of interest to Gyproc. For that reason, Mr [D] telephoned Mr [E] at his Antwerp office each quarter so that he could provide him with the information in question. Mr [E] did so by sending a fax to Mr [D]'s home address in London (Gyproc has not kept copies of those documents). A few days later, Mr [D] would call Mr [E] at his Antwerp office to provide him with data on the French, German and Benelux markets. These information exchanges began in 1998 and were limited to the first three quarters of that year" (130).

(146) Subsequently, Gyproc stated that it had not been aware of any information exchange prior to the beginning of 1998 (131). However, Gyproc stresses that "as from 1997, Gyproc had had suspicions as to the possible divulgation by BPB of its data to third parties" (132). Similarly, on the basis of a quotation from the memo of December 1997 from Mr [E, Gyproc] to Mr [L, Gyproc] according to which "BPB knows our exact figures and, I believe, passes them on to the others" (133), Gyproc claims that "if an information exchange concerning Gyproc data took place with Knauf and Lafarge, this was purely at the initiative of Mr [D, BPB] and certainly not deliberate policy on the part of Gyproc" (134).

(147) Furthermore, Gyproc states that "apart from the above-mentioned limitation in terms of time (June 1996 to 1998) Gyproc Benelux's participation in the information exchange system objected to by the Commission was also limited in terms of the practical arrangements and in terms of the area covered. In terms of practical arrangements: Gyproc Benelux received from Mr [D, BPB] information which the latter presented as estimates and supplied him only with information within the framework or on the margin of the meetings of its board of directors, i.e. in the context of normal relations between directors. It was only quite exceptionally and in very particular circumstances, namely the Versailles meeting (135) and the Opole acquisition negotiations [...] that Gyproc Benelux communicated to the other producers (Knauf and Lafarge) its sales volumes on the German market. In terms of the area covered: with the exception of the period from April to October 1998, the information provided by Gyproc Benelux and BPB related only to the German market. [...] Gyproc Benelux was first contacted by its competitors with a view to concluding an agreement at the Eurogypsum Congress at Versailles. However, the informal contacts engaged in by the producers on the fringes of the Congress related solely to stabilisation of the German market" (136).

(148) Gyproc also states that "it is only as an unavoidable shareholder holding half the seats on the Gyproc Benelux board that BPB obtained information on Gyproc Benelux's turnover. Gyproc Benelux cannot be blamed for this, as in reality it did no more than discharge its company-law obligations. A director simply has to be informed of such basic data as the sales of the company whose business strategy he is to determine and whose activities he is to supervise. Under Articles 527 et seq. of the Belgian Companies Code, directors are responsible for performing the task entrusted to them and liable for any wrongdoing in their management. There is no doubt that to discharge their function properly and manage the company adequately, directors need, among other things, a full detailed overview of the company's business. All documents given to BPB and all discussions with it were in the context of normal relations between directors of a company (preparation of board meetings, discussions at board meetings, information exchanged between board meetings, etc.)" (137).

(149) The Commission notes, firstly, that Gyproc, on its own admission, actively exchanged data with BPB from March 1998 onwards concerning the major European markets on which Gyproc was active - i.e. taking account of the fact that Gyproc was not active on the UK market, the German, French and Benelux markets. Thus, Gyproc provided BPB with sales figures which it knew would be passed on to Knauf and Lafarge; in return, Gyproc received the sales figures of BPB, Knauf and Lafarge.

(150) As regards the period prior to March 1998, Gyproc admits having received sales figure information on the German market from Mr [D, BPB] as from 1995. In addition, Gyproc could reasonably have supposed that BPB did not obtain the information on its competitors on the German market without providing in return the information which BPB had at its disposal.

(151) As will be seen in section 1.4.5.2, Gyproc also received and passed on to BPB, Knauf and Lafarge its data on sales volumes in Germany, as from the meeting of the [high level representative] of these four companies at Versailles in June 1996. Gyproc also participated in the contacts and information exchanges concerning price rises on the German market as from the end of 1994 (see section 1.4.6).

(152) Gyproc's internal memo of December 1997 (138) on the Versailles meeting in June 1996 indicates that the [high level representatives] revealed their sales data for the German market at that meeting, whereas it has been established that information exchanges on European trade including the German market had been taking place for four years, which suggests that Gyproc had not previously been participating in the exchanges of information.

(153) The Commission can therefore conclude that, as from June 1996 at the latest, Gyproc participated in the exchange of information with BPB, Knauf and Lafarge in respect of the German market.

(154) As regards the other markets to which the information exchange initiated in 1992 by BPB, Knauf and Lafarge related, the Commission notes that Gyproc's figures were known to its three competitors from the outset in 1992, witness their inclusion in Mr [D, BPB]'s tables.

(155) It should be noted that BPB's statements to the effect that Gyproc took part directly in information exchanges as from 1994 are not corroborated by other evidence.

(156) Gyproc's statement to the effect that, until March 1998, the information on Gyproc sales in Europe which BPB passed on to Knauf and Lafarge originated from the attendance of BPB representatives at Gyproc's board meetings, but that this took place without the knowledge of Gyproc's management, is plausible.

(157) Gyproc's internal memo of December 1997 referred to in recital (152) also states that "BPB knows our exact figures and I think is communicating them to the others" ("BPB connaît nos chiffres exacts et je pense, les communique aux autres". This statement, which is not restricted to the German market, might confirm that Gyproc was informed of the exchange carried out at European level as from June 1996.

(158) At all events, over and above its proven participation in the exchanges of information on the German market, Gyproc knew or should have known, at least as from the Versailles meeting in June 1996 which its managers attended, that contacts with an anticompetitive purpose were taking place between plasterboard producers with regard to the European markets. Furthermore, the fact that the exchanges were carried out by the top-level executives of the companies concerned, and not just the managers responsible for Germany, should have indicated to Gyproc that the exchange had a wider coverage than just the German market. Moreover, the information exchanged by Gyproc, although relating solely to the German market, formed part of a parallel and more regular communication mechanism than the contacts which followed the Versailles meeting (see the meetings held in Brussels, recital (246), and The Hague, recital (255)). Unless the view were to be taken that such an exchange was a pointless duplication of the exchanges carried out in respect of Germany within the framework of these contacts, it was or should have been clear to the Gyproc managers that such an exchange formed part of a different mechanism.

(159) As from June 1996 at the latest, Gyproc could not therefore have been unaware of the fact that BPB was passing on, or at the very least might be passing on, the information relating to Gyproc's sales on the European markets other than the German market to Knauf and Lafarge and that, by not opposing the supply of information to BPB via its board of directors, it was allowing BPB to pass on its figures to its competitors.

(160) The Commission notes in this regard that, in March 1998, Gyproc stopped supplying such information to BPB at the meetings of the board of directors. Thus, whatever the obligations which Gyproc considered it was under vis-à-vis BPB, it did not hesitate to part from them when it deemed this necessary.

(161) Consequently, the Commission concludes that Gyproc participated in the exchange of information on sales volumes on the four major European markets with BPB, Knauf and Lafarge through the intermediary of BPB as from June 1996.

(162) In conclusion, the Commission notes that BPB, Knauf and Lafarge seek to minimise the scope of the exchange of information on sales volumes on the four major European markets. More specifically, BPB, Knauf and Lafarge claim that the information exchanged was not used (according to BPB and Lafarge) or little used (according to Knauf) in the definition of the companies' strategies.

(163) Even assuming it were proved correct, this argument is irrelevant since all the firms have admitted to the limited use of the information by the [high level representatives] of the companies concerned (139). Moreover, it is the fact that the information might be used which poses the problem. Its non-use could show simply that the conduct of the participants in the exchange does not make any particular reaction on the part of the other participants necessary, precisely because each participant is acting as agreed and because the reciprocal monitoring was effective; or again because the persons receiving the information, the [high level representatives] of the companies, considered that the conduct of the participants, although harmful, did not affect the fundamental objective of the participants in the exchange and that such a situation was better than no coordination at all.

(164) For the rest, these large companies do not explain why their [high level representatives] personally and secretly - or at least (on their own admission) without informing their colleagues - implemented an exchange of information over a period of more than six years. Incidentally, BPB acknowledges this logical contradiction, stating that the information exchange must have been useful, or the competitors would not have persisted with it (140). The companies do not explain why, if the information exchange was regarded as a completely harmless operation, it was surrounded by discretion or even secrecy, with faxes, letters and telephone calls sent or made to each party's private residence and knowledge of it restricted to [high level representatives] [...].

(165) The very fact of carrying out an exchange of information between competitors under the conditions described in this Decision, and by producers representing the whole or virtually the whole supply on a particularly concentrated market such as the plasterboard market is liable to distort competition. In contrast to Lafarge, which argues in substance that "market knowledge" is a necessarily legitimate objective such as to properly justify the conduct of the companies (141), the Commission takes the view that such knowledge may, on the contrary, quite easily lead to conduct whose object or effect is anticompetitive in that it involves a reduction in the uncertainty inherent in the functioning of the market in line with the autonomy requirement on which the economy of the competition system within the Community is based.

(166) Furthermore, the claim made by Knauf or Lafarge that the information exchange was not adequate for the objective of stabilisation is not established. The information provided under the exchange, although global and relating to sales carried out some time before the exchange of information, was nevertheless sufficient to achieve the objective set by BPB and Knauf at the London meeting. The exchange, through its very existence, makes it perfectly possible to achieve the objective of checking that each of the producers remains within the limits of what Knauf has called "commercial reason": each producer participating in the exchange can check that, during the preceding period, none of its competitors tried to disrupt market stability by seeking to gain market share, by cutting prices and by increasing its sales, factors which had hitherto brought about the price war.

(167) Moreover, the managers of BPB, Knauf and Lafarge were not unaware of the risks associated with such practices in the light of competition law and, having been duly warned, they still deemed it necessary to continue with them.

(168) The managers of BPB were fully aware, right from the outset, that they were in danger of infringing Community competition law. On 14 December 1992, Mr [A], Mr [H] and Mr [D] received an internal memo from the executive director on the board, who was apparently unaware of the process in which they were already involved. That memo reminded them of the authorised limits for an official sector-based data collection and distribution system (142). In addition, after having been specifically warned once again in March 1998 of the legal situation in the light of Community law, Mr [D] continued the secret operation without interruption until the date of the investigations.

(169) As for Knauf (143), its legal advisors had informed it of the circumstances in which an information exchange was likely to be tolerated: anonymity and a lack of precise information on competitors' figures, which was clearly not the case with the system set up in 1992.

(170) As far as Lafarge is concerned, Mr [G], who [was a high level representative of] its Plaster Division until [...], stated (see recital (132)) that he suggested in vain to his competitors that they set up a system for the exchange of statistical information "in conformity with the rules" when he first began participating in the exchange mechanism in 1992. In addition, in his statement sent to the Commission following the hearing, Mr [G] says that he had asked his Counsel in 1997 for a legal assessment "of the feasibility of a legally acceptable statistical system" at European level and that he had received an "encouraging" reply (144). The Commission presumes that Lafarge was duly informed of the strict limits to such exchanges on that occasion.

1.4.3. Exchange of sales volume and market share data on the UK market

(171) An information exchange separate from the exchange conducted on the four major European markets - which included the United Kingdom - was carried out by the three companies operating on the UK market, BPB, Knauf and Lafarge. This exchange differs from the first as regards the individuals who organised it, the arrangements made for it and its scope, which was limited to sales on the UK market. However, on that market, its objective was identical to that of the mechanism described in section 1.4.2. in so far as it concerned sales volume and market share data. In addition, it functioned in parallel to the aforementioned mechanism over the same period, i.e. from 1992 to 1998, but apparently stopped functioning in March 1998.

(172) BPB first indicated that market share data appeared to have been exchanged on an informal basis (145). BPB then stated (146) that in 1992, Mr [M], who was then [representative] of British Gypsum, had received instructions from BPB group managers (Mr [H] and Mr [A]) asking him to exchange sales figures with his counterparts at the Knauf and Lafarge subsidiaries. Mr [M] continued his contacts with his company's competitors until he was promoted to [representative] of the BPB group in 1995 but, according to BPB, the practice of exchanging sales figures was continued by his successor at British Gypsum, Mr [N]. These exchanges continued until March 1998, on which date contacts with the company's competitors are said to have ceased.

(173) BPB subsequently stated (147) that "the exchanges in which Mr [N, BPB] was involved until he ceased all contact in 1998 consisted of two parts. The information regarding UK market shares can be analysed in the same way as the exchanges in which others were involved in all four markets, and should be regarded as "less serious". The exchanges were irregular but, when the data was exchanged, it related to monthly data, unlike that received by Mr [D, BPB], where the period varied depending on the frequency of the exchanges. Thus a six month interval would bring forth a market share figure for the six month period; a three month interval would bring forth a figure for the past quarter. The exchanges were never more frequent than that."

(174) The Commission notes that the instructions were given to Mr [M, BPB] shortly after the meeting held in London between [Mr B and Mr C, Knauf] and Mr [A, BPB].

(175) On the basis of BPB's statements, the Commission concludes that the initiative to exchange information on the UK market was taken by BPB, since it instructed one of its employees to start the exchange.

(176) BPB claims (148) that it has not kept copies of all Mr [M, BPB]'s memos concerning figures received and presented as "estimates", but has indicated that two documents dated July and September 1993 respectively and unearthed during checks were examples of such memos. Those documents are two faxes sent by Mr [M] to Mr [D] who, at the time, was still [representative] of BPB [in a European country], but was already conducting an information exchange with [Mr B and Mr C, Knauf] and Mr [G, Lafarge] covering the whole of Europe.

(177) According to BPB, Mr [M, BPB] believes that the first set of data exchanged was for 1992 and that the 1991 data in the documents were purely internal estimates. However, the Commission notes that BPB accepts that Mr [M] took part in exchanges from 1992 onwards (149).

(178) BPB has stated (150) that Mr [M, BPB], who was displeased with these instructions, initially provided his counterparts at Lafarge and Knauf with inaccurate figures on his company's sales. When Mr [M] informed Mr [D, BPB] of this subterfuge, the latter asked him to correct the data for the United Kingdom. Mr [M] gradually amended the figures provided so as not to reveal to Knauf and Lafarge that he had previously provided inaccurate data.

(179) It will be noted that BPB has acknowledged (151) that while Mr [D, BPB] had instructed Mr [M, BPB] to correct the figures provided to his counterparts on the UK market so that they were no longer provided with inaccurate data, he (Mr [D]) was providing inaccurate information to the other [high level representatives] on BPB's sales in France and Germany. This leads to the conclusion that Mr [D] did not hesitate to cheat when he deemed it necessary in order to achieve his strategic objectives. Against that background, it is reasonable to suppose that the provision by BPB to its competitors of reliable figures for the United Kingdom via the UK channel and via Mr [D]'s mechanism would have enhanced the latter's credibility and enabled Mr [D] to deceive his competitors concerning the other two markets.

(180) As regards the practical arrangements for the information exchange, BPB (152) states that when Mr [M, BPB] was in charge, exchanges often took place on the sidelines of business meetings (identified as those of the Gypsum Products Development Association, GPDA) (153) and were conducted verbally "on an ad hoc irregular basis." Exchanges did not take place during the meetings themselves but "rather [...] at separate informal gatherings before or after those meetings." Mr [M] noted the figures down and passed them on to BPB's management as "estimates". Mr [N, BPB] continued these exchanges when Mr [M] was promoted in 1995. Also according to BPB (154), Mr [N] did not receive any specific instructions to continue exchanging information with competitors, and he merely maintained pre-existing contacts. Mr [N] took part in these exchanges "sometimes in person but more generally over the telephone". BPB has specified that the individuals with whom Mr [N] exchanged information were Mr [O] (Knauf) and Mr [P] followed by Mr [Q] (Lafarge) (155).

(181) BPB also submitted to the Commission, further to a request for information (156), a table comprising detailed information on UK sales by "A", "B" and "C". The letters refer respectively to BPB, Lafarge and Knauf (157). The table shows the absolute sales figures, both static and moving, and the trend of the monthly, quarterly and annual combined market shares, as a percentage, to one decimal place, from January 1993 to February 1998. It also shows the trends of sales and market shares as well as sales per working day.

(182) In substance, then, this information is the same as that exchanged by Mr [D, BPB] at European level, except that it concerns the UK market alone and that the most detailed figures, throughout the period under review, are the monthly data.

(183) BPB does not indicate the date on which data collection was interrupted or the reasons for this, but it can be concluded that it corresponds to the date on which BPB's board introduced a compliance programme relating to competition law.

(184) BPB initially stated that the table referred to in recital comprised historic data obtained by Mr [N, BPB] from his counterparts at Lafarge and Knauf. BPB states that "it is possible that the historic figures were the same as those exchanged by Mr [M, BPB]" (158). According to BPB, the table was kept from the point at which Mr [N] became [high level representative] of British Gypsum, i.e. in 1995 (159). BPB subsequently (160) altered its presentation of the facts, stating that Mr [N] had carried out the exchange from November 1996 to about March 1998, and that the figures for the previous period were explained by the fact that he had received a package of historic figures dating back to 1993 from his counterparts at Knauf and Lafarge, which he then stored on computer and updated periodically.

(185) However, nothing in the table supports this claim by BPB, since the figures for the period in which Mr [M, BPB] was responsible for them and those for the subsequent period in which they were the responsibility of Mr [N, BPB] appear in exactly the same form.

(186) In its replies to the Commission's requests for information and its reply to the statement of objections, BPB minimised the scope of the information exchange which took place. Thus BPB states first and foremost that "little use was apparently made of the data" (161). More specifically, BPB seeks to prove that the figures exchanged in this way were not taken into account when decisions were taken on the subsidiary's strategy, on the basis that several internal documents of this subsidiary are based on different figures to those obtained from competitors (162). BPB then states that the objective of this exchange with competitors was to have a better knowledge of the size of the UK market and of British Gypsum's share in it, since no official or trade data were available (163). In addition, BPB justifies the use of this type of exchange by Mr [A, BPB]'s distrust of Knauf in the event of an exchange mechanism being set up via a third party, because Mr [A] suspected Knauf of having supplied the system which had existed previously on the German market with incorrect information (164).

(187) Nevertheless, the fact remains that BPB participated in this exchange on a long-term basis, at the top level of its UK subsidiary, and that the type of exchange opted for was discussed at the top level of the group itself, in so far as different arrangements (exchange mechanism via a third party) were rejected by [a group high level representative]. This discussion is incompatible with the notion of accidental or involuntary activity. Furthermore, in no way can Mr [A, BPB]'s distrust of Knauf in the event of an exchange mechanism via a third party being set up justify establishing a direct exchange between competitors. As already pointed out in recital (125), there is no objective reason to suppose that a direct exchange would necessarily be more reliable than an exchange via a third person.

(188) Moreover, the fact that BPB transmitted incorrect information to its competitors makes it unlikely that it would have accepted data provided by those competitors at face value. As BPB itself acknowledges (165), it is obvious that the company corrected the data before using them. As such, the fact that the figures used for the exchange and those mentioned in the strategy documents of BPB's subsidiary do not tally does not prove that the first set of figures was not used to establish the second set.

(189) Lastly, whatever use is made of the exchanged data - and even if BPB's contention that this exercise was carried out at the highest level for no other purpose than to make a marginal contribution to market knowledge is accepted - the fact remains that this exchange between competitors took place (see recital (163)).

(190) In its reply to the statement of objections166 Knauf acknowledges that it took part in the information exchange described in the statement of objections: "irregular exchanges [took place] between Mr [N] of BPB and his counterparts from Knauf and Lafarge" concerning previous data. These data, which served "internal statistical purposes", made it easier for Knauf "to evaluate market volumes and trends and, accordingly, to plan stocks of raw materials given that all the plaster used by Knauf in its UK plants was imported." However, Knauf has not provided any supplementary information on the arrangements governing this exchange and nor has it disputed BPB's statement.

(191) In its reply to the statement of objections Lafarge does not comment on the exchanges conducted with its competitors in the United Kingdom, except for the following: "Lafarge is unable to comment on the content of Mr [N, BPB]' tables, which are illegible. In any case, Lafarge did not receive those tables" (167).

(192) The Commission, which does not claim that the tables in question were communicated by BPB to its competitors, notes that Lafarge does not acknowledge or dispute having taken part in the information exchange conducted on the UK market, in particular via its employees referred to by BPB, Mr [P] and Mr [Q].

(193) More generally, the Commission notes that the three companies set up, for the UK market, a line of communication between [high level representatives] of their respective subsidiaries in the United Kingdom. The fact that communication took place between the subsidiaries does not alter the fact that these were the same undertakings as those involved in contacts at European level. Nor did any of the three undertakings dispute the relevance of the Commission's presentation in the statement of objections on the grounds that the legal persons in question were not the same.

(194) The Commission also notes that this information exchange remained remarkably consistent over time (seven successive years). BPB's claims to the effect that it was organised on an irregular, ad hoc basis are contradicted firstly by its own statement on the way in which Mr [M, BPB] gradually corrected his initially incorrect data: this would not have been possible if the exchange had been organised on an occasional basis. More fundamentally, BPB's claims are contradicted by the obviously systematic and highly detailed nature of Mr [N, BPB]' table, which was clearly based on a regular flow of information on the other two undertakings' results that was significant in quantitative terms.

(195) The file does not show that the exchange of information between [high level representatives] of the respective subsidiaries in the United Kingdom served a different purpose than the exchange carried out at the same time at European level by [high level representatives] of the undertakings concerned. The exchanges appear similar and there is no evidence to suggest a different purpose. Nor have the undertakings in question argued any such thing. The fact, accepted by BPB, that this undertaking deliberately provided its competitors with reliable data for the United Kingdom via the UK channel while at the same time Mr [D, BPB] was cheating on the other markets, leads to the conclusion that, for BPB, the UK mechanism reinforced the usefulness of Mr [D]'s European mechanism. It is therefore reasonable to conclude that the two exchange mechanisms were linked.

(196) The Commission also notes that the UK market shares of the plasterboard producers to whom this Decision is addressed were extremely stable over the period in question (see the Annex to this Decision).

1.4.4. Exchange of data on price rises on the UK market

(197) Various factors, direct evidence or statements by undertakings show that, over and above the information exchange on sales volumes, contacts took place between the [representatives] on prices and coordinated pricing initiatives. The file also shows that a number of parallel price rises took place on the UK market.

(198) The Commission notes that following an attempted rise in March 1992 which apparently proved unsuccessful (168), on 21 July 1992 British Gypsum announced a new price list which was due to enter into force at end-August 1992, providing inter alia for the reduction of discounts to 5%. The file shows that Redland PLB (Lafarge) and Knauf aligned their price lists and discounts at BPB's initiative169. In that connection, the Commission notes that trade federations complained about these price movements to BPB and the OFT (170).

(199) It also emerges from BPB documents that the company carried out an increase in March 1993 (171) and that in November of that year it announced a 12% rise with effect from January 1994. The same documents show that Lafarge followed this announced rise but that Knauf did not follow it in full (172). Lastly, the Commission notes that on 29 September 1994 Knauf announced a rise of approximately 6.5% with effect from 1 March 1995 (173), that on 2 December 1994 BPB announced a 9% rise with effect from 27 February 1995 (174), and that that rise was followed by the announcement of an identical rise by Lafarge on 6 January with effect from the same date (175).

(200) The Commission notes further that, on 22 September 1995, British Gypsum announced a 12% price rise for standard board to enter into force on 1 January (176). This announcement was apparently followed by Lafarge, which announced the same increase on 13 October with effect from 1 January177 and by Knauf, which announced the same increase on 27 October with effect from the same date (178).

(201) On 7 September 1996, i.e. two days before British Gypsum sent letters announcing a price rise of between 3 and 4%, depending on the products concerned, with effect from 4 November, a regional director of Knauf stated to his counterpart at British Gypsum during the course of a company golfing day that "Knauf would follow BG price increase initiative when British Gypsum intentions are seen in print" (179).

(202) Knauf acknowledges (180) that this discussion took place, but argues that it cannot possibly indicate the existence of an agreement between competitors because the employee concerned was not empowered to implement price rises and that, furthermore, BPB's rise was already under way. As regards Lafarge, the Commission notes that it sent a letter on 20 September 1996 indicating a price rise of approximately 4% with effect from the same date, i.e. 4 November 1996 (181).

(203) The Commission also notes that on 3 June 1997 British Gypsum announced a rise of 3.8% for standard board with effect from 1 August 1997 (182). Lafarge announced a rise of 3.7% with effect from 4 August (183), and Knauf announced a rise of 3.7% with effect from the same date as Lafarge (184).

(204) The Commission also notes that on 27 January 1998 British Gypsum announced a rise of 4.4% with effect from 1 April (185). Lafarge announced a rise of 4.1% with effect from 6 April (186) and Knauf announced an identical rise with effect from 1 April (187).

(205) In the course of another golfing day (before the announcement on 8 September 1998 of British Gypsum's 5% price rise with effect from 1 November 1998), Lafarge's [representative] mentioned to a member of British Gypsum's staff that, for budgetary reasons, Lafarge had decided not to go ahead with the price increase at the beginning of January of the following year (188).

(206) BPB claims that contacts on prices between [representatives] of the three UK suppliers served no purpose because the market was particularly transparent and that customers expected producers to be informed of each others' intentions as part of normal business dealings.

(207) However, BPB (189) acknowledges what it calls "isolated instances" when Mr [N, BPB] telephoned the [representatives] of Lafarge and Knauf in the United Kingdom to inform them of British Gypsum's pricing intentions and the planned range of increases. These telephone calls, of which BPB does not indicate even the approximate date, are termed "pure courtesy calls": according to BPB, Mr [N] did not say anything to his counterparts which they did not already know, and he merely informed them of a decision that British Gypsum had already taken.

(208) Knauf acknowledges that contacts were made between [high level representatives] of the UK subsidiaries, and that it is possible that competitors were provided with information on price increases (190). However, Knauf disputes that an agreement between competitors was concluded for that purpose.

(209) Lafarge states that "the evidence on which the Commission assessment is based lacks precision to say the least and does not appear to be firmly established", that "even if one or two telephone calls had been made by Mr [N, BPB], [Lafarge] had no knowledge of their content and no conclusions can be drawn from the background information available". Without disputing the fact that, on one or two isolated occasions, BPB informed it of its plans to introduce a price rise and the size of the increase, Lafarge claims (191) that as such, "this conduct does not constitute an infringement of Article 81 (nor does the Commission claim that it does). The fact that a competitor notifies two other producers of plans which they are already aware of, or are deemed to be already aware of, is purely unilateral behaviour which is not likely to change the behaviour of those producers. It has already been emphasised above that the producers' customers regularly provided them with accurate information on the other parties' prices and price rises [...]" (192)

(210) As regards most of the price rises referred to in recitals (197) seq, the Commission can only note the existence of parallel behaviour between competing undertakings which, it should be reiterated, were involved in collusion during the period in question.

(211) Nevertheless, the Commission concludes that, as far as certain increases are concerned, contacts took place prior to price movements for the purpose of discussing them, and that those contacts should be seen in the light of the objective of the three producers present on the UK market to end the price war, to ensure that it did not happen again and to stabilise the market.

(212) The Commission also notes that, as BPB (193) and Lafarge (194) have indicated, prices on the UK market tended to rise or, at least, to be stable over the period in question, in sharp contrast to the trend between 1988 and 1992.

1.4.5. Discussions on stabilisation of the German market and exchanges of information on volumes

1.4.5.1. Background

(213) An agreement to share the German market, mentioned in an internal memo of Gyproc's (see recital (221)) and confirmed by BPB (195), Gyproc (196) and Lafarge (197) was struck in Germany in 1984 or 1985, Gyproc's share being [10-15] %. The parties were Knauf, Gyproc the Dutch subsidiary of Norgips (purchased by Lafarge in September 1993). The description given by Gyproc stresses the link between this agreement and the parallel establishment, by the parties, of a system for communicating their sales volumes to a neutral consultancy. The consultancy used the data to calculate the market shares and then told each competitor what its share was (but not the shares of the other parties).

(214) According to Gyproc, the time came when Mr [K], [a high level representative], realised that its competitors were not always supplying the consultancy with accurate figures, in order, according to Gyproc, to minimise the volume of their sales on the German market and hence reduce their share of that market. In this way they made their competitors overestimate their own market shares. Gyproc says that from that moment it no longer felt bound by the agreement. All the parties stopped supplying the consultancy with figures after BPB bought Rigips in 1987. According to Gyproc and the other parties, the competitive situation on the German market was radically altered by BPB's purchase of Rigips in 1987 and Lafarge's penetration of the market through imports from the former Norgips plant in the Netherlands and its factory in Alsace, competition being intensified.

(215) The Commission notes, however, that the wording used in Gyproc's internal memo (see (221) suggests rather that in 1988 the market-sharing agreement in question had not been abandoned, at least formally.

(216) The Commission also notes that, according to Gyproc, from 1988 to 1994, BPB and Knauf regularly accused Gyproc of "aggressive" behaviour on the market (198). The accusations were directed at Mr [E, Gyproc] by Messrs [D, BPB] and [B and C, Knauf] at the Eurogypsum conferences and the board meetings of Gyproc (attended by BPB representatives) or of Gyproc GmbH (attended by Knauf representatives) (199). The market share data exchanged between the [high level representatives] from 1992 provided Messrs [D] and [B and C] with evidence in support of their criticisms, based on accurate knowledge of the market, which they gained from their contacts (see section 1.4.2).

(217) The Commission notes, lastly, that documents discovered at BPB/Rigips contain information indicating that the producers on the German market had been in contact and told each other their sales volumes between 1990 and 1994 (200). BPB stated that it came into possession of these Knauf internal data, because they had been given to one of its distributors in a café by a man he had never met before (201). However, without prejudice to the exchanges of information on the main European markets, including the German market, described in section 1.4.2, the Commission intends to limit its assessment to the period after the Versailles meeting in June 1996 as regards the discussions on stabilisation of the German market.

(218) It should also be noted that, from 1992 onwards, BPB, Knauf and Lafarge regularly exchanged sales volumes data, inter alia on the German market, and that from 1995 onwards, this regular exchange took place on a quarterly basis.

(219) The parties' replies to the statement of objections suggest that the construction of new plants (202) in Germany by all the producers and growth in imports from eastern Europe, especially from Poland, in 1996 were potentially such as to generate major upheaval on the German market and to provoke a new "price war" similar to the one ongoing until 1992203.

1.4.5.2. Discussions on stabilisation of the German market at the meetings in Versailles in June 1996, in Brussels in December 1997 and in The Hague in May 1998

(220) [High level representatives] of the four undertakings ([Mr D, BPB], [Mr B, Knauf], [Mr G, Lafarge] and [Mr E, Gyproc]) met in Versailles in June 1996 to discuss the situation on the German market.

(221) During its inspections at Etex, the Commission discovered, in the offices of the Gyproc [high level representatives], a note204 marked "confidential, to be destroyed" ("confidentiel, à détruire"), which establishes that the meeting was held. Gyproc explained later that the note was "a document drafted by [Mr E, Gyproc], addressed to Mr [L, Gyproc] informing him of the situation on the different Gyproc markets, on the occasion of his appointment as a Gyproc [high level representative] in [1997]" (205).

(222) The note states (206) that: "in 1988, [Mr K, Gyproc] told me that there was a market agreement and that Gyproc's market share was about [10-15] %. From 1988 to 1994, Gyproc's market share, according to our internal estimates, fluctuated between [7-10] % and [10-15] %. This period also saw the strengthening of Lafarge's position. During the same period, BPB and Knauf always reproached Gyproc for being too aggressive. In June 1996 [Mr D, BPB] / [Mr B, Knauf] / [Mr G, Lafarge]/ [Mr E, Gyproc] - at a confidential meeting in Versailles - agreed to disclose their real figures for 1995 and to stick to their market shares. The results were as follows:

<emplacement tableau>

And accordingly the market shares were as follows:

<emplacement tableau>

At the time, Gyproc estimated the 1995 market at [210-220] million m2. I confirmed to [Mr D, BPB] that we were maintaining our previous agreement at [10-15] %. Since 1996, we have had more accurate information on the German market after sending our figures to an independent expert".

(223) There follows a table giving the following results, presented as "estimates" (208):

<emplacement tableau>

(224) The note ends with the following remarks (209): "BPB knows our exact figures and, I believe, passes them on to the others. For the period 1988-1994, we estimate that Gyproc's market share in Belgium declined from [55-60] % to [45-50] %. It has now stabilised. In Holland during the same period, we lost [3-5] % and our market share stabilised at [35-40] %. In France our market share is around [3-5] %".

(225) The Commission notes, moreover, that the Versailles meeting was also mentioned in a document coming from BPB (210). This consists of a table dated 9 October 1997 showing the four producers' market shares from 1995 to 1997 (January to September) in Germany (211). Handwritten annotations have been made to the table, identified by BPB (212) as those of Mr [R], [representative, BPB] and hence in charge of monitoring the German market at BPB : "reminded [Mr S, BPB] that there should be no such thing as "Versailles" and he should walk out if it's brought up. Agreed with [Mr M, BPB]".

(226) The explanation given by BPB (213) for these annotations is as follows: at an internal meeting of the board, Mr [S] of Rigips had reported significant gains in market share in Germany for Rigips and Gyproc, apparently to the detriment of Knauf. To Mr [R, BPB], who asked him how the competition had reacted, Mr [S] replied that a Gyproc representative had told him that "Rigips were in breach of Versailles". According to BPB, neither Mr [R] nor Mr [S] knew what "Versailles" meant.

(227) Whether these two persons were in the know or not, and irrespective of the exact nature of the said contact between Gyproc and BPB, which could be liable to sanction as far as Community law is concerned, the Commission notes that a Gyproc representative clearly used the term "Versailles" in the presence of BPB staff to refer to quotas fixed between competitors, and that that individual would not have used this argument if he had not been convinced that, as far as BPB was concerned, "Versailles" was binding. In addition, the Commission notes that the statements of BPB, Knauf, Lafarge and Gyproc as presented in recitals (229) to (243) prompt the conclusion that the Versailles meeting and the exchange of information after the London meeting in 1992 involved the highest levels of these undertakings and that, consequently, lack of knowledge at the lower levels in no way means that the three undertakings were not party to an agreement.

(228) The Commission notes first of all that all the undertakings concerned (214) confirmed that this meeting was held on 5 and 6 June 1996, at the Eurogypsum conference in Versailles, between [high level representatives] of the four companies - Mr [D, BPB, Mr [B, Knauf], Mr [G, Lafarge] and Mr [E, Gyproc]. However, as regards the Versailles meeting's purpose and result, the parties expressed different opinions.

(229) Thus Gyproc indicated that "with a view to stabilising the German market, Mr [D, BPB] and his colleagues proposed an agreement on volumes. Each disclosed his sales volume in m2 for 1995, and on that basis market shares were allocated. Mr [E, Gyproc] was given a share of about 10 %. He was not satisfied with this and wanted [10-15] %. In support of his claims he referred to the former agreement which gave [Gyproc] a share of [10-15] %. More than 10 % was essential since [Gyproc] had just increased its production capacity with a new factory at Peitz in 1995" (215).

(230) Asked by the Commission about the follow-up to this agreement, Mr [E, Gyproc] stated that "the agreement wasn't working well. On 4 December 1997, the competitors met in Brussels to discuss again the desired stabilisation of the German market. The specific occasion was the new production capacity created by Norgips at Opole in Poland. [...] In 1998 the competitors met again at the Eurogypsum conference in The Hague to discuss the German situation. The talks did not yield any concrete results" (216) (see recitals 246 and ).

(231) The Commission asked Gyproc to explain the reference to the memo referred to in recital (221) in which Mr [E, Gyproc] had "confirmed to [Mr D, BPB] that we were maintaining our previous agreement at [10-15] %". Gyproc stated that "at the end of the Versailles meeting and on a later occasion whose date Mr [E] no longer remembers, the latter told Mr [D] that he did not agree. Mr [D] replied that that agreement had lapsed because, since then, the German market had changed considerably as a result of BPB's purchase of Rigips, Lafarge's purchase of Norgips in the Netherlands and Lafarge's penetration of the German market by exporting products manufactured at its French plant" (217).

(232) Gyproc later backtracked on the memo and Mr [E, Gyproc]' explicit statement by arguing that "the so-called Versailles agreement was only an attempt and was never effectively implemented" and also that "there was never a proper meeting of minds among the participants, and definitely not as far as Gyproc was concerned, about all the details of how the German market should be shared. The parties never agreed about the precise market share that Gyproc should have. [...] So Gyproc scuppered the attempt to conclude an agreement between the four" (218).

(233) The Commission notes, however, that these statements by Gyproc, which in principle have less probative value than the above-mentioned memo and the spontaneous statements of Mr [E, Gyproc], do not negate either the content or the purpose of the talks that were held but only, possibly, the result.

(234) BPB, for its part, states (219) that two separate meetings took place at Versailles, one between Mr [D, BPB], Mr [B, Knauf] and Mr [G, Lafarge], the other between Mr [D] and Mr [E, Gyproc]; it bases this on a statement of Mr [D]'s (220) sent to the Commission.

(235) BPB states that "the purpose of the meeting was to discuss the German market, in which prices continued to fall. A proposal was discussed to stabilise the market at 1995 market shares, but in the event, the proposal did not lead to any agreement or understanding between the producers because it was rejected by Gyproc Benelux" (221). BPB also states that the purpose of these discussions was "to agree to fix shares in the German market" but that "the participants in these discussions never reached any agreement of any form concerning market shares in Germany, mainly because Mr [E, Gyproc] was not willing to accept the market share that it was proposed Gyproc Benelux should have" (222) BPB also indicates that there had been an "attempt, admitted by BPB, of three of the parties to arrive at a market-sharing agreement in Germany at the Versailles meeting" (223).

(236) In addition, BPB emphasises what seems to it to be a contradiction as regards Gyproc's memo and Mr [E, Gyproc]' initial statements between the assertion that an agreement had been reached and Mr [E]'s repeated claim to a market share larger than that which BPB, Knauf and Lafarge were prepared to grant to Gyproc.

(237) The Commission notes firstly that none of the parties refers to two meetings, particularly Gyproc, which refers, on the basis of Mr [E, Gyproc]' statement, to the latter being summoned to Mr [D, BPB]'s hotel room, where he also met "representatives of Knauf and Lafarge" (224), identified (see recital (222)) as Mr [B, Knauf] and Mr [G, Lafarge].

(238) The Commission notes in addition that BPB acknowledges that at Versailles it (BPB), Knauf, Lafarge and Gyproc discussed sharing the German market with a view to stabilising it. However, BPB refers only to an attempt by BPB, Knauf and Lafarge to reach an agreement and does not include Gyproc in that attempt, while stressing that the attempt failed because of Mr [E, Gyproc]' unwillingness to accept the market share proposed for Gyproc; this necessarily implies that Gyproc was involved in this discussion in one form or other.

(239) The Commission also finds that the contradiction alleged by BPB concerns the outcome of the discussion rather than its actual subject or the desire of the parties to reach an market-sharing agreement for Germany.

(240) Knauf, for its part, states that an "unexpected" discussion between the managers of the four undertakings took place whose original purpose was to air questions about raw materials and difficulties facing the builders' merchant [...]. In this context, and without [Mr B and Mr C, Knauf] expecting it, the "unsatisfactory" trend on the German market was mentioned. Knauf states that it was "not interested in sharing the market", but that on the contrary it wanted to "strengthen its position as leader on the German market, its home market and the largest for plasterboard, and to regain lost market shares. This situation as regards its interests did not permit any statements along any different lines - and certainly not of the kind alleged by the Commission - on the part of Knauf. Nor did it make any such statements." Nor was there any agreement about quotas or stabilisation. "Whatever is supposed to have been discussed in detail in these discussions on the German market and whatever Mr [E, Gyproc] may have understood, as far as [Mr B and Mr C] were concerned, at most it constituted more or less clear calls to commercial reason on the part of all the participants". "The discussion did not give rise to a result; in the view of [Mr B and Mr C], the purpose of the meeting was not to reach an agreement. Nor did they interpret the discussion in that way." Besides, after the meeting, no period of calm was observed on the market (225).

(241) The Commission finds that Knauf uses the same expression - "call to commercial reason" - as for the London agreement of 1992 (see recital (61)). Knauf does not explain what "commercial reason" means and it does not appear to be incompatible with the existence of a cartel, which could be perfectly "reasonable" from a purely commercial point of view. The Commission also finds that Knauf's statement confirms that a meeting was held between the competitors, that the situation on the German market was discussed and that information was communicated between them. As regards the fact that the meeting enabled an agreement to be concluded, Knauf's statement is not without ambiguity: the fact that Knauf had no interest in market sharing is not per se incompatible with the fact that this discussion could have taken place in its presence and with its consent; Knauf could have taken part in the meeting against its own interests or could have taken part with the intention of deceiving its competitors.

(242) As regards Lafarge, Mr [G], [a high level representative] of Lafarge's Plaster Division, confirms he was invited "to a special meeting" at that Eurogypsum conference and states that the purpose of the meeting "was to gauge the size of the German market, which at the time was the subject of considerable controversy. At least in our case, in Lafarge Platres, there was a large gap between the estimates [...] which was more than [15-20] %. The subjects discussed at the meeting, therefore, were the overall size of the market and its trend, growth and what was driving it. We suffered terribly from a lack of statistics, when we had all just invested in large production units around Berlin. [...] When we announced our overall sales volumes, it appeared that the German market was even bigger then we thought, by about [30-35] % at least at the reference date selected. Despite disappointment at still being far behind BPB and Knauf, more than we hoped, we could ultimately only be pleased at this finding, since we were rather glad we had invested in a market that was even stronger than we thought. But the reaction of the Gyproc group was quite different. Mr [E], [a Gyproc high level representative], observed on that occasion that its market share was consequently significantly smaller than he thought, whereas he had, it seems, known a time when between the German producers there was an agreement, of which Lafarge Plâtres, which was not present on the German market at that time, was completely unaware. I clearly remember Mr [E] thinking that he had been the victim of a plot between BPB, Knauf and Lafarge Platres, aimed at keeping him at a lower level than the one he claimed, which was based on the old agreement. The fact is that nobody, except Mr [E], had made the slightest claim as regards market share and we were only trying to gauge the size of the market. I maintain that at the Versailles meeting there was no discussion of any allocation of market shares and it is only when I read the Commission's statement of objections that I learned there was a so-called "treaty of Versailles", which never existed except in the head of Mr [E], who was vexed and frustrated at suddenly having to learn that he was now the No 4 and no longer the No 3 player in Germany. While an old agreement on the German market involving Gyproc may have been applied in the 1980s, nobody ever suggested an agreement to me in the period 1992-98 and I never discussed one with my counterparts. Anyway, I would not have wanted one, since its competitors would have been unlikely to give Lafarge Platres, a newcomer to the German market, as large a market share as the one we were aiming at" (226).

(243) The Commission notes that Lafarge's statement confirms in substance that the meeting between the competitors was held and that information was communicated between them, as established by direct proof and statements made by the other parties. As to the purpose of the meeting, Mr [G, Lafarge]'s statement is contradicted by both the Gyproc memo (see recitals (221) to (224)) and by the mutually corroborative statements of Gyproc and BPB.

(244) In the course of the investigations it carried out on Gyproc's premises, the Commission found a note in Mr [E, Gyproc,'s] diary indicating that he had had a conversation with Mr [T], one of the [representatives] of Lafarge's plasterboard operations, on 4 March 1997: "he told me that one month ago he had to halt the negotiations, which had been proceeding well, because he had not received a final mandate from BPB, Knauf and, above all, Gyproc. The problem appears to lie with Gyproc's market share in Germany. Knauf wants to increase its share by 1% to the detriment of Gyproc; 12 ??11 (Germany). BPB appears to have lost interest in the matter and recently refused to inform [Mr U, independent consultant] of its market share. They are on the verge of concluding an agreement for gypsum in Poland and of building a plant. Gyproc could be involved if it wanted but in that case it would pull out of Opole which is still not working. The whole problem lies with the 1% market share in Germany" (227).

(245) The Commission finds that the note in Mr [E, Gyproc,'s] diary proves the existence of contacts between Lafarge and Gyproc in which the state of competition was discussed and in which the opinion of third parties, Knauf and BPB, appeared to be particularly well known. In addition, this note confirms the existence of discussions or even negotiations on each producer's market share and shows a clear link between these discussions and the information exchange system organised by Mr [U, independent consultant] (see recital (271) below). Lastly, the note sheds light on the link between these discussions and the "Opole project". (246) On 4 December 1997 Mr [D, BPB], Mr [G, Lafarge], [Mr B and Mr C, Knauf], Mr [K, Gyproc], Mr [L, Gyproc] and Mr [E, Gyproc] (228) met once again, this time at the premises of Etex in Brussels.

(247) BPB stated that the purpose of the meeting was to study the proposed joint purchase by the four producers of the factory at Opole, Poland, belonging to the independent Norwegian producer Norgips (229).

(248) As already indicated in recital (230), Gyproc stated that, according to Mr [E, Gyproc], "the agreement [on volumes designed to stabilise the German market] wasn't working well. On 4 December 1997, the competitors met in Brussels to discuss again the stabilisation of the German market. The specific occasion was the new production capacity created by Norgips at Opole in Poland. The competitors tried to identify a mechanism whereby one party could purchase the plant while the others contributed to the goodwill (referred to by the competitors as "nuisance value") which would have to be paid by the purchaser" (230).

(249) According to statements by BPB and that undertaking's replies to requests for information from the Commission, none of which was disputed by the other participants in these meetings (231), Norgips had stated that its factory was for sale and had set what was regarded as a ridiculously high price, since, according to BPB, Norgips hoped that one or more of the plasterboard producers would buy it. To enhance the value of the factory in the eyes of one of the four big players, Norgips had apparently made it known that it planned to target the German market in an aggressive pricing campaign when it returned to that market in 1998. Since Norgips was asking much too high a price - the excessive increase representing Opole's "nuisance value" - none of the [high level representatives] wanted to buy it on their own.

(250) BPB stated that in June 1997 it had considered a joint purchase of Norgips together with Knauf, but had decided against it on the advice of its lawyers (232). BPB also stated that the Opole factory had initially been offered to Lafarge, which a few years before had bought Norgips's plant in the Netherlands (233). Lafarge had reportedly asked the other producers their opinion about a possible "joint purchase" of Opole, or indeed of Norgips.

(251) Knauf, for its part, took the view that the Norgips offer was worth discussing and that various models had in fact been debated; not all producers had expressed an interest. It was thus possible, using each party's share of the German market, to estimate each party's share of the purchase in such a way that no party lost turnover (234).

(252) Knauf also stated that the Brussels meeting had provided another opportunity for [Mr B and Mr C, Knauf] to call the competitors to "commercial reason", a term which the Commission notes had already been used to described the 1992 London meeting and the 1996 Versailles meeting (235).

(253) While stating that the joint purchase of Opole was no more than one project among others for its Polish launch, Lafarge explained the Brussels meeting between the four competitors on the basis that Norgips had raised the bidding price, and that it (Lafarge) did not have sufficient funds to purchase the plant by itself. However, no agreement was reached, and at no time was there any question of stabilising market shares in Germany. Lafarge concluded that "the discussion of 4 November was therefore a purely academic, one-off exercise which would not be followed up. Even though the future level of German market shares was mentioned, nothing came of it" (236).

(254) Several discussions took place, evidence for which was discovered by the Commission (237). The Opole project was the subject of several other discussions between Mr [Mr D, BPB] and Mr [G, Lafarge] and between Mr [G] and Mr [L, Gyproc], who had attended the December meeting, and different scenarios were studied up to 1998 (238).

(255) [Four high level representatives] and Mr [K, Gyproc] and Mr [L, Gyproc] met at the Eurogypsum congress held at The Hague from 13 to 15 May 1998.

(256) Mr [E, Gyproc] stated that on this occasion "the competitors met again [...] to discuss the German situation. The talks did not yield any concrete results" (239).

(257) Gyproc subsequently indicated that "the first subject tackled was follow-up to the meeting held in Brussels in December 1997 and concerning the possibility of sharing the 'nuisance' if Norgips was purchased. Mr [D, BPB] stated that he would no longer take part in such discussions and accordingly the matter was closed. The participants then discussed Norgips's potential share of the German market from its Polish factory. Opinions differed and hence no conclusion was reached. The participants also exchanged their sales volumes in Germany for the first four months of 1998. Gyproc and Lafarge provided their figures and BPB and Knauf supplied their provisional figures, which were confirmed a few days later by telephone. Moreover, each person also mentioned the share of the German market he wished to have: Lafarge: [10-15] %; Gyproc: [10-15] %; Knauf: [40-45] %; BPB: [30-35] % [...]. Since this yielded a total of 101%, the others tried to persuade Gyproc to limit its ambitions to [10-15] %. Gyproc refused and the idea was mooted of allocating the extra 1% on a pro rata basis but ultimately no decision was reached" (240). Mr [E, Gyproc]'s notes, which are contemporary with the meeting, take account of the figures in question (241).

(258) The talks on the Opole plant appear to have been unsuccessful. But the analysis does not stop there, since the very holding of the talks and the fact that they were based on mutual information about sales volumes and strategies suggest that they were a consequence of the Versailles meeting. The significance of the Opole purchase does not lie, therefore, in the success or failure of the operation, but in the mere fact that it was possible to discuss it.

(259) The whole premise underlying the talks was that there was a common interest in maintaining the stability of the market, and the [high level representatives] of the four undertakings were in principle prepared to devote substantial sums to this in the form of their respective "contributions". In discussing their possible contributions to Lafarge's purchase of the Opole plant or of Norgips, the four competitors were considering a transaction which would have had the effect, if not the object, of eliminating from the market an independent competitor whose existence was regarded as a nuisance, or even a threat, to all parties.

(260) It should also be emphasised that these talks, as well as mentioning the Opole project, show that the [high level representatives] discussed their respective ambitions as regards market share in Germany and did so in fact as a direct continuation of the talks held at Versailles in June 1996.

(261) Attention should also be paid in this context to Mr [E, Gyproc]'s statement to the Commission, according to which "the agreement wasn't working well" and "the competitors et in Brussels to discuss again the desired stabilisation of the German market." Mr [E] then states that the specific item which prompted this discussion was the Opole proposal and that "the talks were unsuccessful," but he did not specify whether he was referring only to the discussion about Opole - which, it is established, clearly did fail - or also to a general agreement on market shares (242).

(262) BPB maintains (243) for its part that the discussion of market shares in Germany was simply the expression of hypotheses, that there was no overall "agreement" about anything at all: if there was an agreement, it related simply to the actual value of the shares attained in 1997. BPB argued that some of the participants had rejected any restriction of their future market shares in Germany. Since the hypothetical proposed Opole "premium" required constant market shares in Germany, the proposal never succeeded. BPB rejects Mr [D, BPB]'s calculations as mere "notes" without any significance and states that from BPB's point of view the purchase of Opole, or indeed of Norgips, was "bound to fail" (244). BPB's board of directors did not study any of the proposals. Finally, according to BPB, the proposed "joint purchase" collapsed.

(263) In the light of the above, the Commission notes firstly that none of the parties disputes that the Versailles meeting was the occasion for the four operators concerned to disclose their sales figures on the German market to each other. The Commission also notes that BPB, Knauf, Lafarge and Gyproc provided each other with information on their respective positions on the German market, including details of their sales and market shares, at the meetings held in Brussels and The Hague.

(264) Secondly, the Commission regards it as a given that the Versailles meeting proves the existence of a agreement on the principle of sharing the market between BPB, Knauf, Lafarge and Gyproc, although it was not possible to reach agreement on actual percentages either at that meeting or at the subsequent meetings in Brussels and The Hague. The objective of the four undertakings in question was to stabilise the market by laying down a market share for each operator and all had agreed to their market shares being established by joint agreement. The Commission finds that much of the information given in the note referred to in recital (221) as well as the statements by Mr [E, Gyproc] prompts it to conclude that an agreement on the principle of stabilising market shares in Germany was indeed struck at the Versailles meeting. Firstly, according to the note from Mr [E] to Mr [L; Gyproc], the operators' objective is "to stick to their market shares". Similarly, in his statements to the Commission, Mr [E] explicitly mentions an "agreement" designed to "stabilise the German market", whereby the competitors disclose their market shares to each other and are allocated a market share. Mr [E] also stated that "the agreement wasn't working well".

(265) Thirdly, the Commission concludes that after the Versailles meeting, market shares became a constant subject of discussion between BPB, Knauf, Lafarge and Gyproc, as is shown not just by the three meetings, but also by the contacts which took place to prevent the parties' customers being poached. There are also indications that the participants did not dare to aim for a larger market share without discussing it with the others; in other words, when the producers in question wanted to increase their market share, they consulted each other (see recitals (354) to (366)).

(266) Fourthly, the Commission takes the view that, at the Versailles meeting, notwithstanding the position adopted by Gyproc, the three other undertakings, BPB, Knauf and Lafarge, informed each other of the market shares they would agree to and that those market shares corresponded to the shares those undertakings actually held.

(267) Lastly, the Commission notes that the way in which the discussions were organised, first between BPB, Knauf and Lafarge, then between those three and Gyproc (or even, according to BPB's presentation, between it alone and Gyproc) suggests that at times more intensive concerted action took place between BPB, Knauf and Lafarge. This expression of each of these undertakings' will might have been understood by its competitors as an assurance as to future conduct on its market, at least to the extent that it had no intention of seeking to increase its market share.

1.4.5.3. Exchanges of sales volume data on the German market following the Versailles meeting

(268) The situation before the Versailles meeting was that three of the producers (BPB, Knauf and Lafarge) took part in an exchange of information on the four major European markets, i.e. including the German market (see section 1.4.2 above). It is clear that the three operators considered it necessary to make Gyproc take part in the exchange as far as the German market was concerned, where that undertaking, which overall was much smaller than the three others, had a significant market share.

(269) For BPB, Knauf and Lafarge, the exchanges of information continued to follow the pattern established at the London meeting in 1992. In addition to these long-distance exchanges, at the meetings in Versailles, Brussels and The Hague the undertakings disclosed their sales data for the German market to each other with the active additional participation of Gyproc.

(270) Gyproc, from the date of the Versailles meeting, took part in the information exchanges on the German market carried out at the meetings in Versailles, Brussels and The Hague (see recitals (220), (246) and (255)). Gyproc also received information from and supplied it to the other undertakings via BPB in accordance with the procedures followed by the other three undertakings since the 1992 London meeting, but only as regards the German market (see recital (144) et seq.).

(271) In October/November 1996 the four undertakings set up a system for exchanging information through an independent expert, Mr [U, independent consultant] (245). The operation was placed under the aegis of the Plasterboard Industry Group (Industriegruppe Gipskartonplatten, or IGG), a subcommittee of the German Plasterboard Association. [A high level representative] of the IGG at the time was Mr [V], [representative] at Rigips. Each producer gave its figure to Mr [U] on a confidential basis and the results were compiled by the latter's office, giving an aggregate figure, which was then sent to the participants. This figure enabled each producer to calculate its own market share, but not that of the others. Every quarter, the participants - Gyproc, Rigips, Knauf/Dano and Lafarge - supplied their respective sales figures (in square metres) to Mr [U], who collated them and then sent each an aggregate figure for the whole market (246). The producers also supplied Mr [U], on a confidential basis, with the figures for January to December 1995 and for January to September 1996.

(272) The note from Mr [E, Gyproc] to Mr [L, Gyproc](247), which dates from the same period as the events forming the basis of this proceeding, emphasises the link between the contacts among producers and their exchanges of data, on the one hand, and the [Mr U, independent consultant] system, on the other. The link is due to the fact that, chronologically, the system was set up only a few months after the Versailles meeting. Furthermore, the producers did not confine themselves to calculating their own market shares on the basis of the data supplied anonymously. The note clearly suggests the link between the system and the exchange of information on the German market started at Versailles in July 1996. As the table from the document in question shows (248), the participants used the information exchange conducted through Mr [U] to supplement and monitor the exchange of data carried out at Versailles.

(273) The fact that the "[Mr U] system" is not in itself contrary to Community law - emphasised by all the parties - should not be examined in isolation. The Commission considers that the system was set up by the producers only to supplement their direct exchanges, thereby ensuring a minimum of accuracy for the figures exchanged. The Commission notes in this connection the analogy with the system which existed in the 1980s for the German market and which involved some of the same parties (see recital (213)). (274) Rigips withdrew from the system for exchanging information under the aegis of Mr [U, independent consultant] in April 1998, the last data communicated relating to January 1998 (249).

(275) The withdrawal apparently follows the formal instructions of BPB's board of directors and managing board given to all senior managers ordering them to obey competition law. On 24 March 1998, Mr [R, BPB] sent certain BPB managers a memo entitled "competition issues" indicating that "at the Executive meeting this week, Mr [D, BPB] again strongly stressed that there should be no meetings/discussions with competitors other than those that happen at recognised trade association meetings. At these meetings there should be no discussion on pricing or share. While this has been the long standing policy you can expect that as part of BPB's efforts to be seen as clearly abiding by the legislation, more effort will be placed in this area." (250) On 26 March 1998 Mr [D] sent a memo to Mr [S] of Rigips instructing him to cease any communication with the German Plasterboard Association (251). By fax dated 6 April 1998, Mr [V] of Rigips informed Knauf, Lafarge, Gyproc and Mr [U, independent consultant] of his company's decision to withdraw from the exchange of statistics (252).

(276) BPB subsequently stated that following its withdrawal from this information exchange system, it had taken legal advice on the matter, the outcome of which was that the system was in accordance with the law. (253) But BPB preferred, as a matter of caution, not to take part in this system unless it was notified to the Commission - which it never was.

(277) The Commission notes that BPB decided to withdraw on its own initiative for reasons of compliance with competition law, which shows, at the very least, that by that time the undertaking had doubts regarding the lawfulness of the system of exchanging information under the aegis of Mr [U, independent consultant], or was concerned that its involvement in the system might be viewed as an indication of anticompetitive behaviour.

(278) The Commission also notes that although he gave these very strict instructions to his team, Mr [D, BPB] continued to exchange this same competitive information with the other [high level representatives] under the system that had existed since the 1992 London meeting.

(279) Quite clearly, Rigips's withdrawal from the system of exchanging information under the aegis of Mr [U, independent consultant] worried the other producers. On 8 June 1998 Mr [W] of Knauf sent Mr [V], his opposite number at Rigips, a copy of legal advice from Knauf's lawyers saying that the statistics posed "no problem at all from the competition point of view" (254). It should be noted that, just two days before the Commission inspection, Rigips announced at the IGG meeting that it was joining the "[Mr U] operation", and then withdrew again in December (255). The scrupulous concern shown by Knauf for the formalities of complying with competition law should be assessed in the light of the fact that, at the same time, the undertaking was involved in other contacts with competitors.

(280) It is significant that, in a "personal and confidential" memo to Mr [G, Lafarge] of 8 June 1998, Mr [X] of Lafarge Gips, referring to the decision of the other producers (Knauf, Dano, Lafarge, Gyproc) to continue the "[Mr U, , independent consultant] exchanges" without Rigips, remarks: "We took the assumption that Rigips holds a market share of [30-35] %" (256). The fact is that the assumed market share of Rigips for 1998 exactly matches the share obtained by that producer in Germany in 1997, i.e. [30-35] %, as noted by both Gyproc and Lafarge (see recitals (285) and (286)).

(281) Such precise, exact knowledge comes as no surprise given that the undertakings in question and BPB continued to be actively involved in an exchange of information covering inter alia the German market. This confirms the link between the direct exchange of information between competitors and the [Mr U, independent consultant] system.

(282) Knauf also seeks to minimise the scope of the [Mr U] information exchange system by emphasising that Norgips did not form part of the system, which would make it impossible to use it as a control mechanism (257). However, Knauf itself stresses that Norgips was active on the market "until 1993 and as from the end of 1997" (258), which effectively reduces the import of the previous comment, given that the workings of the [Mr U] information exchange system were disrupted from the start of 1998 onwards following BPB's departure.

(283) The figures in the note from Mr [E, Gyproc] to Mr [L, Gyproc] referred to in recital (221) are quoted verbatim in a table in a memo dated 19 January 1998 discovered during the investigation at Gyproc, entitled "Evolution market shares, market potential in Germany." The note states that the figures of each producer, apart from Gyproc, for 1995, 1996 and 1997 are "estimated based on commercial intelligence". The trend of the market shares was commented on as follows:

- "G realised >50 % of market growth, the remaining [5-7] mio went to R;

- K missed [5-7] Mio m2 to keep their market share at [40-45] %;

- L lost [0-1] % market share or [0-3] mio m2. These went most probably also to R"(259)

(284) The same table is also inserted in Mr [E, Gyproc]'s diary (260) together with his notes about a meeting with [an Etex's high level representative] on 27 January 1998. This invalidates the idea defended by certain parties, in particular Lafarge and BPB (261), that the figures in question were part of an attempt by Mr [E] to mislead Mr [K, Gyproc] in order to draw attention to his own work and its importance: the figures were figures used by Gyproc's managers in their strategy and business activity.

(285) In another internal memo also discovered at Gyproc, dated 26 January 1998 and headed "Strictly Confidential" and "corrigendum to memo 19/1/98", some sales figures and market shares have been slightly altered. The memo states that the new figures were considered "as the most reliable currently available" (262). The figures are as follows: 1996 K [40-45]% [95-100] million m² G [10-15]% [20-25] million m² 1997 K [40-45]% [100-110] million m² G [10-15]% [30-35] million m² L [10-15]% [30-35] million m² R [30-35]% [70-75] million m² L [10-15]% [30-35] million m² R [30-35]% [75-80] million m²

(286) It will be noted that the market share percentages given (to two decimal places) for 1996 and 1997 in the Gyproc memo are exactly the same as those that were used in Lafarge Gips's 1998 marketing plan263, an internal document which had not been published by that undertaking and which Gyproc had no reason to have read. The exact match between the Lafarge Gips and the Gyproc data shows that, as a result of contacts between the producers on the German market after the Versailles agreement, the sales figures were known at national level by the executive or commercial directors. The documents indicate that the latter based themselves on the market share information in order carefully to position themselves on the German market and maintain the stability of their market shares. The Commission also notes the accuracy of these data, which provides further evidence that they were the result of a direct exchange between competitors (264).

(287) The Commission concludes from the above that it is apparent that the contacts on the German market are closely linked to the exchanges at European level. The producers, already involved in a European information exchange system, considered it appropriate, from 1996 to 1998, to set up a more sophisticated system concerning the German market and comprising exchanges of information and an anonymous system enabling the accuracy of the data exchanged to be checked.

(288) In order to ensure stability on the German market, it was necessary to strengthen the existing exchanges of information, to supply more accurate and verifiable information ([Mr U, independent consultant] system), and to meet regularly to check on each party's conduct (which was done through an annual meeting on the sidelines of the Eurogypsum conference). Such an arrangement is perfectly consistent with the general mechanism established at European level, the objective being to avoid a price war and to stabilise the market.

(289) The Commission further notes that the market shares in the Germany of the plasterboard producers to which this Decision is addressed showed some stability over the period concerned (see the Annex to this Decision).

1.4.6. Exchanges of data on price increases on the German market

(290) During its investigation on 25 November 1998 at Lafarge Gips GmbH, the German subsidiary of Lafarge's Plaster Division, the Commission discovered a memo (265) dated 7 October 1998, addressed by Mr [X], [representative] of Lafarge Gips, to Mr [F], [a high level representative] at Lafarge's Plaster Division, and Mr [Y], of Lafarge Gips. The memo, which dates from the same period as the investigation, is headed "Prices in Germany (Plasterboard)". It is written in English and describes in turn the "usual process of price increase action", the "price increase for Oct. 1998", the "result of [the] actions" taken on the occasion of that increase, the author's "personal understanding (and [that] of others)" of the price increases and the "next realistic chance" for increasing prices, before going on to describe the recent price "history" of price increases since 1993, the "price differences between competitors", and the "ratio [of] technical [to] standard boards". Handwritten annotations in French have been added mentioning the "effects of the price situation on a restructuring of the sector" as regards "Gyproc" and "Norgips". The memo ends with the hope that it will enable the addressees to "understand the German market mechanism".

(291) The aspects relating specifically to the October 1998 increase are dealt with in more detail in recital (373). As regards the more general description of "prices in Germany", the memo states: "A. Usual process of price increase action one of the two big players (Knauf / Rigips) announces a price increase; the others will follow a bit later; usually the announcement is for 3 months later. [...] C. My personal understanding (and also of others) A price increase could only work, if there are no projects with extra prices on the market. Each competitor needs to be sure that he won't lose market share and that he is happy with his actual share. As soon as one starts to undermine the action, the prices will be at the old, low level again"(266).

(292) The memo goes on to state that slight increases were made in 1993 and 1994, but were quickly brushed aside. A larger increase that came into force on 1 February 1995 was judged "rather successful", but a second one, which according to Lafarge came into force on 1 January 1996 was "very quickly destroyed"; "After the last successful increase on 1.2.97 a second follows in 1.9.97, but with similar circumstances as today (lots of projects, old prices for key clients...). Since then prices collapsed down to today's level"(267). A graph is attached to the memo showing the pattern of prices invoiced for standard boards on the German market.

(293) To explain the "price differences between competitors", the memo states that "Until 93/94 Lafarge tried to conquer market share and was seen as a foreign importer. Prices differed up to -,40/m². 95/96 Lafarge reached a level of Gyproc and Dano and the prices were approx. -,20/m² lower than Knauf /Rigips. Since 97 all companies are producing in Germany and give rather good service. Price difference since then is only [-,10/ -,20]/m². In 98, we were already able to reach Rigips's level as they were coming down from the Knauf level. Actually one could say : Knauf [100-110], Rigips/Lafarge : [90-95], Gyproc/Dano : [90-95], Norgips: [80-85]" (a handwritten annotation in French specifies "before any discounts to appliers (cf. below)")(268). An explanation follows of the complex pricing system used in the sector and its application by each producer.

(294) The memo goes on to describe market developments in relation to standard and specialist boards, referring to a "very good" development "in the East" of specialist boards penetrated "by all players" "in the boom years (90-97)". The share of specialist boards thus reached 40-45 % of that of standard boards, with a share for the whole of Germany of 35-37% (and [30-35] % for Lafarge in 1997, its best year). However, according to the memo, all competitors are now facing the same problem, namely a decrease in the East. Handwritten comments added to the memo specify the importance of specialist boards: this segment of some 15 million square metres, or 6% of the eastern market, is growing fast, with significant added value and is very concentrated (Knauf [70-80] %). (295) The Commission concludes from the wording of this memo that it does not merely concern the October 1998 increase but that its purpose is to provide Lafarge's new Plaster Division manager with general information on the functioning of the German market on the occasion of price increases.

(296) In the Commission's view, the author of the memo describes a customary procedure for increasing prices, whereby Knauf or BPB announces an increase, which is then followed by the other producers. Mr [X, Lafarge] specifies that the success of this scenario depends on each competitor being "sure that he won't lose market share" and "happy with his ... share". The way in which the producers determined and followed the price increase needs to be established, as does the precise scope of the condition set out by Mr [X] in relation to each producer's being satisfied with its current market share.

(297) The Commission notes that Lafarge, in its reply to the statement of objections, failed to make any comment on the conclusions drawn by the Commission regarding the interpretation of Mr [X]'s remarks, apart from an incidental reference to the October 1998 price increase.

(298) BPB commented on the Lafarge memo, claiming that its use by the Commission revealed the latter's penchant for drawing incorrect inferences from innocent documents (269). It claimed that the memo merely described how price increases worked on a market such as the plasterboard market and that "events in 1998 [...] suggest competition rather than coordination in the market." Furthermore, "the memo's description of the history of price increases in Germany also paints a devastating picture for the Commission's case." However, the Commission notes that BPB fails to comment on the other parts of the Lafarge memo, including those relating directly to its own competitive behaviour on the German market.

(299) Knauf regards the memo as a simple description of how the market operates and, like BPB, does not comment on the other parts of the Lafarge memo, including those relating directly to its own competitive behaviour on the German market (270).

(300) Gyproc, on the other hand, uses the memo to stress that it followed the others when it came to price increases (271).

(301) However, it is apparent that, at least from the end of 1994, competitors informed one another of their intentions as regards price rises, and even coordinated them. In particular, it is apparent that the mechanism described in Mr [X, Lafarge]'s memo, where the increase was launched by Knauf or BPB/Rigips, did not correspond merely to a situation in which competitors, informed by customers seeking to negotiate better prices, followed the two market leaders, but rather to an organised procedure whereby the competitors kept each other informed - a procedure which was not always successful, since each of them was tempted to cheat in order to take advantage of the increase to gain market share.

(302) Concerning the "monitoring of market share" mentioned by Mr [X, Lafarge] at point D of his memo, it should be noted that not only were the shares of the German market monitored as part of the "[Mr U, independent consultant]" statistics, but [high level representatives] of the groups - Messrs [D, BPB], [F, Lafarge], [B, Knauf] and [E, Gyproc] - also exchanged sales figures for the United Kingdom, France, Benelux and Germany for the period ending in September 1998 (272) and had moreover confirmed their market share ambitions for Germany at The Hague in May 1998 (273).

(303) At the start of the Commission's investigation, BPB referred generally to the fact that in Germany, the market participants had a good idea of their competitors' price lists, "usually" from their customers (274). BPB did not subsequently enter into what might be the possible alternative source(s) of information thus suggested.

(304) The Commission notes that, at least on several occasions, Knauf informed not only its direct customers but also its competitors that it was planning to increase prices, and that it also justified this practice several times.

(305) A memo sent on 15 November 1993 by Knauf's [representative], Mr [Z], to Mr [B, Knauf] regarding the prospects for the increase planned for 1 January 1994 (275) suggests that this practice was in conformity with internal instructions at the highest level: "Our new price list was sent to all direct customers at the end of October. At the same time, a copy was sent to all our competitors to inform them".

(306) Likewise, on the occasion of the increase scheduled for 1 October 1998, when Knauf sent customers its announcement of a price increase, i.e. on 21 August 1998, it copied it to the respective [high level representatives] of its competitors BPB/Rigips (to Mr [Mr ZA]'s private address), Lafarge, Gyproc (to Mr [ZB]'s private address) and Norgips(276).

(307) Questioned about this, Knauf stated(277) that "since official announcements of competitors' price changes are often sent anyway by customers to other producers - in connection with a request to continue benefiting from old terms of delivery - the practice has long since become established of sending such communications with price lists, etc. not only to customers but also directly to competitors at the same time. It is not unusual on other homogenous product markets for competitors to inform one another of planned price changes".

(308) Knauf subsequently acknowledged that competitors could have been sent letters announcing price increases "occasionally"(278).

(309) BPB stated, on the subject of sending price lists, that "in Germany Knauf has stated that it was its normal practice to send copies of its price increase notifications to its competitors. This may have been unwise but it is difficult to see that it can have affected matters as the competitors would very shortly, in the ordinary course, have received copies through the normal customer channels. For its part BPB denies that it ever sent copies of its own price lists directly to its competitors, and the Commission has adduced no evidence that it did so". BPB adds, "it might also be noted that the August 1998 notification letter sent by Knauf was sent to Norgips as well as the other producers and a number of customers. If Knauf's intention had been to notify its "co-conspirators" why was Norgips included? The Commission has also drawn attention to the fact that the letters were sent to the private addresses of some company representatives. However, in other cases the letters went to office addresses"(279).

(310) Lafarge did not comment on any sending by a producer of its price increase announcements to its competitors, nor did Gyproc make any comment on this issue.

(311) The Commission thus finds that no evidence, and especially no statement by the parties, allows it to conclude that a producer opposed the receipt of information sent by a competitor - or at least by Knauf, which is known to have sent prior notifications of price increases in at least some cases.

(312) The Commission also notes that during its investigations on the premises of BPB/Rigips and Lafarge Gips it found numerous copies of price increase announcements from the competitors of the companies concerned(280).

(313) The Commission takes note of the explanation given by Lafarge and BPB, who held such letters, that they had been obtained from customers. However, it notes the almost systematic presence of such letters from competitors and the above-mentioned statements by Knauf, which contradict this explanation.

(314) The Commission also finds that the presence of these letters must be analysed in the light of the memo from Mr [X] of Lafarge referred to in recital (291), according to which Knauf and BPB, the two biggest producers, announced price increases first. Where a producer which is the market leader (or one of two) gives competitors prior notification of price increases, as Knauf acknowledged having done and justified, this is likely to change competitive behaviour on the market (see section 2.2).

(315) During its investigation the Commission also found other evidence revealing the existence of collusive contacts between competitors on the occasion of price increases on the German market.

(316) Thus, the above-mentioned memo of 15 November 1993 from Mr [Z, Knauf] to Mr [B, Knauf] regarding the prospects for the increase planned for 1 January 1994 also indicates that, given the low prices still charged on the market for "special projects", it will be difficult to get the increase accepted and gives the following advice: "I would like to answer your question as to how the current situation can be improved as follows: a) authority for prices cannot be allocated to too many people b) [high level representatives] themselves should deal with the matter more actively c) limits should strictly be respected d) poaching in foreign territories with low prices must cease e) reciprocal information must correspond to the facts f) without honest contacts with partner undertakings and possibly competitors deteriorating sales revenues are a certainty. [...] The requirement must be therefore: with each other instead of against each other, selfish thinking is no use to us"(281).

(317) This memo reveals an internal discussion on how to achieve a price increase. Mr [Z, Knauf] recommended that [high level representatives] ([Mr B, Knauf] and [Mr C, Knauf] may have been meant) discuss price increases, that the price war in "foreign" territories cease, and that "honest" contacts with partner undertakings and "possibly" competitors be established. Lastly, in his conclusion, Mr [Z] recommended cooperation in general terms, without saying between whom this is supposed to apply.

(318) Knauf commented on this memo, saying that it was purely internal and that the remarks about respecting limits and avoiding poaching were also internal in scope, since employees competed in their respective zones. However, Knauf did not comment on the passage relating to "honest contacts with partner undertakings and possibly competitors"(282).

(319) The Commission considers first that the question of whether Mr [Z, Knauf]'s recommendations were followed is a quite separate issue from that of their interpretation. In this respect, the Commission finds that, given the comment about "honest contacts with ... competitors" and the general background described in this Decision, Knauf's interpretation of the strictly internal nature of the memo can legitimately be ruled out in favour of a more simple reading of these considerations, namely that Mr [Z] considered a form of coordination between competitors and self-restraint on the part of each firm as regards their competitive behaviour to be necessary.

(320) There was a price increase in February 1995. The Lafarge memo of 7 October 1998 referred to in recital (292) considered this increase "rather successful".

(321) An internal memo of October 1994 discovered at BPB/Rigips on the situation on the German market indicates that all the producers prepared the ground for a price increase to be implemented on 1 February 1995. The increase was to be, for everybody, about DEM 0.35/m² net for standard boards (12.5 mm). The producers cancelled the special discounts and stabilised the prices agreed with large retailers at DEM [2.5-3]/m² (Knauf, BPB/Rigips) and DEM [2.5-3]/m² (Lafarge). Gyproc, however, did not do likewise and kept its contracts at about DEM [2-2.5]/m². The memo indicates that Mr [ZB], Gyproc's [representative], complained that his firm had lost market share and had to win it back. The internal memo stresses that "Gyproc is absolutely the biggest problem at the moment!". It ends in the following terms: "It is expected that the above mentioned prices will be frozen on this level and a price increase will take place from 1st February 1995 with an increase of 0,35 DM/ m² net for all boards. It means for Rigips that the prices for 12,5 mm boards will go up from now [2.5-3] DM/m² (for stocks) till [2.5-3] DM/m²"(283). (322) This price increase resulted in March 1995 in Knauf's net prices reaching about DEM [2.5-3], with Lafarge's slightly lower(284). The initial success of the increase was mitigated, however(285). According to an annotation to a table discovered at BPB/Rigips, which compares the net prices announced after the increase with the situation in July 1995, "Gyproc never reached their estimated level and undercut their prices from day one after the price increase"(286).

(323) BPB subsequently stated that Mr [S, BPB]'s comments in the internal memo of October 1994 discovered at BPB/Rigips reflected his innocent assessment of the state of the German market based on the knowledge he had acquired from information collected by the sales staff at BPB/Rigips. BPB also states that "letters announcing list price increases effective from February 1995 were sent out from November 1994 onwards. It is therefore almost certain that by the time Mr [S] wrote the note in mid-October 1994, the producers (Knauf in particular) already would have been consulting with their customers about their proposals and taking steps to facilitate their successful implementation, such as those described in the note. The Commission has offered no evidence to suggest that these actions were not taken independently by the producers, on the basis of the legitimate market intelligence about each other's intentions which therefore would have been available from customers. In BPB's view, there is also nothing suspicious in Mr [S]'s reference to Gyproc Benelux being, 'absolutely the biggest problem at the moment.' On the contrary, Mr [S] states clearly in his note his understanding that Gyproc's [representative] for Germany, Mr [ZB], was under pressure to regain market share which Gyproc believed that it had lost (something which, in itself, appears to be inconsistent with any suggestion of collusion amongst the producers on price, or market share). According to Mr [S], it is self-evident that Mr [ZB] therefore would have been under particular pressure from within Gyproc Benelux at the time to take pricing actions which would be likely to undermine any attempt by any of the other producers to increase its net net prices, hence Mr [S]'s reference to Gyproc Benelux being 'the biggest problem'"(287). BPB adds that the documents cited by the Commission reflected only the typical operation of the market in the event of price increases and that BPB/Rigips's knowledge of price levels was based on a general awareness of its competitors' discounting practices.

(324) Lafarge stated that this increase "reveals no collusion whatsoever, but on the contrary, the fact that Knauf was unable to anticipate the reaction of its competitors"(288). According to Lafarge, the increase failed and does not reveal any form of coordination. It stresses that, in November 1994, Knauf "seems to have been the first" to announce an actual increase on 1 February 1995 (289). Rigips made its own announcement of an increase on 24 November, for the same date (290). Lafarge sent its announcement on 24 November, with the same date for application as the first two (291). Lafarge states (292) that it was "in its interest to stabilise its prices since the other two producers had already announced an increase. In 1995 Lafarge was still importing its boards, which, given the drop in prices, naturally led it to take advantage of the increase announced by its competitors to raise its prices". Lafarge recognises that prices were effectively increased from 1 February but states that they soon fell again very quickly to reach a level barely above what they had been before the increase. It also points out that from March, Knauf noted that its competitors' prices were between DEM [0-0.5] and [0-0.5] lower than its own prices, which had caused it to lose some valuable large projects293, and that Rigips, in May, also notes that "as yet the price increase has not been fully achieved"(294). Commenting on the Rigips document of July 1995 cited by the Commission, Lafarge then states that this document refers to the price situation, hoping to offset it by increasing volumes, and that the author of the memo expects prices to continue falling as a result of the low prices applied by Gyproc (295).

(325) Lafarge contests (296) the argument that the similar announcements and the cancellation of special discounts by all producers in October 1994 in preparation for the price rise announcements point to the existence of collusive contacts. Thus, in respect of the comment "It is expected that the above mentioned prices will be frozen on this level", it remarks that "the fact that Mr [S, BPB] writes "it is expected" that the various producers will not descend below the levels mentioned in no way indicates collusion but merely demonstrates (i) the fact that Mr [S] was well informed ... and (ii) his personal view of how events would unfold".

(326) The Commission would point out that Lafarge, speaking of the failure of the increase, refers in its reply to the statement of objections to the comments made by Mr [X, Lafarge] in his October 1998 memo. It suffices to note that Mr [X], directly involved in the functioning of the market, regarded this price initiative as a success.

(327) However, the Commission notes that it is by no means out of the question that Mr [S] was "well informed" precisely as a result of contacts between competitors.

(328) The Commission thinks it relevant to analyse the events and documents in the light of Mr [X]'s comment in the 1998 memo on the functioning of the German market (see recital (291)) to the effect that: "a price increase could only work, if there are no projects with extra prices on the market. Each competitor needs to be sure that he won't lose market share and that he is happy with his actual share. As soon as one starts to undermine the action, the prices will be at the old, low level again". The Commission notes that events seem to correspond to the scenario described by Mr [X], the increase having been only relatively successful owing to the conduct of Gyproc.

(329) Moreover, the Commission considers that the content of the BPB/Rigips memo of October 1994 referred to in recital (321) and in particular the use of the words "it is expected that the prices will be frozen on this level", to the extent that it is based on a precise knowledge of competitors' strategies and points to direct contact between BPB/Rigips and Gyproc, reveals the existence of collusive contacts between producers with a view to making a success of the increase. Likewise, the use in the table referred to in recital (322) of the words "Gyproc never reached their estimated level" coupled with the close similarity of the producers' conduct might suggest direct collusion on the price increase (297). However, given that there is no direct evidence of such concerted action, the Commission merely concludes that the competitors informed each other of their intentions.

(330) A price increase was planned for 1 December 1995. It should be pointed out that the Lafarge memo referred to in recital (292), giving it the date of 1 January 1996, indicated that it had been "very quickly destroyed".

(331) The increase was announced to customers in September 1995 (298). The new "gross" price for 12.5 mm standard boards was to be DEM [5-5.5] to [5-5.5] /m², which would give a "net" price of about DEM [3-3.5] to [3-3.5] /m².

(332) The failure of this price rise initiative is confirmed by all the parties. Indeed, in 1996, in Germany, net prices fell, from their lowest level of DEM [2.5-3] / m2 in January to DEM [2-2.5] / m2 in August (299). Moreover, no price rise was implemented during 1996.

(333) It should be pointed out that in June 1996, BPB, Knauf, Lafarge and Gyproc, the plasterboard producers operating on the German market, met at Versailles (see recital (220) et seq.).

(334) However, a price rise was planned for 1 February 1997. It was announced in October and November 1996. The October 1996 BPB/Rigips letter informed customers that general rises in costs made an increase in the price of all types of plasterboard inevitable (300). Distribution of the new price list was scheduled for November. The letter warned customers very clearly that all deliveries after 31 January 1997 would be invoiced at the new prices. The other producers sent out letters drafted in broadly similar terms: the Danogips (301) and Lafarge (302) price lists state that the new "net" price (less the distributor's commission of [...] % and the "loading charge" of DEM [...]) was DEM [3-3.5]/m², while their listed price was DEM [4.5-5]/m², as was that announced by Gyproc (303).

(335) Furthermore, during the investigation the Commission found a memo dated 17 December 1996 and marked "strictly confidential and personal!" from Mr [X, Lafarge] to Mr [G], Lafarge's Plaster Division [high level representative] and other [representatives] of the company. The memo recounts the discussions that took place at the meeting of the German Plasterboard Association in Frankfurt on 16 December 1996. There is no trace of these discussions in the official record of the IGG's meeting. Mr [X]'s memo states: "Besides some Marketing and Recycling topics we were discussing again the current situation on the German market. My impression is as follows : 1. BPB/Rigips is again looking more and more for clients on the market which could help them in their distribution chain. Mr [V, BPB] told us, that they have not the same resources for sales and marketing as Knauf and therefore they need to concentrate on key clients and customers which act in the so called Dealer/Dealer- Business. These customers are selling smaller quantities to smaller dealers (1 - 6 pallets). For this they need a margin of 0,40 to 0,50 DM/m². Therefore the price offered to them will be below the current agreed lowest price level (R: [2.5-3] an(d) L: [2.5-3]). Probably they will sell for [2.5-3] DM/m². Surely these customers will use their low prices for projects, too. This will lead to a destabilization again. 2. Knauf Negotiated already with a lot of key clients and gave them prices for projects until May '97 for a lower than agreed price level. With us they insist on discipline for the price increase. 3. Gyproc although price increase is announced everywhere, they reduced their price in some areas again ([2-2.5] DM/m²) strategy against Knauf in some regions, but also against us 4. Present price level for specialists and A-dealer (projects are lower) Knauf: [2.5-3] DM/m² Rigips: [2.5-3] DM/m² Lafarge: [2.5-3] - [2.5-3] DM/m² Gyproc: [2-2.5] - [2-2.5] DM/m² Dano: [2-2.5] DM/m² Our net/net price is probably the highest today, because Knauf and Rigips have high bonus agreement and work as well with contractor payments. 5. To increase the price to the agreed level ([2.5-3] DM/m²) will be very tough again. We will try our best, but parallel we need to achieve our volume targets" (304).

(336) Lafarge interpreted this memo as follows, saying that "the agreed lowest price level" meant the price agreed at Lafarge by [a high level representative] of the Plaster Division and his subordinates at Lafarge Gips and not a price agreed between the producers, and that when it was said about Knauf "with us they insist on discipline for the price increase" (which contrasts with Knauf's apparent negligence) this meant "just as we [Lafarge] do, they [Knauf] insist on discipline by the customers" and not "as far as we [Lafarge] are concerned, they [Knauf] insist on discipline, etc." (305).

(337) Lafarge subsequently stated (306) that, if Mr [X, Lafarge]'s memo is examined in the context of a price rise that had already been announced and where the price which BPB/Rigips intended to apply was known to Lafarge through information from customers, "it is clear that the Frankfurt meeting of 16 December 1996 was far from being an innocent occasion for producers to collude on their price rise". And "the comments [by Mr [X] in his memo] are not conclusions reached by the participants, but the personal impressions of Lafarge's [high level representative] following an exchange of a few ideas on the market with the other participants in the meeting".

(338) Lafarge also states (307) that "the memo reveals that, on the occasion of this conference, Rigips stated that it did not have the same resources as Knauf, which obliged it to concentrate on its 'key' customers. The memo points out that Rigips will probably offer prices below the level of the increase it had announced.... This statement in no way results from the meeting. It is a reference to Rigips's minimum price as announced by the latter to the market in November and a pure supposition on Lafarge's part that this price will not be maintained". And "the word 'agreed' should not be misinterpreted. As Lafarge explained in its reply to a request for information ... the concept of agreed price relates only to Lafarge. The Commission has refused to take account of the document added to the file by Lafarge to show that this was the price level agreed between Lafarge Plâtres and its subsidiary, below which the latter should in principle not descend. However, Lafarge firmly reasserts this explanation, especially because, since the request for information, it has conducted more thorough research which fully confirms it. Lafarge is able to demonstrate that DEM [2.5-3] was precisely the 1997 budget price agreed on by Lafarge Plâtres and its subsidiary. In companies other than Lafarge, this concept of a budget price below which sales staff were not supposed to descend can also be found (Cf. especially Gyproc, p. 5552)".

(339) Lastly, according to Lafarge, the "fact that Knauf insisted to the other producers that it wanted the price rise to work ('with us they insist on discipline for the price increase') adds nothing. This is simply a unilateral statement, to which the other participants did not subscribe. On the contrary, Lafarge's [high level representative] states further on in his memo that it will probably be difficult to maintain the budget price" (308).

(340) Knauf claimed that Mr [X, Lafarge]'s memo was internal to Lafarge in scope, except perhaps as regards the description of discussions between BPB/Rigips and Lafarge (309). Knauf also contests that an agreement between producers on the German market, such as that suggested by the expression "agreed price level" in the memo, was concluded with its participation (310). Knauf refrained from commenting on the sentence in the memo "With us they insist on discipline for the price increase", and affirmed that the memo did not reveal any form of coordination between producers. Mr [X] made no statement to Knauf's Mr [W] during this IGG meeting, it said.

(341) BPB initially denied that Mr [V] of Rigips discussed prices with Mr [X, Lafarge] at the IGG meeting on 16 December 1996, stating that "since the note was written in English, not Mr [X]'s native language, he may have said 'agreed' when he meant 'notified'"(311).

(342) BPB subsequently added (312) the following comments: "BPB has noted that there are at least five other memos written by Lafarge [high level representatives] (one by [Mr ZC] to [Mr ZD] dated 12 January 1996, and four by [Mr X] to [Mr G] and others, dated 5 January, 3 March, 2 April and 5 October 1998) reporting the prices of each competitor to different kinds of dealer. ... The first of these memos states that 'Our sales team gives daily information about the behaviour of our competitors concerning prices. Enclosed please find a survey of the present situation up to today'. The prices given in each memo are precisely stated in deutschmarks and pfennigs, often in a range. The level of detail is similar to that set out in the [Mr X] memo of 17 December 1996 and suggests that pricing information of this level of detail was routinely collected by Lafarge's sales force, and would have been available in the marketplace, as suggested by BPB in its statement of 28 May 1999. The source of Lafarge's information is made even clearer by the chart, evidently compiled by Mr [X], that has been disclosed by the Commission at page 6859 of the file. The chart lists prices given by Lafarge's competitors to particular named customers and seems to have been compiled either for, or from information received at, a Lafarge internal sales meeting. It would appear to set out the raw material for the monthly reports compiled by Mr [X] and sent to Lafarge management".

(343) As regards the explanations given by Lafarge, the Commission finds that its explanation about the internal nature of the discussions referred to in Mr [X, Lafarge]'s memo overlooks the fact that the memo also refers to Knauf and BPB/Rigips.

(344) The Commission takes note of Lafarge's comments to the effect that the meeting "was far from being an innocent occasion for producers to collude on their price rise" and "the comments [by Mr [X, Lafarge] in his memo] are not conclusions reached by the participants, but the personal impressions of Lafarge's [high level representative] following an exchange of a few ideas on the market with the other participants in the meeting". In the Commission's view, this statement by Lafarge confirms the very words of Mr [X]'s memo, according to which competitors "were discussing again the current situation on the German market". The memo, which the Commission has never claimed to be a summary of the conclusions of the meeting between competitors, together with Lafarge's statement, indisputably establish the existence of direct contacts between competitors in the context of which they conveyed analyses and, as the rest of the memo shows, intentions. The above sentence also shows that other similar "discussions" had taken place previously.

(345) The Commission also considers, given the wording of the memo and the general background described in this Decision, that Lafarge's interpretation of the concept of "agreed price level" can legitimately be ruled out in favour of a simpler reading of its content. In this context, the "agreed lowest price level" or "agreed price level" of DEM [2.5-3] for BPB/Rigips and DEM [2.5-3] for Lafarge must be the net price invoiced which the producers had agreed among themselves. It will also be noted that this interpretation of the Commission's is consistent with the information contained in another Lafarge memo313 (see recital 293), according to which there was generally a price differential of DEM [0-0.5] /m² between Knauf and BPB/Rigips on the one hand and Lafarge on the other.

(346) The Commission considers that Mr [X, Lafarge]'s memo referred to in recital (335), against the background of a price increase announced by all producers for 1 February 1997, reflects his concern about the conduct of competitors and the pricing policies, especially discounts, that they applied. Thus the new prices that BPB intended to offer its "dealer/dealer" customers were DEM 0.20 lower than the agreed minimum price of DEM [2.5-3]. The distributors would be able to use their low invoiced prices (DEM [2.5-3], less other deductions) not only to give a margin of DEM 0.40 to 0.50 on the sale to the smallest distributors, but also to offer the product to construction undertakings for their special projects. For special projects up to May 1997, Knauf had in advance negotiated lower prices than the level agreed (DEM [2.5-3]), while urging Lafarge to stick to the lowest price agreed (in its case, DEM [2.5-3]). Gyproc had announced the increase in prices to its customers but was undermining it by proposing to keep the old prices (up to DEM [2-2.5]) in certain sectors. Given the low level of the net prices invoiced - DEM [2-2.5] - still being offered to distributors in December, it was going to be difficult for Lafarge to raise its price to DEM [2.5-3]; Lafarge Gips was going to do its best to achieve this price target while being careful not to lose market share through too rigid an attitude towards the increase.

(347) As regards the explanations given by BPB, the Commission notes that the fact that Lafarge obtained market information from its employees and commercial directors, and that the knowledge thus obtained might explain the level of detail of the information in the Lafarge memos cited by BPB, in no way invalidates the conclusions that the Commission may draw from its reading of Mr [X, Lafarge]'s memo of 17 December 1996. This memo refers explicitly to discussions between competitors. The Commission could even theoretically draw the opposite conclusion from that of BPB as regards the Lafarge memos cited by BPB: these other memos could have been enriched by information collected directly from competitors. The Commission maintains, on the basis of this memo dating from the same time as the events, and despite BPB's denial, that Mr [X] and Mr [V, Rigips/BPB] had a discussion on the situation of the market and prices in particular.

(348) In relation to Knauf's statements, the Commission notes that the company has failed to give an explanation of the phrases "agreed price level" and "with us they insist on discipline for the price increase" other than a general denial that any discussion took place between Knauf and Lafarge or that they participated in any agreement or coordination.

(349) The Commission concludes from Mr [X, Lafarge]'s memo that, having publicly announced the increase, the producers examined, at the meeting of the German Plasterboard Association in Frankfurt on 16 December 1996, how to apply it successfully.

(350) The February 1997 price rise was termed a "success" by Mr [X, Lafarge] in October 1998 (see recital (292)) but a "qualified success of short duration" and even a failure by Lafarge subsequently (314). This assessment is contradicted by the graph attached to Mr [X]'s memo of October 1998 (315) referred to in recital (292), which shows a clear increase in prices, which were not maintained at their highest level but which, despite a gradual falling off, none the less remained at a level consistently higher than that which prevailed at the end of 1996.

(351) The reasons for the drop in prices were assessed by Lafarge (316) or BPB/Rigips (317) as arising from the conduct of Gyproc, which retained, compared with the other producers, a price differential, not revealed to them, of up to DEM 0.30/m² for standard 12.5 mm boards. Gyproc was charging its customers the new prices, but granting them monthly credits and free products, etc. Lafarge noted that Gyproc "gained probably [...] also a bit of market share because they never executed the price increase of February as strict as Knauf and we did".

(352) The Commission concludes from all the documents and statements by the parties, as well as from the close similarity of the producers' conducts, that, on the occasion of the February 1997 price rise, the competitors colluded directly on the price rise and, at the very least, informed one another of their intentions in anticipation of the rise.

(353) The producers fixed the next price rise for September 1997. The letters announcing the September increase were sent out as early as May (318). After 1 September 1997, the new net price was officially fixed at DEM [3-3.5] / m² for Knauf and DEM [3-3.5] / m² for Lafarge, Gyproc and Danogips, but lower prices were indicated for certain customers and special projects (319).

(354) However, as stressed by Mr [X, Lafarge] in his October 1998 memo, the success of the price increases meant that each operator had to be satisfied with its market share. The price increase of September 1997 was the occasion, according to the evidence at the Commission's disposal, of various initiatives illustrating the link between stability of market shares, monitoring of customers and/or distributors, and price increase initiatives.

(355) One example is the contacts designed to avoid any of the big producers "poaching" the major customers of the others, to use the term from Knauf's November 1993 memo (see recital (316)).

(356) An internal Knauf memo320 dated 1 July 1997 shows that Mr [W, Knauf] had a few days previously (321) met with Mr [X, Lafarge Gips]. At this meeting, Mr [X] examined with Knauf, in order to get the latter's go-ahead, a list of customers which it proposed to canvass. Knauf stated, in response to the Commission's request, that the list attached to the Knauf memo on that matter no longer existed (322). However, according to the memo, Mr [X] claimed that the customers concerned were former regular customers of Lafarge's which Knauf had taken from it, and that the reason for its canvassing was that it was trying to increase its market share in southern Germany from 1 to 2%.

(357) After wondering whether Lafarge customers really were concerned, Mr [W, Knauf] concludes that Lafarge's canvassing was a hostile gesture.

(358) Knauf did not at first provide an explanation of the memo (323). Subsequently, it stated that it was "clearly competitive" behaviour illustrating nothing other than the occasional tendency of French firms to show arrogance in their relationships with their competitors: "With such an announcement, Knauf could decide either to enter a price war over the customers, or to 'give in' in the name of business sense. Lafarge clearly thought the announcement was a shortcut ... Obviously, Knauf defended its customers" (324).

(359) Lafarge stated that it "is correct that Lafarge had decided to do what was necessary to recover former customers that Knauf had managed to entice away ... Lafarge wanted to increase its market share in southern Germany ... However, at no time did Lafarge seek to 'negotiate' the return of these customers with Knauf. Lafarge's commercial director merely announced to Knauf his decision to recover key former customers. This action merely showed Knauf that Lafarge had decided not to give in and would firmly defend its position. Any new attack on customers would produce the same reaction. However, the action did not affect Knauf, which confined itself to noting that contacts had been taken up with the largest customers to make them new offers" (325).

(360) Whatever the actual reactions of the two companies following this meeting between Mr [W, Knauf] and Mr [X, Lafarge] - reactions for which there is no evidence other than the statements of the two firms concerned - the Commission is obliged to find that Knauf and Lafarge discussed their key customers in southern Germany. Even supposing that the two firms confined themselves to citing their intentions vis-à-vis their customers, this fact is sufficient to establish a form of coordination between the two competitors.

(361) Another illustration of the link between stability of market shares, monitoring of customers and/or distributors, and price increase initiatives is that, according to the documentary evidence from Lafarge (326), on 9 September 1997, the Hamburg branch of the builders' merchants [ZE] advertised a special offer on 12.5 mm boards at DEM [2.5-3] /m², starting from 1 October 1997. This would have threatened the price increase. Knauf pointed it out on 24 September 1997 to Lafarge, [ZE]'s supplier. On the same day, Mr [X] of Lafarge assured Mr [W, Knauf] in a fax message that Lafarge was as fed up as Knauf: "As discussed at the IGG meeting, this act of [ZE]'s is an individual action that has not been discussed with headquarters. The branch manager must in the meantime fear for his job. According to our information the action has since been cancelled and is not being carried out".(327)

(362) Lafarge stressed that "the reality is quite different [from that described by the Commission]. [ZE]'s special offer in Hamburg was far from being an isolated act but was actually part of a systematic policy at Lafarge. From the beginning of September, it was clear for competitors that Lafarge was not applying the increase it had announced, as illustrated by the internal Rigips memos of 10 September 1997 (p. 2282) and 12 September 1997 (p. 2275). What is more, no measure was taken against [ZE], and its purchases from Lafarge for 1997 increased by [2400-2500] % compared with the previous year (p.5603). Even if Knauf did question Lafarge about [ZE] at an IGG meeting, this proves that Lafarge was following an independent policy and that it did not change its behaviour in any way"(328).

(363) Knauf stated that the contacts over [ZE, merchants] did not constitute evidence of coordination. In Knauf's opinion, this distributor received very favourable conditions from Lafarge. With its special offer, [ZE] threatened to destroy the entire price increase. It is in this context that "Mr [W] from Knauf raised the matter directly with his counterpart from Lafarge. However, it was clearly not a targeted action by Lafarge, as can be seen from its reply. The file shows that Rigips was also following this case closely, see for example page 6.2275, and was also supplying/had also supplied [ZE]. An arrangement, as the Commission suggests, was not made. Lafarge did ensure by virtue of its own business decision that [ZE]'s offer was not implemented. This can be explained by the fact that its own price increase would have been counteracted by [ZE]'s measure" (329).

(364) Whatever the subsequent denials of the two firms, the Commission concludes from this exchange of memos that Lafarge and Knauf were in contact, and did not hesitate to take concerted action over a distributor, at the request of a competitor, in order to ensure compliance with a price increase jointly determined by them.

(365) The Commission points out that an IGG meeting once again provided an opportunity for contact between competitors where they raised issues relating to their competitive situation.

(366) The Commission considers that these two exchanges between Knauf and Lafarge confirm that there was concerted action on prices and considerable monitoring of the prices charged by the respective distributors, in particular at the IGG meetings.

(367) The 1 September 1997 price increase was not a success, as noted by Mr [X, Lafarge] in his October 1998 memo (see recital (292)). Each producer blamed the others and their commercially aggressive behaviour. In November 1997, the "net net" price in Germany had fallen to DEM [2-2.5] /m². (368) In an internal BPB/Rigips memo of September 1997 Lafarge was accused of being "the most aggressive" competitor (330). BPB/Rigips, in another internal memo of 8 November 1997, states "since October 1997 the competition situation in Germany regarding the plasterboard sector has clearly aggravated. At the beginning of the business year Gyproc and Lafarge have tried to gain additional marketshare and have therefore behaved extremely price aggressive. We reacted flexible to these attacks and were able to defend our marketshare. Obviously Knauf was not successful and therefore tries since October to regain lost marketshare. Since this time Knauf decreases sales price in a wide range and tries now at all costs to regain lost marketshare. All attempts on our behalf to stabilize the price have failed so far. The price increase in September has failed and in a view of the present situation we believe, that the further price increase planned for March cannot be realized" (331). In another internal memo dated 10 November 1997, [a high level representative] of BPB likewise stated that Knauf "have been making increasing noise about being unhappy about their losses in market share" and feared a new price war would be triggered, given each producer's determination to keep its market share (332).

(369) In an internal memo from January 1998 providing a report to [a high level representative] of Lafarge's Plaster Division, Mr [X] of Lafarge Gips explains the smaller than expected rise in sales notably by the fact that since October 1997 "Rigips and Knauf came into direct competition - Knauf wants to take back some market share from Rigips - the prices drove from approx. DM [2.5-3] per m² in September down to [2.5-3] per m² right now". Regarding prices, "right now there is a very unstable situation and as long as Knauf and Rigips are not able to fix their own position, we'll have a price pressure here in Germany". However, Mr [X] does not expect a further drop of more than DEM 0.10 to 0.20 per m²: "our hopes are based on a not too hard winter and a stop of the enemy activities between Knauf and Rigips on the market" (333).

(370) According to an internal memo by Knauf's Mr [W] in October 1997, Lafarge insisted that Knauf continue to apply the DEM [3-3.5] price level and reproached Knauf with introducing a retail price of DEM [2.5-3] onto the market (334).

(371) Other evidence confirms that Knauf, during the same period, seems to have been particularly sensitive about the reduction in its market share, which considerably helped to make it more difficult to maintain or even raise prices. Thus, according to annotations in Mr [E, Gyproc]'s diary following a conversation with a Lafarge [representative] (see recital (244)), Knauf was at the time making it known to its competitors that it wanted to increase its share of the German market by 1%, and that it expected to see Gyproc make room for it. It should be pointed out more generally that in 1997, the four producers on the German market were actively discussing, without reaching a specific agreement, their respective market shares (see section 1.4.5.2). This situation may have helped to prevent the price increase being successful.

(372) In May 1998, BPB/Rigips decided to announce a price rise of 4-7%, which was to take effect on 21 September 1998, according to minutes of a management meeting at BPB (335). The standard letter announcing the increase was sent to customers at the beginning of June, with an increase announced for the 12.5 mm boards of net 8.5% (336). The new BPB/Rigips price list was published in July, with the proposed net price at a "nominal" DEM [2.5-3] / m² (337).

(373) In the memo of 7 October 1998, Mr [X] of Lafarge also mentions this increase, dating it in October, and states that "Rigips announced approx. 8 weeks ago, but neither Knauf nor any others follow immediately; furthermore Rigips started to grant old prices to big customers and for bigger projects (> 10 000 m²) until the end of the year for approx. [2-2.5] / m². Only beginning of Sept. Knauf announced an increase, too; but they granted also special prices to key accounts. Lafarge announced end of Sept. for 12.10.98. Gyproc and Dano haven't written anything to the customers" (338). (374) In fact, Knauf warned its customers by letter dated 21 August 1998 that an unspecified increase would take effect on 1 October (339). Lafarge's letter - phrased in similar terms to Knauf's - was sent on 26 August, i.e. a few days later, with an effective date of 5 October 1998 (340). Danogips and Gyproc, too, sent out their letters at the end of August, announcing an increase from 1 October (341).

(375) When Knauf sent its customers its announcement of a price increase, i.e. on 21 August 1998, it copied it to the respective managing or commercial directors of its competitors BPB/Rigips (to Mr [ZA]'s private address), Lafarge, Gyproc (to Mr [ZB]'s private address) and Norgips (342).

(376) On 26 August 1998, Mr [W] from Knauf observed in a departmental memo that he had "also" obtained confirmation over the phone from Mr [ZB] of Gyproc that the letter from his company to customers had been posted the day before: he notes that "the circle is thus now closed" (343) .

(377) On 23 October 1998, Mr [W], Knauf's [representative], faxed Mr [X] of Lafarge at his private address a copy of the instructions dated 22 October which he ([Mr W]) had given to Knauf's sales force so as to impose the increase: no special "actions", no discounts on special projects after 31 December; spread the rumour on the market that prices were going to go up again by DEM 0.50/m² in the first quarter of 1999 (344).

(378) The departmental memo addressed by Mr [X] of Lafarge to the sales force on 21 October 1998 (345) is virtually identical to the instructions given by his counterparts. The handwritten comment "personal-confidential" on one of the two copies that the Commission found at Lafarge Gips could mean that it was faxed outside Lafarge as part of the same series of contacts between producers, since there is no reason for a document circulated internally to a number of addressees to receive such a classification.

(379) On 1 November 1998, Mr [ZB], of Gyproc, sent Mr [X, Lafarge] a copy of his instructions of 25 October along these lines to his company's sales force, likewise to his private address (346).

(380) According to a memo (347) reporting on the BPB/Rigips board meeting of 13 October 1998, BPB/Rigips "has been executing the announced price increase since October 1st 1998 with all determination". It was said that Knauf had tried to impose the desired prices, but with a number of exceptions. Lafarge had informed its customers that it was increasing its prices by DEM 0.15/m², but only from 12 October (and not 1 October, as previously announced) (348). Gyproc was not showing the slightest sign of increasing its prices. The memo adds "It remains our clear target, that Rigips will try with all determination to push through this price increase. If we succeed and when we will have reached a certain stability we should try to execute another price increase during the first three months of 1999".

(381) Another increase of DEM 0.50/m² planned for early 1999 was mentioned. Gyproc's and Lafarge's instructions mention 1 February 1999. This was in the general region of the increase that Mr [X, Lafarge] had outlined in his memo of 7 October 1998 (see recital (291), where he mentioned the date of 1 March 1999: "D. The next realistic chance: Only if point C is OK, we could try it for 1.3.99. Announcement of all players needs to be on the market before Christmas. If 3 need to do it, I would really try an 25-30% up to approx [2.5-3] / m² invoiced (no credit notes !). If good monitoring of our market share. But: E. How will Norgips react? Dano? Gyproc? Will the others accept a 3-4% market share of Norgips? Who gives away how much to N?"(349).

(382) Lafarge stresses the failure of the October 1998 rise, which was no more successful in breaking the sharp fall in prices that had been continuing since September 1997, "and this despite Knauf's desperate attempt to 'pass on the message' to its competitors that it would comply with the increase". "The gap between the dates shows that no collusion took place" (350). According to Lafarge, which refers to its memo (351) of 21 October 1998, the failure can be explained by the absence of any real increase by Gyproc and Danogips and only a partial increase on the part of Rigips and Knauf.

(383) Lafarge also asserts that "Knauf's sending its announcement of a price increase to the other producers was no more than a vain and desperate attempt". "The fact that Knauf sent its letter to the private addresses of other producers, as well as major sellers, does not imply reciprocity on the part of the other producers. The fact that a company makes a unilateral announcement of its next rise is no proof of collusion with its addressee. This unilateral initiative of Knauf's could have only very limited consequences anyway, since on the occasion of each of the previous increases, this information had reached the other producers without Knauf's intervention. The list of addressees includes big merchants such as [...], who are also customers of Lafarge (p. 6946) and would not have failed to inform Lafarge of Knauf's price rise. Lastly, Knauf sent the announcement to the other producers only on 26 although it was sent to customers on 21. It is therefore quite possible that when the producers received Knauf's announcement, they had already been informed of it by their customers" (352).

(384) Lafarge also stresses (353) that "clearly no 'confidence' existed", referring to the concern expressed at the Rigips monthly meeting (354) that their competitors would not follow the rise and to the internal Lafarge memo (355). Moreover, "the fact that Mr [W] (Knauf) obtained confirmation from Mr [ZB] (Gyproc) that it had just announced its rise and thus considered the circle to be closed is not a proof of confidence, but simply of the fact that there had been contact between the two firms and that it seemed on the basis of his information that everyone had announced the increase". And "Gyproc's conduct could not in reality give Knauf grounds for confidence: Gyproc is very probably the producer which consistently followed the upward movements the least".

(385) Lafarge ends, "In this context, the letters sent by Gyproc and Knauf to Lafarge stating that instructions had been given to their sales forces with a view to implementing the rise successfully must be viewed as 'intoxication' designed to deceive competitors a little more, as demonstrated by the Rigips study of the German market. The producer notes that the price of boards fell in the second half of the year by 12%, which means not only that sending the instructions had no effect but also that Gyproc and Knauf appear themselves not to have followed them (p. 4234). The letters sent by Lafarge (p. 6951), Gyproc (p. 6957) and Knauf (p. 6955) to their respective customers stating that they were stopping discounts until the end of the year again proves only one thing: the producers concerned followed the leader of this initiative after being informed of it by their customers. Any hope of seeing prices rise was an opportunity not to be missed." "However, these measures had no effect since Lafarge's prices kept falling to reach DEM [2-2.5] in March 1999" (356).

(386) BPB, for its part, states "the timetable of events ... clearly demonstrates that if there was any collusion between producers - which BPB certainly has no knowledge of - it could not have included BPB. The chronology shows that BPB had sent its price increase letter and published its new prices long before the other companies had notified their customers of their intentions. Any 'confidence building' exercise could not logically have included BPB, who had acted independently when it sent its price increase letters out in June 1998. If there was any 'circle' to be closed, which is for the Commission to show, it could not have included BPB. Against that background it is clear that the quotes that the Commission extracted from the Rigips board meeting of 13 October 1998 are entirely innocent. Not surprisingly, at that meeting the board discussed what Rigips' competitors were doing in the marketplace" (357).

(387) Knauf states that the facts described by the Commission point only to individual decisions by players in the market, who follow or do not follow price initiatives by other players in the market. It confirms that Mr [W, Knauf] sent the instructions for its sales staff to Lafarge in October 1998. However, Knauf says that this was an isolated case which differed from all the other attempts to increase the price and was a personal initiative by Mr [W], unknown to Messrs [B and C, Knauf], and not part of a process of coordination (358).

(388) Gyproc did not contest the Commission's presentation in its statement of objections of the facts relating to the September/October 1998 increase.

(389) As regards the attempted increase of September/October 1998, the Commission notes that the BPB/Rigips initiative of May 1998 was not initially successful with its competitors. However, Knauf finally showed that it was willing to implement the increase and sent its price increase lists directly to its competitors BPB, Lafarge and Gyproc - as well as to Norgips, which had just entered the market and whose presence had a destabilising effect acknowledged by all the producers. Knauf also sent Lafarge its internal instructions to its sales staff. Gyproc sent Lafarge and Knauf its internal instructions to its sales staff.

(390) The Commission therefore finds, on the basis of the evidence at its disposal, that Knauf and Gyproc took active steps to inform their competitors, and that BPB/Rigips and Lafarge received this information. The latter two companies did not claim not to have received the information or to have refused it. Moreover, the fact that the Commission, during its investigations, discovered the documents concerned kept in the offices of the [representatives] of the two companies generally involved in the exchanges suffices to establish that the receipt of the information was not accidental, but forms part of a whole picture of exchange and collusion. Against the general background described in this Decision, the exchange is an additional manifestation of the collusion between BPB, Knauf, Lafarge and Gyproc on the German market.

(391) The fact that this information was sent to Norgips in no way invalidates the Commission's conclusions on the existence of a form of coordination between the four producers already established on the German market: the fact that the information was sent must be assessed in context (see section 2.2). Furthermore, it is not impossible that the four producers might have planned to involve Norgips in their arrangements if the Commission's investigations had not taken place shortly after that undertaking's entering the market.

(392) The exchange between producers of evidence that they were sticking to their course continued after the increase came into force. This exchange appears to have been intended to boost the competitors' confidence: before committing itself, each producer probably wanted to be sure that the others had already moved publicly on prices.

(393) The Commission notes that Gyproc, in its reply to the statement of objections, does not contest the Commission's description of the facts, but confines itself in general to stressing the very limited nature of its participation in the "concerted attempts to raise prices on the German market" (359). The Commission considers that this admission confirms its conclusions.

(394) The Commission concludes from all this information that, on the German market, the four producers, from the February 1995 price increase until the Commission's investigations, informed each other, at least on certain occasions, of their intentions concerning the date and level of the planned increases. They also on occasion contacted one another with a view to ensuring that the coordinated increase was successful. The fact that the information was exchanged between subsidiaries does not change the fact that the same companies as those in contact at European level are concerned. Moreover, none of the four firms argued that the legal persons were different to contest the relevance of the Commission's presentation in its statement of objections.

(395) The Commission would also point out that, as indicated by BPB (360) and Lafarge (361), prices on the German market tended to rise or at least to stabilise over the period concerned, in a clear break with the pattern between 1988 and 1992.

1.5. Economic evidence to demonstrate the absence of an agreement

(396) The participants in the infringement endeavour to demonstrate, in particular through economic analyses showing the functioning of the market between 1992 and 1998, that it is impossible for there to have been a restrictive agreement on the plasterboard market during this period. They conclude from the nature of the market and from the fact that vigorous price and market-share competition was allegedly maintained on the relevant market during this period that there was no collusion between them during the disputed period (362).

(397) The arguments advanced by the parties are very similar, namely that the volatility of "net/net prices", changes in market shares, significant movements in customer levels and efforts by producers to increase efficiency and reduce cost structure demonstrate the existence of healthy competition which leaves no room for a restrictive agreement on fixing prices or sharing out the market (363). The Commission would none the less point out that the descriptions of the structure of the product market by Lafarge and Knauf differ significantly on certain points (364).

(398) The Commission cannot accept these arguments, since they are not relevant to a demonstration that no restrictive agreement existed on the plasterboard market between 1992 and 1998 between BPB, Knauf, Lafarge and later Gyproc.

(399) First, it must be noted that the Commission does not base its argument merely on the similarity of conduct, nor does it use economic evidence to establish the infringement of Article 81(1) of the Treaty but, rather, it draws in particular on direct evidence of the existence of the anticompetitive agreement. The economic analyses supplied by the parties do not shed a different light on the various pieces of direct documentary evidence cited by the Commission to establish the infringement of Article 81(1) of the Treaty.

(400) Moreover, the Commission is not claiming that the existence of an agreement should have had the effect of eliminating all competition from the plasterboard market. The existence of a degree of competition is therefore far from incompatible with the conclusion that collusion took place. The fact that competition existed and that the price rise targets were not always and not fully met does not mean that the participants did not contribute to the implementation of agreements concluded within the cartel or that an infringement of Article 81 cannot be established. Even the members of a cartel cannot act completely freely of the market situation, which can also depend on conditions outside their full control. None the less, by agreeing market share quotas and target prices, the members of a cartel have more room for manœuvre on the market to exploit their customers through higher prices. What is more, the Court of First Instance has already found, in its judgements on the cartonboard cartel, that a company which does not behave on the market in the way agreed with its competitors may simply be trying to exploit the cartel for its own benefit (365).

(401) It must be stressed that since the infringement established by the Commission has an anticompetitive object (366), it is settled case law that there is no need to take account of the concrete effects of an agreement (367). Moreover, where a company which participates with others in meetings at which decisions are taken on prices does not observe the agreed prices, this does not invalidate the anticompetitive object of those meetings and thus the fact that the company participated in the agreement; at most, it would demonstrate that it did not implement the agreements in question (368).

(402) Moreover, it must be stressed that the fact that the parties acknowledge a change between the market situation before 1992 and afterwards establishes the existence of a change on the market (369).

2. PART II - LEGAL ASSESSMENT

2.1. Application of the competition rules

1.1.1. Value of statements made as evidence of the infringement of Article 81(1) of the Treaty

(403) Lafarge challenges the Commission's use, with a view to establishing the existence of certain facts, of statements made by BPB and Gyproc in the context of the Commission Notice on the non-imposition or reduction of fines in cartel cases (370) (hereinafter the "Leniency Notice"), on the ground that they are biased and unsupported by evidence. It takes the judgment given by the Court of First Instance in the Cement cases as its basis (371).

(404) The Commission points out first of all that the paragraphs of that judgment to which Lafarge refers, namely paragraphs 1837 and 1838, concern a case in which the Commission proceeds on the basis of a single document. In the present case, Lafarge criticises the Commission for taking statements, and not documents, as a basis.

(405) As regards the evidential value of the statements at issue, the Commission points out firstly that, in order to establish the existence of the infringement to which this Decision relates, it did not in general base itself on those statements alone. On the contrary, the Commission has in its possession many direct pieces of evidence establishing the facts, and it uses the parties' statements to supplement, clarify or highlight certain facts where it considers it necessary.

(406) It should be pointed out secondly that the evidence must be assessed in its entirety, taking into account all relevant circumstances of fact (372). In the light of the other evidence available, the statements made by BPB and Gyproc, which they made without having knowledge of all the evidence the Commission had in its possession, are perfectly credible and consistent as far as most of the description of the facts is concerned. In this connection the Commission would stress that the disputed statements by BPB and Gyproc are contradicted neither by the facts nor, in the main, by any statements that may have been made by other parties concerned by BPB's and Gyproc's disputed statements.

(407) As regards the veracity and authenticity of the evidence furnished and statements made by Gyproc or BPB in the context of the Leniency Notice, the Commission would point out that the Community courts have not once questioned the legality of the leniency policy and have even, in the exercise of their power of unlimited jurisdiction, reduced fines pursuant to the notice (373).

(408) The Commission would point out further that in the present case the disputed statements by BPB and Gyproc are an acknowledgement of their involvement in an infringement and that they therefore constitute an admission by those undertakings that their conduct was unlawful. Moreover, neither BPB nor Gyproc attempts, in these statements, to exculpate itself by trying to shift responsibility for the infringement on to another undertaking. In these circumstances the Commission concludes that, inasmuch as they are basically contrary to the interests of BPB and Gyproc, the disputed statements are decidedly credible. The Commission accordingly has no reason to believe that they contain exaggerations or inventions, far from it.

(409) It therefore follows from all these considerations that the Commission may refer to the disputed statements by BPB and Gyproc as it does in this Decision.

1.1.2. Agreements and concerted practices

(410) Article 81(1) of the Treaty prohibits as incompatible with the common market all agreements between undertakings, decisions by associations of undertakings and 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, markets, or share markets or sources of supply.

(411) The prohibition of these forms of coordination and collaboration must be understood in the light of the concept inherent in the provisions of the Treaty relating to competition, and in particular Article 81(1), that each economic operator must determine independently the policy which he intends to adopt on the common market. This requirement of independence strictly precludes any direct or indirect contact between such operators, the object or effect whereof is to influence their conduct on the market and to create conditions of competition which do not correspond to the normal conditions of the market in question (374).

(412) Such coordination or collaboration may take any of the forms mentioned in Article 81(1) of the Treaty, including that of an agreement or a concerted practice.

(413) In PVC II375 the Court of First Instance held that "It is well established in the case-law that for there to be an agreement within the meaning of Article 81(1) of the Treaty it is sufficient for the undertakings to have expressed their joint intention to behave on the market in a certain way". The agreement does not have to be in writing, there is no need for formalities and contractual penalties or enforcement measures do not have to be provided for. The existence of the agreement may be explicit or implied from the parties' conduct.

(414) The concept of a "concerted practice" covers a form of coordination between undertakings which, without having reached the stage where an agreement properly so-called has been concluded, knowingly substitutes practical cooperation between them for the risks of competition (376). Such conduct may be caught by Article 81(1) of the Treaty as a "concerted practice" even where the parties have not explicitly expressed their joint intention to conduct themselves on the market in a specific way but knowingly adopt or accept collusive arrangements that facilitate the coordination of their business conduct (377).

(415) Although, under the terms of Article 81(1) of the Treaty the concept of a concerted practice implies not only undertakings' concerting with each other, but also subsequent conduct on the market, and a relationship of cause and effect between the two, the presumption must be, subject to proof to the contrary, that the undertakings taking part in the concerted action and remaining active on the market take account of the information exchanged with their competitors for the purposes of determining their conduct on that market. That is all the more true where the undertakings concert together on a regular basis over a long period (378).

(416) Moreover, a concerted practice is caught by Article 81(1) of the Treaty even in the absence of anticompetitive effects on the market. First, it follows from the actual text of that provision that, as in the case of agreements between undertakings and decisions by associations of undertakings, concerted practices are prohibited, regardless of their effect, when they have an anticompetitive object. Next, although the very concept of a concerted practice presupposes conduct by the participating undertakings on the market, it does not necessarily mean that that conduct should produce the specific effect of restricting, preventing or distorting competition (379).

1.1.3. Concept of complex, continuous agreement

(417) The term "agreement" can be applied both to a general project or conditions expressly agreed on and to the performance of what was agreed on the basis of the same mechanisms and in pursuance of the same common objective. The fact that in law there is an infringement of Article 81(1) of the Treaty as soon as the illegal agreement is concluded does not mean that the "agreement" cannot be continuous; as long as it is performed, it continues to exist until it is terminated.

(418) Such an agreement, where it is characterised as complex, may therefore be regarded as a single continuous infringement throughout its existence. The possibility that one or more components of a series of actions or a continuous line of conduct may, individually and in themselves, constitute an infringement of Article 81(1) of the Treaty does not prevent the conclusion from being drawn that these components may constitute a complex agreement.

(419) As the Court of Justice has held, upholding the judgment given by the Court of First Instance, it follows from the formal terms of Article 81(1) of the Treaty that an agreement may comprise not only an isolated act but also a series of acts or continuous conduct (380).

(420) In the present case, the Commission establishes, as will be seen in recitals (428) et seq, the existence of such a complex agreement.

1.1.4. Concept of a single infringement

(421) A single infringement is a set of individual infringements interlinked by an objective identity (all the component features serve the same purpose) and a subjective identity (same parties, aware of participating in or supporting a common objective) (381).

(422) It is not necessary, particularly in the case of a single infringement of lengthy duration, for the Commission to describe the infringement as being exclusively one or other form of illegal conduct. The concepts of agreement and concerted practice evolve and sometimes overlap. Moreover, it will sometimes be impossible to make the distinction on a realistic basis as an infringement can have the characteristics of each of the forms of illegal conduct all at once whereas certain of its manifestations taken in isolation could be attached with precision to one of them rather than to another. But in analytical terms it would be artificial to subdivide into several discontinuous forms of infringement what is clearly a common continuous project pursuing one and the same overall objective. A restrictive practice can accordingly be at once an agreement and a concerted practice. Article 81 of the Treaty lays down no specific category for a complex infringement of this type (382).

(423) In PVC II the Court of First Instance held that, in the context of a single 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(1) of the Treaty (383).

(424) The reason why Article 81(1) draws a distinction between the concept of concerted practice and that of agreement between undertakings or of decisions by associations of undertakings is to bring within the prohibition of that Article different forms of coordination and collusion between undertakings. A comparison between the concepts of agreement and concerted practice within the meaning of that Article shows that, from the subjective point of view, they are intended to catch forms of collusion having the same nature and are only distinguishable from each other by their intensity and the forms in which they manifest themselves. It follows that, whilst the concepts have partially different elements, they are not mutually incompatible. The Commission is therefore not required to categorise either as an agreement or as a concerted practice each form of conduct found but may characterise some of those forms of conduct as principally agreements and others as concerted practices (384).

(425) It is established case law that a single infringement may be characterised as a complex infringement where it involves at one and the same time elements to be characterised as "agreements" and elements to be characterised as "concerted practices" and that, when faced with an infringement of this type, the Commission is entitled to characterise it as "an agreement and a concerted practice", this dual characterisation having to be understood, not as requiring, simultaneously and cumulatively, proof that each of those factual elements presented the constituent elements both of an agreement and of a concerted practice, but rather as referring to a complex whole comprising a number of factual elements some of which were characterised as agreements and others as concerted practices for the purposes of Article 81(1) of the Treaty (385).

(426) Finally, it should be noted that, although a restrictive agreement may be a common project, each party to the agreement may have his specific role to play. It may be that one or more of the participants play a lead role. There may be internal conflicts and rivalries, and there may even be cheating, but that does not prevent the understanding from constituting an agreement or a concerted practice for the purposes of Article 81(1) of the Treaty, if there is a single continuous common objective.

(427) The mere fact that each party to the agreement may play a role of his own does not mean that he bears no responsibility for the infringement as a whole, including acts done by other parties but sharing the same illicit purpose and the same anticompetitive effect. An undertaking that has taken part in such an infringement through conduct that has contributed to the attainment of the common objective is also responsible, throughout the entire period of its participation in the common project, for the conduct of other undertakings in the commission of the same infringement. That is at all events the case where it is established that the undertaking in question was aware of the unlawful conduct of the other participants or that it could reasonably have foreseen it or been aware of it and that it was prepared to take the risk (386).

2.2. Nature of the infringement in the present case

(428) In the present case, the Commission finds that between 1992 and 1998 representatives of BPB, Knauf, Lafarge and Gyproc contacted each other on a number of occasions, concluded agreements and took part in concerted practices restrictive of competition, affecting their conduct on the plasterboard market at least in Germany, France, Benelux and the United Kingdom, such conduct being the manifestation of a complex, continuous agreement within the meaning of Article 81(1) of the Treaty.

(429) The following should first of all be noted: - representatives of BPB and Knauf met in London in 1992 and expressed the common wish to stabilise the German, French, UK and Benelux plasterboard markets; - representatives of BPB and Knauf set up, starting in 1992, information exchange systems of which Lafarge and then Gyproc became members, covering their sales volumes on the German, French, UK and Benelux plasterboard markets; - representatives of BPB, Knauf and Lafarge informed each other in advance on a number of occasions of price increases on the UK market; - following certain developments on the German market, representatives of BPB, Knauf, Lafarge and Gyproc met in Versailles in 1996, Brussels in 1997 and The Hague in 1998 with a view to sharing or at least stabilising the German market; - representatives of BPB, Knauf, Lafarge and Gyproc informed each other on a number of occasions and engaged in concerted action on the implementation of price increases on the German market between 1996 and 1998.

(430) These various elements combined indicate that the competing undertakings were motivated without interruption from 1992 until 1998 by a common wish to restrict competition on the plasterboard market at least in the four major European markets. In other words, the parties sought to reduce competition between the main operators on those markets, the various types of conduct noted by the Commission being manifestations of this common wish, which, owing to its objective, constitutes an infringement of Article 81(1) of the Treaty.

(431) The Commission considers that between 1992 and 1998 BPB, Knauf, Lafarge and, to a lesser extent (387), Gyproc took part, through their involvement in various anticompetitive acts, whether those acts were committed at the level of the four major European markets or at a more local level and whether they took the form of agreements or concerted practices or both, in the commission of a single, complex and continuous infringement of Article 81(1) of the Treaty on the plasterboard market.

(432) These various manifestations are, moreover, clearly complementary in the light of the functioning of the plasterboard market. The improvement of the economic situation of the undertakings through an increase in prices rendered necessary a coordination of those undertakings at the level of market shares. As the competing undertakings have acknowledged on numerous occasions, for a price increase to be successful, the competitors had to be satisfied with the market share they held. A situation had to be avoided in which the price increases might be used by one of the competing undertakings to gain market share. The various elements of the single infringement are therefore perfectly complementary, this complementarity also being evidence of the identity of object of the various manifestations of this infringement.

(433) The Commission would refer here to the judgment of the Court of First Instance in Rhône Poulenc (388). In that case, in which the various concerted practices observed and the various agreements concluded formed part, in view of their identical purpose, of a series of efforts made by the undertakings in question in pursuit of a single economic aim, namely to distort the normal movement of prices on the market in polypropylene, the Court stated that it would be artificial to split up such continuous conduct, characterised by a single purpose, by treating it as consisting of a number of separate infringements. The fact was that the parties had taken part - over a period of years - in an integrated set of schemes constituting a single infringement, which progressively manifested itself in both unlawful agreements and unlawful concerted practices.

(434) In view of what has just been indicated, the Commission considers that there existed between 1992 and 1998 at least in the four major European plasterboard markets a complex, continuous agreement having as its object the restriction of competition between the addressees of this Decision, namely BPB, Knauf, Lafarge and Gyproc.

(435) The Commission would also point out here that it cannot be ruled out that other manifestations of this complex agreement which existed at the material time may have occurred. It does not, however, have sufficient evidence, if any, to establish the existence of these facts.

(436) The various manifestations of the complex, continuous agreement which have been observed by the Commission are examined in recitals (437) et seq.

(437) When the representatives of BPB and Knauf decided to meet in London in 1992, the situation on the European plasterboard market was so serious that they exchanged views on the subject. Neither the existence of the meeting nor the purport of the remarks exchanged are really contested by the two undertakings. BPB states that, at that meeting, BPB and Knauf "reached an understanding that it was in the interests of BPB, Knauf and the industry as a whole (including, ultimately, the interests of consumers) for the ruinous price war to end and for producers to attempt to compete at more sustainable economic levels" (389). BPB goes on to state that Mr [A, BPB] and the [Mr B and Mr C, Knauf] expressed similar opinions regarding the need for producers to adjust their conflicting strategic goals in relation to market share, but that Mr [A] declared that he did not agree with the [Mr B and Mr C] at that meeting, either expressly or impliedly, that BPB and Knauf would collectively actually attempt to end the price war, increase prices or fix market shares. Knauf, for its part, states that there were calls to commercial reason: "That competition between the parties could not continue in the long term under these ruinous conditions, that consequently the price war had to end, was at that time so clear to everyone that a stated common understanding on this point ("reached an understanding") is simply soapbox oratory" (390). It must be stressed that BPB does not deny that this meeting, among other things, may have been a contributory factor in accelerating the end of the price war on the market (391).

(438) Although the parties deny that a formal agreement was reached in London (392), it is nevertheless certain that the two competing undertakings met there and that they exchanged views on the competitive situation on the plasterboard market as just described. Moreover, it is clear from their statements that there was a consensus of views on the need to adapt their behaviour on the market. The objective was to put an end to a situation in which the strategies pursued, being wholly antagonistic, had resulted in a far-reaching shake-up in the structure of the market and a price war to gain market share. Following the London meeting, the matching strategies pursued by the two undertakings were no longer to enter new markets, gain market shares from competitors or indeed to eliminate competitors. What they wanted was to end the price war and ensure the maintenance of the status quo on the market.

(439) The Commission accordingly considers that it is clear from statements by BPB and Knauf that at the 1992 London meeting BPB and Knauf informed each other that they shared the same opinion regarding the situation on the market and that they told one another that they deemed it necessary to adapt producers' behaviour with a view to adjusting their goals. Although the parties deny having gone as far as reaching an agreement proper, BPB and Knauf had, in the course of that meeting, direct contacts which manifestly had as their object or effect an influencing of their conduct on the market. BPB has, moreover, expressly acknowledged that the contacts between competitors taking place at the meeting were prohibited and that these contacts may be characterised as concerted practices (393).

(440) It is established case law that, for there to be an agreement for the purposes of Article 81(1) of the Treaty, it is sufficient that the undertakings concerned expressed their common wish to behave on the market in a specific manner. Such is the case where there exists between several undertakings a gentlemen's agreement representing the faithful expression of a common wish involving a restriction of competition. In these circumstances, there is no need to examine whether the undertakings considered themselves bound - legally, factually or morally - to adopt the behaviour agreed between them (394).

(441) The Commission considers therefore that these contacts must be characterised as an agreement for the purposes of Article 81(1) of the Treaty, the elements constituting an agreement being present in accordance with the case law on the subject. The fact that the agreement was more of an agreement in principle and that it did not settle every conceivable detail is irrelevant. As is clear from the statements by BPB and Knauf, these London contacts are in reality the first practical manifestation, albeit still vaguely formulated, of the complex, continuous agreement having as its object the restriction of competition on the plasterboard market at least in the four major European markets.

(442) It is also clear from statements by BPB and Knauf that, either at the London meeting (395) or shortly afterwards (396), they agreed to exchange information on their sales volumes on the four major European markets and that this exchange of data continued until 1998. The Commission has a considerable amount of direct evidence of the exchanges in its possession, and in any case the parties do not deny that they took place. BPB, Knauf and then Lafarge exchanged individualised data during telephone conversations or in letters sent to [high level representatives]' private addresses, with increasing frequency (at first annually, then half-yearly, and, finally, quarterly). It should also be noted that cheating seems to have taken place in the context of this information exchange.

(443) According to BPB, the purpose of the information exchange with Knauf was initially to provide a basis on which to assess whether there was a "new mood in the industry". BPB states, moreover, that "such an information exchange, at a high level, would provide the degree of mutual assurance that the price war was ending" (397). It maintains, however, that, while the "[Mr A, BPB] exchanges" were linked to the London talks and the end of the price war, the "[Mr D, BPB] exchanges" were different in character and served a different purpose, namely a simple information exchange for no other purpose than mere market knowledge. As the Commission has already indicated (398), it cannot accept this reinterpretation of the facts. It should be noted, moreover, that BPB is the only party to invoke the change in the purpose of the exchange.

(444) Various factors support the Commission's position that BPB and Knauf concluded an agreement setting up an anticompetitive system for the exchange of information in order to monitor the market and thus to prevent any competition considered too aggressive by the parties on the four major European markets.

(445) First of all, the Commission would point out that the exchange was set up between the two competitors, namely BPB and Knauf, who had expressed at the London meeting the common wish to put an end to the price war between them. Next, it would point to the existence of the two statements by BPB concerning the initial purpose of the information exchange. Moreover, it is clear from statements by Knauf that this exchange was intended to establish market transparency, in a direct form by an exchange of individualised figures, and that it was an internal information instrument for assessing the volume of the market in a fairly reliable manner, enabling several parameters to be planned (including raw material stocks) (399). Furthermore, there is evidence that the competitors exchanged individualised data for almost seven years, first of all annually, then, in 1993, half-yearly, and, finally, between 1995 and 1998, quarterly, in a multilateral manner and by telephone or by letter sent to the private addresses of the persons concerned.

(446) The fact that the data exchanged were "historical" does not suffice to exclude the possibility that this data exchange system may have permitted the existence of mutual monitoring by the parties to the exchange. In the Beams case, the Court of First Instance found that the mutual monitoring operated by reference to past figures, with the parties tending towards the maintenance of "traditional flows" of trade. The exchange thus tended to partition markets by reference to those traditional flows (400). In fixing the fine, the Court considered that an exchange of information on past quantities was similar to a sharing of markets (401). The only plausible explanation for an information exchange which, as in the present case, is meant to be secret and which is based on figures of no major value when it comes to defining a strategy for the future is that there is, between the parties, a tacit agreement to respect traditional flows.

(447) The argument of some of the parties that the system could not serve as a monitoring system because the data exchanged were not actually used is not pertinent (402). The Commission would point out here that Knauf has admitted using the data in its planning and that it therefore used the exchanged data actively.

(448) Moreover, as the Commission has just shown, the purpose of the system was to monitor the market, which means that for the system to achieve this object knowledge of competitors' market shares was sufficient. Each undertaking knew that it was being kept under close surveillance by its competitors and that it could, if necessary, react to the conduct of its competitors in order to ensure compliance with the common wish to restrict competition. It follows therefore that the information which the undertakings received under the exchange system was capable of appreciably influencing their conduct (403). It should be pointed out in this connection that, in view of the highly concentrated structure of the market and its transparency, a commitment in relation to volumes was tantamount to renouncing aggressive competition. The fact of having to notify any penetration would have elicited an immediate, targeted response from competitors, which would have thwarted the attempt.

(449) The Commission would point out that the case law has confirmed the particularly harmful character of any information exchange in an oligopolistic market (404). An information exchange constitutes per se an infringement of Article 81(1) of the Treaty if the requirement of independence according to which each trader must determine independently his conduct on the market is undermined as a result. This requirement of independence will without a doubt be affected if the exchange takes place in a highly concentrated market and if it reduces the risk of uncertainty for the trader.

(450) These two conditions are manifestly met in the present case inasmuch as the market is an oligopolistic one405 and the uncertainty has disappeared since the parties agreed in London to put an end to the aggression reigning on the market, this wish being manifested, moreover, on a number of occasions on the relevant markets. The Commission considers therefore that the system constitutes an infringement of Article 81(1) of the Treaty.

(451) Lastly, the fact that the parties at certain times knowingly exchanged incorrect information confirms that this exchange was actually used as a monitoring mechanism (406) and does not affect the anticompetitive nature of the exchange. In this connection, it must be pointed out that according to established case law non-observance by an undertaking of commitments entered into does not change the fact that the conduct was anticompetitive and that, therefore, the undertaking concerned participated in the restrictive practice (407).

(452) This body of evidence permits the Commission to assert that the London agreement concerning the exchange of information which began in 1992 and continued until 1998 was intended to enable participants therein to monitor the conduct of their competitors at least on the relevant markets and constitutes a manifestation of the common wish of the parties to restrict competition on the plasterboard market in the four major European markets. In so doing, the Commission characterises the agreement as restrictive of competition within the meaning of Article 81(1) of the Treaty, being a particular manifestation of the complex, continuous agreement having as its object the restriction of competition on the plasterboard market at least in the four major European markets.

(453) If this system were to be characterised as a concerted practice, the Commission is of the opinion that, in view of the factors outlined, it must be considered that this practice was intended to enable participants therein to monitor the conduct of their competitors at least in the four major markets concerned.

(454) It is established case law, moreover, that, subject to proof to the contrary, which it is for the economic operators concerned to adduce, there must be a presumption that the undertakings participating in concerting arrangements and remaining active on the market take account of the information exchanged with their competitors when determining their conduct on that market. Furthermore, it is clear from the case law that a concerted practice falls under Article 81(1) of the Treaty even in the absence of anticompetitive effects on the market. First, it follows from the actual text of that provision that, as in the case of agreements between undertakings and decisions by associations of undertakings, concerted practices are prohibited, regardless of their effect, when they have an anticompetitive object. Next, although the concept of a concerted practice presupposes conduct of the participating undertakings on the market, it does not necessarily imply that that conduct should produce the concrete effect of restricting, preventing or distorting competition (408).

(455) As regards Lafarge's participation in this agreement, the Commission notes the existence of various factors which enable it to establish that Lafarge knew or should have known that, by taking part in the exchange, it was not only involved in an exchange whose object was the exchange of reliable individualised data but that such an exchange served as a market monitoring mechanism (409).

(456) In those circumstances, the Commission concludes that Lafarge subscribed in mid- 1992 to the London agreement on the exchange of information between [high level representatives] on the four major markets. Lafarge's argument that it could not have subscribed to such an agreement as the agreement's existence has not been established is not pertinent, the Commission having duly established its existence.

(457) As regards Gyproc's participation in the agreement on the exchange of information initiated in London in 1992 by BPB and Knauf, to which Lafarge subscribed in mid-1992, the Commission notes the existence of various factors enabling it to establish that Gyproc knew or at least should have known that the anticompetitive conduct in which it was taking part extended beyond the German market (410).

(458) In those circumstances, the Commission considers that Gyproc subscribed to the agreement setting up the system for the exchange of information on sales volumes on the four major European markets as from the Versailles meeting in June 1996.

(459) As to the exchange of information on volumes and market shares in relation specifically to the UK market, the Commission notes that starting in 1992 BPB, Knauf and Lafarge set up an information exchange system of seven years' duration, the systematic and detailed nature of which is apparent from the documents contained in the file (411). It is clear from the factual circumstances, namely the fact that the instructions to set up this exchange system were given soon after the London meeting, that the undertakings involved are the same as those involved in the exchange initiated in London, that cheating took place and that this information exchange system had an identical object to that initiated in London, namely market monitoring. There is no reason to suppose, moreover, that the exchange on the UK market, which was similar to and more regular even than the exchange that took place at the level of the four European markets, had an object or an effect different from the latter.

(460) In those circumstances, the assertion that the exchanged data were not used is not a pertinent factor for the purpose of establishing that the exchange could not have performed a monitoring role, the reasons being the same as those already set out above (412).

(461) The Commission concludes therefore that this exchange must be characterised as a concerted practice for the purposes of Article 81(1) of the Treaty in that it is made up of contacts between competitors likely to influence their conduct on the market and that it constitutes a particular manifestation of the complex, continuous agreement having as its object the restriction of competition on the plasterboard market at least in the four major European markets.

(462) As to the meeting that took place between BPB, Knauf, Lafarge and Gyproc in Versailles in 1996 and the meetings in Brussels in 1997 and The Hague in 1998, these formed part of the general drive to stabilise markets and in so doing to restrict competition, and they constitute a more regional manifestation linked to the situation on the German market at the time. At these meetings, the competing undertakings exchanged data on their market shares with a view to reaching agreement on the sharing of the German market (413). It is clear from the facts that one of the specific features of the German market in 1996 was the construction of new plants by all producers and the growth of imports from eastern Europe, which were capable of causing a major upset on the German market and sparking a new price war like the one in 1992 (414). Another specific feature of this market was the fact that BPB and Knauf resented the behaviour of Gyproc, which held a significant market share on the German market, viewing it as particularly aggressive (415).

(463) At the Versailles meeting, BPB, Knauf and Lafarge expressed the wish to retain the market shares they held on the German market and they tried to conclude an agreement to that effect with Gyproc, which rejected the market share it was offered while accepting the principle of a sharing of the market. At that meeting, the competitors exchanged information on their situation on the German market, as they did subsequently at the meetings in Brussels and The Hague devoted to the purchase of Opole.

(464) As regards its participation in the discussions that took place on the occasion of the Versailles meeting, Knauf maintains that it was surprised at the content of these discussions and in particular at the fact that they concerned the sharing of the German market between the competitors. In this respect, the Commission would point out that it is established case law that, where an undertaking participates in meetings between undertakings with an anticompetitive object and does not publicly distance itself from what occurred at them, thus giving the impression to the other participants that it subscribes to the results of the meetings and will act in conformity with them, it may be concluded that it is participating in the cartel resulting from those meetings (416).

(465) Even if the discussions which took place in Versailles and as part of the talks on the purchase of the Opole factory did not lead to the conclusion of an agreement, the Commission nevertheless observes that BPB, Knauf, Lafarge and Gyproc consulted one another on several occasions on the sharing of the German market with a view to sharing it among them or at least to stabilising it, an objective which they pursued by and large on the four major European markets. It is clear from the factual circumstances and from the contacts already existing at that time between the parties that, even if they did not succeed in concluding a specific agreement on sharing the German market between them, they again expressed on that occasion their common wish to conduct themselves on that market in a specific manner, in other words to restrict competition. The existence of this common wish is again confirmed, if that were necessary, by the terms of a statement by Mr [E, Gyproc], according to which "the agreement wasn't working well" and "the competitors met in Brussels to discuss again the desired stabilisation of the German market" (417).

(466) BPB, Knauf and Lafarge informed each other at the Versailles meeting what market shares they would agree to, these being the shares the undertakings actually held. Such a reciprocal declaration of intentions is manifestly anticompetitive as it is contrary to the requirement of independence of economic operators, which 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. The mere disclosure by an undertaking of the fact that it does not want a larger market share than the one it already holds is sufficient to inform competitors of an essential element of its strategy.

(467) Contrary to what some of the parties maintain, the facts to which the Commission refers cannot be characterised as a mere attempt to reach agreement as a result of which the parties cannot be accused of behaving anticompetitively. In the present case, through the various meetings at which the participants were present, the competing undertakings established a continuous process of commercial "diplomacy" aimed at reconciling their respective interests, notably on the German market. The question of market shares on the German market became a constant subject of discussion and it is clear that the undertakings did not dare aim at a larger market share without discussing it with their competitors (418).

(468) The Commission would refer in this connection to the judgment of the Court of First Instance in the HFB case. In that judgment, the Court begins by stating that the facts invoked cannot be characterised as mere attempts to reach agreement. It then states that it is clear from the series of meetings at which the parties discussed the division of market shares that, at least at a certain moment, the undertakings expressed their common wish to conduct themselves on the market in a specific way and that it must therefore be found that, even if there was no agreement on all the elements forming the subject-matter of the negotiations, a common wish to restrict competition determined the course of the negotiations (419).

(469) The Commission concludes therefore that, even if BPB, Knauf, Lafarge and Gyproc did not actually conclude an agreement on the division of market shares in Germany, they again (or for the first time in the case of Gyproc) expressed their common wish to restrict competition on the plasterboard market. The Commission accordingly considers that an agreement was concluded between the parties, who aimed to divide up the German market between them or at least to stabilise it, this agreement being a particular manifestation of the complex, continuous agreement having as its object the restriction of competition on the plasterboard market at least in the four major European markets.

(470) As regards the exchange of sales volumes data which took place on the German market from the Versailles meeting onwards and the "[Mr U, independent consultant] system" set up by the parties in October/November 1996, the Commission would point out that these are also a manifestation of the parties' common wish to restrict competition, this particular manifestation being justified by specific developments on the German market. It is inconceivable, in view of all the manifestations by the parties of their common wish to stabilise the European, and in particular the German, market - this wish having been restated at the Versailles meeting and in the course of the discussions on Opole - that this information exchange did not, in one way or another, serve their purposes. The existence of an apparent link between the information exchange and the parties' wish to stabilise the market has, moreover, been confirmed (420). The Commission concludes therefore that a concerted practice existed, constituting a particular manifestation of the complex, continuous agreement having as its object the restriction of competition on the plasterboard market at least in the four major European markets.

(471) The Commission also notes the existence of an exchange of data on price increases on the German market between BPB, Knauf, Lafarge and Gyproc. It is clear from the facts that the four producers, from the February 1995 price increase until the Commission's investigations, informed each other, at least on certain occasions, of their intentions concerning the date and level of the planned increases. This state of affairs must also be viewed in the context of a period which, need it be repeated, was characterised by a series of anticompetitive manifestations of a common wish of the competitors to stabilise the plasterboard market in the four major European markets.

(472) Some of the parties have acknowledged having, at least on certain occasions, informed each other of their intentions concerning the dates and levels of price increases. However, even if the price increase (or the price itself) was not always decided in advance, and even if the informing was not always entirely mutual (that is to say, all the competitors did not always notify the others directly of their price increases), the fact that the leaders on the relevant market informed their competitors of their intentions, and this on various occasions at least, is not indicative of purely unilateral conduct. The case law states that the condition of reciprocity of a concerted practice is met where one competitor discloses its future intentions or conduct on the market to another when the latter requests it or, at the very least, accepts it (421). The Commission concludes from this that the continuous transmission of information by the price leader, without any counter-reaction by its competitors, fulfils the condition of reciprocity.

(473) It should also be pointed out here that the Court of First Instance held in a case in which there was direct contact between competitors that an undertaking which merely received information concerning the future conduct of its market competitors had not only pursued the aim of eliminating in advance uncertainty about the future conduct of its competitors but could not have failed to take into account, directly or indirectly, the information obtained in order to determine the policy which it intended to pursue on the market (422).

(474) Regarding the fact that the price information transmitted was known by the customers of the undertaking concerned before it was notified to competitors and that, therefore, the information revealed could already have been gathered by the latter on the market, the Court held that this argument had no relevance, for the following reasons : first, even if the price leader did first notify its customers, individually and on a regular basis, of the prices which it intended to charge, that fact does not imply that, at that time, those prices constituted objective market data that were readily accessible. Second, participants became aware of that information more simply, rapidly and directly than they would via the market. Third, the systematic participation of undertakings in the meetings in question allowed them to create a climate of mutual certainty as to their future pricing policies (423).

(475) The Commission accordingly concludes that in the present case the system for the exchange, between BPB, Knauf, Lafarge and Gyproc, of data on price rises on the German market constitutes a concerted practice forming part of the general framework of the parties' common wish to stabilise the market and constituting a particular manifestation of the complex, continuous agreement having as its object the restriction of competition on the plasterboard market at least in the four major European markets.

(476) As regards the exchange of data on price increases on the UK market, the Commission concludes that contacts took place, as BPB, Knauf and Lafarge acknowledge, which accompanied certain price increases. It notes further that, as on the German market, these increases were coordinated with respect to their practical arrangements (424). The Commission accordingly analyses these exchanges on the basis of the principles already set out above.

(477) In view of the simultaneous existence on the UK market of various anticompetitive manifestations between BPB, Knauf and Lafarge, namely the exchange of information at [high level representatives] level, the exchange organised at subsidiary level and the exchange of data on price increases, the Commission concludes that this exchange of data cannot be examined outside the existing context. In this connection, the existence of concerted action on price increases on the German market during the same period confirms that the UK exchange was not an isolated act but that it formed part of a necessarily broader strategy. In those circumstances, the Commission considers that it constitutes a concerted practice forming part of the particular manifestations of the complex, continuous agreement having as its object the restriction of competition on the plasterboard market at least in the four major European markets.

2.3. Restriction of competition

(478) Article 81(1) of the Treaty specifically describes as anticompetitive agreements which: - directly or indirectly fix purchase or selling prices or any other trading conditions; - limit or control production, markets or technical development; or - share markets or sources of supply.

(479) As the Commission has stated above (425), the set of agreements and concerted practices in the present case form part of a series of actions by the undertakings in question pursuing a single economic aim, namely the restriction of competition, and constitute the various manifestations of a complex, continuous agreement the object and effect of which was to restrict competition. In view of the fact that the above-mentioned agreements and concerted practices gave, without interruption from 1992 until 1998, substantive shape to the parties' common wish to stabilise, and hence restrict competition on, at least the German, French, UK and Benelux plasterboard markets, the Commission characterises the infringement as single, complex and continuous.

(480) Given the manifestly anticompetitive object of the single, complex and continuous infringement, there is no need to demonstrate a negative effect on competition.

(481) The existence of the anticompetitive object of the infringement enables the Commission to affirm that the intent and motives of each undertaking organising or taking part in the information exchanges are factors without relevance. The Commission would refer here to the case law, which states that, in order to determine whether an agreement has as its object the restriction of competition, it is not necessary to inquire which of the two contracting parties took the initiative in inserting any particular clause or to verify that the parties had a common intent at the time when the agreement was concluded. It is rather a question of examining the aims pursued by the agreement as such, in the light of the economic context in which the agreement is to be applied (426).

2.4. Effect on trade between Member States

(482) Article 81(1) of the Treaty applies to agreements that are liable to impede the attainment of a single market between the Member States either by fragmenting national markets or by affecting the structure of competition in the common market.

(483) The Court of Justice has held that "in order that an agreement between undertakings may 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 it may have an influence, direct or indirect, actual or potential, on the pattern of trade between Member States" (427). In any event, Article 81(1) of the Treaty does not require these agreements to have actually affected trade between Member States but rather that they should be capable of having that effect (428).

(484) There is a substantial volume of trade between Member States on the plasterboard market. It is significant that producers buy supplies to cover their requirements without regard for national barriers while endeavouring to control or prevent parallel dealings in the product. However, the application of Article 81(1) of the Treaty to restrictive agreements is not confined to the part of the participants' sales that involves transfers of goods from one Member State to another. For that provision to apply, there is no need to show that the individual conduct of each participant in relation to the agreement as a whole, has a substantial effect on trade between Member States (429).

(485) In the present case, the illegal agreements covered trade on the four major markets in the Community in this important industry. The existence of collusive arrangements inevitably had the actual or potential effect of automatically deflecting trade flows from their normal pattern (430).

2.5. Duration of the single infringement

(486) The collusion began at the beginning of 1992 with the London agreement between Knauf and BPB, extended to include Lafarge in mid-1992. Considered as a single, complex and continuous infringement, the infringement committed by these undertakings continued at least until the time of the investigations by the Commission at the end of 1998.

(487) For the purpose of calculating the duration of the infringement, although the relevant meeting probably took place earlier, the Commission will proceed on the basis of the last day of the first quarter of 1992. BPB said that it was possible that the two undertakings had not agreed to set up the exchange of information at the initial meeting but "at some point later in 1992, probably around the second quarter of 1992". The Commission concludes that the initial meeting took place earlier, in the first quarter and therefore no later than 31 March 1992, and it accordingly selects this date of 31 March 1992 as the date of the beginning of the infringement.

(488) As regards Lafarge, the Commission selects as the date of the beginning of its participation in the infringement the middle of 1992, i.e. 31 August 1992 at the latest (431).

(489) As regards Gyproc, the available evidence does not support a finding that it participated actively in the collusion begun in 1992 by Knauf and BPB before the Versailles meeting on 5 and 6 June 1996 (432). However, from that date onwards, Gyproc knew or should have known that BPB, Knauf and Lafarge were engaged in an anticompetitive relationship the object of which extended beyond the German market (433).

(490) In this connection, it is established case law that an undertaking may also be held responsible for an overall cartel even though it is shown that it participated directly only in one or some of the constituent elements of that cartel, if it is shown that it knew, or must have known, that the collusion in which it participated was part of an overall plan and that the overall plan included all the constituent elements of the cartel (434). Similarly, an undertaking that took part in a single, complex infringement through conduct of its own which was intended to help bring about the infringement as a whole may also be responsible, throughout the entire period of its participation in that infringement, for conduct put into effect by other undertakings in the context of the same infringement where it is established that the undertaking in question was aware of the offending conduct of the other participants or that it could reasonably have foreseen it and that it was prepared to take the risk. 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 (435).

(491) The Commission accordingly considers Gyproc to have been responsible for the commission of the single, complex and continuous infringement as from 6 June 1996.

2.6. Responsibility for the infringement

2.6.1. General considerations

(492) The subject of the Treaty competition rules is the "undertaking", which is not necessarily synonymous with the forms of legal personality familiar to company law or national tax law. The term "undertaking" is not defined in the Treaty but has been extensively analysed in the case law (436). It can refer to any entity which engages in business activity. In the case of a major multinational, the myriad subsidiaries, the complex network of shareholders and capital structure and the organisation, for management purposes, of the group's activities into separate operational or functional divisions or geographical areas not necessarily corresponding to the undertaking's structure can create complications. Depending on the circumstances, it may be appropriate to treat the whole group, individual sub-groups or subsidiaries as addressees for the purposes of Article 81 of the Treaty.

2.6.2. BPB

(493) The Decision is addressed to BPB, directly involved in the infringement and never having denied its direct involvement or pleaded structural arguments prompting the Commission to modify its analysis in that respect.

(494) In the case of BPB, given the active personal involvement of its top-level executives in the infringement, the Decision must undeniably be addressed to the company. The fact that the product covered by the agreement is only one of the items in the group's range is not material, especially as plasterboard is one of BPB's main products.

2.6.3. Knauf Westdeutsche Gipswerke

(495) It is established that Knauf actively participated in all the anticompetitive conduct described in this Decision and that [Knauf group's high level representatives], [Mr B and Mr C], were personally involved in that conduct.

(496) The Decision is addressed to Knauf Westdeutsche Gipswerke in view of the particular structure of the Knauf group. The Commission is unable to identify one individual at the head of the group of companies constituting the undertaking. Consequently, there is not one legal entity which, at its head, could, as the body charged with coordinating the group's activities, be held responsible for the infringements committed by the various companies composing it.

(497) Knauf Westdeutsche Gipswerke, of which [Mr B] and [Mr C] are [high level representatives], is the most representative company in that undertaking. In particular, as regards Gebrüder Knauf Verwaltungsgesellschaft KG, the function of which is to administer other companies of the Knauf group, it must be noted that it is dependent, at least partly, on Knauf Westdeutsche Gipswerke both for its premises and for its staff.

(498) In those circumstances, and in order that matters of pure form do not stand in the way of a finding of conduct on the plasterboard market by Knauf for purposes of the application of the competition rules, the Commission considers that Knauf Westdeutsche Gipswerke must be held responsible for all the actions of Knauf. Knauf Westdeutsche Gipswerke has not objected to the fact that the Commission sent it the statement of objections, despite the fact that this made clear that the Commission intended to hold it responsible for all of Knauf's conduct (437).

(499) The Commission considers (438) that, with a view to the possible imposition of a fine (see section 3), the turnover to be taken into account for the purposes of this Decision is that of the "undertaking" within the meaning of Article 81(1) of the Treaty, that is to say in this case the worldwide turnover achieved by all the companies of the Knauf group as communicated (439) by Knauf to the Commission.

(500) The fact that the product covered by the agreement is only one of the items in the group's range is not material, especially as plasterboard is one of Knauf's main products.

2.6.4. Lafarge

(501) The Decision is addressed to Lafarge. The plaster business, including plasterboard, is one of the five branches of Lafarge and is incorporated as Lafarge Plâtres International SA (440). Regarding its involvement, it must be concluded that Lafarge Plâtres International did not act independently of Lafarge.

(502) In this connection, the Commission would refer to the judgment of the Court of Justice in the Stora case ("Cartonboard"), according to which, as the subsidiary in question was wholly owned, the Commission could legitimately assume that the parent company in fact exercised decisive influence over its subsidiary's conduct, and that in those circumstances it was for the appellant to reverse that presumption by adducing sufficient evidence (441).

(503) In the present case, it is incontestable that Lafarge Plâtres International SA is 99.99% controlled by Lafarge and that its accounts are consolidated with those of Lafarge.

(504) The Commission notes, moreover, the existence of other factors strengthening the presumption referred to in recital (502). First of all, it is clear from the commercial documentation that Lafarge Plâtres International SA is presented as the plaster branch of Lafarge (442). Secondly, in a statement to the Commission following the hearing, Mr [G], who was [high level representative] of Lafarge Plâtres International until [...], describes himself as "[a high level representative] of the Lafarge group's plaster business" from 1992 to 1998, the period covered by this proceeding (443). Thirdly, it is clear from the facts that Mr [F], Mr [G]'s successor as [high level representative] of Lafarge Plâtres International since [...], is [high level representative] of Lafarge and took part in the exchanges of information at dispute with the competing undertakings (444). The fact that he performed two functions for only a limited period, that he did not take part in collusive meetings and that he put an end to this exchange following the Commission's investigation is not a factor which permits the conclusion to be drawn that Lafarge Plâtres International was independent of Lafarge (445).

(505) Lastly, it is clear from the file that in 1992 BPB contacted Mr [J], who was [high level representative] of the Lafarge group between 1989 and 1994 and then [high level representative] of the group, with a view to organising the exchange of information (446). This fact is not disputed by Lafarge, which merely states that Mr [J] has no recollection of such a meeting (447), without, however, going so far as to assert that Mr [J] never contacted BPB about this exchange of information. It is therefore incontestable that Lafarge exerted a decisive influence over its subsidiary Lafarge Plâtres International.

(506) In its reply to the statement of objections addressed to Lafarge, Lafarge Plâtres International tries to overturn the presumption that Lafarge actually exerted a decisive influence over its subsidiary Lafarge Plâtres International by submitting to the Commission a letter from Mr [G] to Mr [J] dated 27 November 1997 (448). This letter is in fact an investment dossier dealing with the proposal to purchase the Opole business, setting out the advantages and disadvantages of the plan and recommending the purchase under certain conditions. The mere fact that Mr [G] addressed the letter to Mr [J] shows once more, if that were necessary, the subsidiary's dependence on the parent company. The existence of the letter is proof of the part played by the parent company in the policy pursued by its subsidiary.

(507) It is therefore incontestable that Lafarge Plâtres International is legally and economically fully integrated in the structure of Lafarge.

(508) In those circumstances, the Commission is therefore entirely justified in addressing this Decision to Lafarge inasmuch as it is responsible for the actions of its subsidiary Lafarge Plâtres International.

(509) The Commission notes that Lafarge states that is "a holding company" having "as its sole object the acquisition and management of shareholdings in other companies. As such, it does not achieve any turnover of its own" (449). In those circumstances, the Commission considers that, with a view to the possible imposition of a fine (see section 3), the turnover to be taken into account for the purposes of this Decision is that of the "undertaking" within the meaning of Article 81(1) of the Treaty, controlled by Lafarge, this being the undertaking that committed the infringement. The turnover of that undertaking is the worldwide turnover achieved by all the companies of the group controlled by Lafarge, as communicated (450) by Lafarge to the Commission (451).

2.6.5. Gyproc

(510) Etex had a majority shareholding in Gyproc (54%, as against 46% held by BPB). However, the involvement of [high level representatives] of Etex, Mr [K] and Mr [L], in conduct covered by this Decision is proved only for the meetings in Brussels at the end of 1997 and The Hague in May 1998, which they attended in their capacity as [high level representatives] of Gyproc (452).

(511) BPB, although only a minority shareholder, was heavily influential in the commission of the infringement by Gyproc, because of both structural links (capital, trademark and patents) and the specific conduct of BPB's managers, who among other things took advantage of their membership of the Gyproc board in their collusive arrangements with Knauf and Lafarge and had influence over Gyproc. However, Gyproc's managers enjoyed a certain discretion, which while difficult to determine was not negligible.

(512) It must be stressed, moreover, that Gyproc has not supplied the Commission with evidence to refute allegations of its own responsibility or to make BPB or Etex liable for its conduct. Gyproc should accordingly be regarded as solely responsible for its own conduct.

2.6.6. Etex

(513) The Commission considers that, in the light of the evidence in its possession, the involvement of Etex managers in directly collusive conduct is proven only for the meetings in Brussels at the end of 1997 and The Hague in May 1998. That being so, the Commission considers that this evidence is insufficient to establish Etex's involvement in the single, complex and continuous infringement. Etex is therefore not an addressee of this Decision.

3. CORRECTIVE MEASURES

3.1. Article 3 of Regulation No 17

(514) By virtue of Article 3 of Regulation No 17, where the Commission finds that there is infringement of Article 81(1) of the Treaty, it may require the undertakings concerned to bring it to an end.

(515) In the present case, the members of the cartel made considerable efforts to conceal their unlawful actions. Practically all documentary traces of their activities were destroyed: there is little or nothing in the way of records of meetings, attendance lists or invitations. It is therefore impossible to state with absolute certainty that all those involved have terminated the infringement. The Commission must consequently demand that the undertakings to which this Decision is addressed put an end to the infringement, if they have not already done so, and desist in future from any agreement, concerted practice or decision by an association having an identical or similar object or effect.

3.2. Article 15(2) of Regulation No 17

3.2.1. General considerations

(516) Under Article 15(2) of Regulation No 17, the Commission may by decision impose on undertakings fines of from EUR 1 000 to EUR 1 000 000, or a sum in excess thereof but not exceeding 10% of the turnover in the preceding business year of each of the undertakings participating in the infringement where, either intentionally or negligently, they infringe Article 81(1) of the Treaty.

(517) The cartel was a deliberate infringement (453) of Article 81(1) of the Treaty: the main producers were perfectly aware that their conduct was anticompetitive and illegal and they agreed to establish a covert institutionalised system to restrict competition in an important industrial sector (see recitals (167) to (170).

(518) To determine the amount of the fine, the Commission must have regard to all relevant circumstances, and in particular the gravity and duration of the infringement, which are the two criteria explicitly referred to in Article 15(2) of Regulation No 17.

(519) The role played by each of the undertakings involved in the infringement will be assessed case by case. In particular, in setting the amount of the fine, the Commission will have regard to aggravating or mitigating circumstances and if appropriate will apply the Leniency Notice.

3.2.2. Amount of fines

(520) The amount of the fines is determined by calculating a basic amount and increasing it to take account of aggravating circumstances or reducing it to take account of mitigating circumstances.

3.2.2.1. The basic amount

(521) The basic amount is determined on the basis of the gravity and duration of the infringement.

Gravity of the infringement

(522) To assess the gravity of the infringement, the Commission has regard to its specific nature, its practical impact on the market where this is measurable, and the scale of the relevant geographic market. It also has regard to the economic capacity of the perpetrator of the infringement to cause other operators, particularly consumers, to suffer serious damage, and the need to set the amount of the fine at such a level that it has adequate deterrent effect.

(523) Gyproc considers (454) that, in the light of the criteria of gravity, impact and scale of the relevant geographic market, the infringement it committed must be considered to be minor. According to Gyproc, its infringement meets the definition of minor infringements according to 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 (455) in the case of restrictions which, although aimed at restricting trade, have a limited market impact and which affect only a substantial but relatively limited part of the Community market.

(524) Gyproc also indicates that its involvement in the infringements to which the Commission objects was "radically different from that of the other producers concerned" and "particularly limited". It bases this conclusion on the premiss that "it did not participate from 1992 to 1998 in the global agreement between [high level representatives] of the plasterboard producers on certain national European markets as described in the statement of objections" (456).

(525) The Commission considers, however, that this undertaking did participate from June 1996, albeit in a manner and to a degree which set it apart, in the single, continuous infringement committed by all the parties to the agreement. Consequently, it is not pertinent to single out a different infringement committed by this undertaking. As to the extent of this undertaking's involvement in the said infringement and the consequences to be drawn therefrom for the purpose of fixing the amount of the fine, the Commission would refer to the examination of mitigating circumstances (see below, recitals (570) et seq.).

(526) BPB considers that the Commission should conclude that the infringements concerned were minor, taking into account in its analysis only that conduct for which it acknowledges that the Commission might adopt the characterisation of infringement. It thus distinguishes between several types of exchange: the exchanges initiated by Mr [A, BPB], BPB affirming that the contacts with competitors were confined to a very narrow group of individuals, that the first contact was initiated by the [Mr B and Mr C, Knauf], and that the data exchanges were no more than annual; the exchanges conducted by Mr [D, BPB], of which the only unlawful element was, in its view, the exchange of individual company data, but which had little or no perceptible impact on competition and should therefore be considered less serious than any discussion, however vague, on prices; as to the exchanges relating to the United Kingdom, they were irregular and must be considered minor; the same goes for the "one or two" advance warnings of price increases which Mr [N, BPB] acknowledges having given to competitors (457).

(527) However, the Commission does not base its considerations on the same legal characterisation or the same assessment of the gravity of the facts as BPB. Even viewed separately, none of the conduct referred to deserves to be assessed as leniently as BPB would have it assessed, nor with all the more reason can the conduct's characterisation claimed by BPB, when that conduct is regarded as a manifestation of a complex, continuous agreement, be accepted.

Nature of the infringement

(528) It can be seen from the facts described in Part I that this infringement took the form of a complex, continuous agreement aimed at restricting competition, consisting of various manifestations by which the competitors sought to put an end to the price war and to stabilise the market and carried out exchanges of confidential information between competitors on a long-term basis and, in the United Kingdom and even more so in Germany, exchanges of information on certain price increase initiatives. Such an agreement belongs by its nature to the category of the most serious violations of Article 81(1) of the Treaty.

(529) The arrangements constituting the cartel have involved all the main operators in the EEA and were conceived, directed and encouraged at the senior levels of the hierarchy in each of the undertakings participating. By its very nature, the implementation of this type of cartel automatically generates a serious distortion of competition to the exclusive benefit of the participating producers and does great harm to their customers and, therefore, the final consumer.

(530) The Commission therefore considers that this infringement by its nature constitutes a very serious infringement of Article 81(1) of the Treaty.

Practical impact of the infringement

(531) BPB has asserted that the Commission has no proof of an effect on competition, and in particular that the information exchange could have little or no effect on competition inasmuch as the information exchanged related to the past, was highly aggregated and was not disseminated within the undertakings concerned (458). It also states that the price increase announcements made in advance to competitors, which it has acknowledged as far as the United Kingdom is concerned, could scarcely have any impact on competition given the divergence between official list prices and "net/net" prices and the process of prior consultation of customers (459).

(532) Lafarge has likewise minimised the scope and impact of the information exchange, claiming that "the information exchange has not had the effect of maintaining a system of limited competition". "Throughout the period 1992-98, competition was effective at all levels. The cut-throat price competition is the most visible element. ... If producers had enjoyed the luxury of limited competition, prices would have risen on every market back to their 1989- 1990 level. The same applies to the cut-throat competition based on volumes, which is what caused such price competition. The fact that market shares could remain fairly stable over short periods is not in the least surprising and is in no way a sign of limited competition. The highly mobile nature of some customer segments means that producers are constantly struggling either to profit from it or to find a new outlet once a customer is lost (but in that event it is more often than not from one of the other three large producers that a customer is poached). Thus an analysis of the customer base in Germany carried out by Lafarge in 1998 (pp. 10633 et seq.), noting the variations between 1996 and 1997 in the 200 leading customers, reveals that 45% of customers move on each year. Lafarge's belligerent attitude on the market is extensively documented in the file and shows just how impossible it is that Lafarge could have accepted any market share stability in the other producers' favour. The competition based on capacity and raw materials is also manifest. ... Lastly, the competition based on innovation which took place in the 1990s also means that the market is highly competitive" (460).

(533) Gyproc for its part has stated that "account must also be taken of the lack of any noticeable effect on the market attributable to the practices objected to, especially during the period when Gyproc Benelux was involved in the infringement. In fact, both operators' market shares and price levels on the German market underwent significant change between 1996 and 1998. But concerted attempts to increase prices almost always proved fruitless and never had a lasting effect on the level of plasterboard selling prices in Germany. Quite the contrary, as can be seen from the Commission's file, selling prices on the German market fell significantly during the period in question. ... Desirous of acquiring market shares, operators, and above all Gyproc Benelux, constantly developed aggressive pricing policies. ... More generally, it is clear that stiff competition continued to reign on the plasterboard market during the period in which Gyproc Benelux participated in the practices objected to" (461).

(534) In reality, the Commission considers that the infringement, committed by undertakings which, during the period covered by this Decision, accounted for all or almost all plasterboard supply on the four national markets to which the cartel extended, had a major practical impact on those markets inasmuch as almost all supply there was subject to anticompetitive conduct. Thus, whatever the intensity of that conduct, its effect was felt on almost all the markets concerned. Moreover, this impact was strengthened by the fact, which has been established (and, moreover, in no way challenged by the undertakings concerned), that plasterboard constitutes a market with a highly concentrated and oligopolistic structure owing to the simple fact that competition tends to be more limited on such a market. The cartel therefore inevitably had an effect on the conduct of the operators and hence on the markets. Lastly, this impact is also due to the fact that the various manifestations of the cartel were put into practice in that the undertakings effectively modified their conduct after the London meeting (see recital (60)); that prices effectively tended to rise or, at least, stabilise (see recitals (212) and (395)); that market shares were relatively stable during this period (see recitals (71), (196) and (289) and the Annex to this Decision), or at least more so than during the period before 1988-1992, characterised by the parties as a "price war"; that that stability was the result of the parties' pursuit of that objective; that the information exchanges decided on were effectively implemented throughout the period, both on the main markets and, more specifically, on the UK and German markets; and that the contacts relating to price increases are effectively linked to the publication of price lists subsequently taken into account in the prices invoiced to customers.

(535) This clearly holds true for the wish to stabilise the main European markets expressed as from 1992. It likewise holds true for any form of coordination of price increases in the United Kingdom or in Germany, or for the wish to stabilise the German market as from 1996 on the basis of existing market shares. The very purpose of the collusive conduct at issue was to modify the behaviour of undertakings active on the market by avoiding competition considered destructive and by seeking to achieve an increase in the price of plasterboard, which was possible if, and only if, each undertaking made do with its market share. It also applies to the data exchanges on an oligopolistic market such as the plasterboard market. The communication of data on volumes was, in view of the highly concentrated structure of the market and its transparency, tantamount to renouncing aggressive competition (see, for instance, recitals (449) and (450)). As the Court of Justice held in the British tractors case (462), the exchange of information on the market is such as to enable traders to know the market positions and strategies of their competitors and thus to impair appreciably the competition which exists between traders. The fact of having to notify any penetration would have elicited a response from competitors, which would have negated the attempt.

(536) A system for the exchange of confidential information may, moreover, have not only an object (see recitals (446) and (448)), but also an effect similar to a sharing of markets by reference to traditional trade flows in so far as an information exchange of the type at issue is equivalent to an agreement for the sharing of quotas and indicative of the existence of a tacit agreement to respect traditional flows.

(537) With regard to the information on price increases, as the Court of First Instance held in Cartonboard (463), the fact that the undertakings actually announced the agreed price increases and that the prices so announced served as a basis for fixing individual transaction prices suffices in itself for a finding that the collusion on prices had both as its object and effect a serious restriction of competition. As already indicated, this fact has been established in the present case. It is not necessary therefore to determine whether the variations in the transaction prices obtained evolved in parallel with those of the prices announced in order to show that the cartel had a practical impact on the relevant plasterboard markets. As already stated in section 1.5, a cartel is compatible with a certain degree of competition, and the Commission does not have to prove the non-existence of competition. Moreover, the fact that some price increases were a failure or had only limited success in no way invalidates the finding that the cartel produced certain effects.

(538) Lastly, the Commission considers that it is inconceivable, in view of the risks incurred of which the participants were well aware (see, for example, recitals (167) to (170)), that the parties should have agreed repeatedly to contact one another, to meet and to exchange information in a manner sufficiently secret to justify sending messages from and/or to private residences for such a long period if they had considered that the cartel had no, or only a limited, impact on the relevant plasterboard markets (464).

Extent of the relevant geographic market

(539) The cartel covered the four major markets in the Community, namely Germany, the United Kingdom, France and the Benelux.

(540) Gyproc states (465) that "only a small part of the Community market was concerned by the conduct of Gyproc Benelux, i.e. mainly Germany. With the exception of the period from April to October 1998, the sensitive information received or transmitted by Gyproc Benelux in fact concerned only the German market. Similarly, the price increases in which the other producers suggested Gyproc Benelux might like to participate covered only Germany".

(541) In fact, the exchanges in which Gyproc participated actually covered the four markets despite that undertaking's doing business only in Germany, France and the Benelux.

(542) The Commission considers that Germany, the United Kingdom, France and the Benelux, which are the four major plasterboard markets in the Community, constitute a large part of the Community market, both geographically and in terms of value (approximately 80% of the market's total value).

Commission conclusion on the gravity of the infringement as a whole

(543) Given the nature of the conduct under examination, its practical impact on the plasterboard market and the fact that it targeted the four major plasterboard markets in the Community, which had a total value of EUR 1 210 million in the last full year of the infringement, the Commission considers that the undertakings to which this Decision is addressed committed a very serious infringement of Article 81(1) of the Treaty.

(544) In the category of very serious infringements, the range of fines available enables undertakings to be treated differently depending on their real economic capacity to seriously harm competition, and fines can be set at a level that ensures an adequate deterrent effect.

Classification of those involved in the cartel

(545) In the circumstances of this case, which involves several undertakings, the specific weight of each of the undertakings involved and the real effect of its unlawful behaviour on competition must be borne in mind in determining the basic amount of the fines. The undertakings concerned may basically be divided into three categories according to their respective importance on the relevant market, subject to adjustment in the light of other factors, if any, and of the need to ensure an adequate deterrent effect.

(546) For the purpose of comparing the relative importance of the undertakings on the market affected by this agreement, the Commission considers it appropriate in this case to take as a starting point the market shares of the undertakings based on sales turnover for the product on the four principal markets of the Community, that is, Germany, the United Kingdom, France and Benelux. This approach is justified by the fact that the agreement concerned extends to these four markets. The comparison is made on the basis of the sales turnover figures for the product in these four markets in 1997, the last full year of the infringement. The Commission also takes into account the large size of the market in question. The necessary data are contained in table 3 in recital (25).

(547) With a share of around [40-45] % of the total of these four markets, BPB is the largest producer and is, therefore, placed in the first category. Knauf and Lafarge, with shares of around [25-30] % and [20-25] % of the market respectively, are placed in the second category. With a market share of around [7-10] %, Gyproc is therefore placed in the third category.

(548) Gyproc has pointed to its status as a "very modest" operator as a factor to be taken into account in determining the gravity of the infringement committed (466). The Commission considers that the present classification of the participants in the cartel is of such a character as to meet Gyproc's concern.

(549) On this basis, the Commission has determined that the basic amounts of the fines should be as follows on the basis of gravity:

- BPB: EUR 80 million

- Knauf Westdeutsche Gipswerke: EUR 52 million

- Lafarge: EUR 52 million

- Gyproc: EUR 8 million

Adequate deterrent amount

(550) To take into account the size and aggregate resources of the undertakings and ensure that the fine has an adequate deterrent effect, the Commission considers that this starting amount should be adjusted for one or other of the undertakings.

(551) As the Court of First Instance held in the ABB - District heating pipes case (467), "the fact that, in fixing such a multiplier, the Commission took into account the deterrent effect that fines must have, is wholly consistent with the established principle that the gravity of infringements has to be determined by reference to numerous factors, such as the particular circumstances of the case, its context and the dissuasive effect of fines, although no binding or exhaustive list of the criteria which must be applied has been drawn up. ... In that regard, the Commission's power to impose fines on undertakings which, intentionally or negligently, commit an infringement of the provisions of Article 85(1) of the Treaty is one of the means conferred on the Commission in order to enable it to carry out the task of supervision conferred on it by Community law. That task ... also encompasses the duty to pursue a general policy designed to apply, in competition matters, the principles laid down by the Treaty and to guide the conduct of undertakings in the light of those principles".

(552) In these circumstances, the Commission considers that in the case of Lafarge (468) the starting amount of the fine should be increased to take account of its size and aggregate resources.

(553) On this basis, the Commission considers that to ensure that the fine has adequate deterrent effect, the starting amount of the fine determined at recital (549) should be increased by 100 % to EUR 104 million in the case of Lafarge.

Duration of the infringement

(554) The Commission has established that Knauf and BPB infringed Article 81(1) of the Treaty from 31 March 1992, at the latest, until 25 November 1998. For the purpose of calculating the fine, the Commission takes into account complete months and therefore fixes as the duration of the infringement by these two undertakings a period of six years and seven months.

(555) Lafarge committed the same infringement from 31 August 1992, at the latest, until 25 November 1998. For the purpose of calculating the fine, the Commission takes into account complete months and therefore fixes as the duration of the infringement by this undertaking a period of six years and two months.

(556) Gyproc was actively involved in the infringement from 6 June 1996, at the latest, until 25 November 1998. For the purpose of calculating the fine, the Commission takes into account complete months and therefore fixes as the duration of the infringement by this undertaking a period of two years and four months.

(557) The Commission concludes that the infringement was of lengthy duration (more than five years) for Knauf, BPB and Lafarge and of average duration (from one to five years) for Gyproc and accordingly increases by 65 % the basic amount of the fine imposed on BPB and Knauf Westdeutsche Gipswerke, by 60 % the basic amount of the fine imposed on Lafarge and by 20 % the basic amount of the fine imposed on Gyproc.

Conclusion as to the basic amount

(558) The Commission accordingly sets the basic amounts of the fines as follows:

- BPB: EUR 132 million

- Knauf Westdeutsche Gipswerke: EUR 85.8 million

- Lafarge: EUR 166.4 million

- Gyproc: EUR 9.6 million

3.2.2.2. Aggravating circumstance: repeat offence

(559) As stated above in recitals (13) and (14), BPB and Lafarge have already been subject to previous Commission measures in cartel cases in, respectively, Decision 94-601-EC and Decision 94-815-EC.

(560) The facts presented in this Decision establish that BPB and Lafarge continued for more than four years to participate actively in a cartel in the plasterboard sector after those decisions had been notified to them. The fact that they repeated the same type of conduct in a sector other than that in which they had incurred a penalty shows that the first penalty did not prompt them to change their conduct and is therefore, for the Commission, an aggravating circumstance.

(561) As regards BPB, the fact that Decision 94-601-EC was addressed to a BPB subsidiary, BPB De Eendracht NV, is not of such a character as to prevent the application of such an aggravating circumstance in this case. As BPB De Eendracht NV was a subsidiary of BPB at the time of the said Decision, the Commission considers that they are one and the same undertaking for the purposes of Article 81 of the Treaty and that, when an undertaking has been censured by the Commission, it has a responsibility not only to put an end to its anticompetitive conduct but also to ensure that its commercial policy throughout the Community conforms to the individual Decision notified to it; BPB did not do this, quite the reverse, as this Decision attests (469).

(562) BPB has stated in this connection that "the following needs to be emphasised. The Commission found that the subsidiary was a trivial and unwilling component of the conspiracy, and the company's representatives withdrew from meetings where unlawful matters were under discussion. There was therefore no active participation. The subsidiary was at fault, procedurally, in not clearly expressing its complete detachment from the common purpose of the others. In short, it was found to have infringed Article 85 (now Article 81) by virtue of its passive participation. This was reflected in the reduced fine on appeal. The duties placed on undertakings faced with such a situation were not, it is submitted, legally clear at the time the infringement was committed". The judgment could not, in BPB's opinion, be considered proof "of any recidivism or corporate neglect or indifference to the competition rules. The latter example, in particular, showed great sensitivity to the rules on the part of the company but it underestimated what was required to extricate the company from the taint of cartel membership" (470).

(563) It is not even necessary to discuss BPB's assessment of the Commission's decision or the judgment of the Court of First Instance, the Commission being of the opinion that the mere fact that the same undertaking has already formed the subject matter of a finding of infringement and that, despite that finding and the penalty imposed, it continued its involvement in another, similar infringement of the same Treaty provision, is proof of recidivism.

(564) The Commission accordingly considers that it must augment the penalty imposed by increasing the basic amount by 50 % in the case of BPB and Lafarge.

3.2.2.3. Mitigating circumstances

Exclusively passive or follower role under pressure from another undertaking

(565) Gyproc states that it played an exclusively "follower" role in the development of the collusive practices and that it often did not implement them. It claims that, in view of its size, it clearly "could not make a stand and openly rebel by refusing to talk with the three big producers who control nearly 95% of the European plasterboard market." It also claims that, in practice, it "continued to develop an independent and aggressive commercial policy with a view to winning new customers and increasing its market share. Moreover, Gyproc Benelux has never taken the initiative in exchanges of confidential information or concerted price increases. In some cases, Gyproc Benelux merely followed moves by the three big producers while refraining, in very many cases, from complying with the objectives sought. Gyproc Benelux has thus often engaged in independent commercial conduct which flew in the face of initiatives adopted by its competitors". Gyproc indicates lastly that it was "limited in its room for manœuvre by the contractual and structural links which tie it to BPB. Besides its substantial participation in the company's capital, BPB enjoys significant prerogatives within Gyproc's board of directors. BPB also owns the 'Gyproc' trademark used by Gyproc for marketing the vast majority of its products. Moreover, Gyproc is dependent on the technical and technological assistance provided by BPB. Obviously, Gyproc could not ignore this influence by BPB when deciding what attitude to adopt towards its competitors, prominent among which was a major shareholder" (471).

(566) It must be observed in reply to the arguments advanced by Gyproc that Gyproc was not subject to the other undertakings, or even only to BPB, to the extent of being unable to distance itself from them in practice by cheating in relation to the collusive arrangements in which it was involved. Passive or follower conduct would in fact have meant not breaking free from the other participants in the conduct at issue. Moreover, everything points to Gyproc having played an active part in the Versailles meeting and the meetings that followed.

(567) In those circumstances the Commission concludes that it must reject this claim by Gyproc as a mitigating circumstance. Termination of infringements upon the Commission's first intervention

(568) Gyproc states that it put an end to the infringement the moment the Commission intervened in that it terminated all information exchanges with Mr [D, BPB] and all collusive practices on the German market upon the Commission's first intervention, i.e. the inspection carried out at Gyproc's head office on 25 November 1998 (472).

(569) However, the flagrant and manifestly unlawful character of the conduct at issue being established beyond question, the Commission is entitled to expect it to cease immediately after the inspections, without this being capable of constituting a mitigating circumstance. The Commission accordingly considers that Gyproc's claim must be rejected.

Other mitigating circumstances

(570) Gyproc stresses the limited character of its involvement in the various manifestations of the cartel both at the European level of [high level representatives] and at the German level, and its lack of involvement in the cartel as far the UK market is concerned. It claims not to be one of the instigators of the collusive practices engaged in from 1992 onwards by the three big producers to end the price war and stabilise markets and that "it was only from 1996 onwards that Gyproc gradually and very partially became integrated in the practices of the other producers" (473).

(571) The Commission would point out first that it takes into account, in its assessment of the duration of Gyproc's participation in the infringement, only the period after June 1996.

(572) The Commission would point out secondly that, for the period from June 1996 to November 1998, Gyproc was an active party to the complex, continuous agreement described in this Decision inasmuch as its managers took part regularly in meetings aimed at stabilising the German market, knew or should have known as from that time that the information exchange of which Gyproc was a member had a collusive purpose, and actively participated in every aspect of that information exchange between March and November 1998. The Commission accordingly holds Gyproc responsible for the whole infringement for the period during which it participated in it, that is to say, from June 1996 to November 1998.

(573) It is true that Gyproc did not participate in the cartel on the UK market, but this is due to the fact that it was not present on that market. It is also true, in the light of the evidence available, that Gyproc Benelux did not take the initiative in seeking concerted price increases on the German market and that it compromised their success by not implementing them or not implementing them fully. However, this does not establish the lack of any participation in the cartel in so far as the fact that the undertaking received prior announcements of competitors' price increases, even if it did not implement them, affected its own pricing policy, which was thus pursued with full knowledge of the facts. In reality, this is a frequent occurrence in secret cartels, where one undertaking uses the cartel for its own benefit without observing its constraints.

(574) During a substantial period of its participation in the cartel, Gyproc does seem to have had difficulty in preventing Mr [D, BPB] from obtaining and transmitting information concerning it, owing to his status as member of its board of directors; it did act as a constant destabilising element, which helped limit the impact of the cartel on the German market and it was absent from the UK market, where the manifestations of the cartel were most frequent. This set of circumstances places Gyproc in an objectively different situation such that the Commission recognises that Gyproc did not play the same part in the cartel as the other undertakings.

(575) The Commission considers that all these elements must be taken into account as mitigating circumstances in the assessment of the gravity of Gyproc's infringement.

(576) Gyproc also stresses that it introduced a scheme to bring it into line with competition law and organised training seminars for its staff with the help of specialised consultants (474).

(577) While the Commission views such a scheme as a positive initiative, it does not regard it as a mitigating circumstance, since it came too late and does not exempt the Commission from performing its task of punishing the infringement of the competition rules committed by Gyproc. The Commission accordingly considers that there is no ground for accepting this element as a mitigating circumstance in Gyproc's case.

(578) BPB claims that "due weight must be given to the timely steps taken by BPB, upon receipt of the anonymous letter, to convene a board meeting and appoint independent external legal advisers, [...]. This firm was given free rein thoroughly to investigate the accusations contained in the anonymous letter (save those of a personal nature against Mr [D, BPB]), and all employees were required to cooperate and make available all papers. The report (known internally as 'the Alpha' report) reported Mr [D]'s statements regarding market data exchanges and those also of Mr [N, BPB]. Material weight must be placed upon the fact that BPB, on receipt of this report, immediately reinforced its compliance measures and specifically drew every senior manager's attention to the absolute necessity to abide by all relevant competition rules. Messrs [D] and [N] were told to cease all contacts with competitors forthwith" (475).

(579) It states, furthermore, "before this, BPB instructed Rigips to withdraw from its information sharing arrangements with [Mr U, independent consultant], though, on any reasonable legal assessment, such contacts were conducted by [Mr U] in accordance with strict propriety. In fact, both Knauf and the German Gypsum Association furnished BPB with legal opinions they obtained as to the legality of the scheme in the hope that BPB would rejoin, and BPB obtained its own advice, but BPB chose, given the sensitivity, to remain outside the exchanges" (476).

(580) Moreover, "the seriousness with which BPB regarded compliance is highlighted by the immediate decision to part company with Mr. [D] once it had become clear that Mr [D] had disobeyed express instructions to have no further contact in relation to information sharing with competitors. [Mr D] was a long serving employee of BPB and at the time of his departure had been on the main board for [some] years and [high level representative] for [some] years. We should add that [Mr D]'s sudden departure placed BPB in significant difficulties as there was no obvious successor to him" (477).

(581) Lastly, "after the Commission's inspection BPB instructed its solicitors, [...], to embark upon a further inquiry to investigate the matters contained in the affidavit sworn by Mr [..., Commission official], and generally. Once again the lawyers were given free rein over all persons and papers. The willingness to expose itself to two external enquiries and to cooperate fully with the Commission are actions which are inconsistent with a company which has something to hide; to the contrary, these actions are consistent with a general corporate culture of seeking to ensure compliance with all relevant laws wherever BPB trades" (478).

(582) The Commission can only find that the first scheme set in place by BPB evidently did not achieve the desired result.

(583) As for BPB's withdrawal from the [Mr U, independent consultant] information sharing arrangements, while it demonstrates a willingness to avoid conduct which could give rise to suspicion, it was not accompanied by other measures designed to put an end to the collusive arrangements, as illustrated by the continued exchange of information, or, for example, the discussions between competitors at the Hague. The Commission is therefore unable to accept BPB's request as a mitigating circumstance.

(584) Lastly, as regards the decisions to change the company's management and the measures taken after the inspection, the Commission does not view these initiatives as a mitigating circumstance, since they came too late and do not exempt the Commission from performing its task of punishing the infringement of the competition rules committed by BPB. The Commission accordingly considers that there is no ground for accepting this element as a mitigating circumstance in BPB's case.

(585) Gyproc Benelux stresses that it has never in the past been convicted of or prosecuted for a breach of Community or national competition legislation.

(586) The Commission finds that such a conviction might have constituted an aggravating circumstance, but its absence does not conversely constitute a mitigating circumstance. The Commission accordingly considers that there is no ground for accepting this element as a mitigating circumstance in Gyproc's case.

(587) BPB states that it "has secured no financial benefit arising from the matters under investigation". It also states that "after the ending of the price war, between January and December 1992, BPB's net net prices in Germany and the UK increased by [3-5]% and [7- 10] % respectively" but that "as the level of prices in the price war was, by common consent, very low and unsustainable, the Commission cannot argue that prices would not have gone up at some point in 1992. At worst, the [Messrs B and C, Knauf] / [Mr A, BPB] discussions, if they had had any impact at all, accelerated an inevitable increase in prices. But neither BPB nor the Commission can begin to quantify the increase". Likewise, according to BPB, "the Commission cannot isolate any material benefit accruing to BPB as a result of the market exchanges ... the market position was unlikely to have been any different if the information exchanges had not taken place". BPB none the less recognises that "the knowledge of the total size and the participants' positions in it, had commercial value to its [high level representative] - and as such, it is at least possible, that the competitive situation may have been slightly different in some respects. As a result of knowledge of competitor shares, competition might even have become fiercer. It is, however, very difficult to attribute, with confidence, any financial benefit. This is a factor which must be taken into account by the Commission in assessing the penalty" (479).

(588) The Commission does not consider the fact, even supposing it was proven, which BPB's own description leaves open to doubt, of not having benefited from an agreement to constitute a mitigating circumstance in setting the penalty, or that the companies can claim a right to have the penalty reduced for an infringement, not only the effect, but also the object, of which was anticompetitive.

(589) In the light of the above, the Commission considers it appropriate to grant Gyproc a reduction of 25 % of the basic amount of the fine.

3.2.2.4. Conclusion on the amounts of the fines prior to application of the Leniency Notice

(590) Having taken account of the aggravating and mitigating circumstances, the Commission determines the amount of the fines prior to application of the Leniency Notice as follows:

- BPB: EUR 198 million

- Knauf Westdeutsche Gipswerke: EUR 85.8 million

- Lafarge: EUR 249.6 million

- Gyproc: EUR 7.2 million

3.2.2.5. Application of the Leniency Notice

(591) Certain of the addressees of this Decision have cooperated with the Commission, at different stages of the investigation and in conjunction with the different periods of the infringement under examination, in order to enjoy the favourable treatment referred to in the Leniency Notice. To meet the relevant undertakings' legitimate expectations as to immunity from fines and reduction of fines in return for their cooperation, it is necessary to consider whether those parties meet the conditions set out in the Leniency Notice.

Major reduction in the amount of the fine

(592) BPB was the first member of the cartel to provide evidence in addition to that discovered during the inspections and which confirmed the existence of the cartel. After the Commission sent BPB a request for information, BPB supplied statements and documents relating to the case.

(593) However, contrary to what is claimed by the company (480), BPB did not supply the Commission with decisive information, which would in any event have been in a position to initiate proceedings with a view to reaching a decision in this case. Following its investigations, the Commission had enough information to prove the existence of the cartel.

(594) The Commission accordingly considers that the cooperation afforded by BPB does not meet the conditions of point B(b) of the Leniency Notice and it is therefore not eligible for a major reduction in the amount of its fine under Title C of the Notice.

Significant reduction in the amount of the fine

(595) Before the Commission adopted its statement of objections, BPB and Gyproc supplied it with information and/or documents. However, the extent and quality of their cooperation with the Commission differed.

(596) The Commission finds that BPB was the first member of the cartel to send it, after the Commission's request for information but going beyond what was requested, evidence in addition to that discovered during the inspections, confirming the existence of the cartel (see recital (593)). As pointed out by BPB, this information includes details regarding the meetings concerned, especially that in London, and the sharing of information on the major European markets, and the UK market in particular (481).

(597) BPB also admitted some of the facts set out in the statement of objections (482). However, it disagrees that certain facts described in the statement of objections and also identified in this Decision rank as infringements.

(598) The Commission finds that Gyproc also provided evidence that helped to establish proof of the infringement. As stressed by the company (483), Gyproc supplied the Commission, after the Commission's request for information but going beyond what was requested, with information on cartel meetings, specifying the periods during which meetings took place in the different Community Member States and the undertakings attending. During an interview requested by Gyproc with the Commission departments, Mr [E, Gyproc] voluntarily provided verbal explanations of his handwritten notes. Gyproc subsequently provided written confirmation of his explanations, which help to show that the plasterboard producers sought to share out the German market, or at least to stabilise it. Gyproc also voluntarily supplied the Commission with notes handwritten by Mr [E], which the Commission had not identified during the inspection and which contained, in particular, information on the exchange of sales volumes during the meeting at Versailles.

(599) Gyproc does not contest the facts or their status as infringements of Community competition law.

(600) The Commission accordingly considers it justified to reduce the amount of the fines by 30 % for BPB and 40 % for Gyproc.

3.2.2.6. Final amount of fines imposed in this proceeding

(601) By way of conclusion, the amounts of the fines to be imposed in accordance with Article 15(2)(a) of Regulation No 17 should be set as follows:

- BPB: EUR 138.6 million

- Knauf Westdeutsche Gipswerke: EUR 85.8 million

- Lafarge: EUR 249.6 million

- Gyproc: EUR 4.32 million

HAS ADOPTED THIS DECISION:

Article 1

BPB PLC, the Knauf Group, Société Lafarge S.A. and Gyproc Benelux N.V. have infringed Article 81(1) of the Treaty by participating in a set of agreements and concerted practices in the plasterboard business.

The duration of the infringement was as follows:

(a) BPB PLC : from 31 March 1992, at the latest, to 25 November 1998

(b) Knauf: from 31 March 1992, at the latest, to 25 November 1998

(c) Société Lafarge S.A. : from 31 August 1992, at the latest, to 25 November 1998

(d) Gyproc Benelux N.V. : from 6 June 1996, at the latest, to 25 November 1998.

Article 2

The undertakings referred to in Article 1 shall put an end to the infringement referred to in that Article, if they have not already done so. In their plasterboard business they shall refrain from any agreement or concerted practice that might have an identical or similar object or effect to the infringement.

Article 3

In respect of the infringement referred to in Article 1, the following fines are imposed on the following undertakings :

(a) BPB PLC: EUR 138.6 million

(b) Gebrüder Knauf Westdeutsche Gipswerke KG: EUR 85.8 million

(c) Société Lafarge S.A.: EUR 249.6 million

(d) Gyproc Benelux N.V.: EUR 4.32 million

The fines shall be payable within three months following notification of this Decision, by transfer to bank account No 642-0029000-95 (SWIFT Code: BBVABEBB - Code IBAN BE76 6420 0290 0095) held by the European Commission at

Banco Bilbao Vizcaya Argentaria (BBVA) SA, Avenue des Arts , 43 B-1040 Brussels.

After expiry of that period, interest shall automatically be payable at the interest rate charged by the European Central Bank for its main refinancing operations on the first day of the month in which this Decision is adopted plus 3.5 percentage points, i.e. 6.79%.

Article 4

This Decision is addressed to:

BPB PLC

Park House

15 Bath Road

UK - Slough SL1 3UF

Gebrüder Knauf Westdeutsche Gipswerke KG

Am Bahnhof 7

D-97346 Iphofen

Société Lafarge S.A.

61, rue des Belles Feuilles

F-75782 Paris cedex 16

Gyproc Benelux N.V.

Merksemsebaan 270

B-2110 Wijnegem

This Decision shall be enforceable pursuant to Article 256 of the Treaty.

ANNEX:

SALES VOLUMES OF PLASTERBOARD FOR 1992 TO 1998 OF THE ADDRESSEES OF THE DECISION

The first figure represents the sales volumes in metres squared, as disclosed by the undertakings and rounded to the nearest million.

The second figure, in brackets, represents the percentage for the undertaking concerned relative to the total sales of the four undertakings.

<Emplacement tableau>

(*) The data replaced by [ ] with or without an approximate value in their place, are of a confidential nature or business secrets.

(**) The names of the persons involved have been replaced by letters according to the order in which they appear in the description of the infringement (section 1.4 "Details of the infringment"). The detailed descriptions of their functions have been replaced by "representative" or "high level representative" of their undertaking.

NOTES

1 OJ 13, 21.2.1962, p. 204/62.

2 OJ L 148, 15.6.1999, p. 5.

3 OJ L 354, 30.12.1998, p. 18.

4 "Knauf" means all the companies in the Knauf group, see section 1.3.3.2.

5 pp. 49.10665-10718 of the Commission file. Unless otherwise specified, the pages referred to below are those of the Commission file, with the first figure referring to the number of the volume in the file and the second to the page number.

6 Hereinafter referred to as "BPB's first reply".

7 Hereinafter referred to as "BPB's memo".

8 Hereinafter referred to as "BPB's statement".

9 Hereinafter referred to as "BPB's second reply".

10 Hereinafter referred to as "BPB's third reply".

11 Hereinafter referred to as "Gyproc's statement".

12 Replies submitted by Lafarge on 26 August, Gyproc on 6 September, BPB on 6 September, Etex on 9 September and Knauf on 19 September and 9 October 2002.

13 Case IV/C/33.833 - Cartonboard, OJ L 243,19.9.1994, p. 1.

14 [1998] ECR II-1129.

15 Cases IV/33.126 and 33.322 - Cement, OJ L 343, 30.12.1994, p. 1.

16 [2000] ECR II-491.

17 The figures for each company are the rounded figures provided by it; the figure under "Other", for Germany in 1998, is based on BPB's estimate (reply to the statement of objections, p. 29) and relates to Norgips' sales in 1998; Norgips resumed sales on the German market only in or around September 1998 (see recital (47)).

18 In Tables 1, 2 and 3, DE = Germany; UK = United Kingdom ; FR = France; BNL = The Netherlands, Belgium and Luxembourg.

19 The figures provided are based on the companies' replies to the information requests and on Commission estimates. Sales in 1997 and 1998 are provided as examples. The relative level of each company's annual sales may have varied from one year to another during the period covered by the infringement.

20 All the turnover figures shown in this Decision are given in ecus or in euros, as appropriate. The conversion rates used in calculating the amounts in euros are the annual (or, where appropriate, monthly) average official rates published by the Commission for the calculation of turnover; for BPB, 2001 corresponds to the period from 1.4.2001-31.3.2002 (exchange rate obtained from the average monthly exchange rates from April 2001 to March 2002), 1997 corresponds to the period from 1.4.1997-31.3.1998 (exchange rate obtained from the average monthly exchange rates from April 1997 to March 1998), 1998 corresponds to the period from 1.4.1998-31.3.1999 (exchange rate obtained from the average monthly exchange rates from April 1998 to March 1999).

21 Monopolies and Mergers Commission, Report on the supply of plasterboard in the United Kingdom, Cm1224, October 1990, p. 35.

22 Knauf's reply of 19 September 2002 to the Commission's request for information, Zusammenstellung 7.

23 Knauf's website: http://www.knauf.de/home.htm.

24 Source: Verband der Vereine Creditreform e.V. Creditreform-Datenbank-Dienste; EcoRegister: German Company Registrations (ECCO) - Bundesanzeiger Verlagsgesellschaft mbH.

25 Source: Verband der Vereine Creditreform e.V. Creditreform-Datenbank-Dienste; EcoRegister: German Company Registrations (ECCO) - Bundesanzeiger Verlagsgesellschaft mbH; Dun & Bradstreet European Standard Report.

26 Knauf's reply of 19 September 2002 to the Commission's request for information, p. 1, 3.

27 Lafarge's reply to the statement of objections, p. 6.

28 BPB's reply to the Commission's request for information, p. 36.7987: Knauf's reply to the statement of objections, pp. 42-43.

29 Lafarge's reply to the statement of objections, p. 37.

30 Ibid., p. 40.

31 BPB's reply to the statement of objections, p. 28.

32 Knauf's reply to the statement of objections, p. 23.

33 Knauf's reply to the statement of objections, p. 28, concerning what Knauf describes as the main purpose of the talks held with BPB in London in 1992.

34 Second reply to the request for information, pp. 36.8128-8130.

35 Knauf's reply to the statement of objections, p. 29.

36 BPB's second reply to the request for information, p. 36.8129.

37 BPB's reply to the statement of objections, p. 45.

38 Ibid., pp. 45-46.

39 Ibid., p. 46.

40 Ibid., p. 4, p. 108.

41 Ibid., p. 108.

42 BPB's reply of 28 October 1999, p. 36.8129.

43 Reply to the statement of objections, p. 46.

44 Reply to the statement of objections, pp. 29-33.

45 Knauf's reply to the statement of objections, p. 29-30.

46 Ibid., p.32.

47 Ibid., p. 30.

48 Ibid., p. 31.

49 In recitals 71 et seq., according to Knauf (p. 31 of the reply to the statement of objections).

50 Reply to the statement of objections, p. 20.

51 Original German: "Verständigung", p. 32.

52 p. 16.4175.

53 See recital (61), referring to p. 29 of Knauf's reply to the statement of objections; witness also the inability of [Mr B and Mr C, Knauf] to recall who initiated the meeting, see recital (53).

54 BPB's reply to the statement of objections, pp. 30-31.

55 Knauf's reply to the statement of objections, p. 14.

56 Annex I to Knauf's reply to the statement of objections, conclusion, p. 68.

57 Knauf's reply to the statement of objections, p. 18.

58 Ibid., p. 31.

59 p. 36.7946, paragraph 2; see also Gyproc's statement, p. 35.7802.

60 p. 36.7946, paragraph 5.

61 p. 1.42.

62 BPB's first reply to the request for information, p. 36.8001.

63 p. 1.64.

64 BPB's first reply to the request for information, p. 36.8002.

65 "Fin septembre 98" p. 1.65.

66 BPB's first reply to the request for information, p. 36.8003.

67 p. 25.5970.

68 p. 25.6010.

69 p. 25.5938.

70 pp. 1.76-77.

71 Letter dated 1 September 1999 concerning the oral statement of 23 July 1999, pp. 35.7791-7792.

72 Reply dated 2 November 1999 to the Commission's request for information , p. 39.8623.

73 Reply to the statement of objections, p. 18.

74 pp. 1.76-79, 1.43-44 and BPB's first reply to the request for information, p. 36.7999.

75 Provided by BPB at the Commission's request following the investigations, during which it was found that documents had been removed from BPB's premises and transferred to the premises of its lawyer, pp. 35.7846-7851.

76 Reply by [Mr D] to the questions asked by the Commission officials, p. 1.2 of the file, point 3.

77 Reply by BPB dated 17 March 1999 to the Commission's request for information, p. 36.7999.

78 Ibid., pp. 36.8000-8001.

79 Reply by BPB dated 28 October 1999 to the information request, p. 36.8132.

80 Statement made by [Mr D] during the investigations, p. 1.2; reply by BPB dated 17 March 1999, p. 36.7999, reply dated 1 October 1999, p. 36.8131.

81 BPB's second reply, p. 36.8132.

82 Gyproc's reply to the Commission's request for information, p. 39.8623.

83 Knauf's reply to the Commission's request for information, p. 37.8310; Knauf's reply to the statement of objections, pp. 32-34.

84 Lafarge's reply to the statement of objections, pp. 60-70.

85 Reply dated 28 October 1999 to the Commission's request for information, p. 36.8130. 86 See recital (79).

87 It should be noted that, in this table, Lafarge includes Epysa and Norgips, even though Norgips is included in "Others" as regards the Nordic countries.

88 It should be noted that, in this table, Knauf includes Danogips and Isolava.

89 It should be noted that for BPB itself, the so-called "real" data do not tally with the estimated data on Spain and Portugual for 1995, on Austria, Spain and Portugal for 1996, or on France, Germany, Italy, Spain, the United Kingdom and the Czech Republic for 1997. It may be concluded from this that [this BPB's high level representative] did not believe the figures received from certain of their partners or subsidiaries, which is corroborated by BPB's statements to the effect that [MrD] checked the results of his colleagues (see reply from BPB, pp. 2-3).

90 Reply dated 28 October 1999, p. 36.8131.

91 BPB's second reply, p. 36.8131.

92 Reply to the statement of objections, p. 110.

93 Reply to the statement of objections, Annex 50, p. 3.

94 BPB's second reply, p. 36.8130.

95 Ibid., p. 36.8122.

96 Reply to the statement of objections, Annex 50, p. 2.

97 Reply to the statement of objections, p. 46.

98 Reply to the statement of objections, p. 108.

99 pp. 4 and 22, 48, 108-110.

100 Reply to the statement of objections, p. 65.

101 Reply to the statement of objections, p. 48.

102 Reply to the statement of objections, Annex 50, p. 2.

103 See recital (133).

104 Reply to the statement of objections, pp. 48, 112.

105 Reply to the statement of objections, p. 105.

106 BPB's first reply, p. 36.8000.

107 Knauf's reply to the statement of objections, p. 33.

108 Ibid., p. 20.

109 Ibid., p. 33.

110 Ibid., p.22.

111 Ibid., p. 21, p. 33.

112 Knauf's reply to the statement of objections, p. 33.

113 p. 5. 1256-1259.

114 p. 36. 8130.

115 Second reply to the request for information, p. 36.8129.

116 Reply to the statement of objections, pp. 51, 61.

117 Ibid., p.61.

118 Reply to the statement of objections, p. 62-68.

119 p. 2-3 of [MrG]' statement.

120 Reply to the statement of objections, p. 47.

121 Letter from the Commission's Directorate-General for Competition dated 12 June 2002. Lafarge's reply dated 25 June 2002.

122 pp. 27.6259-6260.

123 www.lafarge.com

124 BPB's second reply to the Commission's request for information, p. 36.8129.

125 [high level representative] of Gyproc in 1992.

126 BPB's reply to the statement of objections, p. 47.

127 Reply to the statement of objections, p. 41.

128 BPB's reply of 28 October 1999 to the Commission's request for information, p. 36-8133.

129 p. 35.7791.

130 Reply to the Commission's request for information, p. 39.8623.

131 Reply to the statement of objections, p. 17.

132 Ibid.

133 Quoted p. 32.6969.

134 Reply to the statement of objections, pp. 21-22.

135 See recital (220).

136 Reply to the statement of objections, p. 20.

137 Reply to the statement of objections, pp. 16-17.

138 See recital (221).

139 See the statements by [Mr D] and [Mr G] and Knauf's reply to the statement of objections.

140 Memo of 17 March 1999, p. 36.8027.

141 Lafarge's reply to the statement of objections, pp. 60-61.

142 pp. 5.1258-1259.

143 As will be shown in recital (271) in connection with the exchange of information on the German market set up in 1996 under the aegis of [Mr U].

144 p. 3 of [Mr G]'s statement.

145 Commission memo attached to BPB's first reply to the request for information, p. 36.8020.

146 BPB's statement, pp. 36.8061-8062; BPB's second reply to the request for information, p. 36.8122-8128.

147 Reply to the statement of objections, p. 124, original English.

148 Reply of 28 October 1999, p. 36.8125.

149 See recital (172) and reply by BPB to the statement of objections pp. 51.

150 Statement of 28 May 1999, p. 36.8048.

151 Second request for information, p. 36.8123.

152 Statement, p. 36.8048.

153 See reply of 28 October 1999, p. 36.8124.

154 Reply of 28 October 1999, p. 36.8125.

155 Ibid, p. 36.8125-8126.

156 Reply of 28 October 1999 to the Commission's request for information, pp. 36.8153-8154.

157 BPB's reply to the statement of objections, pp. 52-53.

158 Reply of 28 October 1999 to the Commission's request for information, pp. 36.8125.

159 Reply of 28 May 1999, p. 36.8048.

160 BPB's reply to the statement of objections, p. 52.

161 Statement of 28 May 1999, p. 36.8048.

162 BPB's reply to the statement of objections, p. 51-53.

163 Reply of 28 October 1999, p. 36.8124.

164 Ibid.

165 Reply to the statement of objections, p. 57.

166 Reply to the statement of objections, p. 45.

167 Reply to the statement of objections, p. 86.

168 P. 16.4142; memo from Mr [...] to Mr [...] of 7 September 1992; draft reference memo for the OFT from BPB; file p. 16.4204; BPB's draft corporate plan.

169 Pp. 16.4157 and 16.4163.

170 Pp. 16.4136-4137; letter from the Federation of Plastering and Drywall Contractors dated 27 August 1992 and p. 11.2531, article from the Financial Times dated 26 August 1992, "Builders complain to OFT over plasterboard prices".

171 Pp. 20.5083 and 16.4205.

172 Pp. 2.175 and 20.5060-5061.

173 P. 20.5023.

174 P. 36.8159.

175 P. 33.7375.

176 P. 16.4319; BG's letter of 22 September 1995 announcing the rise.

177 P. 33.7339; Lafarge internal memo of 13 October 1995, including to the text to be sent to customers.

178 P. 17.4479; Knauf's letter of 27 October 1995 announcing the rise.

179 P. 18.4502.

180 Reply to the statement of objections, p. 45-46.

181 P. 33.7317, Price list 15 - effective November 1995.

182 P. 17.4468.

183 P. 18.4501, Price list 16 - effective 4 August.

184 Pp. 17.4484-4485, Price increase August 1997.

185 P. 16.4296, Price increase merchants and distributors, 27 January 1998.

186 P. 18.4601, Price list 17 - effective 6 April 1998.

187 Pp. 18.4573-4574.

188 p. 18.4614.

189 Statement, pp. 36.8048-8049 ; second reply, pp. 36.8125-8127.

190 Reply to the statement of objections, p. 47.

191 Ibid., p. 96.

192 Reply to the statement of objections, p. 93.

193 Reply to the statement of objections, pp. 33-34.

194 Reply to the statement of objections, p. 35.

195 BPB's reply to the statement of objections, p. 70.

196 Statement by Gyproc, pp. 35.7787-7788; Gyproc's reply to the request for information, p. 39.8618.

197 Statement by [Mr G], forwarded to the Commission following the hearing, pp. 3-4.

198 See the memo referred to in recital (224).

199 Gyproc's reply to the request for information, p. 39.8619.

200 See Commission's request for information of 30 March 2000, especially Annexes 2 and 3, pp. 36.8210-8211 and 36.8206.

201 BPB's reply of 28 April 2000 to the Commission's request for information, p. 36.8234.

202 i.e. Lafarge's 20 million m² plant near Berlin, Lafarge's reply to the statement of objections p.104, BPB's 50 million m² plant in the same region, BPB's reply to the statement of objections p.87.

203 BPB's first reply to the Commission's request for information, p. 5: " probably because of opportunities in eastern Germany and Poland, the German market never stabilised" ; Lafarge's reply, p.104.

204 p. 32.7050.

205 Gyproc's reply to the request for information, pp. 39.8618-8620; see also Gyproc's statement pp. 35.7789-7790.

206 Ibid. p. 32.7050.

207 Gyproc.

208 Invited to comment on this document, Gyproc stated that the figures concerning Knauf, BPB and Lafarge indicated as "estimates" in the note had been obtained from [Mr D] by [Mr E]: Gyproc's reply to the Commission's request for information, p. 39.8619.

209 Ibid.

210 p. 36.7886.

211 The data for Rigips (BPB) and the total market were said to be "based on [Mr U]'s data", while the data for Knauf/Dano, Lafarge and Gyproc were said to be "own estimates".

212 BPB's reply to the Commission's request for information, pp. 36.8003-8005.

213 BPB's reply to the Commission's request for information, pp. 36.8003-8004.

214 BPB, reply to the statement of objections, inter alia p. 21 ; Lafarge, see statement of [Mr G], pp. 3-4 ; Knauf, reply to the statement of objections, p. 34.

215 Statement by [Mr E], p. 35.7801.

216 Ibid., pp. 35.7801-7802.

217 Gyproc's reply to the Commission's request for information, p. 39.8620.

218 Gyproc's reply to the statement of objections, pp. 28-31.

219 BPB's reply to the statement of objections, pp. 21 and p. 70.

220 Statement by [Mr D] sent to the Commission (Annex 50 to BPB's reply to the statement of objections, p. 4).

221 BPB's reply to the statement of objections p.70.

222 Ibid., p. 3.

223 Ibid., p. 103.

224 Statement by [Mr E], p. 35.7800.

225 Knauf's reply to the statement of objections, pp. 34-42.

226 Statement by Mr [G], forwarded to the Commission following the hearing, pp. 3-4.

227 p. 25.5969.

228 Mr [K] was [a high level representative] of Etex and had preceded Mr [E] at Gyproc; Mr [L] is [a high level representative] of Etex.

229 Reply to the first request for information, pp. 36.7987-7990.

230 Statement by [Mr E], p.35.7801.

231 Knauf's reply to the statement of objections, pp.42-44. Gyproc's reply to the statement of objections pp.35-36. Lafarge's reply to the statement of objections pp.105-108. Etex's reply to the statement of objections, pp.24-26.

232 Reply to the statement of objections, p. 69.

233 First request for information, pp. 36.7987.

234 Reply to the statement of objections, pp. 42-44.

235 Ibid., pp. 52-53.

236 Reply to the statement of objections, p. 105-109.

237 pp. 1.39-41 (note written by [Mr D] about the meeting and other conversations on the same subject); p. 1.97 (separate set of notes and calculations made by [Mr D]); pp. 25.5950-5951 (detailed notes written by [Mr E]); p. 32.7053 (note written by [Mr L]).

238 pp. 1.39-41; BPB's first reply to the request for information, p. 36.7991.

239 Statement by [Mr E], p. 35.7791.

240 Gyproc's reply to the request for information, pp. 39.8621-8622.

241 p. 25.5989 and explanations provided by Gyproc: Gyproc's reply to the request for information, pp. 39.8622-8624.

242 Reply to the statement of objections, p. 35.7790.

243 First reply to the request for information, pp. 36.7989-7996.

244 Memo to the Commission dated 17 March 1999, p. 36.8022.

245 pp. 10.2301-2302, 31.6853.

246 for example p. 2.477.

247 p. 32.7050.

248 Ibid.

249 BPB's reply to the statement of objections, p. 74. Some confusion as to the exact date will be noted: a memo from Mr [S] dated 25 June 1998 states that Rigips had withdrawn from the "[Mr U]" system at the end of 1997; p. 5.1211.

250 p. 5.1337.

251 p. 5.1299.

252 p. 31.6864.

253 Reply to the statement of objections, pp. 74-75.

254 pp. 5.1212-1215, 10.2251.

255 pp. 31.6792, 31.6883.

256 p. 31.6880.

257 Reply to the statement of objections, p. 39.

258 Ibid., p.19.

259 p.23.5605.

260 p. 25.5966; the date of 27 January 1997 indicated in the memo is wrong and should read 27 January 1998.

261 Lafarge's reply to the statement of objections, pp. 101-104; BPB's reply to the statement of objections, pp. 71-73.

262 pp. 23.5603-5604.

263 p. 31.6906 ; see also p. 31.6786, also from Lafarge Gips.

264 See in this respect, Joined Cases T-305/94 Limburgse Vinyl Maatshappij v Commission ("PVC II") [1999] ECR II-931 , paragraphs 613 and 633.

265 pp. 31.6764-6766.

266 Ibid.

267 Ibid. The date of 1 January 1996 is a mistake, in fact it was 1 December 1995.

268 Ibid. Handwritten annotation.

269 Reply to the statement of objections, p. 84.

270 Reply to the statement of objections, p. 57.

271 Reply to the statement of objections, p. 31.

272 pp. 1.42, 1.64-65, 25.5970, see recitals (76) et seq.

273 See recital (255).

274 Memo to the Commission attached to BPB's first reply to the Commission's request for information, p. 36.8020.

275 pp. 21.5143-5144.

276 p. 22.5431. 277 Knauf's reply to the Commission's request for information, pp. 37.8307-8308.

278 Reply to the statement of objections, p. 55.

279 Reply to the statement of objections, pp. 60-61.

280 BPB/Rigips had Knauf's letter announcing the February 1995 price increase and Knauf's price list (pp. 12.3319-3320); Lafarge Gips had Knauf's letter announcing the February 1995 price increase and Knauf's price list (pp. 31.6820-6822), as well as copies of Gyproc's and BPB/Rigips's letters and price lists (pp. 31.6816-6817, 31.6827-6828); BPB/Rigips had Knauf's letter of 7 September 1995 announcing the price increase planned for 1 December 1995 (pp. 12.3311), and copies of letters from Danogips (a Knauf subsidiary), Lafarge and Gyproc (the heading "[Mr V] - Vertraulich" on the Knauf letter suggests that it came into the hands of its addressee through a channel that was not part of normal commercial practice; BPB affirmed that it had come from a customer (reply to request for information, pp. 36.8240-8241)); Lafarge Gips had letters from all its competitors announcing price increases on 1 December 1995 (pp. 31.6830-6831, 31.6833-6836); it also had a letter from October 1996 from BPB/Rigips on the price increase of 1 February 1997 (p. 31.6839) and a letter from Gyproc (pp. 31.6837-6838) and Danogips (pp. 31.6809-6810, 31.6837-6838); it had letters from BPB/Rigips and Gyproc on the September 1997 price rise (pp. 31.6840-6845); BPB/Rigips had letters from Gyproc and Knauf on the September 1997 price rise (pp. 10.2293-2294, 10.2298-2299); Knauf had a letter from BPB/Rigips on the September 1998 rise (p.22.5446); and BPB/Rigips had letters from Knauf, Lafarge, Danogips and Gyproc on the September 1998 increase (pp.12.3262-3265).

281 See recital (305).

282 Reply to the statement of objections, pp. 58-59.

283 pp. 12.3328-3329.

284 p. 21.5152.

285 pp. 21.5160, 3.737, 3.739.

286 pp. 12.3203, 12.3207.

287 Reply to the statement of objections, pp. 63-64.

288 Reply to the statement of objections, pp. 118-120.

289 p. 31.6825.

290 p. 31.6827.

291 p. 31.6826.

292 Reply to the statement of objections, pp. 118-119.

293 p. 21.5160.

294 p. 50.11182.

295 p. 12.3203.

296 Reply to the statement of objections, p. 119.

297 pp. 12.3203, 12.3207.

298 pp. 12.3294, 12.3300-3305, 12.3307-3308, 31.6829.

299 pp. 21.5108, 21.5116.

300 pp. 31.6839.

301 p. 31.6809.

302 p. 31.6810.

303 p. 31.6837-6838.

304 pp. 29.6481-6482.

305 Reply to request for information of 21 September 1999, p. 38.8576.

306 Reply to the statement of objections, p 124.

307 Ibid., pp. 124-125.

308 Ibid.

309 Reply to the statement of objections, pp. 60-63.

310 Reply to the statement of objections, p. 64.

311 Statement of 28 May 1999, pp. 36.8062-8063.

312 Reply to the statement of objections, pp. 62-63.

313 pp. 31.6764-6766.

314 Reply to the statement of objections, pp. 122-123.

315 P.31.6942.

316 pp. 31.6876-6877.

317 p. 5.1270.

318 pp. 31.6840-6845, 10.2293-2294, 10.2298-2299.

319 pp. 10.2277-2278, 10.2282, 10.2286-2287.

320 p. 22.5206.

321 Perhaps during the German Plasterboard Association (IGG) meeting held on 26 June 1997 in Frankfurt, p. 37.8415.

322 Knauf's reply to the request for information, p. 37.8275.

323 See its reply of 14 September 1999 to the request for information, pp. 37.8297-8299.

324 Reply to the statement of objections, p. 65.

325 Reply to the statement of objections, p. 127.

326 pp. 31.6801-6803.

327 p. 31.6803.

328 Reply to the statement of objections, pp. 127-128.

329 Reply to the statement of objections, p. 66.

330 p. 10.2275.

331 pp. 5.1293-1294.

332 p. 5.1298.

333 pp. 31.6781-6782.

334 p. 22.5218.

335 pp. 10.2198, 10.2201.

336 p. 12.3259.

337 p. 22.5446.

338 See recital (291).

339 p. 22.5432.

340 p. 34.7467.

341 pp. 12.3261-3265, 26.6224-6225.

342 pp. 22.5431.

343 pp. 22.5438.

344 pp. 31.6954-6955.

345 pp. 31.6951-6952.

346 pp. 31.6956-6957.

347 pp. 10.2144, 10.2147.

348 See also Lafarge Gips internal memo on this increase, p. 31.6962.

349 p. 31.6764.

350 Reply to the statement of objections, pp. 129-131.

351 p. 31.6953.

352 Reply to the statement of objections, p. 130.

353 Ibid., p. 130.

354 p. 41.9113.

355 p. 31.6964.

356 Reply to the statement of objections, p. 131.

357 Reply to the statement of objections, p. 66.

358 Reply to the statement of objections, p. 67-68.

359 pp. 31-35.

360 Reply to the statement of objections, p. 35.

361 Reply to the statement of objections, p. 39.

362 BPB's reply to the statement of objections, p. 6 et seq. and especially NERA study; Gyproc's reply to the statement of objections, p. 29 et seq.; Lafarge's reply to the statement of objections, p. 29 et seq.; Knauf's reply to the statement of objections, p. 4 et seq. and especially the market analysis "Gutachterliche Stellungnahme" by University Professor [Mr ...] and Business Graduate [Mr ...].

363 BPB's reply to the statement of objections, pp. 6, 32 et seq.; Lafarge's reply to the statement of objections, pp. 9, 16 et seq.;. Knauf's reply to the statement of objections, p. 17 et seq.; Gyproc's reply to the statement of objections, p. 29 et seq.

364 Lafarge's reply to the statement of objections, p. 9 and Knauf's reply to the statement of objections, p. 4 and above-mentioned market analysis.

365 Case T-327/94 SCA Holding v Commission [1998] ECR II-1373, paragraph 142 and Case T-348/94 Enso Española v Commission [1998] ECR II-1875, paragraph 306.

366 See recitals (478) to (480).

367 Joined Cases 56/64 and 58/64 Consten and Grundig v Commission [1966] ECR 299, 342; Case C-49/92 Commission v Anic Partecipazioni [1999] ECR I-4125, paragraph 99 and Case C-199/92 P Hüls v Commission [1999] ECR I-4287, paragraph 178; Joined Cases T-39/92 and T-40/92 CB and Europay v Commission [1994] ECR II-49, paragraph 87.

368 Case T-148/89 Tréfilunion v Commission [1995] ECR II-1063, paragraph 79.

369 See recital (71).

370 OJ C207, 18.7.1996, p. 4

371 Joined Cases T-25-95 etc. Cimenteries CBR and others v Commission [2000] ECR II-491, paragraphs 1837 and 1838.

372 Case T-141-94 Thyssen Stahl v Commission [1999] ECR II-347, paragraph 175 and Case T-311-94 BPB de Eendracht v Commission [1998] ECR II-1129, paragraph 201.

373 Joined Cases T-202/98 etc. Tate & Lyle v Commission [2001] ECR II-2035, paragraphs 164 to 166. The Commission would also draw attention here to the various judgments relating to the cartonboard cartel, where the Commission's decision was based mainly on statements by Stora, which had its fine reduced, and where the court did not accept the arguments questioning the credibility of these statements advanced by the parties who were seeking a larger reduction in their fine; see Case T-347-94 Mayr-Melnhof v Commission [1998] ECR II-1751, paragraphs 82 et seq.

374 Case T-9-99 HFB Holding für Fernwärmetechnik Beteilungsgesellschaft mbH & Co. KG and others v Commission[2002] ECR II-1487, paragraph 212; Joined Cases 40-48-73 etc. Suiker Unie and others v Commission [1975] ECR 1663, paragraphs 173 and 174; and Case C-199-92P Hüls v Commission, cited above, paragraphs 159 and 160.

375 PVC II judgment, cited above, paragraph 715; upheld by the judgment of the Court of Justice of 15 October 2002 in Joined Cases C-238/99P etc. LVM and others v Commission, not yet reported.

376 Case 48-69 Imperial Chemical Industries v Commission [1972] ECR 619, paragraph 64 and Case C- 199-92P Hüls v Commission, cited above, paragraph 158.

377 Case T-7-89 Hercules v Commission [1991] ECR II-1711, paragraphs 255 and 258-261.

378 Case C-199-92P Hüls v Commission, cited above, paragraphs 158 to 166, and Case T-9-99 HFB v Commission, cited above, paragraphs 213 and 216.

379 Case C-199-92P Hüls v Commission, cited above, paragraphs 163 to 165, and Case T-9-99 HFB v Commission, cited above, paragraph 217.

380 Case C-49-92P Commission v Anic Partecipazioni SpA, cited above, paragraph 81.

381 Joined Cases T-25-95 etc. Cimenteries CBR and others v Commission, cited above, paragraphs 4025 et seq.

382 Case T-7-89 Hercules v Commission, cited above, paragraph 264.

383 Joined Cases T-305-94 etc. PVC II, cited above, paragraph 696; upheld by the judgment of the Court of Justice of 15 October 2002 in Joined Cases C-238-99P etc. LVM and others v Commission.

384 Case C-49-92P Commission v Anic Partecipazioni SpA, cited above, paragraphs 131 and 132.

385 Ibid., paragraph 43.

386 Ibid., paragraph 83.

387 See recital(489).

388 Case T-1-89 Rhône Poulenc SA v Commission [1991] ECR II-867, paragraph 126.

389 BPB's second reply to the request for information, p. 36.8129.

390 Knauf's reply to the statement of objections, pp. 29-30.

391 BPB's reply to the statement of objections, point 6.2.3.

392 See recitals (57) and (61).

393 BPB's reply to the statement of objections, p. 108.

394 Case T-9-99 HFB v Commission, cited above, paragraph 200.

395 Knauf's reply to the statement of objections, p. 31.

396 See recitals (58) and (59).

397 See recital (106). 398 See recitals (111) and (113).

399 Knauf's reply to the statement of objections, p. 22.

400 Case T-141-94 Thyssen Stahl AG v Commission , cited above, paragraph 404.

401 Ibid., paragraph 675.

402 See recitals (162) and (163).

403 Case T-141-94 Thyssen Stahl AG v Commission, cited above, paragraphs 403 et seq.

404 Case C-7-95 John Deere Ltd v Commission [1998] ECR I-3111, paragraphs 88 to 90.

405 This characteristic of the market is confirmed by the undertakings to which this Decision is addressed; see BPB's reply to the statement of objections, p. 26.

406 See recital (139).

407 Case C-89-85 Ahlström v Commission [1993] ECR I-1307, paragraph 175 and Case T-148-89 Tréfilunion SA v Commission, cited above, paragraph 79.

408 Case T-9-99 HFB v Commission, cited above, paragraphs 216 and 217.

409 See recitals (127) et seq., in particular (136) to (140).

410 See recitals (158) to (161); the Commission would point out that there are certain factual elements which indicate that Gyproc took part in conduct which may be characterised as anticompetitive before the Versailles meeting took place (see recital (471)). However, with regard to this conduct, the Commission is not able to furnish the necessary proof that Gyproc knew or should have known that it was subscribing to the London agreement on the exchange of information between [high level representatives] on the four major markets.

411 See recitals (171) et seq.

412 See recitals (447) and (448).

413 See, in particular recitals (229), (251) and (260).

414 See recital (219).

415 See recital (216).

416 Case T-23-99 LR AF v Commission, [2002] ECR II-1705, paragraph 39.

417 Gyproc's statement to the Commission, p. 35.7790.

418 See recital (265).

419 Case T-9-99 HFB v Commission, cited above, paragraph 206.

420 See recital (272).

421 Joined Cases T-25-95 etc. Cimenteries CBR and others v Commission, cited above, paragraph 1849.

422 Joined Cases T-202-98 etc. Tate & Lyle v Commission, cited above, paragraph 58.

423 Ibid., paragraphs 60 and 61.

424 See recital (211).

425 See in particular recitals (441), (452), (461), (469), (470), (475) and (477).

426 Joined Cases 29-83 and 30-83 CRAM & Reihnzink v Commission [1984] ECR 1679, paragraph 26 and Case 96-82 IAZ v Commission [1983] ECR 3369, paragraphs 22 et seq.

427 Joined Cases C-215-96 and C-216-96 Bagnasco and others [1999] ECR I-161, paragraphs 47 and 48.

428 Case C-306-96 Javico International and Javico AG v Yves Saint Laurent Parfums SA [1998] ECR I- 1983, paragraphs 16 and 17, and Case T-374-94 European Night Services Ltd and others v Commission [1998] ECR II-3141, paragraph 136.

429 Case T-13-89 Imperial Chemical Industries v Commission [1992] ECR II-1021, paragraph 304.

430 Joined Cases 209 to 215 and 218-78 Van Landewyck and others v Commission [1980] ECR 3125, paragraph 170.

431 Decision, recitals (136), (455) and (456).

432 On the fact that Gyproc participated in anticompetitive conduct before that date, see footnote 410.

433 See recitals (158), (161) and (457).

434 Case T-9-99 HFB v Commission, cited above, paragraph 231; Case T-295-94 Buchmann v Commission [1994] ECR II-813, paragraph 121; and Case T-310-94 Gruber + Weber v Commission [1998] ECR II-1043, paragraph 140.

435 Case C-49-92P Commission v Anic Partecipazioni SpA, cited above, paragraphs 83 and 203.

436 See notably Case T-9-99 HFB v Commission, cited above, paragraphs 54 et seq.

437 Ibid., paragraphs 66-67; see also Case T-156-94 Aristrain v Commission [1999] ECR II-645, paragraphs 119 et seq.

438 See Joined Cases T-25-95 etc. SA Cimenteries CBR and others v Commission, cited above, paragraph 5040.

439 Replies of 19 September and 9 October 2002 to the Commission's request for information.

440 Lafarge Plâtres International SA (formerly known - until 10 August 1995 - as Compagnie du Plâtre) changed its name on 15 August 2000 to Lafarge Gypsum International.

441 Case C-286-98P Stora Kopparbergs Bergslags v Commission [2000] ECR I-9925, paragraph 29; Case 107-82 AEG v Commission [1983] ECR 3151, paragraph 50; Case T-65-89 BPB v Commission [1993] ECR II- 389, paragraph 149; and Joined Cases C-238-99P etc. LVM and others v Commission,cited above, paragraphs 960-961.

442 See recital (43) and Lafarge's Internet site: www.Lafarge.com..

443 Note from [Mr G] dated 8 August 2001 addressed to Mr [..., executive in Legal Department of Lafarge] and sent to the Hearing Officer under cover of a letter from Lafarge's legal adviser.

444 See recitals (80), (290) and (302).

445 Lafarge's reply to the statement of objections, p. 174.

446 See recital (133).

447 See recital (134).

448 Lafarge's reply to the statement of objections, p. 174.

449 Reply of 26 August 2002 to the Commission's request for information, p.10.

450 Ibid.

451 Joined Cases T-25-95 etc. SA Cimenteries CBR and others v Commission, cited above, paragraph 5040.

452 Etex's reply to the statement of objections, p. 24.

453 See Case 82/96 IAZ v Commission , cited above, paragraph 45, and Joined Cases T-45/98 and T-47/98 Krupp Thyssen v Commission [2001] ECR II-3757, paragraph 200.

454 Reply to the statement of objections, pp. 40-47.

455 OJ C 9, 14.1.1998.

456 Reply to the statement of objections, pp. 41-42.

457 Reply to the statement of objections, pp. 123-125.

458 Reply to the statement of objections, pp. 8, 49 and 58-59.

459 Ibid., 125.

460 Reply to the statement of objections, pp. 68-69.

461 Reply to the statement of objections, pp. 42-44.

462 Case C-7/95P John Deere Ltd v Commission, cited above, paragraphs 88-90.

463 See, for example, Case T-354-94 Stora Kopparbergs Bergslags AB v Commission, [1998] II-2111, paragraph 170; see also the judgment in PVC II, cited above, paragraphs 745-748.

464 See the judgment in PVC II, cited above, paragraph 748.

465 Gyproc's reply to the statement of objections, p. 44.

466 Reply to the statement of objections, pp. 46-47.

467 Case T-31-99 ABB Asea Brown Boveri Ltd v Commission, [2002] II-1881

468 As indicated in recital (509), the turnover of Lafarge taken into account for purposes of the imposition of a fine is that of the "undertaking" within the meaning of Article 81(1) of the Treaty controlled by Lafarge, this being the undertaking that committed the infringement, that is to say in this case the worldwide turnover achieved by all the companies of the group controlled by Lafarge.

469 See also Commission Decision 2002-405-EC of 20 June 2001 relating to a proceeding pursuant to Article 82 of the EC Treaty - Michelin, OJ L 143, 31.5.2002, p. 1, recitals 361 - 363

470 Reply to the statement of objections, pp. 127-128.

471 Reply to the statement of objections, pp. 48-49.

472 Ibid., p. 49.

473 Reply to the statement of objections, pp. 14-23 for the cartel at the European level of [high level representatives], pp. 27-35 for the infringement in relation to the German market, pp. 25-26 for the UK market, p. 48.

474 Ibid., p.50.

475 Reply to the statement of objections, p.122-123.

476 Ibid., p. 123.

477 Ibid.

478 Ibid.

479 Ibid., p.126.

480 Ibid., pp. 121-122.

481 Ibid., pp. 121-123, 125-126.

482 Ibid., pp. 19-23.

483 Reply to the statement of objections, pp. 51-53.