EC, December 7, 1988, No 89-93

THE COMMISSION OF THE EUROPEAN COMMUNITIES,

Having regard to the treaty establishing the European Economic Community, Having regard to Council Regulation No 17 of 6 February 1962, first Regulation implementing Articles 85 and 86 of the treaty (1), as last amended by the Act of Accession of Spain and Portugal, and in particular Articles 3 and 15 thereof, Having regard to the application within the meaning of Article 3 of Regulation No 17 made to the Commission on 31 October 1986 by Industria Vetraria Alfonso Cobelli of Reggio Calabria asking the Commission to find that Società Italiana Vetro-SIV SpA, Fabbrica Pisana SpA and Vernante Pennitalia SpA had committed infringements of the Community competition rules, Having regard to the decision taken by the Commission on 15 October 1987 to initiate own-initiative proceedings and proceedings upon application in this case, Having given the undertakings concerned the opportunity of being heard on the matters to which the Commission has taken objection, in accordance with the provisions of Article 19 (1) of Regulation No 17 and with Commission Regulation No 99-63-EEC of 25 July 1963 on the hearings provided for in Article 19 (1) and (2) of Council Regulation No 17 (2), Having consulted the Advisory Committee on Restrictive Practices and Dominant Positions, Whereas :

I. THE FACTS

Introduction

(1) This decision follows on from investigations undertaken in July and October 1986, pursuant to Article 14 (2) of Regulation No 17, into three Italian flat-glass producers and one wholesaler and from investigations undertaken in January 1987, pursuant to Article 14 (3) of Regulation No 17, into the three Italian producers. Between the investigations carried out in 1986 and those carried out in 1987, Industria Vetraria Alfonso Cobelli of Reggio Calabria made application to the Commission on 31 October 1986, pursuant to Article 3 of Regulation No 17, for a finding that infringements of the competition rules had been committed by Società Italiana Vetro, Fabbrica Pisana and Vernante Pennitalia.

A. The market

1. The product

(2) The product to which this decision relates is flat glass in all its varieties. There are three types of flat glass produced using different manufacturing processes :

- drawn glass, which takes the form of a colourless, transparent sheet, used for the manufacture of window panes,

- cast glass, which takes the form of an uneven, translucent but not transparent sheet, obtained by rolling,

- plate glass, in the form of a transparent sheet whose surfaces are almost perfectly parallel,

which may be obtained by continuous casting or by the float-glass process. The float-glass process is the one most widely used in producing glass.

(3) Float glass began to be produced on an industrial scale in the early 1960s; because of its characteristics and its relatively low production costs in relation to its quality, it rapidly replaced the other types of glass. At present, more than 90 % of flat glass is manufactured using the float-glass process.

(4) Flat glass may be used unprocessed (for example, window panes) or processed (for example, automotive glass, building glass and mirror manufacturing); between 70 % and 80 % of glass produced is processed, either directly by the producers or by specialists. Flat glass intended for the motor vehicle industry, which is exclusively processed glass, is treated only by the glass producers, whereas flat glass intended for building and furnishing is processed either by the glass producers, if they are vertically integrated, or by independent processors.

(5) Within the industry, two markets may be distinguished : the automotive, or more generally the transport, market and the non-automotive or building market. The first market comprises essentially glass for use in the motor vehicle industry and, to a lesser extent involving small quantities, glass intended for the railways, ships and travelling cranes. The second market comprises glazing for use in building, furniture glass, mirrors, glass for use in household electrical goods, etc.

The automotive market is supplied directly by the glass producers; the non-automative market is supplied to a lesser extent by the glass producers and more usually through wholesalers, processing wholesalers and processors.

(6) The tables in Annex 1 present figures for the Italian flat-glass market in recent years.

Italy is, together with Germany, one of the largest markets in Europe : the Italian market accounts for some 20 % of the European automotive market and roughly the same percentage of the European non-automotive market.

2. Supply

(7) During the period under examination, an average of 79 % of Italian demand for non-automotive glass and an average of 95 % of Italian demand for automotive glass were met by the three Italian producers to whom this decision is addressed.

(8) Fabbrica Pisana SpA (FP) is a subsidiary of the Saint-Gobain Group, which is one of the largest industrial groups in the world. In Italy, through the intermediary of Fabbrica Pisana or other group companies, Saint-Gobain owns the following companies involved in the glass industry : Luigi Fontana SpA, Balzaretti e Modigliani SpA, Home Glas SpA, Saint-Gobain Italiana Auto SRL, Toscana Glas SpA and Flovetro SpA. Luigi Fontana SpA is the largest processing wholesaler on the Italian market. Through the intermediary of Toscana Glas, FP owns a float-glass plant at Pisa and a float-glass plant shared with SIV and operated by Flovetro at San Salvo and has remained the only producer of cast glass in Italy.

(9) Società Italiana Vetro SpA (SIV) is a company controlled by the State holding company EFIM. It owns a float-glass plant at San Salvo and a float-glass plant shared with FP and operated by Flovetro at San Salvo. Since taking over control of Veneziana Vetro in 1986, SIV has closed its drawn glass furnaces and replaced them by a float-glass plant, which began production at the end of 1987.

SIV owns the following companies operating in the glass industry: Vetro Europa SpA in Italy and SIVESA in Spain, which produce automotive glass; Società Vetri Speciali at San Salvo, which produces reflective glasses; and the glass marketing companies SIV-Deutschland in Frankfurt and SIV-France in Paris. Since 1986 it has controlled the Belgian automotive glass producing company Splintex SA, a joint subsidiary of SIV and Glaverbel, and has transferred to Glaverbel control of its mirror producing company ILVED.

(10) Vernante Pennitalia SpA is a subsidiary of the American group PPG-Industries Inc., Pittsburgh.

It owns a float-glass plant at Cuneo and another at Salerno and controls the company Pennitalia Securglass, which produces automotive glass. In 1982, PPG-Industries Inc. bought Boussois from BSN; Boussois has a long tradition in glass manufacture and a strong position on the French market for automotive and non-automotive glass.

(11) The market shares of the three companies, calculated on the basis of the figures set out in Annex 2, are as follows:

(in %)

1.2,3.4,5.6,7.8,9.10,11 // // // // // // // // 1982 // 1983 // 1984 // 1985 // 1986 1.2.3.4.5.6.7.8.9.10.11 // // Auto- motive // Non-auto- motive // Auto- motive // Non-auto- motive // Auto- motive // Non-auto- motive // Auto- motive // Non-auto- motive // Auto- motive // Non-auto- motive // // // // // // // // // // // // FP // [...] (1) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // SIV // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // VP // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // // // // // // // // // // //

(1) In the published version of the decision, some information has hereinafter been omitted, pursuant to the provisions of Article 21 of Regulation No 17 concerning non-disclosure of business secrets. At European level, the market shares of the three companies or of the groups to which they belong are estimated as follows:

(%)

1.2.3 // // // // // Auto- motive // Non-auto- motive // // // // Saint-Gobain // [...] // [...] // SIV (including Splintex) // [...] // [...] // PPG-Industries // [...] // [...] // // //

(12) At European level, the flat-glass market is dominated by a tight oligopoly comprising the abovementioned groups together with Pilkington, Glaverbel (Asahi group) and Guardian. These are groups with integrated activities upstream from glassmaking (refractory products, siliceous sand pits, basic chemicals) and downstream (processing of glass fibre and flat glass).

The geographical spread of the six groups in Europe is as follows:

- Saint-Gobain operates in France, Belgium, Germany, Italy, Spain and Austria and owns a total of 10 1 / 2 float-glass plants,

- Pilkington operates in the United Kingdom, Germany, Sweden and Austria and owns a total of seven float-glass plants,

- PPG operates in Italy and France and owns a total of four float-glass plants,

- Glaverbel operates in Belgium, the Netherlands and Spain and owns three float-glass plants,

- SIV operates in Italy, France, Germany and Spain and owns two-and-a-half float-glass plants,

- Guardian owns one float-glass plant in Luxembourg and another in Spain.

According to the figures of the European flat-glass producers' association, GEPVP, Community production was amply sufficient to meet demand in the period from 1980 to 1987, with exports to non-Community countries having consistently exceeded imports from outside the Community. According to the GEPVP's forecasts, drawn up in June 1986 for the Community of Ten and valid until 1996 and drawn up in June 1987 for the Community of Twelve and valid until 1989, Community demand for glass will, on the most optimistic assumptions, increase by only 1 to 3 % a year. The GEPVP thus anticipates that excess production capacity, which was high in the early 1980s, will persist in the years ahead, albeit at a lower level.

According to the information provided by FP, the average investment involved in a float-glass plant of 150 000 tonnes per year amounts to some ECU 70,5 million to 86 million on an existing site and twice that amount on a new site. The possibility of other producers entering the market in the foreseeable future can therefore be ruled out.

(13) With regard to the automotive market, processing in all its aspects (laminated glass, vertically toughened glass and horizontally toughened glass) requires different production lines for each technique (side windows, laminated windscreens, heated rear windows), in addition to the tools needed for the initial cutting and shaping and in addition to moulding and toughening. Each production line incorporates a wide variety of operations which differ from product to product. The production lines are specially designed in the light of the shapes and technical specifications required by demand. Technical obsolescence is therefore fairly rapid and occurs after seven to eight years, according to technical developments in motor vehicle manufacturing.

The shapes and technical specifications required by motor vehicle manufacturers involve considerable investment both at production level and in terms of research and development. According to the information provided by FP, the cost of a processing line is difficult to calculate, in view of the specific character of each plant, but may be put at ECU 40 million for the manufacture of 650 000 motor vehicle fittings a year. Few firms are therefore capable of bearing the costs and risks of processing glass for motor vehicles.

3. Demand

(14) The customer of non-automotive flat-glass producers are wholesalers and processors. Some 40 % of demand is accounted for by processors purchasing directly from producers, while the remaining 60 % is accounted for by wholesalers. Wholesalers themselves process at least half of the glass they purchase, with the bulk of the remainder being sold directly to final consumers and a smaller proportion to small processors.

Processing consists in the manufacture of safety glass, insulating glasses, mirrors, etc. Processors are often in competition with flat-glass producers who process glass themselves. Sometimes, processors are dependent on the transfer of technology from glass producers and thus manufacture processed products under licence granted by their suppliers.

(1) OJ No 13, 21. 2. 1962, p. 204/62.

(2) OJ No 127, 20. 8. 1963, p. 2268/63.

(15) The customers of automotive-glass producers are the car manufacturers. Before being sold to the motor vehicle manufacturers, the glass is processed by the glass producers or by their subsidiaries in the light of the designs and technical specifications required by the motor vehicle manufacturers. Collaboration between glass producers and motor vehicle manufacturers takes place as follows :

When a manufacturer's design department designs a new model, one or at most two glass suppliers are asked to produce individual parts. The geometry of the glass parts evolves constantly, firstly at the prototype stage, and subsequently when the model is manufactured in limited numbers in the pilot plant. At this stage, the glass suppliers are paid on the basis of a 'prototype' rate, which is indepedent of the model concerned and is calculated solely by reference to the type of glazed surfaces concerned.

When it is decided to launch the new model commercially, the manufacturer uses other glass suppliers, replacing in full or in part the initial glass suppliers. The new suppliers then take over the technical and quality specifications developed by the initial suppliers. It is at the stage where the new model is marketed that glass suppliers and motor vehicle manufacturers negotiate prices and quantities, normally on an annual basis.

(16) The non-automotive glass market saw a fall in demand in the period 1979 to 1983, reflecting the recession in the European economy. As from 1984, demand picked up again, allowing producers, particularly as from the second half of 1985, to increase their prices considerably. As was stated in paragraph 12, demand for glass is expected, according to the industry's forecasts, to increase by between 1 and 3 % a year over the next decade.

(17) The market in automotive glass is closely bound up with the trend of car production, which, following the second oil shock, went through a period of recession in Italy and the rest of Europe up to 1984. Only in 1985 did production begin to pick up again somewhat, with the recovery gathering further momentum in 1986 and 1987. According to the forecasts, the growth in demand for automotive glass over the next decade is expected to be lower than that for non-automotive glass.

B. The behaviour of the firms

4. The market in non-automotive glass

(a) Identical price

(18) The three Italian producers communicated identical price lists to their Italian customers on dates which were close to one another and in some cases on the same days (see Annex 3). The initiative in altering the price lists was not always taken by the same producer, but sometimes by one and sometimes by another of the three producers.

(19) The closeness or identity of the dates on which the prices were communicated to customers was not, as claimed, fortuitous, since the three producers sent at least half of the price lists examined on the same day or within a short period of time.

The uniformity of the upward adjustments in prices is not due to the homogeneity of the products and the existence of an oligopoly situation on the market. As FP and VP have admitted in their replies, any producer deciding of its own accord to increase prices always runs the risk of not being followed by its competitors and thus of losing orders and market shares. To avoid such a risk and to avoid what happened to VP - which, in November 1981, had to withdraw the price increase announced the previous September, since FP and SIV did not apply VP's price increase - each producer takes the precaution of checking whether the others are prepared to increase prices as well. Following VP's experience in 1981, there has never again been a case of a price rise not being immediately matched by all the other producers, and this despite the fact that, at least until the end of 1983, demand was falling and that, in the years thereafter, excess production capacity continued to persist, despite the growth of the market.

(b) Identical discounts

(20) Identical discounts on the listed prices were granted in accordance with the categories or levels in which customers were classified. Whereas prices were communicated to all customers without distinction, the classification of customers by category or level and the list of discounts were not divulged.

The percentage discounts granted on the listed prices were as follows:

1.2.3 // - 1983: // A: // 10+5+15+5 // // B: // 10+5+15 // // C: // 10+5+8 // - 1984: // A: // 10+5+15+5 // // B: // 10+5+15+3 // // C: // 10+5+15 // - 1985 and 1986: // A: // 10+5+15+5+5+4 // // B: // 10+5+15+5+5+2,5 // // C: // 10+5+15+5

(21) The fact that the scales of the discounts granted to customers by the three producers coincide is not due to objective factors, such as, for example, the quantities bought from a producer or the terms of payment. As was admitted at the hearing, each customer normally buys the bulk of its needs from one supplier and spreads the remainder of its requirements amongst the other producers, in the knowledge that, whatever the quantities it purchases from the various suppliers, it will always obtain the same terms. The fact that a customer which purchases large quantities from a supplier obtains the same discounts as a customer which purchases small quantities cannot be regarded as normal business practice, especially where the products in question are homogeneous and commonplace, with the only possibly competition, since differences in quality or other differences are not possible, being price competition and competition in respect of terms of sale.

(c) Identical classification of the main customers by category or level

(22) The main customers, i.e. those which account for more than half of demand, were classified in the same category or level, whenever they obtained their supplies from any of the producers (see Annex 4 (1). It may be seen from the table in Annex 4 that there are two exceptions amongst the first 20 customers and a number of exceptions amongst the smaller customers classified in lists A and B. The exceptions are due to the fact that some producers, such as VP, aim to give preferential treatment to processors or to the fact that each producer tries to give preferential treatment to certain customers in certain regions, as may be seen from a handwritten note on a meeting between SIV and FP on 30 January 1985 (see paragraph 27 below).

The classification of customers by category or level was not dependent on their purchases from a given producer, but on each customer's total purchases from all producers. The classification was updated several times during the period under consideration (see table in Annex 5, which it was possible to draw up only with reference to the years as from 1984).

(23) VP denies that it classified customers into categories. The fact that there was no formal classification by categories does not mean that VP did not put the system into practice. What matters is not the form of the behaviour, but whether it actually took place. Furthermore, VP admits in its reply to the Statement of objections not only that its price lists and basic discounts were the same as those of its competitors, but also that it applied the same pricing policy to its customers as did its competitors.

The three producers argue that the classification of customers is based on objective criteria specific to each firm, such as the volume of purchases, solvency, management and development.

If each firm asserts that it has its own internal and secret criteria for classifying customers, customers cannot be the means of circulating information from one producer to another, since customers cannot know the criteria, but only the discounts granted to them. It is therefore not probable that each customer will pass on internal, secret information to each producer in order to allow the other producers to adapt their customer lists within a brief or relatively brief period of time, or the same day, as was the case with the changes in the lists of FP and VP on 1 January 1984, 1 January 1985 and 1 August 1985.

The tables contained in FP's answer to the Statement of objections and VP's comment that it granted high discounts to most customers confirm what the Commission asserted in its Statement of objections, namely that the differences of classification between FP and SIV are minimal (one customer out of 23 in the case of List A dating from 1985, and three customers out of 25 in the case of List A dating from 1986) and that most of the difference evident in the case of VP is due to the fact that VP, which unlike FP and SIV, does not have any processing activity, gives preferential treatment to a number of processors, whereas in the case of the main wholesalers it applies the same pricing policy as its competitors.

The table in Annex 4 shows that the main wholesalers are classified in the same category by the three producers.

FP's claim that the main classification criterion is the turnover achieved with it is contradicted by the following facts :

Comparison of the tables provided by FP in its reply shows that customers which purchased larger quantities from FP are classified in category B, whereas customers which purchased smaller quantities are classified in category A. In addition, there are nine wholesalers out of 23 in 1985 [...] and eight wholesalers out of 25 in 1986 [...] classified in category A, even though their purchases do not fall within the turnover brackets indicated by FP in the tables in its reply.

(d) Elements of concerted practices between producers

(24) The uniformity of prices and of discount scales and the uniform classification of the main customers by category or level are the result of concerted practices between the producers agreed on directly during talks, meetings or contacts or through the intermediary of the spokesman of the main customers.

The documentary evidence of such concerted practices, which was communicated with the Statement of objections to the firms to which this decision is addressed, is set out below.

(25) During the meeting held on 12 July 1983 with FP, Socover complained to FP's representative that SIV and VP had not fulfilled the undertakings to wholesalers regarding discounts and terms of payment (handwritten note of Socover dated 12 July 1983). The parties have given three different interpretations of the note. According to Socover's written Statement, produced by VP, Socover was complaining that SIV and VP had not fulfilled the undertakings entered into to try and obtain better terms from FP. According to FP, by contrast, it was FP which was complaining because SIV and VP had not complied with the undertakings. According to SIV, lastly, the undertakings referred to in the note are undertakings entered into by each producer vis-à-vis Socover.

The note clearly mentions undertakings on discounts ('discount scales and super credit') by the three producers vis-à-vis all wholesalers. Socover, acting as spokesman for the wholesalers, refers to non-compliance not with the undertakings given to itself, but with those given to all wholesalers, as is confirmed in FP's reply : 'Fabbrica Pisana had learned, through glass operators, that SIV had, at that time, glass in stock with its main wholesale customers: the latter where thus able to obtain glass at the old price in May and June. By way of response, Vernante Pennitalia had lowered its price for Planilux 4 mm to Lit 4 640, selling it as "second choice"'.

(26) On 30 October 1984, a meeting took place between SIV and FP in Rome. During the meeting, the following decisions were adopted : cast glass was not to be sold below FP's price; a common policy was adopted for triple glass (handwritten note by SIV dated 30 October 1984).

(27) During the meeting held in Rome on 30 January 1985, SIV and FP mutually noted that they were complying with the agreements in substance and in particular that they were complying with the prices for clear glass. However, as regards coloured and laminated glass, SIV accused FP of breaching the provisions of the cartel by resorting to various devices, such as supplementary discounts granted to certain wholesalers. SIV proposed that each producer should have its preferential customers to which to grant supplementary discounts (FP's handwritten note dated 30 January 1985). For its part, SIV undertook, with regard to discounts and customers classified in category 'super A', to apply the same terms and conditions as the other producers (SIV's handwritten note dated 30 January 1985).

(28) On 28 March 1985, FP undertook, during a meeting, to sell SIV 1 000 tonnes a month of cast glass. The two firms agreed that SIV would not sell the cast glass to 16 listed customers reserved for FP (FP's handwritten note dated 28 March 1985).

These 16 reserved customers are not the Fontana affiliates, as FP claims in its reply to the Statement of objections, because there are eight Fontana affiliates (Ovest, Est, Quentin, Centro, Vesuviana, Adriatica, Sud, Sarda) and not 16, and because it is illogical to reserve customers that are by definition already reserved since they are totally controlled.

(29) On 12 April 1985, Socover discussed with FP the possibility of appointing sales agents in each region. At the same meeting, Socover and FP also discussed the apportionment of orders among the producers on the historical basis of the last two years (Socover's handwritten note dated 12 April 1985).

(30) At the meeting between Socover and FP held on 10 July 1985, an FP manager informed Socover that FP was in the process of consulting with the other producers on a price increase of 7 to 8 % for clear glass as from October and that the prices for coloured and cast glass would not be changed (Socover's handwritten note dated 10 July 1985).

A 7,5 % price increase was indeed announced to customers by the three producers in October.

(31) During the meeting between Socover and SIV held on 23 July 1985, SIV's commercial director informed Socover that the producers were 'in the process of deciding on' a 10 % price increase for float glass as from August and that they would grant an end-of-year premium of 3 % to a number of preferential customers (Socover's handwritten note dated 23 July 1985). An 8 % price increase was indeed notified to customers by the three producers towards the end of July. The 3 % annual premium was indeed granted by the three producers to a number of preferential customers.

(1) This Annex refers to 1985 and 1986. In the case of earlier years, it was not possible to draw up tables, since not all the producers were able to provide full information.

(32) On 12 March 1986, VP's commercial director and director of sales for Italy met Socover in Milan and announced that VP would also adopt the price increases envisaged by FP and SIV as from 1 April 1986 (Socover's handwritten note dated 12 March 1986).

In its reply to the Statement of objections, VP produced a written Statement by Socover on the interpretation of this note by Socover. According to the Statement, Socover informed VP of the fact that SIV and FP had increased their prices with immediate effect and VP replied that it would probably follow suit. It should be noted, firstly, that, on 12 March 1986, Socover could perhaps have had knowledge of FP's decision, since FP had sent out the new price lists on 10 March 1986, but it could not have had knowledge of a decision that had not yet been taken by SIV, which sent out its price lists on 14 March 1986. Then again, VP could not react immediately on the basis of two pieces of information one of which could perhaps have been known and the other not : alignment can be announced only where the reference data are known.

More generally, VP argues that it is concerned only by the references contained in Socover's notes of 12 July 1983 and 12 March 1986, on which it has given its interpretations. The Commission cannot share this view. Socover's notes are explicit, since they always mention the decisions adopted by the three producers. As far as the notes of SIV and FP are concerned, SIV's note dated 30 January 1985 does not mention FP, but the others, i.e. the three producers, as a reference parameter for the discounts ('we will apply the same terms as the others'); FP's note dated 30 January 1985 does not refer to agreements between FP and SIV, but to the existence of a cartel between the producers. The only conclusion that may be drawn from the above is that VP participated less directly than FP and SIV in the concerted practices, but not that it did not take part in such practices.

(33) As will be explained in greater detail under heading VI, the exchanges of products between the three Italian manufacturers also provide an opportunity for knowing the prices charged by competitors or for agreeing on the conduct to be pursued on the market, as may be seen from the facts set out below.

The prices of the products exchanged were always set and subsequently adjusted on the basis of the price adjustments of the transferring producers.

By letter dated 6 March 1985, FP accused SIV of not having adhered to the undertakings agreed on in Rome regarding the conditions of sale of cast glass transferred by FP to SIV.

FP's handwritten note on the meeting between SIV and FP held on 23 and 30 April 1985, during which product exchanges were discussed, contains the following Statement : 'Prices which we will charge you free at customer's destination ... . Distribution (by region) ... . meeting on 4th for compensation'.

FP's handwritten notes on the meetings between SIV and FP held on 16 December 1985 and 3 February 1986, at which the apportionment of the production of the joint subsidiary Flovetro was discussed, contain the following statements '... . two elements: transfer prices - apportionment of profits' and 'whoever draws off more must undertake not to disrupt the market'.

(34) The prices and discounts agreed on under the concerted practices were actually applied. Examination of certain invoices shows that one and the same customer who purchased widely differing quantities from the three producers was charged identical prices and granted identical discounts. The invoices examined were as follows :

- for 1983 : SIV invoices Nos 00866, 01450, 02885, 03912, 09701; FP invoices Nos 00594, 01208, 02824, 08883, 09580; VP invoices Nos 110040, 210321, 210475, 210629,

- for 1984 : SIV invoices Nos 04397, 11612, 11619, 11984, 11985; FP invoices Nos 01356, 10473, 12041; VP invoices Nos 110236, 110475, 11/3142, 11/3189, 11/3253, 11/3265,

- for 1985 : SIV invoices Nos 212, 325, 1752; FP invoices Nos 508, 773; VP invoices Nos 11/0733, 11/0742, 11/0778.

(e) Relations between producers and wholesalers

(35) The three producers took care to ensure that their prices and discounts were also applied downstream.

In the Statement of objections, on the basis of the written Statement of the complainant Cobelli, the Commission accused the three firms of having met on several occasions, indicating the instigators, dates and places of the meetings, with wholesalers in order to get them to accept and pass on the price increases. Since the complainant confirmed such meetings at the hearing only in general terms, the Commission does not have any direct evidence on them except the Catania meeting held on 17 April 1986 between FP, SIV, Fontana Sud, Callipo, Tortorici and ISV, this meeting having been admitted to by FP and SIV, its purpose having been to introduce the new admitted to by FP and SIV, its purpose having been to introduce the new administrator of Fontana Sud, the presence of the SIV representative not having been anticipated.

However, certain documents show, firstly, that some meetings betwen wholesalers were arranged on the initiative of the producers and that, given their identical prices and discounts, the producers managed to guide the commercial choices of the wholesalers and, secondly, they confirm that the customers expected producers' prices to be identical.

(36) Socover's handwritten note dated 12 July 1983 refers to discussions between the wholesalers' spokesman and FP on discounts and on the failure of the other producers to fulfill the undertakings given.

This note shows that Socover is the channel for passing messages from the wholesalers to the producers and from the producers to the wholesalers. Socover does not discuss with FP the terms granted it, which would have been of particular interest to it, but discusses the undertakings of the producers vis-à-vis all the wholesalers.

(37) On 10 October 1984, a meeting was held between a number of wholesalers at the Sheraton Hotel in Rome. After the meeting, the following telex was sent to producers on 11 October 1984 : 'The undersigned confirm the orders noted and will place new orders on the new terms taking account of all sources of supply ... The undersigned confirm their willingness to cooperate in improving prices on the market ... (the names of 28 customers follow)'.

Also on 11 October 1984, VP's director of sales met Socover in Milan, as may be seen from the FP manager's report on his mission, annexed to the reply to the Statement of objections.

According to the producers, the telex simply reflected the intention of a group of customers to pursue a line of conduct imposed by circumstances.

The telex of 11 October 1984 demonstrates clearly the signatories' intention of cooperating with the producers. The telex affirms that wholesalers will not exert pressure on the producers in order to get the date of application of the new prices postponed, that they will allocate orders in such a way as to take account of the various sources of production and that they will cooperate in improving prices on the market. Wholesalers cannot spontaneously embark on such conduct when it should be in their interest to obtain the best prices, to obtain their supplies from suppliers who grant them the best terms of sale and to achieve a good profit margin.

(38) Socover's letter dated 19 October 1984, found in VP's offices, contains the following Statement: 'At the request of the producers, we have drawn up the annexed price list, which will enter into force on 5 November. We have called a meeting to which we have invited all operators in the industry on Wednesday 7 November at 10.30 at the premises of Vetrounione in Milan ...'. By letter dated 20 November 1984, VP denied to Socover that it had requested any meetings on marketing.

With regard to the letter, the contents of which were not disputed by FP and SIV in their answers, VP claimed that it had obtained a copy through a customer, whose name was deleted, and produced a written Statement from Socover which included the following Statement: 'As for the letter referred to under (iii), it was sent to a number of wholesalers/distributors, but not at VP's request. As far as I can remember, the letter was never sent, even in the form of a copy, by Socover to VP, and VP did not take part in the meeting mentioned in the letter'. Since Socover states merely that the letter was not sent at VP's request, the obvious deduction is that it was sent only at the request of the other two producers, and this despite the fact that the letter states that 'at the request of the producers (without excluding any of them), we have drawn up the annexed price list, which will enter into force on 5 November'. Socover states that VP, and only VP, did not participate at the meeting held in Milan on 7 November 1984. However, in order to demonstrate that it never participated in the meetings, VP, in Annex B to its answer, produced certain monthly reports on the activities and whereabouts of its staff. One of the reports shows that VP's commercial director was on mission in Milan on 7 and 8 November 1984, i.e. on the day of the meeting called by Socover.

(39) Socover's handwritten note dated 12 April 1985 mentions discussions between Socover and FP on the appointment of sales agents by region and on the apportionment of orders on the historical basis of the last two years. None of the producers indicated what interpretation should be given to the phrase 'apportionment on the historical basis of the last two years' used in the note, or on the advantage to wholesalers of apportioning their orders not on the basis of the terms offered by each supplier, but solely on the basis of past statistical figures.

(40) Socover's handwritten note dated 23 July 1985 deals with discussions between Socover and SIV on the granting of an end-of-year premium of 3 % to a number of preferential customers.

With regard to this note, on which neither FP nor VP adopted a position, SIV argued in its answer that it was 'the usual practice to grant a discount at the end of the year to customers which have attained a given level of purchases and which have been punctual in their payments'. The note itself refers to preferential customers and not customers that have purchased certain quantities and pay punctually for their purchases. In addition, it states that a customer, described by FP as the wholesalers' spokesman, is informed not of the discounts granted to it, but of the discount granted to preferential customers. Usual and normal business practice seems hardly an appropriate term to describe the fact that suppliers confide in a customer the business relationships maintained with other customers.

(41) The report dated 16 September 1985 by Fontana Est to its parent company FP on the meeting of wholesalers held on 31 July 1985 refers to the fact that all the producers communicated the price increases and that, on the basis of such increases, the wholesalers decided on the scale of resale prices.

The report dated 23 September 1985 by Fontana Ovest to its parent company FP contains the following remarks: 'The customers seem prepared to accept the price increases ... . The producers want the prices to be strictly complied with so as to allow greater market stability'.

The other report from Fontana Ovest, dated 31 October 1985, on the meeting of wholesalers held on 29 October 1985, deals with the wholesalers' adoption of the scale of resale prices on the basis of the new prices charged by the producers and contains the following Statement: 'As far as the producers are concerned, the increase is in the process of being introduced ... In any case, everyone has communicated the new prices ... even if the terms are somewhat elastic ...'.

Far from being vague and unspecific or from revealing the wholesalers' general position, as the parties claim, the Fontana reports deal with precise facts such as the willingness of the wholesalers to accept the producers' price increases, the publication by the producers of identical price lists, and the producers' desire that the prices be complied with so as to stabilize the market. This means that the producers directly or indirectly exert an influence over the market downstream, since destabilization of that market could have a harmful effect on their business policy.

(42) During the meeting held on 10 April 1986, VP and Socover discussed the setting up of a club, consisting of 12 wholesalers, for the distribution of glass products. The two firms stated their intention of discussing the matter with FP and SIV the following week (VP's handwritten note dated 10 April 1986).

5. The market in automotive glass

(43) The company documents discussed below indicate that SIV and FP agreed on prices and the allocation of quotas at least as from 1982. VP also participated in these restrictive practices from 1983 at least, albeit less strictly than the other two producers.

(a) The agreements and concerted practices concerning the Fiat group

(44) By handwritten note dated 26 October 1982, FP forwarded to SIV an internal memo of the same date indicating the average percentage increases in prices obtained from Fiat from 1978 to 1982 and forecasts of average increases under the 1983 and 1984 contract signed between Fiat and FP on 14 June 1982.

SIV's internal memo dated 11 November 1982 shows that the average percentage increases obtained by SIV are the same as those obtained by FP. As regards SIV's quotas for 1983, the memo emphasizes that their calculation did not take account of the possible subsequent 2 % quota indirectly granted in Paris to SIV for 1983 and 1984, that a strict monitoring of actual quotas was necessary for original equipment, and that following the agreements entered into there was gradual alignment with the competition in respect of small batches.

(45) As a result of pressure from Fiat on its suppliers for a reduction in prices, SIV's commercial director wrote the following in internal memo No 532 dated 11 May 1983: (5) Inevitable involvement of FP in the granting of these discounts and acceleration of the reduction process without any concession in terms of penetration quotas and with exacerbation of business relations with the competition ... (B) ... In order to act in this direction, we must ensure parallel, reliable and strict conduct on the part of our more qualified competitors ...'. The price reduction granted by the three producers FP, SIV and VP was uniform, at 8 %, and was effective from 1 January 1984.

(46) SIV's handwritten note dated 12 October 1983 contains the following: 'Subject: IVECO - Following recent events connected with the aforementioned customer, I think it appropriate to communicate the following, which relates to the SIV-Saint-Gobain-Penny price level, on the two vehicles produced in larger quantities, i.e. Gamma S and Gamma Z (according to 1983 contract):

1.2,4 // // // // Price level / % allocation // // 1.2.3.4 // // Saint-Gobain // Penny // SIV // // // // // Gamma S // [...] [...] % // [...] [...] % // [...] [...] % // Gamma Z // [...] [...] % // [...] [...] % // [...] [...] % // // // //

N.B. For the moment, at any rate, we have succeeded unofficially and without the knowledge of the department responsible for purchases in Turin in maintaining our quota at average levels equal to [...] of supply.'

(47) At the beginning of 1985, discussions began on how to get price increases accepted by the Fiat group.

SIV's handwritten note on the meeting held in Rome on 30 January 1985 between SIV and FP contains the following Statement: 'Fiat problem - increase in holes and brackets as Trojan horse in Fiat for increase in prices'. The problems concerning FIAT are also mentioned in point 3 of FP's handwritten note on the same meeting.

FP's handwritten note on the meeting held on 7 May 1985 between FP and SIV refers to the following agreement: 'Automobile meeting. Meeting (Chariman of SIV) - from 1 July 1985 to 31 December 1985 + 7 % real // from 1 January 1986 to 30 June, 5 % real // from 1 July 1986 ? ... ending up with the same increases + 0 - We will share 50 % of Fiat's reaction at least at the production level. Joint reduction in costs'.

This agreement between the two producers was confirmed, according to FP's handwritten note, at the meeting between SIV and FP held on 23 May 1985, at which VP's agreement was mentioned: 'Agreement Fiat + Alfa Romeo + 7 % immediately; + 5 % 1 January 1986; + X second half 1986. Agreement of Pennitalia'. During the investigation, the question was put in writing on 15 January 1987 to the director of automotive glass sales whether VP had been contacted or had had any meetings with SIV and FP to agree on or follow suit on price increases for the types of automotive glass referred to above. VP's director of automotive glass sales asked for time to think before replying; the next day, on 16 January 1987, he replied in the negative.

FP's handwritten note dated 20 June 1985 confirms the agreement in the following terms: '(commercial director of SIV) ... Aid for Fiat - 7 % - reduction 3 % (according to him 1,5 %). Except for new models 2 %'.

SIV's internal memo No 090 dated 24 June 1985 refers indirectly to the implementation of the abovementioned agreements: 'Thursday 20 June, we concluded the negotiations with Fiat Auto for the second half of 1985 ...

We give a brief reminder of the terms of the agreements. (1) Increase of 7 % for all vehicles, increase of 2 % for new models, i.e. Y 10, Thema and 154. Since the latter models will have an impact over the next few months amounting to around 14,5 %, the theoretical weighted increase is 6,3 %. (2) Date of application of this increase: 1 July to 31 December 1985'. The average increase of 6,3 % was applied by SIV and FP, while VP applied an increase of 6 %, i.e. 0,3 % less.

Other uniform percentage increases in prices were introduced on the following dates: on 15 December 1985 by SIV and VP and 20 December 1985 by FP; on 1 May 1986 by FP, 15 May 1986 by SIV and 1 September 1986 by VP; on 1 December 1986 by SIV and FP.

With Alfa Romeo too, a uniform 2 % increase for the second half of 1985 was obtained, with subsequent uniform increases which FP and SIV applied with retroactive effect as from 1 September 1985 and which VP applied 10 months later.

(48) The three producers carried out reciprocal sales or purchases of products (which will be discussed in greater detail under heading 6) with the objective of maintaining their respective penetration quotas or of achieving the quotas agreed with their competitors.

- SIV-FP. According to the figures provided by the two producers, FP sold SIV the following quantities of products not processed for the motor vehicle industry: [...] tonnes in 1982, [...] tonnes in 1983, [...] tonnes in 1984, [...] tonnes in 1985 and [...] tonnes in 1986, while SIV sold FP [...] tonnes in 1985 and [...] tonnes in 1986. In the case of products processed for the motor vehicle industry, the following transfers were made: FP sold SIV [...] tonnes in 1983, [...] tonnes in 1984, [...] tonnes in 1985 and [...] tonnes in 1986; SIV sold FP [...] tonnes in 1982, [...] tonnes in 1983, [...] tonnes in 1984 and [...] tonnes in 1985. - FP's handwritten note dated 25 June 1985 contains the following Statement: 'SIV exchanges products in particular those we purchase in France'.

- FP's internal memo dated 31 October 1985 states the following: 'It appears that SIV might need about 500 000 side glasses preferably in the first half of 1986. The vehicles in question might be: Uno, Ritmo, Regata, Thema Croma, Alfa 33 ... The order might reach us in November (handwritten note in the margin: yes). I checked that there would not be any difficulties in supplying us with 40 000 rear windows a year for the Y 10'. The contents of this memo are confirmed by another internal memo by FP dated 8 November 1985. FP's internal memo dated 4 March 1986 mentions the quantities being produced in 1986 on behalf of SIV.

- FP's handwritten note dated 17 December 1985 contains the following Statement: 'Relationships with Fiat SIV hypothesis FP [...], SIV [...], others [...]. SIV has no production capacities. FP has capacities for [...] Fiat: [...] direct, [...] for SIV as subcontractor'. The main points of this note are confirmed by another handwritten note by FP dated 23 January 1986: 'Policy meeting motor vehicle sector: today us [...] SIV [...] ... FP [...] guaranteed (today [...]). The decreases are offset by sales to SIV at Fiat price free at destination ...'.

- VP-SIV. According to the figures provided by the two producers, the exchanges involved the following quantities. VP sold SIV products not processed for the motor vehicle industry amounting to [...] tonnes in 1984, [...] tonnes in 1985 and [...] tonnes in 1986, while SIV sold VP [...] tonnes in 1983, [...] tonnes in 1984, and [...] tonnes in 1985. In the case of products processed for the motor vehicle industry, VP sold SIV [...] tonnes in 1984 and [...] tonnes in 1985.

- VP's internal memos dated 14 February 1986, 28 April 1986, 3 June 1986, 13 June 1986 and 13 October 1986 and the telexes exchanged between VP and SIV on 28 March 1986, 15 May 1986, 20 June 1986, 25 June 1986, 11 July 1986 and 25 August 1986 refer to agreements under which VP, with the aid of its French fellow group member Boussois, was to produce for SIV [...] tonnes of green automotive glass and [...] tonnes of green plus automotive glass in 1987, and [...] tonnes of green automotive glass and [...] tonnes of green plus automotive glass in 1988.

- FP-VP. The figures provided by the two producers do not tally; both producers' figures will therefore be indicated, with VP's figures being shown in brackets. FP sold VP products not processed for the motor vehicle industry amounting to [... (...)] tonnes in 1982, [... (...)] tonnes in 1983, [... (...)] tonnes in 1984, [... (...)] tonnes in 1985 and [... (...)] tonnes in 1986. VP's internal memos dated 13 December 1985, 7 January 1986 and 18 February 1986 and the telexes exchanged between FP and VP of 19 December 1985, 20 December 1985, 7 January 1986, 9 January 1986 and 10 January 1986 refer to the agreement under which FP was to produce for VP [...] tonnes of clear 2,5 mm automotive glass, [...] tonnes of clear 2 mm automotive glass and [...] m2 of standard green automotive glass.

(49) FP has in its possession tabulations for the first and second halves of 1985, for 1986 and for 1987 showing, model by model and in overall terms, the quantities which each producer supplied in 1985 and 1986 and will supply in 1987 to the Fiat group and the percentage quotas represented by such supplies. The tables for 1985 are undated, while those for 1986 and 1987 are dated 20 October 1986. The overall percentage quotas for the years in question are as follows: first half of 1985: FP [... %], SIV [... %], PPG (VP) [... %], Splintex [... %]; second half of 1985: FP [... %], SIV [... %], PPG [... %], Splintex [... %]; 1986: FP [... %], SIV [... %], PPG [... %], Splintex [... %]; 1987: FP [... %], SIV [... %], PPG [... %], Splintex [... %].

(50) The three producers dispute that there was any collaboration between them on prices and on quotas, claiming that the Commission's accusations are based on misinterpretations of the documents. TEXT CONTINUED UNDER DOC.NUM : 389D0093.1

The automotive-glass market in Italy, they argue, is dominated by Fiat. Like any other motor vehicle manufacturer, Fiat allocates overall supply percentages to each supplier on the basis of considerations relating to prices, its technical capacity and the service offered. Each overall percentage is reviewed bilaterally when negotiations are carried out on the updating of prices. The overall allocation percentages thus defined are converted into allocation percentages for each model of car. The confirmations sent out by Fiat to its automotive-glass suppliers explicitly mention the quota allocated by it to the supplier concerned in respect of the model in question.

During the negotiations, Fiat is in the habit of revealing the most favourable quotation, with the aim of getting the competing suppliers to fall into line. This leads inevitably to an alignment of prices. Contrary to the Commission's supposition, Fiat frequently informs each of its suppliers of the supply quotas allocated to competitors. All in all, the system results in total transparency of the market. Consequently, it is argued, concerted practices between the producers are not necessary.

With regard to the exchange of products, it is argued, the Commission has not provided any proof that such exchanges are deliberately arranged in order to share the market between the producers. In addition, the products exchanged are primary glass and only in exceptional cases do the exchanges involve processed products.

(51) The Commission cannot accept the arguments advanced by the producers, for the following reasons:

(i) It is true that Fiat enjoys a position of almost total monopoly in Italy as a purchaser of glass for motor vehicles; however, its contractual power is greatly limited by the restrictive nature of alternative offers on the market. The supply of motor vehicle glass in Europe is controlled by an oligopoly comprising, as well as Saint-Gobain, PPG SIV, the present suppliers for Fiat, Pilkington and a number of small producers.

(ii) On the allocation quotas, it is true that Fiat, like any other motor vehicle manufacturer, specifies in its confirmations of orders the quota allocated to the supplier concerned. However, the quota allocated is merely an indication and not a final allocation, it commits neither the motor vehicle manufacturer nor the supplier, and changes may be made during the period of the contract either on the initiative of the motor vehicle manufacturer or on that of the supplier. That the manufacturer is not bound is shown by two facts reported by VP in its reply to the Statement of objections: when the 8 % price reduction was negotiated with Fiat, Fiat on its own initiative reduced VP's quota for the current contract (Annex P to VP's answer); Fiat's letter to VP dated 3 July 1985 confirming its order (Annex E to VP's answer) contains the following Statement: '... with regard to the short-term supply programmes, as you have been informed, it is not possible to define allocation percentages which are binding on us ...'.

That the supplier is not bound is shown by FP's Statement in its reply: 'The suppliers are limited by their technical processing capacities, according to their best estimates. It may happen that their estimates are not correct, or that the glass supplier cannot supply the volumes represented by its percentage quota, where sales of a model exceed forecasts'. Moreover, if it were really the case that the motor vehicle manufacturer allocated fixed and final quotas for the period of the contract, the suppliers would not have to worry about any danger of non-compliance with the quotas allocated and would not have to feel the need to collaborate on this subject or to exchange products in order to maintain their quotas. The documents referred to in paragraphs 44 to 48 show the contrary: 'Strict monitoring of actual quotas for original equipment is necessary ... we are moving towards gradual alignment with the competition in respect of small batches' (SIV's note dated 11 November 1982): ' ... acceleration of the reduction process without any concession in terms of penetration quotas ... we must ensure parallel, reliable and strict conduct on the part of our more qualified competitors' (SIV's note dated 11 May 1983); SIV can show that despite its higher prices it was able to maintain its quotas, without the knowledge of Fiat's purchasing department (and yet it is asserted that Fiat allocates the quotas) (SIV's note dated 12 October 1983); the exchanges of processed and unprocessed products show the concern for mutual assistance in order to maintain the quotas. As for the tabulations, referred to in paragraph 49, the Commission considers that they do not come from Fiat and that they were not compiled on the basis of the figures provided by Fiat either. As VP states in its reply, negotiations with Fiat are conducted in percentages of sets and not in square metres; even if the tabulations for 1985, undated, and for 1986, dated 20 October 1986, could conceivably have been compiled by means of an estimate which, having been carried out ex-post, could be fairly precise, the same could not be said for the 1987 tabulation, since, on the date it was compiled, 20 October 1986, nothing was final and definite for the following year.

(iii) The Commission does not dispute that a buyer may, during business negotiations, claim to have had a better offer, whether real or imaginary, in an attempt to get the other suppliers to fall into line. However, this does not mean that all suppliers will apply the same terms and that all suppliers are on an equal footing vis-à-vis a given buyer, since each supplier's position depends on what it has to offer in terms of production and business strength (wider range, special products, better service). Nor does it mean that a dominant buyer can oblige all its suppliers to align themselves on the most favourable offer or that it can easily turn to foreign suppliers in order to force local suppliers to reduce their prices, since, as was admitted at the hearing, motor vehicle manufacturers increasingly demand to be supplied 'just on time'; this sort of service can more easily be provided by local glass manufacturers, which have an on-the-sport structure, than by a foreign glass manfacturer, who has to set up such a structure, and, amongst the glass manufacturers having the necessary structure, this requirement confers an advantage, in terms of quantities to be supplied and prices, on thos which have a more firmly established and, at the same time, more flexible structure.

At all events, whatever the types of relationships that develop between a dominant buyer and its supplier, it is established that SIV, FP and VP collaborated in order to decide on the attitude to be adopted towards the Fiat group. Thus, FP sent SIV, by note dated 26 October 1982, the forecast price increases contained in the contract signed between FP and Fiat on 14 June 1982. The 8 % price reduction referred to in the note by SIV dated 11 May 1983 was aggreed on and applied by the three producers. VP claims that the Statement in SIV's note to the effect that 'we must ensure parallel, reliable and strict conduct on the part of our more qualified competitors' does not concern it, since, with [...] of supplies to Fiat, it had to be regarded as a marginal and not a 'qualified' competitor.

VP's interpretation is contradicted by that of SIV, the author of the note, which does not refer to qualified or marginal competition, but simply to competition. In SIV's reply to the Statement of objections, we read the following: 'Whereas the expression ''ensure parallel conduct on the part of the competition" is merely a very logical resolution on the part of the commercial director reflecting the intention to resist as far as possible excessive pressures for reduction'.

The price increases for the second half of 1985 and the first and second halves of 1986 were discussed and agreed on between the three producers. According to VP, the fact that FP's handwritten note dated 23 May 1985 mentions 'Agreement of Pennitalia' does not mean anything and at any rate there is no proof that the reference is reliable. In addition, the correspondence with Fiat dated 28 June 1985 and 3 July 1985, it claims, contradicts FP's handwritten note. Lastly, the coincidence of the date of the 15 December 1985 price change with that of SIV is, it is argued, the consequence of what was agreed with Fiat through the correspondence dated 28 June 1985 and 3 July 1985. In the Commission's opinion, the reference to 'Agreement of Pennitalia' is reliable, since VP's accession to the agreement on the price rises for the three half-years concerned was essentially put into practical effect. The letters cited by VP have a quite different content to that mentioned by VP: VP's letter to Fiat dated 28 June 1985 does not contain any undertaking on prices on the part of VP, but merely indicates a willingness to leave prices for 1985 unchanged, provided that VP can double its supplies; Fiat's letter to VP dated 3 July 1985 similarly does not contain any undertakings but Fiat on supply quotas or on prices, but simply notes the possibility for VP to keep its prices competitive (not lower); these letters do not contain any mention of the dates of price changes nor do they mention any agreement on prices between Fiat and VP.

(iv) The facts which demonstrate that the exchanges of products are aimed at market sharing are essentially as follows:

The exchanges are considerable every year and, as shown by the documents referred to in paragraph 48, they are not simply intended to make good shortages. In the case of unprocessed products, the exchanges involved a total of [...] tonnes in 1982, [...] tonnes in 1983, [...] tonnes in 1984, [...] tonnes in 1985 and [...] tonnes in 1986, i.e. between [...] and [... %] of the three producer's total retained consumption in the automotive sector; in the case of processed products, the exchanges totalled [...] tonnes in 1982, [...] tonnes in 1983, [...] tonnes in 1984, [...] tonnes in 1985 and [...] tonnes in 1986.

The notes and documents referred to in paragraph 48 explicitly mention the fact that FP is performing subcontracting work for SIV in respect of processed products.

The exchanges relate in particular to thicknesses and colours that are not manufactured by another given producer and are intended to allow the three producers to have the whole range of products available. However, the competitive advantage for a producer of having the whole range available is annulled if the three producers act in such a way as to ensure that there is no difference between them in this respect. The fact that a fuller range confers a competitive advantage is confirmed by FP in its reply: 'The suppliers can resist this pressure (from the purchaser) only if they have an objective competitive advantage over their competitors (fuller range, special products, better service) that may be decisive for the manufacturer'.

(b) The agreements and concerted practices concerning the Piaggio group

(52) SIV and FP reached agreement, at least as from 1983, on the sharing of supplies and on prices charged to Piaggio.

FP's internal note dated 12 December 1984, found during the investigations on the premises of FP and SIV, refers to the following agreements: 'The situation at the end of 1982 was as follows: Quotas: Saint-Gobain ± [... %], SIV ± [... %]. Prices: (there follows the table of Saint-Gobain prices and SIV prices) ... following contacts with (the SIV official) at the end of that year, we had agreed to proceed along the following lines: (1) SIV would continue not to supply the 6 011 model. (2) The objective would be to share [...] the quotas for the other models. (3) Differentiated increases were agreed so as to get Piaggio to spread the orders for each part in a non-uniform manner in such a way as to achieve the objective in point (2). In short, the argeement provided for ... the following situation (there follows the table of prices agreed on and to be charged as from 1 March 1983 and 1 September 1983) ... The agreement was largely complied with, but the situation hardly changed (at the end of the year ± [... %] S. G., ± [... %] SIV). In view of this fact, at the end of 1983, it was agreed with (the SIV official) still with a view to the sharing of the quotas [...] that for 1984 Saint-Gobain would increase prices by 4 % as from 1 march and by 3,5 % as from 1 September, while SIV would introduce differentiated price increases as from 1 July (there follows the table showing the increases agreed) ... In actual fact, this agreement was never complied with ... SIV had already agreed to postpone the increase to September-October ... We were thus obliged to agree to delay the 1 March increase to 1 June ... It should be pointed out that I always kept SIV informed of all my initiatives vis-à-vis the customer ...'.

Evidently, FP's internal note was sent to SIV, since the SIV official concerned wrote, on 28 December 1984, the following note: 'With reference to Saint-Gobain's note dated 12 December 1984 on the same subject ... (1) No particular comments on the contents of pages 1, 2 and 3, which summarize exhaustively and precisely our previous agreements, which SIV has always complied with. (2) With reference to page 4 - frist paragraph - I must point out that (the SIV official) ... has always been in contact with (the FP official) and has pointed out that Piaggio pressure had forced us to agree to postpone until November 1984 the increase requested and that, in any case, the increase (± 4 %) would have had to be considered solely as a basis for the 1985 contract. Our telex No 3048 of 6 December 1984 to Piaggio ... confirms totally what has been said above ... To conclude, I confirm that there have always been contacts between (two SIV officials and two FP officials). Lastly, I must point out that, during a telephone conversation at the beginning of December between (an FP official and an SIV official) the latter communicated to the former the text of our telex No 3048 ...'.

The agreements in question were implemented, as is evident from SIV's memo. SIV and FP altered their prices: for 1983, they did so on 1 March and 1 September; for 1984, FB did so on 1 March (postponed to 1 June, as is stated in the note dated 12 December 1984) and on 1 September, SIV on 1 January (as is stated in the note dated 28 December 1984) and on 1 September, SIV on 1 November (as is stated in the note dated 28 December 1984); for 1985, FP did so on 1 March and SIV on 1 May; for 1986, FB did so on 1 April and SIV on 1 May.

6. Exchanges of glass between the producers

(53) Large quantities of glass are exchanged under contracts between the three producers. The purpose of the exchanges is to enable each producer to have available the full range of products, even those which it does not manufacture, and to maintain its market quotas. They also provide a means of sharing markets and customers and of knowing the prices charged by competitors, as may be seen from the documents and facts described below.

(54) SIV-VP exchanges. According to the information provided by the two producers, automotive and non-automotive exchanges involved the following quantities. Since the figures provided do not tally, those supplied by VP will be shown in brackets. SIV sold VP [... (...)] tonnes in 1982, [... (...)] tonnes in 1983, [... (...)] tonnes in 1984, [... (...)] tonnes in 1985 and [... (...)] tonnes in 1986. VP sold SIV [... (...)] tonnes in 1984, [... (...)] tonnes in 1985 and [... (...)] tonnes in 1986.

By letter dated 16 September 1986, following a series of meetings and telexes between SIV and VP, SIV placed the following orders with VP: for 1986, [...] tonnes of green automotive float glass, [...] of clear, thick float glass, and [...] tonnes of green plus float glass; for 1987, [...] tonnes of green automotive float glass, [...] of green plus float glass and [...] tonnes of clear, thick float glass; for 1988, [...] of green automotive float glass and [...] tonnes of green plus float glass. As regards non-automotive glass, the letter contains the following Statement: 'Prices will be readjusted on the basis of the changes in the national price list'.

(55) VP-FP exchanges. According to the figures provided by the two producers, with VP's figure being shown in brackets, automotive and non-automotive exchanges involved the following quantities: VP sold FP [... (...)] tonnes in 1983, [... (...)] tonnes in 1984 and [... (...)] tonnes in 1986. FP sold VP [... (...)] tonnes in 1982, [ ... ( ... ] tonnes in 1983 [... (...)] tonnes in 1984, [... (...)] tonnes in 1985 and [... (...)] tonnes in 1986. The telexes dated 19 December 1985, 20 December 1985 and 28 February 1986 and VP's internal note dated 3 April 1986 specify exactly the exchanges for 1986: VP sold FP [...] tonnes of bronze building glass; FP sold VP [...] tonnes of grey building glass, [...] tonnes of clear automotive glass and [...] m2 of green automotive glass. Prices for glass not intended for the motor vehicle industry were readjusted on the basis of the changes in the respective price lists.

(56) FP-SIV exchanges. According to the figures provided, FP sold SIV the following quantities of cast glass, SIV's figures being given in brackets: [... (...)] tonnes in 1983, [... (...)] tonnes in 1984, [... (...)] tonnes in 1985 and [... (...)] tonnes in 1986.

Exchanges of automotive and non-automotive float glass involved the following quantities (SIV's figures given in brackets). FP sold SIV [... (...)] tonnes in 1982, [... (...)] tonnes in 1983, [... (...)] tonnes in 1984, [... (...)] tonnes in 1985, and [... (...)] tonnes in 1986. SIV sold FP [... (...)] tonnes in 1982, [... (...)] tonnes in 1983, [... (...)] tonnes in 1984, [... (...)] tonnes in 1985 and [... (...)] tonnes in 1986.

The exchanges of memos, the minutes and the handwritten notes show that the prices of the products sold are established and subsequently readjusted on the basis of the sellers' price changes, that the prices are established by reference to the destination of the products sold on the national market and on specific foreign markets and that, even in the case of sale for the national market, the destination of the products is broken down by individual regions and uses.

- Telexes of 18 October 1984 and 20 October 1984 and letter from FP to SIV dated 18 December 1984: 'Normal thick clear float glass intended for export ... [...] tonnes; amber exclusively intended for silvering [...] tonnes; bronze exclusively intended for silvering [...] tonnes; foreign automotive green plus [...] tonnes ...'.

- The contract signed on 18 January 1985 contains the following clause: 'The abovementioned selling price will be automatically increased during the year by the same increases as those applied to selling prices and made public by Fabbrica Pisana's price lists'.

- SIV's handwritten note on a meeting between SIV and FP held on 30 Octber 1984: 'Problems resolved: Clear float, normal thicknesses for export [...] tonnes 1 September 1984 to 31 December 1985. Bronze for silvering ... grey Italy ... automotive green plus ... Cast glass ... Must not sell at a price lower than FP. Laminated glass (especially flashed glass) - (director-general of FP) - Wish (concerning) a policy of apportionment in the factory instead of having glass processed - in the case of triple glass, adopt a common policy'.

- FP's letter to SIV dated 6 March 1985: '... I am obliged to inform you that I am not satisfied with the way in which your firm is managing the marketing of the cast glass supplied by us. At the last meeting in Rome, you agreed that SIV's salesmen should not for any reason apply terms that could adversely affect the proper rules of normal competition ...'.

- Handwritten note by FP on a meeting between SIV and FP held on 28 March 1985: 'Cast glass - (1) OK for supplies. (2) Price + 3 % on the product as from 1 April. (3) [...] tonnes/month. (4) Customers - 16 customers reserved for us in accordance with list in our and their possession'.

- Following the meetings held on 23 and 30 April 1985, FP's handwritten note dated 30 April 1985 refers to the following decisions adopted by FP and SIV: 'Prices which we will charge you free at customer's destination ... Total quantity [...] tonnes/month. Distribution: Piedmont 3 %, Lombardy 1 %, Trentino 1 %, Emilia-Romagna 8 %, Tuscany 10 %, Abruzzi-Milise 3 %, Lazio 4 %, Compania 24 %, Apulia 14 %, Calabria 10 %, Sicily 5 %, Sardinia 3 %, meeting on 4th for compensation'.

- FP's handwritten note dated 4 June 1985: '(commercial director of SIV) - [...] tonnes/month - continuity - invoicing directly abroad. GV all 4 mm, price FF 84 = 209,63 - 370 Lit/kg, any thickness + 4,8 % for the FF exchange rate he will say (? ) a change of 2 % we apply automatically ... Laminated glass he would like (? ) + 3 % ...'.

- FP's handwritten note dated 16 December 1985 (meeting with SIV): '(the director-general of SIV) speaks of an extra [...] tonnes for them ... two elements: transfer price - apportionment of profits'. With reference to these transfers, FP's handwritten note on a meeting between FP and SIV held on 3 February 1986 contains the following Statement: 'Whoever draws off more must undertake not to disrupt the market'.

(57) According to the producers, supplies of primary glass are economically necessary for reasons inherent in the structure of the industry. Each producer cannot produce all colours and thicknesses, since switching from one colour to another and from one thickness to another means downtime, which costs money. It is therefore normal for producers fo exchange products, so that each producer can have the full range of products available. In addition, account must be taken of the technical need to shut down and maintain furnaces periodically, which involves downtime and hence the need to buy in supplies of the primary product.

With regard to exchanges of processed products, it would be mistaken to believe that the transferor could take advantage of a shortage of products on the part of the transferee in order to place products with the consumer, since, in the event of refusal by the transferor, the transferee will try to obtain supplies from other competitors.

The exchanges are not systematic in character, since they are carried out in response to ad hoc requirements.

Lastly, the geographic destinations or intended uses referred to in certain documents are essential in defining the quality of the product and in setting the transferor's selling price.

(58) (i) As it said in the Statement of objections to the firms, the Commission does not intend to call into question exchanges of products to help out firms facing temporary shortages (renewal of production plant, shutdown for maintenance of furnaces, fulfilling occasional orders), but, as stated in the present case, only the systematic exchanges of products agreed over long periods and which are the result of industrial and commercial policy decisions made by the manufacturers in the context of other agreements restricting competition.

The Commission cannot agree with the producers' view that such exchanges are economically necessary. As the producers affirm, the purpose of the exchanges is to allow each of them to have a full range of products available at any time, thus cancelling out the economic advantage deriving from each producer's specialization.

Whether the specialization relates to a given thickness or a given colour, the exchanges cancel out the advantage of specialization and artificially place all the producers on an equal footing, thus preventing customers from benefiting economically from the productive and commercial edge enjoyed by individual producers. As the uniformity of the price lists and discounts of the three producers shows, the exchanges result in practice in a flat and uniform market. Furthermore, so as to prevent customers from identifying the producer of the product transferred and thus applying directly to that producer, the transferee rarely reveals the origin of the product sold, and then only at the express request of the customer, as VP admitted at the oral hearing.

(ii) With regard to exchanges of processed products, the argument put forward by the producers does not tally with the facts. If what the producers claim is true, namely that each processed product, notably those for the motor vehicle industry, must meet certain technical and design specifications imposed by the consumer, it is difficult to see how a producer lacking a given product could obtain the product outside the restricted circle of the suppliers of the product in question, since those which are not suppliers of the product do not have the right equipment for its manufacture. It is thus within the group of current suppliers that the exchanges are able to be carried out. Consequently, the transferor of the products gives up its ability to increase its market share and allows the transferee to maintain its share, since, as was stated in paragraph 51, the quotas allocated by the consumer are for guidance purposes and are binding neither on the suppliers nor on the consumer.

(iii) Contrary to what the producers claim, the product exchanges are, in the Commission's view, systematic. Although alongside multiannual exchange contracts there are also annual contracts, such contracts must be seen in the context of a commercial policy which, as the producers admit, is known and is always applied. This means that a producer which requires a product can always be sure, in view of the established practice, that it will be able to obtain supplies from its competitors.

This is all the more true as the exchanges apply not to marginal quantities, but to considerable tonnages. However, such large quantities could not be exchanged unless there were an established system. A formal framework agreement is not necessary for the system to function, as the parties claim, since the fact that a business practice is current and repeated is sufficient to give the exchanges a systematic character. The significance of the exchanges of float glass by the three producers is shown by the percentage which they accounted for each year in relation to each producer's production:

(in %)

1.2,3.4,5.6,7 // // // // // // FP // SIV // VP 1.2.3.4.5.6.7 // // Sales // Purchases // Sales // Purchases // Sales // Purchases // // // // // // // // 1982 // [...] // [...] // [...] // [...] // [...] // [...] // 1983 // [...] // [...] // [...] // [...] // [...] // [...] // 1984 // [...] // [...] // [...] // [...] // [...] // [...] // 1985 // [...] // [...] // [...] // [...] // [...] // [...] // 1986 // [...] // [...] // [...] // [...] // [...] // [...] // // // // // // //

Over the five years, the average for reciprocal sales by the three producers was equivalent to [... %] of their overall production, while the average for their reciprocal purchases over the five years represented [... %] of their production.

In the case of cast glass, FP, which since the beginning of 1984 has been the only producer in Italy, sold to its Italian competitors quantities which accounted for the following percentages of its total sales of this product: [... %] in 1983, [... %] in 1984, [... %] in 1985 and [... %] in 1986.

(iv) On the need to indicate geographical destinations or intended uses, it should be noted that glass was defined, by these very producers, as a homogeneous and commonplace product, so homogeneous that it is impossible to identify the individual manufacturer. If this is the case, and given the fact that the product is transferred to another professional who is able to detect its faults and its qualities, it is difficult to see why there should be any prior indication of a destination, all the more so since, the identification of the manufacturer being impossible on the basis of the product, the transferor need not fear any liability for the product. Furthermore, if the product, whatever the manufacturer, is always homogeneous, its transfer price cannot be dependent on the geographical destination or intended use, but solely on whether the product involved is of top quality or second quality.

II. LEGAL ASSESSMENT

A. Article 85 (1)

(59) Article 85 (1) of the treaty prohibits as incompatible with the common market all agreements between undertakings or concerted practices which may affect trade between Member States and which have as their object or effect the prevention, restriction or distortion of competition within the common market, and in particular those which directly or indirectly fix pruchase or selling prices or any other trading conditions and those which share markets or sources of supply.

(60) The agreements and concerted practices described below are covered by the prohibition laid down in Article 85 (1), and the undertakings in question are undertakings within the meaning of that Article, since they are engaged in economic activities in the production, processing and sale of flat glass.

7. The agreements and concerted practices concerning glass not intended for the motor vehicle industry

(61) The publication of identical price lists within a short period of time, or indeed on the same date, and the existence of identical discount scales and of identical lists of categories of customers qualifying for such terms are the result of agreements and concerted practices between the producers in question. The documents and notes on the meetings and discussions referred to under heading 4 show that FP, SIV and VP agreed or at any rate colluded, at least from 1983 and up to 1986, in charging uniform prices and applying uniform terms of sale, and that the argument that the identity of the price lists and terms of sale was merely the consequence of the homogeneity of the products and of an oligopoly situation on the market does not correspond to the facts.

(62) The Commission cannot accept the view of certain producers on the concept of agreement.

For a restriction to constitute an agreement within the meaning of Article 85 (1), it is in no way necessary for the agreement to be legally binding on the parties. An agreement exists once the parties agree on a practice which restricts or is liable to restrict their commercial freedom by laying down the lines of their action or of their mutual abstention from action on the market. No contractual sanction is required. It is not necessary that the agreement be established in writing.

In the case in point, as was explained in paragraphs 18 to 33, Socover's notes dated 12 July 1983, 12 April 1985, 10 July 1985, 23 September 1985 and 10 March 1986 and SIV and FP's notes dated 30 January 1985 refer to agreements between the three producers on price lists and terms of sale, and SIV and FP's notes and documents dated 30 October 1984, 6 March 1985, 28 March 1985, 12 April 1985, 23 and 30 April 1985, 16 December 1985 and 3 February 1986 refer at least to agreements between SIV and FP on prices and terms of sale agreed in connection with exchanges of products, agreements whose point can be explained only in a market which is already cartelized.

(63) The producers claim that the Commission has not provided any proof of meetings or agreements between the three producers in question intended to ensure unifrom price lists or terms of sale, but solely proof of bilateral contracts between a given producer and a wholesaler. As has been evident above, the Commission has provided proof of meetings between the producers and has shown that Socover's notes and SIV's and FP's notes refer to prior agreements between the three producers. However, even if the producers' claim were true, the fact remains that the publication of identical price lists over a long period, the existence of the same discount scales and the application of uniform terms of sale to the same customers could only be the result of concerted practices either directly between the three producers or through the intermediary of the spokesman for the wholesalers.

Although the concept of agreement is distinct from that of concerted practice, collusion may involve elements of both forms of illicit cooperation. By treating concerted practices as a separate concept, the treaty aims to prevent undertakings from getting round the application of Article 85 (1), by reaching an understanding on anti-competitive arrangements that do not amount to an agreement, for example, by obtaining information from one another in advance on the attitude which each proposes to take, so that each undertaking can decide on its business conduct in the knowledge that its competitors will act in the same way (1).

In its judgment delivered on 16 December 1975 (2), the Court held that the criteria of coordination and cooperation laid down by the case law of the Court must be understood in the light of the concept inherent in the provisions of the treaty relating to competition that each economic operator must determine independently the policy which he intends to adopt on the market. Although it is correct to say that this requirement of independence does not deprive undertakings of the right to adapt themselves intelligently to the existing and anticipated conduct of their competitors, it does however strictly preclude any direct or indirect contact between them, the object or effect whereof is either to influence the conduct on the market of a 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.

Even if it is not wished to describe the content of SIV's and FP's notes and documents and of Socover's notes as amounting to agreements, there is no doubt that the notes and documents reflect collusion between the three producers, whatever the precise form which the collusion took, since the producers could not fail to be aware of the meaning of the words they were using when they stated 'we will apply the same terms as the others', 'we are breaching the provisions of the cartel' and '... must undertake not to disrupt the market', or that Socover, defined by the producers themselves as the spokesman of the wholesalers, could be used, as it often was, as the medium for conveying planned or anticipated conduct on the market from one producer to another.

Even if the three firms were able, which they are not, to argue that they obtained knowledge of their competitors' price lists through their customers, the price lists being public, they could not legitimately rely on such an argument to justify their knowledge of documents such as the discount scales and the lists of customers qualifying for the discounts, such documents being defined by the firms themselves as being internal documents jealously kept secret. Add to this the fact that the firms changed the classification of customers by category or level sometimes on the same day and in most cases within a brief period of time, and the obvious conclusion is that the parties' argument is without foundation.

(64) As may be seen from the facts set out in paragraphs 35 to 42, the meetings between the principal wholesalers that were instigated and/or organized by the producers are the result of agreements or concerted practices between FP, SIV and VP designed to orientate, along lines that suited their concerted interests, the purchasing and sales policies of the wholesalers, who, because of their economic dependence, are unable to assert themselves against the power and manipulations of the producers. If the wholesalers were not obliged to take their business decisions in corporate meetings in which the free choice of each wholesaler is impeded or prevented by the fact that the meetings are instigated by the producers, who wish thus to underline the fact that they are acting by common accord, and by the direct or indirect presence (the marketing companies controlled by the producers) of the producers and of the other wholesalers, who are their natural competitors, they could individually exert pressure on the producers and purchase from the producer or producers that offered the best terms of sale, thus upsetting the market equilibria desired by the producers. The purpose of the corporate meetings is precisely to avoid such pressures, to enable the producers to monitor one another and adopt uniform conduct and, consequently, to consolidate the market equilibria and structures agreed.

(65) The fact that from time to time individual producers are accused of not complying with their undertakings and of breaching the cartel and the fact that some producers are late in applying agreed prices and terms are reflections of the existence of the abovementioned agreements or concerted practices, which, by their nature, constitute infringements of Article 85. In addition, even if the agreements are not fully complied with, this does not mean that they do not have a significant influence on the conduct of the firms.

(66) These agreements and concerted practices between firms manufacturing indentical products constitute serious restrictions of competition within the meaning of Article 85 (1). Through the agreements and concerted practices, the undertakings in question have committed themselves to restricting substantially their independence of conduct vis-à-vis their customers, and they have acted in such a way as to ensure that their principal customers fall into line with their decisions and are unable to take their own business decisions freely. The effects of these restrictions are all the more appreciable as FP, SIV and VP control some 79 % of the Italian home market. Through the conduct at issue, the undertakings have reduced the scope for purchasers to benefit from competition between the local producers, given their overall market share even after imports are taken into account. In addition, in view of the risks involved in not being able to rely on regular supplies, it is difficult for the main purchasers, wholesalers and processors to dispense with supplies from the producers established in Italy.

8. The agreements and concerted practices relating to automotive glass

(67) The facts and documents described under heading 5 show clearly that FP and SIV, from 1982 to 1986, and FP, SIV and VP, from 1983 to 1986, entered into agreements or at least concerted their behaviour on the prices to be charged to the Fiat group and, from 1982 to 1987, on the sharing of the market, thus removing any uncertainty as to their mutual conduct.

The agreements and concerted practices between FP and SIV, which involve the setting of prices, down to the retail stage, the apportionment of supplies, even down to particular items, and the apportionment of profits or losses deriving from the customer's reaction constitute much more serious infringements than those committed through the cooperation with VP.

However, VP's conduct also constitutes an infringement. There is no doubt that VP participated in the agreements or concerted practices relating to prices, as is evident from SIV's and FP's notes and the fact that VP applied the price list changes without fail, albeit with a difference of 0,3 % less in one case and with some delay in others. Nor is there any doubt that VP participated in the agreements relating to the apportionment of supplies, as is evident from the tabulations in FP's possession and the fact that it actively produced and continues to produce non-processed and processed products on behalf of its competitors.

(1) Case 48-69, ICI v. Commission [1972] ECR 619.

(2) Joined Cases 40 to 48, 50, 54 to 56, 111, 113 and 114-1973 Suiker Unie v. Commission [1975] ECR 1663.

It is unlikely that a supplier will learn of increases in his competitors' prices through a customer, since it is in the customer's interests to communicate price reductions, but not price increases. It is in the customer's interests to break up the suppliers' common front so as to obtain different prices reflecting the contractual strength of each supplier. Even if it were true that, during negotiations with a customer, a supplier could obtain information on its competitors' terms, the fact remains that the terms applied by each supplier would depend on its contractual strength, the range of products it can supply and the service offered and that even a customer in a dominant position could not, in determining the terms of purchase, fail to take account of these objective facts. At all events, it is clear that FP, SIV and VP collaborated prior to any negotiation with the Fiat group either on price reductions or on price increases.

Lastly, it is improbable that the customers were the source of the quotas or were the means of circulating information on the quotas, since it is difficult to see how it could be in the customers' interests to pass on information on the origin and percentages of their supplies. On the contrary, such information is jealously guarded as a business secret. The fact that, in confirming orders, the customer mentions the quota allocated for the relevant model to the supplier concerned does not mean that the quota is final or that the customer will communicate to each supplier the quotas allocated to the other suppliers. The quotas allocated individually to each supplier by the customer are solely for guidance purposes and are not binding on the customer or on the supplier, changes always being possible. In addition, the percentages negotiated with the customer are expressed in terms of units, whereas the tabulations in FP's possession contain percentages expressed in terms of square metres. It follows from this that the allocation of final and detailed quotas is not the customer's doing, but the result of the apportionment of supplies agreed between the three producers.

(68) The documents referred to under heading 5 show that FP and SIV agreed or collaborated, from the end of 1982 to 1986, on the prices to be charged to Piaggio and on the quantities and items which each of them would supply.

Through these agreements and practices, which constitute clear infringements, the two producers developed a long-term strategy designed to get the customer in question to apportion its orders in accordance with what they had decided, thus depriving Piaggio, through the system of differentiated prices, of any economic scope for choosing its own sources of supply.

Such conduct is all the more serious as FP and SIV actually put their agreements into effect, adapted them to changing circumstances in the course of their implementation and extended them beyond the period initially provided for.

(69) The above agreements and concerted practices constitute restrictions of competition within the meaning of Article 85 (1). Through these agreements and concerted practices, the producers in question created a market situation such as would exclude or, at the very least, reduce to a minimum any form of competition between them. The agreements and concerted practices allowed those concerned to seek and achieve an equilibrium in prices and outlets at a different level than that which would have occurred in a normal competitive situation and to crystallize their respective market positions. The effects of the restrictions applied by the producers in question are appreciable, since FP and SIV control more than 80 % of the Italian automotive glass market and since FP, SIV and VP control some 95 % of that market. Because of the conduct at issue, consumers were deprived of the possibility of benefiting from competition between local producers, in view of the preponderance of sales by local producers on the market in question even after imports are taken into account. In addition, it must be borne in mind that, in order to be able to rely on regular supplies, consumers are unable to dispense with supplies from producers established in Italy.

9. The agreements relating to exchanges of glass between the producers

(70) The agreements and contracts described under heading 6 and relating to systematic exchanges of glass between the three producers constitute appreciable restrictions of competition within the meaning of Article 85 (1), since they deprive the parties of their independence of conduct and of their ability to adjust individually to circumstances. Through the agreements and contracts, each producer gives up the right to take advantage, through increased direct sales to customers, of the other manufacturers' lack of products, of its own productive capacity, its specialization and its technical processing capacity, being in turn protected from such a risk where the situation is reversed.

At the very time when it would be particularly easy for one party to encroach on the market of another, since the products are identical, or when one party could take advantage of its specialization to establish itself more strongly in the market segments which are of particular interest to it, it must instead refrain from such types of conduct and deflect part of its production from normal sale in order to supply a competitor. Moreover, the recipient competitor is in a situation of dependence on the transferring producer which prevents him from using the products received under the agreements and contracts for any competitive activity, as certain documents specified under heading 6 show.

As may be seen from the documents referred to under heading 6, the ultimate purpose of the agreements and contracts is to share markets and customers between the producers and to prevent any change in their respective positions in the various market segments and any pressure from the consumers. In the automotive glass sector, the sharing of the market and customers is sometimes very far-reaching: certain producers perform subcontracting work for their competitors, which have the appropriate technology and processing facilities, simply in order to achieve a given apportionment of supply quotas for each customer.

10. Affect on trade between Member States

(71) The restrictions of competition described under headings 7, 8 and 9 are liable to have an appreciable effect on intra-Community trade. The agreements on prices also concern products imported by SIV from other Member States, by FP from other companies belonging to the Saint-Gobain group and by VP from its fellow group member Boussois. The agreements on prices therefore also have as their object products of Community origin.

The agreements relating to exchanges of glass also involve products from the three firms intended for export. They therefore have the effect of preventing each producer from developing freely its sales within the Community.

In addition, the agreements on exchanges of products and the agreements on prices and on quota and market sharing influence sales achievable in Italy through imports of articles produced in neighbouring countries. The practices in question establish a structure of uniform business terms differing from the structure of differentiated conditions that would normally have obtained if competition had not been restricted, and they therefore deflect flows of trade between countries from the course they would otherwise have followed. By establishing these uniform terms, the addressees of this decision, which account for some 79 % of the Italian market for non-automotive glass and some 95 % of the market for automotive glass and which control a large part of the external sources of supply, have affected the structures of effective competition. It follows that the agreements appreciably affect trade between Member States by consolidating national compartmentalizations which obstruct the economic interpenetration desired by the treaty.

B. Article 85 (3)

(72) The agreements and concerted practices to which this decision relates cannot be exempted under Article 85 (3), since they were not notified in accordance with Article 4 (1) of Regulation No 17 and since they do not fall within the exemptions provided for in Article 4 (2) of that Regulation.

(73) The Commission considers that, even if the agreements and practices had been notified, the conditions for applying Article 85 (3) would not be met. It is difficult to see how agreements on prices and on market sharing can contribute to improving production and distribution or contribute to technical progress, or what benefit the agreements can allow consumers. In addition, the agreements afford the undertakings the possibility, on the Italian market, of eliminating competition in respect of a substantial part of the products in question.

C. Article 86

(74) Article 86 of the EEC treaty states that any abuse by one or more undertakings of a dominant position within the common market or in a substantial part of it shall be prohibited as incompatible with the common market in so far as it may affect trade between Member States.

(75) FP, SIV and VP are undertakings within the meaning of Article 86. 11. The relevant market

(76) The market covered by this decision is the market in flat glass. Unlike hollow glass, which is in competition with other products such as aluminium cans, tin cans, specially treated cardboard cartons and plastic bottles and containers, flat glass is not interchangeable, except in the case of two uses, with other products in respect of the various applications for which it is intended. Leaving aside use for greenhouses and verandas, in respect of which flat glass is in competition with plastic, unless there is a particular need to avoid heat dispersion, there are no competing products for the other uses of glass. In the case of automotive glass, building glass, mirrors, reflecting glass, insulating glass, laminated glass and armoured glass, flat glass cannot be replaced by other products.

Flat glass must consequently be considered to be a specific market, since, given its mechanical, thermal, optical and decorative properties and the ratio of quality to price, it is able, in its various uses, to meet constant requirements, and it is not interchangeable with or replaceable by other products, except in the case of two uses.

(77) Italy, which represents a substantial part of the common market, must be considered to be the relevant geographical market in which to assess competition. The geographical location of production facilities is a vital factor in the glass industry. Although the cost of transport is not an insurmountable obstacle in marketing flat glass beyond national frontiers, it is certainly a very important factor, since, the greater the distance between the production centre and the points of supply, the more the logistic system becomes critical and the more the competitiveness of the product decreases. Consequently, although each producer exports some of its production, the proportion is limited compared with the quantities sold on the home market, since, if the profitability of the firm is to be maintained, only quantities produced at marginal cost can be sold for export. Local producers are therefore aware that, while they may face some competition from abroad, such competition is quantitatively limited and that they remain the masters over most of the home market. If in addition, as in the present case, most of the sources of supply outside Italy are controlled by groups to which the local producers belong, the risk of competition in respect of the abovementioned limited quantities is further reduced.

The logistic and economic importance of the geographical location of production facilities means that, in order to have regular supplies, consumers are essentially forced to rely on local producers. Consequently, the pressure which consumers can exert on local producers is limited either in terms of quantities or in terms of time: in terms of quantities, because, as has been pointed out, the availability of foreign products is limited; in terms of time, because, even if a consumer manages to purchase a significant proportion of the foreign products available, it will be able to dispense with national products for a very limited period of time and only if it does not need continuity of supplies, but, once the stock of foreign products is finished or if it does need regular supplies, it will be forced to turn to the home market to obtain its supplies.

It follows from the above that the Italian market is the relevant geographical market, since it is the arena of supply and demand in which at least four-fifths of domestic consumption has to be met.

12. The collective dominant position

(78) FP, SIV and VP, as participants in a tight oligopoly, enjoy a degree of independence from competitive pressures that enables them to impede the maintenance of effective competition, notably by not having to take account of the behaviour of the other market participants.

(79) The collective dominant position of FP, SIV and VP derives from the following factors.

The joint market shares of some 79 % for non-automotive glass and some 95 % for automotive glass are sufficient in themselves to give FP, SIV and VP a dominant position on the Italian market in flat glass. These market shares have been fairly stable for several years.

FP, SIV and VP form part of groups having multinational dimensions which control more than half of Community production and supply of automotive and non-automotive flat glass. They are thus to a large extent protected from the competition which, within the limits explained in paragraph 76, might come from abroad. Their direct control of domestic supply and their indirect control of supply from abroad enable the three undertakings to pursue a commercial policy that is not dependent on market trends and the conditions of competition.

The fact that the three undertakings have had to face competitive actions, even continuous ones, from some Community producers and individual competitive actions from non-Community producers does not contradict the above conclusion. As the Court of Justice held in the United Brands and Hoffmann-La Roche cases (1), a dominant position does not preclude some competition nor does it presuppose that the producer or producers holding it have eliminated any possibility of competition. Other producers can actively compete with them without such behaviour adversely affecting them and without the undertakings losing their dominant position. The fact that, despite their efforts, competing undertakings have not managed to weaken the position of the three undertakings on the Italian market and the fact that FP, SIV and VP have thus consolidated their market shares are significant indices of dominance.

The scale of the investments required in order to operate in this industry and the forecast of a small increase in demand over the next decade do not suggest that there will be any structural change in market conditions or that any new producers will enter the market.

The undertakings present themselves on the market as a single entity and not as individuals.

As may be seen from what was said in paragraphs 35 to 42, the three producers jointly maintain special links with a group of wholesalers who are the main glass distributors in Italy, they instigate the meetings, and they do everything possible to get them to accept price list changes and to ensure that the changes are passed on downstream in a consistent manner, so as to prevent any individual decisions by the wholesalers from creating commercial pressures on each producer leading to changes in market equilibria.

The business decisions taken by the three producers display a marked degree of interdependence with regard to prices and terms of sale, relations with customers and business strategies.

The three undertakings have in addition established among themselves structural links relating to production through the systematic exchange of products, as described under heading 6. The exchanges are, firstly, the result of some undertakings' structural lack of primary products or of certain processed products and, secondly, the expression and instrument of their desire to prevent this situation from resulting in changes in their relative positions on the market and in existing relations between them.

13. Abuse of the collective dominant position

(80) FP, SIV and VP have abused their collective dominant position on the Italian market in flat glass, which constitutes a substantial part of the common market.

The business conduct of FP, SIV and VP described in paragraphs 18 to 34, and paragraphs 43 to 47, 49 and 52 constitutes an abuse of a collective dominant position, because it restricts the consumers' ability to choose sources of supply and limits the market outlets of the Community's other flat-glass producers.

The conduct of the undertakings has involved methods other than those on which normal competition in products or services between economic operators is based, thus further weakening the degree of competition on a market in which, precisely because of the collective dominant position of these undertakings, the degree of competition is already reduced.

(81) The conduct adopted by FP, SIV and VP, described in paragraphs 18 to 34 and paragraphs 43 to 47, 49 and 52 also constitutes an abuse within the meaning of Article 86 of the EEC treaty, since it is incompatible with the objective enunciated in Article 3 (f) of the treaty, which provides for a system of undistorted competition within the common market.

With regard in particular to Article 86 (a) and (b), the three producers prevented customers from getting the suppliers to compete with one another on prices and terms of sale and limited outlets through the setting of sales quotas for automotive glass, thus crystallizing established market positions and restricting competing producers' access to the market.

(82) The conduct of FP, SIV and VP, which has the effect of adversely affecting the structures of competition in the flat-glass market, is, for the reasons stated under heading 10, liable to affect trade between Member States within the meaning of Article 86.

D. Article 15 (2) of Regulation No 17

(83) On the basis of the above, the Commission considers that it must impose fines within the meaning of Article 15 (2) of Regulation No 17 on the three undertakings which have infringed Article 85 (1) through the agreements and concerted practices relating to prices and terms of sale, market sharing

and exchanges of products applied in Italy in respect of automotive and non-automotive glass and which at the same time, by abusing their collective dominant position, have infringed Article 86 (a) and (b).

(84) In fixing the amount of the fines, the Commission has had regard to the following factors:

(a) the simultaneous infringement of two provisions of the treaty by the same conduct raises the problem of how fines should be imposed in such a case. In the absence of any express provision in Community law and, in particular, in Article 15 (2) of Regulation No 17, the Commission considers that the principle that fines should not be applied cumulatively in respect of the same set of facts should apply and that therefore only the fines for the more serious infringement should be imposed on the undertakings. In the case in point, in view of the fact that the concept of collective dominant position is being used for the first time under Article 86, the Commission considers that no fines should be imposed under Article 86;

(b) the duration of the infringements. The infringements have been of relatively long duration.

In the case of non-automotive glass, the Commission considers that the agreements and concerted practices on prices and terms of sale and those intended to influence the business decisions of the main wholesalers existed, at least in their most serious forms, from 1 June 1983 to 10 April 1986.

In the case of automotive glass, the Commission considers that: the agreements and concerted practices on prices relating to the Fiat group existed, in their most serious forms, between FP and SIV from 26 October 1982 to 1 December 1986 and between FP, SIV and VP from 11 May 1983 to 1 December 1986; the agreements and concerted practices on quotas relating to the Fiat group existed between FP, SIV and VP from 1 January 1982 to 30 June 1987; the agreements and concerted practices on prices and quotas relating to the Piaggio group existed between FP and SIV from 1 January 1983 to 1 May 1986 and, in their most serious forms, from 1 January 1983 to 28 December 1984. In the case of the exchanges of products, the Commission considers that the agreements between FP, SIV and VP existed from 1 January 1982 to 31 December 1986;

(c) gravity of the infringements. The nature of the infringements, which are of the traditional type and which are clearly covered by Article 85, the economic importance of the undertakings concerned and the position which they hold on the Italian market point to the conclusion that the agreements and concerted practices must be regarded as particularly serious.

The Commission cannot be sure that the infringements have been terminated. However, even if the undertakings have put an end to the infringements, this has not happened spontaneously, but in response to the Commission's action.

Furthermore, the three undertakings have committed infringements in the past. Under Commission decision 81/881/EEC (1), they were found to have infringed Article 85;

(d) attenuating circumstances. As a factor making for a reduction in the amount of the fines, the Commission has taken account of the fact that from 1979 to 1983, in the case of non-automotive glass, and from 1979 to 1984, in the case of automotive glass, there were periods when demand fell and that, consequently, the undertakings suffered losses.

(85) In fixing the amount of the fines to be imposed on the various undertakings, the Commission has taken account of the role each undertaking played in the agreements and concerted practices (the role played by VP having been much less important than that played by FP and SIV), the period of time during which they participated in the infringement, their respective supplies of glass and the total turnover of each undertaking,

HAS ADOPTED THIS DECISION:

Article 1

Fabbrica Pisana SpA, Società Italiana Vetro-SIV SpA, and Vernante Pennitalia SpA have infringed the provisions of Article 85 (1) of the EEC treaty by participating in the following agreements and restrictive practices:

(a) Fabbrica Pisana, SIV and Vernante Pennitalia, from 1 June 1983 to 10 April 1986, in agreements and concerted practices on prices and terms of sale and in agreements and concerted practices designed to influence the purchasing and selling policies of the main wholesalers in the non-automotive flat-glass sector;

(b) Fabbrica Pisana and SIV, from 26 October 1982 to 1 December 1986, and Fabrica Pisana, SIV and Vernante Pennitalia, from 11 May 1983 to 1 December 1986, in agreements and concerted practices on the prices to be charged to the Fiat group in the automotive flat-glass sector;

(c) Fabbrica Pisana, SIV and Vernante Pennitalia, from 1 January 1982 to 30 June 1987, in agreements and concerted practices relating to the apportionment of quotas for supplies to the Fiat group in the automotive flat-glass sector;

(d) Fabbrica Pisana and SIV, from 1 January 1983 to 1 May 1986, in agreements and concerted practices relating to the prices to be charged and supply quotas to be applied to the Piaggio group in the automotive flat-glass sector;

(e) Fabbrica Pisana, SIV and Vernante Pennitalia, from 1 January 1982 to 31 December 1986, in product exchange agreements in the flat-glass sector designed to achieve market sharing.

Article 2

Fabbrica Pisana, SIV and Vernante Pennitalia have infringed the provisions of Article 86 of the EEC treaty by abusing their collective dominant position through conduct whereby they deprived customers of the opportunity of getting suppliers to compete on prices and terms of sale and whereby they limited outlets through the setting of quotas for automotive glass:

(a) Fabbrica Pisana, SIV and Vernante Pennitalia from 1 June 1983 to 10 April 1986 in respect of non-automotive flat glass;

(b) Fabbrica Pisana and SIV from 26 October 1982 to 1 December 1986 and Fabbrica Pisana, SIV and Vernante Pennitalia from 11 May 1983 to 1 December 1986 in respect of prices for automotive flat-glass intended for the Fiat group;

(c) Fabbrica Pisana, SIV and Vernante Pennitalia, from 1 January 1982 to 30 June 1987, in respect of supply quotas for automotive flat glass intended for the Fiat group;

(d) Fabbrica Pisana and SIV, from 1 January 1983 to 1 May 1986, in respect of prices and supply quotas for automotive flat glass intended for the Piaggio group.

Article 3

Fabbrica Pisana, SIV and Vernante Pennitalia shall immediately put an end to the 6. established in Articles 1 and 2 (if they have not already done so) and shall in future refrain, in their flat-glass activities, from entering into any agreement or concerted practice that may have an identical or similar object or effect, including any exchange of information of the type generally covered by professional secrecy such as would allow them to monitor the implementation of any express or tacit agreement or any concerted practice relating to prices or to market sharing.

Article 4

The following fines are hereby imposed on the undertakings to which this decision is addressed, on the grounds of the infringements established in Article 1:

- Fabbrica Pisana SpA, a fine of ECU 7 million,

- Società Italiana Vetro-SIV SpA, a fine of ECU 4,7 million,

- Vernante Pennitalia SpA, a fine of ECU 1,7 million.

Article 5

These fines imposed in Article 4 shall be paid into the Commission's accounts Nos:

(a) Account No 9.130.707 - Commission of the European Communites, Brussels - ECU (for payment in ecus), Istituto Bancario S. Paolo di Torino, 156, Piazza S. Carlo, 10121 Torino

(b) Account No 26952/018 - Commission of the European Communities, Brussels (for payment in Italian lire) Cassa di Risparmio delle Provincie Lombarde, Via Monte di Pietà, 8, 20121 Milano within three months from the date of notification of this decision.

After three months interest shall automatically be payable at the rate charged by the European Monetary Cooperation Fund on its ecu operations on the first working day of the month in which this decision was adopted, plus three-and-a-half pecentage points, i. e. 11 %.

Should payments be made in the national currency of the addressees, the exchange rate applicable shall be that prevailing on the day preceding payment.

Article 6

This decision is addressed to:

- Fabbrica Pisana SpA - Direzione Generale - Via E. Romagnoli, 6 - I-20146 Milano

- Società Italiana Vetro-SIV SpA - I-66050 S. Salvo (Chieti)

- Vernante Pennitalia SpA - Corso Aurelio Saffi, 37 - I-16128 Genova

It shall be enforceable, in accordance with Article 192 of the EEC treaty.

(1) Case 27-76 United Brands v. Commission [1978] ECR 207.

Case 85-76 Hoffmann-La Roche v. Commission [1979] ECR 451.

(1) OJ No L 326, 13. 11. 1981, p. 32.

ANNEX 1

1. Primary transparent glass

(in tonnes)

1,2.3.4.5.6.7 // // // // // // // // 1982 // 1983 // 1984 // 1985 // 1986 // // // // // // // Output: Float glass (producers' figures) Drawn glass (Assovetro figures) // [...] [...] // [...] [...] // [...] [...] // [...] [...] // [...] [...] // Total // [...] // [...] // [...] // [...] // [...] // 1.2.3.4.5.6.7 // + Imports (Istat): // Float glass // 121 409 // 103 979 // 124 579 // 107 238 // 105 070 // // Drawn glass // 39 644 // 42 666 // 50 134 // 40 396 // 61 370 // // Total // [...] // [...] // [...] // [...] // [...] // Exports (Istat): // Float glass // 124 309 // 134 288 // 122 639 // 155 336 // 158 747 // // Drawn glass // 5 901 // 5 982 // 14 087 // 16 513 // 8 182 1,2.3.4.5.6.7 // A. Apparent consumption // [...] // [...] // [...] // [...] // [...] // Imports (Istat) (Float glass + drawn glass) // 161 053 // 146 645 // 174 713 // 147 634 // 166 440 // - Imports from France (1) // 44 475 // 37 875 // 29 320 // 22 805 // 16 310 // - Imports from other countries by the three producers // [...] // [...] // [...] // [...] // [...] // B. Adjusted imports // [...] // [...] // [...] // [...] // [...] // B/A // [...] % // [...] % // [...] % // [...] % // [...] % // // // // // //

2. Primary cast glass

1,2.3.4.5.6.7 // // // // // // // Output (producers' figures) // [...] // [...] // [...] // [...] // [...] // + Imports (Istat) // 39 080 // 34 689 // 57 937 // 58 707 // 66 876 // Exports (Istat) // 14 484 // 9 831 // 5 370 // 1 542 // 2 320 // A. Apparent consumption // [...] // [...] // [...] // [...] // [...] // Imports (Istat) // 39 080 // 34 689 // 57 937 // 58 707 // 66 876 // - Imports from France (1) // 500 // 211 // 1 632 // 2 505 // 3 298 // - Imports from other countries by the three producers // [...] // - // - // [...] // [...] // B. Adjusted imports // [...] // [...] // [...] // [...] // [...] // B/A // [...] % // [...] % // [...] % // [...] % // [...] % // // // // // //

3. Total flat glass (1 + 2)

1,2.3.4.5.6.7 // // // // // // // Output // [...] // [...] // [...] // [...] // [...] // + Imports // 200 133 // 181 334 // 232 650 // 206 281 // 233 316 // Exports // 144 694 // 150 101 // 142 096 // 173 391 // 169 249 // A. Apparent consumption // [...] // [...] // [...] // [...] // [...] // Imports (Istat) // 200 133 // 181 334 // 232 650 // 206 281 // 233 316 // - Imports from France (1) // 44 975 // 38 083 // 30 953 // 25 312 // 19 609 // - Imports from other countries by the three producers // [...] // [...] // [...] // [...] // [...] // B. Adjusted imports // [...] // [...] // [...] // [...] // [...] // B/A // [...] % // [...] % // [...] % // [...] % // [...] % // // // // // //

(1) Imports from France have been deducted, since there are only two producers in France, namely Saint-Gobain, which is the parent company of Fabbrica Pisana, and Boussois, which is an affiliate of Vernante Pennitalia. Imports from these two French companies are mostly intended for the Italian producers. When quantities are sold direct on the Italian market by one of the French companies, they cannot be considered to be sales by competitors.

ANNEX 2

1. Non-automotive market

(in tonnes)

1.2.3.4.5.6 // // // // // // // // 1982 // 1983 // 1984 // 1985 // 1986 // // // // // // // Sales of float glass + drawn glass on the Italian market - GEPVP's figures (1) // [...] // [...] // [...] // [...] // [...] // Automotive self-supply FP-SIV-VP // [...] // [...] // [...] // [...] // [...] // Total market float glass + drawn glass // [...] // [...] // [...] // [...] // [...] // + Apparent consumption Cast glass (Annex 1, point 2) (2) // [...] // [...] // [...] // [...] // [...] // A. Total market // [...] // [...] // [...] // [...] // [...] // B. Total non-automotive sales of FP-SIV-VP // [...] // [...] // [...] // [...] // [...] // B/A // [...] % // [...] % // [...] % // [...] % // [...] % // Market shares of Vetrocoke // [...] % // [...] % // [...] % // [...] % // [...] % (3) // Market shares of the Italian producers (2) // ± 84 % // ± 82 % // ± 79 % // ± 84 % // ± 77 % // // // // // //

(1) GEPVP's figures have been used, because they are more reliable then Assovetro's figures. The GEPVP figures comprise only sales, from which market shares can be calculated, whereas Assovetro's figures comprise sales and stocks.

(2) Since GEPVP does not published any figures on cast glass, Assovetro's figures for this product given in Annex 1 had to be used. Since the figures also include stocks, their gives rise to a number of discrepancies (though not important ones) especially for 1985 and 1986 between the share which the adjusted imports shown in this table would have had and the shares resulting from the table in Annex 1.

(3) As from 1986, Vetrocoke has been a subsidiary of SIV. Vetrocoke's market shares for 1986 will therefore be included with SIV.

2. Automotive market

(× 1 000 m2)

1.2.3.4.5.6 // // // // // // // // 1982 // 1983 // 1984 // 1985 // 1986 // // // // // // // FB + SIV + VP's sales + imports // [...] // [...] // [...] // [...] // [...] // Aggregate shares of FP-SIV-VP // 95 % // 95 % // 95 % // 94,5 % // 95 % // // // // // //

ANNEX 3

Price lists - dates sent to customers (in brackets, dates of entry into effect)

1,6.7,12 // // // Clear building glass // Coloured building glass // // 1,2.3,4.5,6.7,8.9,10.11,12 // Fabbrica Pisana // SIV // Vernante // Fabbrica Pisana // SIV // Vernante // // // // // // 1.2.3.4.5.6.7.8.9.10.11,12 // 26. 6. 1981 // (25. 7. 1981) // 14. 7. 1981 // (14. 9. 1981) // 1. 9. 1981 // ( 1. 10. 1981) // 26. 6. 1981 // (25. 7. 1981) // 14. 7. 1981 // (14. 9. 1981) // // // // // // (cancelled) // // // // // // // // // // // 1. 2. 1982 // ( 1. 3. 1982) // // // // // // 7. 5. 1982 // (15. 6. 1982) // 20. 5. 1982 // (30. 6. 1982) // 24. 6. 1982 // ( 1. 9. 1982) // 7. 5. 1982 // (15. 6. 1982) // 20. 5. 1982 // (30. 6. 1982) // // 7. 3. 1983 // (11. 4. 1983) // 7. 3. 1983 // ( 7. 4. 1983) // 17. 2. 1983 // ( 5. 4. 1983) // 2. 9. 1983 // ( 5. 9. 1983) // 7. 3. 1983 // ( 7. 4. 1983) // // // // // // // // // // 19. 9. 1983 // ( 2. 11. 1983) // // 26. 9. 1983 // ( 2. 11. 1983) // 19. 9. 1983 // ( 2. 11. 1983) // 21. 7. 1983 // (10. 10. 1983) // 27. 12. 1983 // (13. 2. 1984) // 28. 12. 1983 // (20. 2. 1984) // begins to sell coloured glass 1.2.3.4.5.6.7.8.9.10.11.12 // 27. 12. 1983 // (13. 2. 1984) // 28. 12. 1983 // (20. 2. 1984) // 16. 12. 1983 // (15. 2. 1984) // 3. 7. 1984 // (20. 8. 1984) // 7. 5. 1984 // (23. 5. 1984) // 27. 7. 1984 // ( 3. 9. 1984) // // // 7. 5. 1984 // (23. 5. 1984) // 20. 4. 1984 // (21. 5. 1984) // // // // // // // 3. 7. 1984 // (20. 8. 1984) // 3. 7. 1984 // ( 3. 9. 1984) // 27. 7. 1984 // ( 3. 9. 1984) // 12. 11. 1984 // (16. 1. 1985) // 15. 11. 1984 // (15. 1. 1985) // 22. 11. 1984 // (14. 1. 1985) // // // // // 25. 10. 1984 // ( 9. 11. 1984) // // // // // // // 12. 11. 1984 // (16. 1. 1985) // 15. 11. 1984 // (15. 1. 1985) // 22. 11. 1984 // (14. 1. 1985) // 00. 9. 1985 // ( 7. 10. 1985) // 28. 8. 1985 // ( 4. 10. 1985) // 8. 8. 1985 // ( 2. 9. 1985) // // // 11. 3. 1985 // (11. 3. 1985) // 7. 3. 1985 // (14. 3. 1985) // // // // // // // 13. 5. 1985 // (18. 6. 1985) // 13. 5. 1985 // (15. 6. 1985) // 27. 5. 1985 // (24. 6. 1985) // February 1986 // (10. 3. 1986) // 14. 3. 1986 // (17. 3. 1986) // 6. 2. 1986 // ( 1. 4. 1986) // 28. 7. 1985 (1) // ( 1. 8. 1985) // 26. 7. 1985 // (29. 7. 1985) // 31. 7. 1985 // ( 3. 8. 1985) // // // // // // // 25. 10. 1985 (2) // (25. 10. 1985) // 21. 10. 1985 // (28. 10. 1985) // 25. 10. 1985 // ( 4. 11. 1985) // // // 9. 9. 1986 // (29. 9. 1986) // 20. 10. 1986 // (15. 11. 1986) // 10. 3. 1986 // (17. 3. 1986) // 14. 3. 1986 // (24. 3. 1986) // 21. 3. 1986 // ( 1. 4. 1986) // // // // // // // 17. 9. 1986 // ( 1. 10. 1986) // 9. 9. 1986 // (29. 9. 1986) // 30. 9. 1986 // (31. 10. 1986) // // // // // // // // // // // // // // // // // //

(1) On page 43 of its answer to the Statement of objections, Fabbrica Pisana stated that the price increase was announced by telegram: no copy of the telegram was produced during the investigations or attached to the answer to the Statement of objections.

(2) On page 43 of its answer to the Statement of objections, Fabbrica Pisana stated that the price increase was announced by telegram and attached (Item 16) the copy of the telegram. The copy of the telegram, which does not bear any stamp of the post office of departure, is dated at the top 19 October, whereas at the bottom, alongside the signature of the sender, the date indicated is 21 October and the stamp of the post office of arrival is dated 21 October 1985.

ANNEX 4

Analysis of the main Italian wholesale operators and their classification

1.2 // // // Estimate by Fabbrica Pisana (Annex 12 to its answer to the Statement of objections) of total purchases from Italian and foreign producers in 1986: tonnes per year // Classification in categories by FP-SIV-VP 1.2.3.4,6.7,9 // // // // // // Names of wholesalers // Total purchases // Progressive % // 1985 // 1986 // // // // 1.2.3.4.5.6.7.8.9 // // // // FP // SIV // VP // FP // SIV // VP // // // // // // // // // // 1. Vitarelli (W) (1) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 2. Gruppo Sangalli (PW) (2) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 3. Socover (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 4. SAVAS (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 5. Salento (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 6. Marchigiana (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 7. Laborvetro (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 8. D'Adda/Multiglass/Sacilese (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 9. Checchin (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 10. Foschi (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 11. VIC (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 12. Riccardi (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 13. Co. Vetro (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 14. Laziale/Vetralcomi (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 15. Cilvea (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 16. Vetro Brianza (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 17. IVAD (W) // [...] // [...] // [...] // [.. .] // [...] // [...] // [...] // [...] // 18. Bini Vetro (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 19. Tortorici-Fanara (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 20. Covel (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 21. Barbato (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 22. ISV (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 23. Camaeti (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 24. IVAM (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 25. ILVA (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 26. D'Amico (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 27. Callipo (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 28. G.V.A. Valentini (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 29. Sicilglass (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 30. Marotta (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 31. Versari (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 32. VAM Restelli (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 33. Marafiotti (unknown) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 34. Longoni (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 35. Piavevetro (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 36. Scordino (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 37. Ravera (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 38. Sardavetri (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 39. Landi (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 40. Nova Vetro (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 41. Rubei/Tekne (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 42. Cafiero (PW) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // 43. Covet (W) // [...] // [...] // [...] // [...] // [...] // [...] // [...] // [...] // // // // // // // // // // Total // 303 600 // // // // // // // // // // // // // // // //

(1) W = wholesaler.

(2) PW = processing wholesaler.

ANNEX 5

Date of changes in the classification of customers by category or level

1.2.3 // // // // SIV // FP // VP // // // // ? // 1. 1. 1984 // ? // ? // 1. 8. 1984 // 1. 7. 1984 // 20. 11. 1984 // 1. 11. 1984 // 1. 11. 1984 // 8. 2. 1985 // 1. 1. 1985 // 1. 1. 1985 // 13. 3. 1985 // // 1. 3. 1985 // 3. 4. 1985 // 1. 4. 1985 // 1. 6. 1985 // // 1. 8. 1985 // 1. 8. 1985 // 7. 11. 1985 // // 1. 11. 1985 // 20. 1. 1986 // 1. 1. 1986 // // 8. 5. 1986 // 1. 5. 1986 // 6. 6. 1986 // // // 5. 9. 1986 // // //