EC, March 14, 2000, No M.1663

THE COMMISSION OF THE EUROPEAN COMMUNITIES,

Having regard to the Treaty establishing the European Community, Having regard to the Agreement on the European Economic Area, and in particular Article 57 thereof, Having regard to Council Regulation (EEC) No 4064-89 of 21 December 1989 on the control of concentrations between undertakings (1), as last amended by Regulation (EC) No 1310-97 (2), and in particular Article 8 (2) thereof, Having regard to the Commission decision of 10 November 1999 to initiate proceedings in this case, Having given the undertakings concerned the opportunity to make known their views on the objections raised by the Commission, Having regard to the opinion of the Advisory Committee on Concentrations (3), Whereas:

(1) On 20 September 1999, the Commission received notification of a proposed concentration by which Alcan Aluminium Limited (hereinafter "Alcan") would acquire sole control of the undertaking Alusuisse Lonza Group AG (hereinafter "Alusuisse").

(2) On 24 September 1999, the notification was declared incomplete. The parties submitted the missing information on 6 October 1999. The notification became complete on 7 October 1999.

(3) By decision dated 10 November 1999, the Commission found that the notified operation raised serious doubts as to its compatibility with the common market and the functioning of the EEA. The Commission accordingly initiated proceedings in this case pursuant to Article 6 (1) (c) of Regulation (EEC) No 4064-89 (hereinafter "the Merger Regulation").

1. THE PARTIES AND THE OPERATION

(4) Alcan is a Canadian corporation involved in all aspects of the aluminium industry. Its activities include bauxite mining, alumina refining, power generation, aluminium smelting, manufacturing and recycling, and research and development.

(5) Alusuisse is a Swiss corporation involved in all aspects of the aluminium industry. Its activities comprise bauxite mining, alumina refining, aluminium smelting, manufacturing and recycling, research and development, and packaging.

(6) On 11 August 1999 Alcan, Alusuisse and the French integrated aluminium producer Pechiney entered into a Memorandum of Understanding ("MoU") concerning the combination of their respective aluminium and packaging businesses. According to the MoU and to the combination agreements signed by the three companies, Alcan will make two separate and independent share exchange offers, one for the shares of Alusuisse and the other for the shares of Pechiney. Each of the two offers are stand-alone transactions; they are not conditional upon each other and it is possible for one transaction to happen without the other. On the basis of the information given by the parties, Alcan's offer for Alusuisse will be made prior to the launch of Alcan's offer for Pechiney. Each offer could thus be regarded as constituting a separate concentration. This decision concerns Alcan's acquisition of sole control over Alusuisse.

2. COMMUNITY DIMENSION

(7) The undertakings concerned have a combined aggregate worldwide turnover of more than EUR 5 billion (4). Alcan and Alusuisse each have a Community-wide turnover in excess of EUR 250 million, but they do not achieve more than two-thirds of their aggregate Community-wide turnover within one and the same Member State. The notified operation therefore has a Community dimension.

3. COMPETITIVE ASSESSMENT

(8) The merging parties are vertically integrated companies, with activities at all stages of the aluminium supply chain. They mine bauxite, refine it into alumina and smelt alumina into primary aluminium. Primary aluminium (5) is transformed into semi-finished products (flat-rolled products or extrusions). Semi-finished products may be further processed into finished products. Semi-finished and finished aluminium products are used in a wide range of end-use applications such as the transportation industry, building and construction industry, the container and packaging industry, the electrical industry, and so on.

(9) There are various types of intervening operators in the aluminium markets. The merging parties are both vertically integrated producers. Other integrated producers that are active in the EEA market include Pechiney, VAW, Alcoa, Norsk Hydro and Hoogovens. There is a large number of aluminium transformers/manufacturers, which means non-integrated aluminium companies using primary aluminium of flat-rolled aluminium as a raw material in order to produce finished aluminium products. The merging parties may be both supplying and competing against smaller, downstream, non-integrated aluminium transformers or manufacturers on the various final applications markets.

(10) The Commission concluded that the proposed concentration will lead to the creation of dominant positions in the following markets: alumina trihydrate, lithographic sheet, and semi-rigid containers.

3.1.1. ALUMINA TRIHYDRATE

Product market definition

(11) Alumina is produced by refining bauxite. The vast bulk of alumina is used in the smelting of primary aluminium metal (smelter-grade alumina or SGA). Alumina trihydrate ("ATH") is an intermediate product from the SGA production process. Its most notable use is as a flame retardant. A special type of fine precipitated densified ATH is used in this application because it contains a large amount of moisture (normally about 35 %). Its flame-retardant properties stem from its water content. Another type of specially de-watered ATH is used as paper pigment or in the production of detergents. In these applications ATH has a larger particle diameter and is used because of its whitening qualities or to improve gloss and smoothness of high quality papers. Finally, a different type of specialty, activated ATH, is used in the ceramics, refractory and surface treatment industries.

(12) The investigation carried out by the Commission has led it to conclude that ATH used as flame retardant constitutes a distinct product market from other grades and types of aluminas.

Supply-side considerations

(13) The supply-side substitutability of flame-retardant ATH is very low. Only a few alumina refiners are currently supplying it in the EEA. Owing to their focus on metallurgical alumina, most refineries do not have the mechanical installation necessary to intercept ATH between filter and calciner, or they may produce ATH containing a high level of residues from bauxite organic compounds, which makes it unsuitable for use as a flame-retardant. The major suppliers of flame-retardant ATH in the EEA are Alcan, Alusuisse, Nabaltec and Alcoa.

(14) The market investigation has shown that a 5 % to 10 % increase in the price of flame-retardant ATH would not induce immediate new entry, as it would not justify the opportunity cost and the necessary capital investment. Nor could any increase in capacity by incumbent suppliers be only the result of a small, albeit non-transitory, price increase, as capacity increments in alumina production are heavy and costly. In the short to medium term, a price increase in flame-retardant ATH would be profitable.

Demand-side considerations

(15) ATH used as a flame retardant is a distinct product market from other types of ATH and other types of flame retardants. Flame retardants are additives that can be mixed or applied as a treatment to organic materials such as plastics, textiles or timber. They reduce the statistical likelihood of a fire starting by providing improved resistance to ignition in a wide range of potential fire sources. The increasing use of flame retardants in flammable materials, mainly plastic compounds such as cables and wiring, is driven by government regulation and consumers' awareness of fire safety standards.

(16) As was mentioned in recital 11, other types of speciality ATH are used as fillers in the paper industry and as oxides in the production of detergents, ceramics, refractories, etc. The different types of ATH correspond to differences in particle size, morphology, >ISO_7>á->ISO_1>alumina content (6) and impurities. Depending on the end-use requirements, only a given alumina type can match any one intended use. Thus, the fine precipitated densified type of ATH, used as a flame retardant, could not be replaced by ATH used in any of the abovementioned applications, and vice versa.

(17) The effectiveness of a flame retardant is specific for each flammable product. As growing concerns come from the fire-safety requirements for buildings and other indoor public areas, plastic compounds used in such areas need to match specific flame-retardance characteristics.

(18) ATH is low in toxicity and functions as a smoke suppressant. Its largest application is polyethylene and its copolymers, followed by unsaturated polyesters, rubber and epoxies. End-use markets are carpet backings, electrical applications, as well as cables and wires, conveyor belts and extruded plastic products. The growing use of ATH in the EEA is mainly due to the growth of various types of polyolefins, particularly polyethylene. Certain copolymers such as ethylene/vinyl acetate copolymers in mixtures with polyethylene and polypropylene are used increasingly in sheathing and coating for cables and wires and for telecommunication cables placed in underground locations. A trend toward replacing PVC floors with synthetic rubber floors in public buildings is further increasing the use of ATH.

(19) There is another category of alternative flame-retardant substances for the above applications, namely, halogenated flame retardants. Halogenated substances may be efficient flame retardants; however, they are toxic and generate more smoke than inorganic flame retardants. Because of growing environmental concerns - which are expected to lead to the phasing-out of those substances - halogenated phosphorous substances are being replaced by inorganic flame retardants, in particular ATH. As plastic compounds are manufactured under the LSOH label (low smoke, zero halogen), halogenated flame retardants are not substitutable for ATH. There is a separate demand for LSOH compounds, stemming from producers specialising in cables and other plastic products used in the construction industry<. Such producers are identifiable by suppliers and suppliers can, in setting prices, discriminate for or against them. Therefore, a price increase charged to manufacturers of LSOH plastic compounds would be profitable for the merged firm.

(20) As regards the demand for LSOH flame retardants, the notifying parties have pointed to another alternative flame retardant, with properties similar to ATH, namely, magnesium di-hydroxide (MDH). The Commission does not consider that MDH is part of the same market as ATH. There are technical and economic differences that severely limit the demand-side substitutability between ATH and MDH. The cost of using MDH may be twice as high as that of ATH (or three to four times higher according to some users). This is partly due to its relative price, and also to the fact that larger volumes of MDH have to be used in order to achieve high-performance flame retardance. The large amounts of MDH required may in turn affect negatively some key mechanical properties of the compound, such as malleability and impact strength. It is also claimed that it is difficult to achieve uniform particle distribution of MDH within the plastics. MDH is more expensive because it can be used at higher processing temperatures than ATH. While ATH begins to decompose at 230 °C, releasing one third of its original mass as water vapour, MDH decomposes at 340 °C. The difference in the temperature of its thermal decomposition may make MDH more suitable for use in certain polymeric materials such as thermoplastic resins (for example, polypropylene), exceeding processing temperatures of 200 °C. Below such processing temperatures, ATH is the most advantageous filler, in terms of cost and technical properties.

(21) ATH is therefore the preferred filler, to the exclusion of other flame retardants, in LSOH copolymer compounds used in several applications such as cables, wires and floors. Manufacturers of such products replied to the Commission's market investigation that in the event of a 5 % to 10 % increase in the price of ATH, they would not switch to other flame retardants. Some stated that if they were to switch, this would entail high costs, as ATH would have to be replaced by combination of various other retardants that would require the development of new formulations and the approval of the relevant final products by the customer.

Conclusion on the product market definition

(22) On the basis of the preceding points, the Commission reached the conclusion that there exists a distinct product market for fine precipitated densified ATH used as a flame retardant in the manufacture of LSOH copolymer compounds.

Geographic market definition

(23) The geographic scope for ATH is not larger than the EEA. The handling and logistics as well as the relative transport costs of ATH appear to be the main reasons for this determination. Customers of ATH in the chemical and plastics industry require just-in-time deliveries of small consignments which cannot be delivered economically over long distances. The high water content of flame retardant ATH makes transportation over long distances even more costly. Finally, bulk long-distance transportation always involves the danger that the products will deteriorate and become useless. At present, only minor imports into the EEA have been recorded, corresponding to 4 % of EEA consumption.

(24) The parties argued that the ATH market is global and that buyers of ATH would find alternative suppliers outside Europe. These would be Alcoa, which could ship more ATH from Australia and mill it at its milling plant in Moerdijk, Rotterdam; Reynolds, which could ship moist ATH from the USA; Alunorte, which could ship high-quality moist ATH from Brazil; Huber (Solem), which already ships high quality milled ATH from the USA; Sumitomo and Showa Denko, which can sell flame-retardant grade ATH from Japan to Europe; and Hungalu, which is actively promoting itself within the EU and has recently established additional capacity of 10 kt of precipitated ATH per annum.

(25) The Commission market-tested the above alternatives and came to the conclusion that buyers would not consider them as substitutable for local supplies. Some of the major customers said that suppliers of moist ATH from the USA and Brazil are not viable alternatives because they sell only moist ATH, which cannot be used as a flame retardant. As to the suppliers located in Hungary and the Far East, they said that sourcing ATH from these countries would increase their operating and final costs and, as such, those suppliers were not substitutable for the EEA-based ones. Other customers reiterated their initial views, given during the market investigation, that additional costs related to imports of ATH would increase their final price, which would, in turn, decrease the competitiveness of their products (namely, cables) in markets outside the EEA. They would not consider it worthwhile to import ATH in the event of a 5 % to 10 % price increase.

(26) For those reasons, the Commission considers that the market for flame-retardant grade ATH is not wider than the EEA.

Competitive assessment

(27) In the European market, there are a few major ATH suppliers: Alcan, Martinswerk-Alusuisse, Nabaltec, and Alcoa. They represent more than 95 % of EEA production, the remainder being held by some traders (Eurallumina SpA, Giulini Chemie GmbH, Incemin AG, Superfos Biosector a/s).

(28) The following table shows the production capacities of EEA suppliers of flameretardant grade ATH:

<emplacement tableau>

Source:

Commission's inquiry.

(29) The combined market share of the merging parties is [65 % to 75 %]*, the other main competitor being Nabaltec. Alcan and Alusuisse are the world's largest producers of fine precipitated densified ATH. Alusuisse is Europe's largest supplier. Its strongest market position is in Germany, the largest consumer of halogen-free flame retardants in Europe. Alusuisse is also a large producer of MDH, through its 50 % Magnifin Magnesiaprodukte subsidiary located in Austria. Alcan is the second largest European producer, supplying ATH from its Burntisland alumina refinery on the Scottish coast. Nabaltec is an independent producer of specialty aluminas. Since it closed its Bayer process facility in 1994, Nabaltec has been buying raw chemical alumina from third parties for further processing into ATH.

(30) The high market share of the merging parties is already indicative of market power in the relevant flame retardant markets. In addition, there are relevant factors other than the high combined market share that confirm that the proposed operation will lead to the creation of a dominant position held by the merging parties in the market for fine precipitated densified ATH used as a flame retardant.

(31) In the foregoing product-market analysis, the question was considered whether a price increase in ATH would be defeated by either the reaction of other suppliers of various alumina grades or the replacement of ATH by other types of flame retardants. It was concluded that neither supply nor demand-side substitutability could occur in the medium to long term.

(32) The in-depth market investigation considered whether, in the event of a sustained price increase in ATH, customers would take measures enabling them, in the long run, to replace ATH with other flame retardants that came into play only recently.

(33) In response to the statement of objections, the parties argued that flame retardant materials other than ATH would constrain the price of ATH. These can be: phosphorous; magnesium di-hydroxide; naturally occurring MDH (brucite) or MDH derived from seawater; new materials such as "MagShield" or "Ultracarb"; other inorganic products such as zinc borate, zinc stannate, zinc hydroxystannate, tin oxide, molybdenum trioxide and melamine.

(34) The Commission market-tested these alternative materials with manufacturers of LSOH plastic compounds that use fine densified precipitated ATH as a flame retardant. Their answers suggested that no switch to such materials would occur in the long run, even in the event of a sustained price increase.

(35) One of the major buyers of ATH stated that all of the materials referred to in recital 33 are known and that several are used, generally as the lesser constituent when their synergistic properties are exploited. However, none could totally replace ATH, particularly on an equal cost basis. Certain phosphorus-based compounds are effective flame retardants, but are currently over five times the cost of precipitated ATH. In addition, during a fire, acid gases are created with phosphorus systems, this being unacceptable and not the case with ATH. In addition, materials containing phosphorus are prohibited by most client requirements. Moreover, all precipitated MDH is three to four times more expensive than ATH. One customer uses a modest quantity of this material for high temperature, specialist compounds, but price precludes its use in the majority of its applications. As to "Magshield" and "Ultracarb" users said that both materials have significant technical drawbacks and that they may be satisfactory in small amounts only to supplement predominantly ATH-based formulations. As to the inorganic systems, customers said that they may be useful, again as minor components within synergistic ATH-based formulations, but in general do not offer high levels of flame retardance. In any case their cost is much higher than that of ATH. Melamine is an organic compound, which is mainly used in synergistic combinations. However, an important drawback of this material is its release of toxic gases in a fire. This prevents it from qualifying as a perfect demand-side substitute. In the view of another customer, "none of the above materials can directly replace ATH and consequently, they have a limited value in constraining ATH prices. Because of what is in general their very high cost premium, ATH price could rise significantly (perhaps to double its existing level) before [a user]* could consider them as possible replacement materials. Even then a number of technical problems would have to be tackled, some of these probably being insurmountable". The Commission therefore concludes that, in the long run, the abovementioned materials could not offer sufficient potential competition in the event of a sustained price increase in ATH. They cannot be viewed as alternatives but rather as complementary materials to ATH.

(36) The parties also argued that ground ATH can be used instead of fine precipitated densified ATH. A price increase in ATH would induce customers to switch to lower grades of precipitated ATH or to fine ground ATH. In this respect, there are many grinders in Europe which could start grinding and selling ATH as a flame retardant.

(37) The Commission market-tested these alternative materials with users of fine densified precipitated ATH. In summary, their reactions were negative to the above alternatives. More particularly, they considered the statement that "ground ATH can be used instead of fine precipitated densified ATH" to be incorrect. In general, ground materials, for example ATH, MDH, and calcium carbonate, have flame retardance properties similar to precipitated grades, but because of their much larger particle size, they cannot match their physical properties. An example is tensile strength and surface finish. If, in fact, this were not the case there would be no market for precipitated products, which are always several times more expensive than ground products. Recently a number of technically modified precipitated grades, such as the Alcan ESD range, have been introduced to give improved processing characteristics. These features are unavailable from conventional precipitated, let alone ground grades.

(38) As to the statement that "many grinders in Europe could start grinding and selling ATH as a flame retardant", the reaction of customers was also negative. Customers pointed to the fact that grinding ATH is not a low-tech operation that any "grinder" could easily start up. Moreover, from a technical point of view, the shortcomings of ground products preclude their use for sophisticated applications.

(39) The parties also argued that, in the long run, a supra-competitive pricing of ATH would induce entry of new ATH capacity in the market. The Commission considers, however, that only limited entry might be prompted and that, as such, it would not be sufficient to curtail the market power of the parties. The possible entry scenarios and their consequences on competition in the ATH market are discussed in the recitals 40 to 44.

(40) According to the parties, the incumbent producer, Nabaltec, has available capacity of fine precipitated densified ATH of 4 kt to 5 kt. This producer would be prompted to place it on the market in the event of a supra-competitive price increase in ATH. Such an additional tonnage would correspond up to a mere 2,7 % of the current ATH market and will not have any meaningful effect on the price of ATH.

(41) The parties argue that Alcoa could also react to a price increase. It would then need to produce the material at low cost at its Australian refineries and ship it to Europe, store it in large inventories, process it at its milling plant in Moerdijk, the Netherlands, and start its distribution from there. In this way, Alcoa could bring some additional 10 kt of ATH to the EEA, corresponding to 6,5 % of the current market. The ATH was defined as being an EEA market, precisely because imports are not economical. This also applied to Alcoa. The possible cost advantage stemming from importing material from Australia would fade away on account of the costs involved in such imports (additional investment and time for imports to be logistically handled, customs duties, transportation costs, and so on). Therefore, the competitiveness of Alcoa's imports could not be considered a threat to the exercise of market power by the merged firm.

(42) The parties also argue that a sustainable price increase might also prompt the entry of the Hungarian producer Hungalu. It has recently developed an additional capacity of 10 kt of fine precipitated ATH, corresponding to 6,5 % of the current market. The parties believe that this additional capacity could be made available for sale in the EEA. Finally, the parties argue that companies with grinding facilities, not currently active in the sale of ATH, could start selling milled ATH within a few months of a price increase.

(43) The Commission considers that such companies, whose core business is not in the ATH market, could hardly become established suppliers of ATH. They would not have control of the production/supply chain, as they would be dependent on producers of commodity ATH for their supplies. They would then have to acquire know-how and technology to convert the raw ATH material into fine precipitated densified ATH that can be used as a flame retardant. This is likely to increase their cost basis, therefore removing any price advantage that they could offer compared to a supra-competitive price of the merged firm. This may also pose market entry problems. Buyers of ATH are sensitive to the quality of the material they buy. They are themselves accountable to their customers, and each time that they change supplier or their flame retardant mix, they need to inform, and obtain approval from, their own customers. It is highly unlikely that such entry prospects could discourage an anti-competitive price increase by the merged firm.

(44) However, even assuming that all of the above potential suppliers entered the market at the same time and with the maximum quantities projected above, this would reduce the market share of the parties from [65 % to 75 %]* to approximately [55 % to 65 %]*. The merged firm would still be by far the largest local producer and supplier of ATH in the European flam retardant market. As these quantities would be the maximal possible competitive response to a price increase, the loss of [5 % to 15 %]* market share would not be sufficient to make an anticompetitive price increase unprofitable.

(45) The parties argued that excess capacity in ATH would prevent any supra-competitive price increase. However, buyers of ATH stated to the Commission that prices of ATH in Europe have recently increased and thus that these price increases run counter to the argument that prices are constrained by excess capacity. In addition, several customers stated that they had difficulties in satisfying the whole of their demand for ATH. This had not, however, induced them to seek alternative sources outside the EEA.

(46) As ATH does not compete with magnesium hydroxide or any other alternative filler in LSOH compounds manufactured in lower processing temperatures, users of ATH would be deprived of any countervailing power. They would still depend on EEA-based ATH suppliers for those applications where ATH is the most suitable filler. In their replies to questionnaires, most users stated that, as ATH could not be replaced for such uses, they would have to accept the increase and, where possible, pass it on to their customers. Their customers would also accept the increase in cost, as the fire safety requirements they have to observe would prevail over such a cost variation.

Conclusion

(47) On the basis of the foregoing, the Commission concludes that the notified operation will result in the creation of a dominant position in the EEA market by the parties for ATH used as a flame retardant in the manufacture of LSOH copolymer compounds.

3.1.2. FLAT-ROLLED PRODUCT MARKETS

(48) Flat-rolled products (FRPs) are semi-finished aluminium products that are produced in aluminium rolling mills. They constitute the raw material for the manufacture of several finished aluminium products (ranging from industrial aluminium plate to household foil). FRPs are produced in a number of steps, involving hot and cold rolling in hot and cold rolling mills, respectively.

(49) FRPs comprise over 15 categories of products, some of which correspond to a specific end application, whereas some others may be multipurpose products (a large part of the latter is stockist material which cannot be allocated to specific sectors). The categories of FRPs that mills can produce include the following: common material alloy (1xxx, 3xxx and 5xxx); building sheet; foil stock; beverage can body stock; can end stock; food can stock; bright sheet; brazing sheet; plate and shate (sheet and plate); lithographic sheet; fin stock; other thin gauge (3xxx and 5xxx); autobody; and heat treat.

(50) The parties have argued that FRPs constitute one single relevant market as there is a high degree of supply-side substitutability. In previous Commission decisions concerning the aluminium sector, FRPs had been considered as one product market(7). The Commission notes that these decisions concerned aluminium producers with limited FRP activities in the EEA, and accordingly the notified operations did not require any thorough analysis of the FRP markets in the EEA since no competition problems resulted from the operations.

(51) The market investigation in this case has shown that the conditions of competition are not the same across all types of FRPs. First, the degree of supply-side substitutability varies from one FRP type to another. In general, aluminium rolling mills produce a range of FRPs, the "product mix". Different types of FRPs sell at different prices and their relative profitability is reflected in the rolling margin (that is, the profit margin resulting from rolling a particular type of FRP). Each rolling margin "contributes" to the overall profitability of the mill. As the ultimate goal of a mill is to maximise its profitability, aluminium producers try to optimise the product mix of their mills, by producing the highest-margin types of FRPs within the availability of their rolling and finishing equipment. However, not all the mills are equipped to produce all types of FRPs. Thus, when it comes to several specific types of FRPs - in this case lithographic sheet - only a few mills in the EEA are capable of producing them. Those types of FRPs are products relatively difficult to make and, as a result, they require a high degree of management commitment, workforce discipline and different operating practices. In addition, the rolling margins achieved in producing these types of FRPs may be lower than in other, more standardised types. The high quality requirements and the relatively small rolling margins have dissuaded several mills from setting up production processes that would enable them to include those products in their product mix. This is particularly so because the buyers of such types of FRPs require their suppliers to pass stringent qualification tests that may last several years before a long-term commercial relationship is set up. Therefore, although the supply-side substitutability argument seems relevant in respect to the standard FRPs categories (standard sheet, plates, foil stock, etc.), it was not supported by the market investigation with respect to products such as those dealt with in the present decision. Rolling mills not currently active in the production of those types of FRPs could not start competing in those markets in a timely and immediate fashion, even in case of a non-transitory price increase of 5 % to 10 %. Therefore, on the supply side, the presence of a large number of mills cannot influence the competitive conditions prevailing in those types of FRPs which can be defined as distinct product markets.

(52) Secondly, the conditions of competition in those distinct markets are not influenced by the existence of aluminium mills outside the EEA. The parties have argued that the FRPs' geographic market is not limited to the EEA but comprises eastern European countries and Turkey, as well as Russia and the CIS. In earlier decisions in the aluminium sector, the Commission had defined the markets for FRPs as being EEA-wide. In this case, the investigation on this specific type of FRP (lithographic sheet) showed that competition takes place at the EEA level. Import duties vary from 7,5 % to 12 %. Even if some non-European countries have duty-free access to the EEA market, for the customers the geographic market for those types of FRPs is determined more by the quality and technological guarantees of the producers and by the need for just-in-time deliveries and short lead times. Long distance transportation requires a very good control of the supply chain management, stocks and logistics. As non-EU producers, but also some smaller EU-based producers, reportedly have a lower level of supply reliability, customers are required to hold larger inventories for their supplies from eastern European countries and Turkey, relative to the average rate of sales of their finished goods, than for their supplies by incumbent EU-based suppliers, including the parties. The larger inventories increase the customers' working-capital related costs and therefore their unit costs. For this reason, almost no imports whatsoever have been recorded in the EEA from Russia, Turkey or the eastern European countries, in particular for the two types of products mentioned above. Moreover, for the same reasons mentioned above, in case of a non-transitory price increase of 5 % to 10 %, customers would not shift their purchases to these countries. Therefore, mills located there could not be considered to be part of the EEA market.

(53) As far as the remaining types of FRPs are concerned, imports from Russia and the CIS in 1998 amounted to 0,7 % of EEA consumption, whilst those from eastern European countries and Turkey came to less than 5 %. The categories of FRPs that are produced at rolling mills in those countries comprise the following: standard 1xxx, 3xxx and 5xxx sheet; building sheet; foil stock; plate and heat treat - that is, seven out of at least 15 categories of FRPs.

(54) As a conclusion, the Commission does not accept the argument that FRPs should constitute one single product market nor that its geographic scope should include the EEA, eastern Europe, Russia and Turkey. There is at least one type of FRP, examined in detail in this Decision, which constitutes a distinct product market of EEA geographic dimension. This is lithographic sheet. The proposed concentration will result in the creation of a duopolistic dominant position held by the merging parties and VAW in the market for lithographic sheet.

(55) However, the merger will also bring about a high degree of concentration at the overall FRP production level in the EEA. The notifying parties' rolling capacity in the EEA will be [25 % to 35 %]*, whereas their combined share in terms of actual production will be of [20 % to 30 %]*. Moreover, the merging parties will have the largest overcapacity among all the FRP producers in the EEA [( > 30 %)]*. Finally, the merging parties will have the largest number of mills (13 out of 36) and therefore the largest range of possible output, hence the best opportunity to optimise their product mix across all the FRP types. This is in sharp contrast with the position of the remaining competitors. VAW, the immediate competitor of the merging firms, will have a rolling capacity of [< 20 %])*, that is, almost three times smaller than that of the merging firms, and a capacity utilisation of [ > 80 %]*. Other competitors lag even further behind. Their capacity and production figures are the following: Alcoa (12 % - 12 %); Hoogovens (9 % - 9 %); Elval (3 % - 3 %); Norsk Hydro (3 % - 3 %), Granges (2 % - 2 %).

(56) The combined market shares of the merging parties in flat-rolled products will be, in themselves, indicative of market power, in particular taking into account the significantly lower market positions of the immediate competitors. If one adds to that the effects of the concentration on the merging firm's immediate competitor, VAW, then the overall competitive situation on FRPs applications will be seriously distorted. As will be seen in recitals 58 to 84, this is so because the merged firm and VAW will share the largest aluminium rolling mill worldwide, Alunorf in Germany, which is currently a 50:50 production joint venture between Alcan and VAW.

(57) In order to explain better the competitive assessment in those markets a number of preliminary considerations are necessary concerning the links between Alcan and one of its main competitors, VAW.

3.1.2.1. THE NORF JOINT VENTURE BETWEEN ALCAN AND VAW

(58) Aluminium Norf GmbH ("Norf") is a joint venture company owned 50:50 by Alcan and VAW. It was established in 1965 to build and operate the aluminium rolling plant at Norf, Germany. It is a production joint venture operating on a tolling basis for both its parents.

(59) As far as the corporate structure of Norf is concerned, the GmbH acts as a joint venture vehicle. Its capital share is owned 50:50 by Alcan and VAW. The GmbH owns legal title to the property and the equipment at the rolling plant and employs its own work force. The GmbH is run by two chief operating officers (Geschäftsführer) who are employees of the GmbH. Each parent company is entitled to propose one of the Geschäftsführer; however, the selection of each requires the approval of both parent companies. The Geschäftsführer are responsible for the day-to-day management of Norf.

(60) The GmbH has a board of directors, composed of two directors appointed by Alcan, two appointed by VAW and two employee representatives. A shareholder meeting is convened twice a year.

(61) A partnership company (Aluminium Gemeinschaft Norf - "BGB") is set up to act as a steering committee. It is composed of three Alcan representatives, three VAW representatives and the Geschäftsführer. Certain important decisions require approval of the BGB ([...]*).

(62) The legal framework for the cooperation of the parent companies of Norf relies on a certain number of agreements. Those agreements show that the parents have agreed on a consensus basis for everything that has been carried out at Norf.

(63) [This paragraph describes the provisions of the constituent acts of the joint venture, such as the partnership agreement and other covenants. In summary, the two parent companies have joint control over Norf; business is jointly run by both shareholders and management is exercised by representatives of the parent companies; day-to-day business by two representatives, each appointed by one parent company; representatives decide unanimously and follow instructions of parent companies, otherwise the matter is referred to shareholders' partnership; important decisions need the consent of the parent companies; parent companies grant Norf access to their know-how; parent companies may use each other's patents at no cost; for additional investments, the parent companies have to decide on unanimously]*.

(64) The production operations of Norf are conducted according to the following principles. Each parent company is entitled to 50 % of the available capacity or "time of the mill". This time of the mill corresponds to the time required to produce a tonne of standard FRP. When the parent companies want to produce other types of FRP (i.e., beverage can stock, lithographic sheet, etc.), usually more time is required, as the number of passes from the mills increases. An "equivalency factor" corresponds to the extratime needed for the production of every other type of FRP compared to the standard FRP. The Norf management sends each parent company the equivalency factors by product or type of FRP that could be produced at Norf. Each parent company then sends to the management of Norf its forecast in standard tonnes. If the forecast of the parent companies does not meet the 50 % capacity entitlement in the month, the parent companies must both accept the deviation. Fixed costs are shared equally by the parent companies. Variable costs are attributed by means of a "tolling fee" and are charged according to the utilisation of the mill.

(65) The Norf management informs the parent companies of the capacity utilisation of each parent companies, aggregated across all FRP types and the aggregate weighted "equivalency factor" applicable to each parent company's production. The parent companies are also informed of their share of the fixed costs as well as of the total amount of variable costs to be borne by each parent company, aggregated across all FRP types and calculated on the basis of standard tonnes.

(66) As for the production process, each parent company informs the Norf management of the production process to be used (such as the number of passes). The Norf management is not supposed to disclose this information to the other parent company.

(67) The parties argue that the Norf joint venture is not a classical production joint venture but a time-sharing facility. Consequently, they claim that Norf will not restrict competition between the parent companies, as a result of the merger. It has been structured in such a way, the argument goes, as to minimise the information flow between the parent companies and not to allow competitively sensitive information to be passed between the parent companies. Thus, according to the parties, on parent company could not anticipate the competitive actions of the other. Moreover, the parent companies do not share competitively sensitive information, whereas information on the parent companies' own customer bases, selling prices, quantities of finished product delivered, and such like is beyond the scope of the joint venture, and therefore remains out of the parent companies' reach. As the final product differentiation may take place at a later stage for certain products and even performed at other mills, outside Norf (as in the case of lithographic sheet), ultimately one parent company could not know the exact product mix nor the differing cost basis of the other parent company. Finally, the parties argue that since the Norf arrangements do not involve profit-sharing between Alcan and VAW, it is not rational for the parent companies to engage in collusive behavior.

(68) The Commission considers that the degree of interdependence stemming from the joint operation of Norf prevents it from qualifying as a simple time-sharing facility. In substance, Norf operates as a production joint venture and any examination of it should take into account the cooperative aspects of its joint operation. These include the governing principle of consensus between the parent companies, resulting from the various agreements concerning the operation of Norf. The large amount of industrial cooperation to be achieved between the parent companies is an indication that the degree of integration of the parent companies in Norf goes beyond what could be described as a time-sharing facility. Norf is the only rolling mill joint venture world-wide (8). When it was set up, it was considered to be a rather risky project as it was already generally accepted that it could only be operated at a sufficient level of efficiency if both parent companies were to pool resources. At Norf, there are joint production facilities and a single labour force serving the needs of both parent companies in the same way. In terms of the costs pertaining to the operation of Norf, [...]*. At Norf, there is one common department responsible for determining the production sequences for both parent companies, which makes its technical findings available to both parent companies. According to one of the parent companies, VAW, it would not have made economic sense to jointly build and operate Norf if any given FRP type produced by one of the parent companies were to be produced differently by the other parent company.

(69) At Norf, there are several dedicated production lines (including both the hot- and cold-milling process and the finishing machines, such as the slitters) for the various types of FRPs that Norf can produce. The dedication of lines stems from the common interest of both parent companies in reducing their common costs. Consequently, [...]*.

(70) Investments in Norf have, so far, been jointly agreed by both parent companies. For instance, the determination of the proper technology has been decided on jointly. Technical specifications of investments are jointly discussed and agreed upon at the BGB.

(71) For the foregoing reasons, it can be concluded that the Norf joint venture is not a time-sharing facility. Consequently, the Commission considers that the presence of the two largest EEA producers of FRPs in a joint structure which manufactures products in which the merged firm will acquire important market positions, may pose competition problems. These will stem from the legal structure of Norf and the resulting de facto cooperation between the parent companies; from the strong dependence of VAW on Norf and its capacity constraints in its other rolling mills; from the right of Alcan to veto investments proposed by VAW; and from the ability of the merged firm to raise VAW's costs and hence decrease its ability to compete. Overall, the asymmetrical capacity constraints, in favour of the merged firm, and the credible threat of retaliation against VAW will create a situation where collusion among the parent companies of Norf will be sustainable in the long run. This situation is analysed in recitals 72 to 81.

(72) First, the legal structure at Norf requires a constant consensus-based cooperation between the parent companies. For Norf to be able to continue its operations effectively after the merger, the parent companies will have to agree on several aspects of the production process. This will increase the transparency and predictability in the competitive strategies of the parent companies. For instance, technical and financial cooperation between the parent companies will be necessary in order to [...]*. Not only will major investments require the consent of both parent companies, but also routine operations will create opportunities for cooperation (such as [...]*). The argument of the parties on the arrangements concerning the management and operation of Norf, in particular those related to the exchange of information, disregards the difference between a pre-merger situation and a post-merger one. The question is whether the competitive situation would change following the proposed merger. In that respect, the parties may argue on an ex post basis, that in the pre-merger situation, the exchange of information did not prevent the two parent companies, Alcan and VAW, from competing effectively with each other. By contrast, the post-merger relation between the merged firm and VAW is a matter of the future, to be assessed in the context of the market structure likely to prevail following the notified acquisition. After the merger, for Norf to be able to effectively continue its operations, Alcan and VAW will have to agree on several aspects concerning the production process.

(73) Secondly, the merger will alter VAW's incentives to compete. Ultimately, VAW would have more incentives to align its competitive behaviour on that of the merged firm rather than compete aggressively against it. This stems from the combination of two factors: VAW's strong dependence on the output of Norf and its capacity constraints, and the right of Alcan to veto expansion or other investments (the latter in conjunction with the overcapacity of the merged firm, which will have the largest underutilised rolling capacity in the EEA).

(74) VAW is totally dependent on Norf for a number of FRPs (beverage can body stock and end stock, lithographic sheet, brazing sheet, etc.). Before the merger, Alcan was also dependent on Norf; however, after completion of the merger, the combined entity will no longer depend on Norf, as additional hot-rolling capacity will be brought along by the other merging partner. If VAW decides to compete fiercely against the merged firm in view of capturing additional market share, it will need to increase its rolling capacity. Thus, VAW will need to reach an agreement with the merged firm. The latter may block such an investment if it perceives that this will enable VAW to compete against it.

(75) The parties have argued that VAW has announced investments aiming at the expansion of its capacity in other rolling mills (Cisterna and Hamburg) and that such expansions would reduce VAW's dependence on Norf and would provide it with the needed capacity in order to compete against the merged firm. The Commission does not agree with the argument of the parties. The planned capacity expansions concern VAW's cold-rolling mills in these plants. These are, and will continue to be, supplied with hot re-roll band (input) coming from Norf. Those mills could not be supplied by other hot mills of VAW as there are technological differences between the hot-rolling processes between Norf and those other mills (different product dimensions make the Norf system incompatible with other mills). Therefore, contrary to the argument of the parties, the planned expansions will increase VAW's dependence on the hot mills of Norf.

(76) Thirdly, economic theory suggests that the firms' asymmetrical capacity constraints affect the degree of collusion in the short run as well as in the long run. This is further enhanced by the possibility of a credible threat of retaliation. In this case, the existence of asymmetric capacity in favor of the merging entity and the existence of de facto punishment mechanisms, stemming from the functioning of Norf, will affect VAW's incentive for engaging in a price war.

(77) The condition relating to the asymmetry of capacity constraints is illustrated by the merged firm's capacity utilisation of [< 70 %]* compared to that of VAW which is [ > 85 %]*. The condition relating to the credible threat of retaliation is illustrated by the asymmetry in the degree to which VAW and the merged firm are dependent on Norf. The merger will enable the merged firm to raise its rival's costs, should the merged firm and VAW remain the parent companies of the Norf joint venture. Because of the additional rolling mills it will acquire from its merging partners, the merged firm will be more flexible on the supply side to shift production to and from Norf than is currently the case for Alcan. Consequently, the merged firm could increase VAW's costs at Norf while reorganising its production around the merged firm's new rolling system (it was announced that such a post-merger reorganisation would result in cost savings of several hundred million dollars). Ultimately, the benefit of raising VAW's costs is that VAW would be forced to charge higher prices to customers, and this would increase the demand for the merged firm's output, thus enabling it to raise the prices. Alternatively, if VAW decided not to increase its prices, it would suffer a decrease in its profitability.

(78) The way in which the merged firm may increase VAW's costs whilst reorganising its product mix at Norf is described as follows. As was mentioned in recitals 58 to 77, the functioning of Norf is based on the common optimisation of the parent companies' product mix. Any change in the product mix is likely to alter the cost situation at Norf. Under certain circumstances, the change in the cost structure may be more harmful for one parent company than for the other. For instance, as the costs to be borne by each partner are based on the factual level of utilisation of the facilities, should the merged firm decide to redistribute some FRPs away from Norf or to replace certain FRPs with others, VAW would have to bear a disproportionate share of the costs arising out of the unused capacity. Any change in the product mix of Norf will have repercussions on the operation of the production process and of the accompanying equipment (cast house, types of alloys, scrap chain, number of passes, finishing equipment and so on). This will disturb the cost-sharing equilibrium on the basis of which Norf has been able to function optimally to date. Additionally, in the day-to-day operation of Norf, the parent companies will have to collaborate to work efficiently. The diminution of the relative commitment by one parent company will have negative repercussions on the profitability of the other. Given VAW's strong dependence on Norf, it is more likely that a relaxation of the merged firm's commitment to Norf will raise VAW's costs and affect its ability to compete.

(79) [...]*, owing to the way in which the cast house costs are shared at Norf and to the way in which the costs of the rolling mills are allocated, it is technically feasible for the merged firm to raise VAW's production costs at Norf, either by reducing the amount of activity at Norf or by changing its product mix at Norf. [...]*. The parties consider that a strategy consisting in raising rivals' costs would be sound if the deterioration of VAW's cost basis were greater than that of the merged firm. In this case, a reallocation of the product mix of Norf would not make economic sense if it had the result of leaving the merged firm's share of Norf unutilised. According to the parties, this would amount to a severe opportunity cost. Thus, a change in the product mix of Norf, or the redistribution of certain products away from Norf, would have to be offset by keeping Norf reloaded. [...]* the impact on VAW's and the parties' marginal cost as a result of the reallocation of lithographic sheet and beverage can body stock away from Norf and the reloading of Norf with other types of FRPs. In the case of lithographic sheet, the parties have estimated that both parent companies' marginal costs will increase and that VAW's costs will increase by [%]* more than those of the merged firm. In the case of other FRP types, and in particular beverage can body stock, the increase will be sharper for VAW, as the reallocation of that product away from Norf will result in cost savings for the merged firm ([- %]*) and a cost increase for VAW ([+ %]*). Overall, the resulting cost difference to the detriment of VAW will be close to [ > 5 %]*.

(80) The Commission cannot agree with the parties that a cost penalty borne by VAW will inflict a dissuasive cost increase on the merged entity. As seen in the preceding paragraph, under both reallocation scenarios analysed by the parties, VAW's cost penalty is bigger that the merged firm's. Nor can the Commission accept that VAW's cost variation is too small to prevent it from competing aggressively. As has been stated several times by the parties, the margins achieved in the rolling business are minimal. Accordingly, a small scale variation in the cost basis may have significant consequences on the competitiveness of a producer. As a result, VAW will not be keen to engage in a price war. It will, on the contrary, align its behaviour on that of the merged firm, in the expectation of maximising its profits at a given output through higher prices rather than higher market share.

(81) In addition to that, and quite apart from the current dependency of VAW on Norf, the fact remains that any attempt on the part of either the new entity or VAW to gain market share at the expense of the other will necessarily have an immediate effect on the rate of capacity utilisation/product mix at Norf and consequently on the cost structure of both parties. The very existence of Norf thus modifies the parties' incentive to compete. Furthermore, the very fact that the cost basis of all FRP types manufactured at Norf will be the same or at the very least very similar for both parties, will in itself diminish the amount of price competition that the parties can realistically practice against each other.

(82) Furthermore, the merged firm and VAW compete with each other in a multitude of other markets. The prospects for retaliation in such other markets, stemming from multi-market contacts, can provide a further rational reason for the alignment of strategies between the merged firm and VAW.

(83) VAW explained in public, at the oral hearing held on January 31 and February 1 2000, how it views its future competitive position as a result of the merger. VAW believes it will be under a permanent threat of retaliation through measures which will fundamentally affect its supplies from Norf and therefore its competitiveness. For each activity, VAW will have to anticipate the possible reaction of its joint venture partners and will tend to refrain from activities which could trigger a reaction by the merged firm. This will reduce VAW's willingness to engage in fierce competition.

(84) As a consequence of the foregoing, it is the opinion of the Commission that the existence of Norf will reduce competition between the merged entity and VAW. In accordance with the approach which must be followed in the context of the merger control review, the Commission is to assume the most rational course of action which, in the markets analysed in recitals 85 to 105, would be the alignment of the competitive strategies of the merged firm and VAW.

3.1.2.2. LITHOGRAPHIC SHEET

Product market definition

(85) Lithographic sheet is a type of bright and well-surfaced flat-rolled product used in the manufacture of photographic and printing plates. Aluminium rolling mills produce this special type of product and sell it to photographic companies that have facilities to surface-treat it, slit it, and resell it to end users, such as their subsidiaries or third parties (other photographic companies and printers of newspapers, magazines, etc.).

(86) There is no demand-side substitutability between aluminium lithographic sheet and any other material, whether made of metal or not. The parties claim, however, that from the supply side there exists a high degree of substitutability.

(87) The market investigation has indicated that this is not true. The crucial technical requirement for lithographic sheet is that it shall have a very flat surface which must be carefully de-greased. Given the specific know-how and special machine capabilities involved in producing that flat surface, customers would not consider all established aluminium mills as their potential suppliers, in the event of an increase in the price of lithographic sheet.

(88) Customers require their suppliers to pass a stringent qualification process before they will start puchasing from them. Such a process can take up to two years before qualification and generally includes a multi-phase testing programme aimed at determining the final product specification required by the customer. For a full entry, in addition to the certification procedure, producers have to undertake investments of between EUR 10 million and EUR 20 million, and must constantly keep up with improvements in quality. The example of Elval, referred to in the following paragraphs, illustrates how long it can take an aluminium mill to establish itself in the market place as a qualified supplier of lithographic sheet. As barriers to entry in this market are high, only a limited number of aluminium producers are active (Alcan, Alusuisse, VAW and Alcoa; Elval is at the stage of entering the market). A price increase of 5 % to 10 % would not result in any timely successful entry by non-existent suppliers, as the degree of commitment, the capital cost and the time lag would make prospects of the entry uncertain. Overall, a supra-competitive price increase of that magnitude would remain unchallenged. For these reasons, the Commission considers that lithographic sheet is a distinct relevant product market.

Geographic market definition

(89) The geographic market for lithographic sheet is restrained to the EEA. All of the EEA-based customers purchase lithographic sheet from local suppliers. One of the reasons limiting imports into the EEA is the custom duty of 7,5 %. Another is the long lead times for imported material. Transport times for locally purchased material may be of one to two days, as opposed to two to three weeks for imported material. Imports compel customers to buy large quantities and set up and follow large inventories. Another reason is the uncertainty associated with imports. Remote suppliers cannot provide the same quality of after-sales technical assistance to EEA-based customers; moreover, the quality of the sheet may differ substantially for lithographic sheet imported into the EU. This is in particular so for the thickness and the width of the sheet. Customers have given examples of differing sizes of sheet in every different delivery. This entails an extra cost or a loss of product for the customer. When the material is larger in size, it has to be trimmed off to fit in the customer's machinery, which results in losses of output. When it is narrower, nothing can be done and the product needs to be returned or spoiled (that is to say, resold at a lower price to be used as a standard flat-rolled product).

(90) One customer has imported some limited quantities from a Japanese supplier. This customer is the European subsidiary of a Japanese photographic company, and owing to the existing supply agreements between the parent company and its local Japanese supplier, it was made possible to divert to Europe part of the volume purchased in Japan. These imports accounted, at the maximum, for 20 % of that customer's supply requirements. They were made under an outward-processing customs system. This consists in importing a limited quantity of a product in order to process it within the Community with a view of re-exporting it as an added-value product - that is, under a different customs product number. Thus the amount of duties paid on importation is refunded when the product is re-exported.

(91) For the rest, imports accounted in 1998 for 2 % of EEA consumption. The parties argue that imports accounted for approximately 5 % in 1994/95 and that the reduction in imports was due to lost volumes from the USA, which was in turn due to the strong appreciation of the US dollar. However, even assuming that the US dollar depreciates against the euro zone, and that imports from the USA will rise, this would not be sufficient to point to the conclusion that the market is consistently broader than the EEA. Such imports would depend on business cycles and currency fluctuations. They would not represent a structural feature of the market, as they would not constitute a stable and reliable source of supply. Therefore, for the present analysis, it can be concluded that the market for lithographic sheet is EEA-wide.

Competitive assessment

(92) The EEA production for lithographic sheet was of 189 kt in 1998. The merged firm accounted for [kt]* or [45 % to 55 %]* of the EEA production (Alcan: [kt]* or [%]*; Alusuisse: [kt]* or [%]*). Only a few other competitors produce and supply litho in the EEA. These include VAW ([35 % to 45 %]*), Alcoa ([5 % to 15 %]*) and Elval ([< 5 %]*).

(93) The degree of concentration in the litho market is significant. The merger will reduce the number of present suppliers from five to four. Of those, VAW is the most likely to constitute a serious competitive threat to the possible exercise of market power by the merged firm. Before the merger, VAW was the biggest producer and supplier of litho in the EEA. However, after the merger, the Commission can no longer consider VAW as a genuine competitive threat to the merged firm. VAW finishes its lithographic sheet production at the plants of Grevenbroich and Cisterna, which account respectively for 90 % and 10 % of VAW's lithographic sheet production. These plants are entirely dependent on Norf for the supply of hot re-roll band of lithographic sheet (see above, chapter on the Norf joint venture). There are no alternative sources of supply of hot re-roll bands with the necessary lithographic quality, as the only producers of such bands, Alusuisse, Alcan and Alcoa (Inespal), are competitors and therefore unlikely to supply VAW. Norf is a 50:50 joint venture between Alcan and VAW. In the foregoing analysis of the Norf joint venture, it was explained why the merger will alter the incentives of VAW to compete with the new entity, should they remain as the parent companies of Norf. The individual market characteristics of the merged firm and VAW, as well as the general features of the lithographic sheet market, suggest that as a result of the merger the economic interests of the two competitors might be aligned.

(94) The market positions of the merged firm and VAW are almost symmetrical ([...]*). The same goes for the production costs, since, on the one hand, the initial steps of production (hot rolling) take place at Norf, whereas, on the other hand, the individual processes followed at separate plants involve the same technology and thus production costs. There is a degree of transparency prevailing in the joint production unit, and as a result the two partners could, with no difficulty, track each other's actions. This adds to the predictability required for a competitor to detect in time what action competitors are going to take and therefore to align his behaviour on that of competitors. This similarity in the production process and the functioning of the joint venture strengthens the relation of interdependence between the parent companies, whose ultimate goal would be the optimal functioning of their venture. A breach of this optimum would act on both parents as a sufficient disincentive against competing head-on.

(95) Lithographic sheet is a fairly homogeneous product. Once the know-how and technology is acquired, a producer cannot much differentiate his lithographic products from those produced by competitors. Price is therefore the relevant factor in competition in this market. The lithographic market is relatively transparent. The prices are negotiated on a yearly basis. There are a few customers only and both the merged firm and VAW are supplying them. When negotiating with buyers, the suppliers get to know the price, or at least the price range offered by their competitors. In that way, buyers try to bring different suppliers to competition with one another in the expectation of achieving a fair market price. As a result, bidding suppliers tend to know what the market price is, at any given moment.

(96) The lithographic sheet market is stagnant. This is due to the fact that the demand from the offset industry is declining, due to the development of new forms of paper-free media, such as electronic and Internet media. Stagnant markets are more conducive to parallel behaviour than high-growth markets, as in the latter there are more incentives for entry and higher incentives for existing players to compete for an increase in market share. In contrast, in the lithographic sheet market, the incentives to compete are different, as a competitor can only increase his customer base at the expense of another competitor. Price is again the relevant competitive factor.

(97) Finally, the above market characteristics are corroborated by the existence of a substantial structural link between the two main competitors, that is, the Norf joint venture. The existence of Norf will change significantly the incentives of both parent companies to compete. In particular, VAW might consider it more profitable to follow a price increase in lithographic sheet initiated by the merged firm. In the absence of sufficient competitive response from existing local suppliers or from imported material, a parallel price increase would lead to joint profit maximisation. As was explained in recitals 58 to 84 on the Norf joint venture, the need for the two parents to optimise the product mix of Norf and keep the cost basis at the lowest possible level makes cooperation between the parent companies inevitable. Any future investment or change in the product mix will have to be decided jointly. In sum, decisions concerning the output of Norf will involve both parent companies. Moreover, VAW would have no incentive to compete in order to steal customers from the other partner. As was explained in the relevant paragraphs on Norf, if VAW is to compete it will be necessary to increase its capacity at Norf. The merged firm may block such an investment. Additionally, the merged firm could dissuade VAW from competing aggressively through the credible threat of retaliation. This can be achieved through raising VAW's costs at Norf, therefore rendering it non-competitive. The parties have calculated that even a simple reorganisation of the merged firm's product mix at Norf may result in VAW's having costs [%]* higher than the merged firm. Therefore, should the merged firm and VAW continue to jointly control Norf, VAW could not be regarded as an effective actual competitor that would be keen to defeat a supra-competitive price increase in lithographic sheet.

(98) Alcoa is the only real competitive force in the market. However, its lithographic production capacities in the EEA are limited. Given that it only accounts for 9 % of the market, in order to defeat a price increase it will have to either convert a large part of its current product mix into lithographic sheet (assuming the relevant opportunity costs), or to increase its capacity. Capacity increases do not take place just to satisfy demand for one product. They entail a big investment, they are managed as a system and are motivated by more global considerations than a snapshot price increase in one product market, such as lithographic sheet. If Alcoa wished not incur an expansion cost, it would then have to import material from its US mills. However, the transportation costs and the import duties would erode any price competitiveness of imported material. Under these circumstances, Alcoa might consider it more profitable to follow the price increase, thus maximising its profits per given quantity sold.

(99) Finally, Elval is the smallest of the existing suppliers. Elval has attempted to enter the lithographic market since 1990. Even today, it does not consider itself able to produce and supply quality material in a consistent way. Elval has produced material on an experimental basis. Its first test shipments of 500 tonnes were made in 1993 - that is, three years after the decision to enter this market was taken. The qualification process with EEA customers is not yet terminated and Elval believes that further investment and time will be required until full acceptance by customers is achieved. Elval cannot, therefore, be considered a competitive threat in the EEA lithographic market.

(100) The example of Elval also illustrates the significant barriers that prevent entry in the lithographic market. On top of the EUR 10 million to EUR 20 million investment in setting up a finishing line, aluminium mills not active in lithographic sheet will have to undertake a significant investment in time and resources in order to qualify with customers. The time lag is such that any competitive threat from their side would not be meaningful before a long period of time, which could amount to more than five years.

(101) Hoogovens, one of the largest aluminium producers and currently part of the Corus group (British Steel and Hoogovens), was active in the production of lithographic sheet but quit the market recently. Hoogovens stated that when it was still active, it was not able to keep up with the increasing quality requirements set by customers, nor could it envisage the necessary investments of EUR 20 million, in order to catch up with the fact that its main competitors set up finishing lines specifically geared to the manufacture of this product.

(102) That being so, it can be concluded that actual or potential competition is not likely to alleviate the market power that the merged entity will acquire as a result of the concentration.

(103) The parties have argued that customers have considerable buying power, as they are large corporations in the photographic industry, which is as concentrated as the aluminium industry.

(104) Nevertheless, the market investigation has indicated that customers do not necessarily have countervailing power. The lithographic sheet market is highly concentrated, as is the photographic industry. Before the merger, there are no indications that customers have buying power. After the merger, the concentration on the part of the lithographic sheet producers will increase substantially. Considering the effects of Norf on the competitive behaviour of the merged firm and VAW, and the low competitive prospects of Elval, the available EEA suppliers of lithographic sheet will change from four to two, of which Alcoa is substantially smaller. The rapprochement of the largest and most efficient pre-merger suppliers is expected to adversely affect the existing equilibrium in terms of countervailing power. The loss of countervailing power will be accentuated by the impossibility of switching to other alternative materials and the long time period necessary for potential new entrants to establish operations and acquire know-how and recognition. Imports from Japan or the USA will increase the operating costs of customers and will affect their profitability. As a result, imports could not constitute a credible threat to local suppliers. Overall, customers will not be able to exercise any countervailing power after the merger.

Conclusion

(105) For the reasons outlined in the preceding paragraphs, the Commission considers that the concentration will result in the creation of a duopolistic dominant position, held by the merging parties and VAW, in the EEA market for lithographic sheet.

3.1.3. PACKAGING MARKETS

3.1.3.1. SEMI-RIGID CONTAINERS

Relevant product market

(106) Semi-rigid aluminium containers are put to a number of end-uses such as petfood containers, frozen food containers, take-away food trays, airline food containers and the like. Semi-rigid aluminium containers are either coated (smoothwall container) or uncoated (wrinklewall container). The parties are of the opinion that the relevant market comprises both types of aluminium containers, as well as containers made of other materials such as multilayer plastic, steel, or pouches. The market investigation, however, has not supported that view.

Demand side substitutability

(107) In the view of all customers contacted, semi-rigid smoothwall containers constitute a separate product market from wrinklewall containers. Wrinklewall containers cannot be sealed but only closed with a lid of aluminium or metallised carton. Wrinklewall containers are used for frozen food and an extensive variety of bakery and catering products which do not need a long shelf life. Any food in a wet state, however, needs to be packaged in a sterilised and sealed container in order to guarantee a safe and hygienic long shelf life. Coated aluminium provides the heat resistance (129 °C) needed for sterilisation. Only the smoothwall container has the sealing edge on to which a lid can be sealed and which then provides a shelf life of more than two years. This difference in properties is reflected in a considerably higher price for the smoothwall container. The price differential between coated and uncoated aluminium containers is in the order of 25 % to 50 %.

(108) According to the parties the multilayer plastic container provides a viable alternative to the aluminium smoothwall container. However, the shelf life of plastic containers is only about six months - one quarter of the sterilised smoothwall aluminium container. This fact puts plastic containers at a considerable disadvantage vis-à-vis the aluminium container. More than 60 % of coated semi-rigid aluminium containers are used for the packaging of petfood. Another 10 % to 15 % contain sterilised food for human consumption. Therefore, for more than 70 % of the smoothwall containers a long shelf life is guaranteed which leads to a completely different supply chain. It is therefore not uncommon that sterilised food arrives on the shelves at the point of sales weeks or even months after it has been produced.

(109) Plastic containers suffer also from a marketing point of view. The parties claim that Mars uses multilayer plastic containers for its products in the USA. However, Mars does not use this container in Europe, despite the fact that it is 20 % cheaper. The reason is that the appearance of plastic does not square with the premium brands packaged in the aluminium container. The parties pointed to a Swiss company, Säntis, which sold such a container for packaging pet food to the largest retailer in Switzerland, Migros, following allegations that aluminium containers are more harmful to the environment than such a plastic container. However, as it turned out the multilayer plastic container is more difficult to recycle than the aluminium container. Accordingly, Migros switched back to aluminium in 1997, and the multilayer container is not in use any more.

(110) Another alternative proposed by the parties is the steel container. Rigid steel is considerably heavier than aluminium, which makes it unsuitable for the 100 g single-serve container. It is therefore predominantly used for the multiserve segment with a content of 400 g or more. As a general marketing point, customers point to the fact that aluminium has certain properties in terms of marketing which are not achieved by steel. These properties include greater convenience in opening and, particularly important for premium brands, appearance. There is also a semi-rigid steel container. However, the price for this laminated semi-rigid steel container is, according to the parties, 15 % to 20 % higher.

(111) Lastly, the parties point to the pouch as an alternative to smoothwall aluminium containers. Pouches have a thin inner aluminium layer of 6 to 9 µ and can be sealed. Pouches guarantee a similar long shelf life as the smoothwall container. They are available in the 100 g single-serve size. This product was able to gain market share in recent times. The Commission underlines, however, that this gain in market share was not at the expense of the smoothwall containers but rather at the expense of the tinplate can. The growth forecast for the pouch is, according to one market player, 20 % for the coming years, whilst the aluminium container will grow at a rate of 5 % but the tinplate can will be stagnant or even decline. This reflects the fact that demand for the single-serve size is growing fast, a size where tinplate plays a very minor role.

(112) Moreover, pouches contain a product with a semi-liquid consistency, namely chunks in gravy whereas smoothwall containers contain a jellied or a paté-like product. Jellied and paté-like products cannot be packaged in pouches. Therefore, a switch to pouches would require a reformulation of the product. Consequently, pouches are not considered by all customers as an alternative to the smoothwall container. Moreover, switching to pouches would also require new filling lines and an advertising campaign. It has been estimated by one customer that these switching costs are as high as EUR 10 million. Therefore, there is only limited demand-side substitutability between semi-rigid coated aluminium containers and pouches.

(113) The parties claimed that many food products are packed in both plastic and smoothwall semi-rigid containers. These products include single-serve jams and marmalades, coffee creamers or cheese spreads. However, these products are not sterilised when packed in aluminium and do not have the same shelf life. Similarly, ready meals for use on board airplanes are packed in smoothwall aluminium containers but are not sterilised. Therefore, for these products and applications demand-side substitutability exists. However, as was mentioned above, these applications account for less than one third of the demand for smoothwall containers.

(114) For the above reasons, demand-side substitutability for the packaging of sterilised food products, between smoothwall aluminium containers, and multilayer plastic containers, steel containers and pouches seems to be very limited.

Supply-side substitution

(115) The parties claim that there is a high degree of supply-side substitutability between producers of wrinklewall and smoothwall containers. Smoothwall containers are made of laminated aluminium foil, referred to as laminated strip. This strip is the material which is pressed into the shape of a container and to which the lid is added to close the container. The strip is standard container foil to which a polypropylene strip is added on a laminating machine. The parties claim that there is only limited know-how and no intellectual property rights are required for producing the laminated strip. The only equipment necessary is a laminator to glue the polypropylene strip to an aluminium FRP. According to the parties the costs of such a machine is around [EUR]*.

(116) A producer of wrinklewall containers can produce smoothwall containers on the same container presses, if capacity is available. A new container press with a production capacity of 100 million containers (5 % of the market) costs between EUR 160000 and 350000. However, in order to actually produce smoothwall containers such a producer of wrinklewall containers would have to change the tooling. Such a forming tool for smoothwall containers costs, according to the market investigation, up to EUR 125000. Since customers usually order several different shapes and sizes, various tools are necessary to meet such demand. In addition, presses for the lid are necessary, which cost around EUR 150000 including tools. Therefore, a producer of wrinklewall containers would have to invest between EUR 1 million and EUR 2 million in new machinery. The total investment necessary to convert aluminium foil into laminated strip and to start producing smoothwall containers could amount to as much as EUR 12 million. Accordingly, the Commission considers that smoothwall and wrinklewall containers belong to separate markets.

Conclusion

(117) For the above reasons, the conclusion of the Commission is that there is a separate market for semi-rigid coated aluminium (smoothwall) containers for the packaging of sterilised products, as was suggested in the Commission Decision of 14 April 1993 in Case IV/M.322 - Alcan/Inespal(9).

Relevant geographic market

(118) In the view of the parties, semi-rigid aluminium containers can be transported at low cost throughout Europe and prices are similar throughout Europe. It follows, according to the parties, that the relevant geographic market is at least the EEA. The market investigation revealed that imports into Europe are marginal and that logistical considerations, such as just-in-time delivery, are very important. Consequently, the relevant geographic market is, for the purposes of this Decision, limited to the EEA.

Competitive assessment

(119) The market share of the parties concerning smoothwall containers amounts to [50 % to 60 %]*. Alusuisse would add [%]* to Alcan's [%]*. Other competitors include deSter ([10 % to 20 %]*), Alupack ([5 % to 15 %]*), Pluspack ([< 10 %]*) and Teich ([< 10 %]*).

<emplacement tableau>

(120) Only one competitor is vertically integrated, namely Teich. Teich is part of the Teich/Constantia group which has a stake in AMAG, a hot- and cold-rolling mill in Austria, and also produces the laminated strip. All other competitors depend on the supply of the laminated strip from third parties. According to market players, such non-integrated producers as Pluspack in fact do nothing more than die-cutting and pressing aluminium into containers, whereas the technology is developed by the supplier of the aluminium foil and the laminated strip. According to the parties [ > 70 %]* of the final price of a smoothwall container is determined by the cost of the aluminium foil and the lamination.

(121) The parties have a very strong position for this material, but claim that there are many independent suppliers of laminated strip. However, apart from VAW, these companies like Feron or Omial are small firms with small capacities. The parties claim that Feron could easily increase its capacity by [between 2000 and 8000 tonnes]* by increasing its number of shifts or changes in the product mix. However, this allegation was rejected by Feron. In the case of Omial, customers were of the opinion that Omial does not produce the required high quality. In a letter submitted by the parties Omial even asks the parties for technical cooperation. Moreover, these companies are pure converters who have to buy the aluminium foil from integrated companies such as VAW, or the parties which are direct competitors on the downstream market for the laminated strip.

(122) However, in one instance even the integrated company Teich was facing severe problems. Some years ago demand shifted towards container lids which were coated by an extrusion process. Teich did not have this extrusion capability in-house. The parties to the merger refused to supply this type of lid to Teich, which led to a very difficult situation for Teich.

(123) According to the parties, the demand side is highly concentrated. However, even the largest customer for smoothwall containers, Mars, accounts for 25 % of the total market as opposed to the market shares of the parties of [%]*. Moreover, no individual customer of Alcan accounts for more than [%]* of its sales, whereas the biggest customer of Alusuisse accounts for only [%]* of its sales. In the case of the market leader Alcan, the top three customers account for [ > 50 %]* of its sales, the figure for Alusuisse is [ > 50 %]*. In turn, top customers such as Mars, Nestlé or Saturn source between 90 % and 100 % from the merging parties. Consequently, there is an imbalance in bargaining power between the merging parties and customers of coated semi-rigid aluminium containers used for the packaging of petfood.

(124) The parties claim that prices for smoothwall containers have decreased by 10 % to 20 % over recent years. However, this reflects a market situation where the two leading firms competed. The merger will unite the top two suppliers of the smoothwall container, which will be six times bigger than the nearest rival. While it seems to be correct that customers could switch to another supplier in a time frame of up to one year, it will be a problem to find a supplier which has the spare capacity to meet this demand. If in the event of a post-merger price increase a company like Mars were to shift considerable parts of its demand to the two smaller suppliers of the petfood container, Teich and Alupack, they would have to invest. However, any investment in new capacity carries the risk that the parties will apply a limit-pricing strategy so that it would be economically more viable for small producers to follow an increase in prices by the merging parties.

(125) Such a rise in prices could be passed on to the final consumer by Mars which is the undisputed market leader in petfood with a share of almost half the market and no close second. Mars has the largest collection of premium brands. There is only one other petfood producer with similar brands albeit much smaller: Nestlé. The price of the smoothwall container accounts for 30 % of the final sales price of the petfood. An increase of 5 % would result in an increase of 1,5 % of the final product which Mars would not find difficult passing on to the final consumer. The other petfood suppliers such as Saturn and Arovit produce private labels. They would most likely follow the lead, since the price gap between branded products and the private labels would remain the same.

Conclusion

(126) On the basis of the foregoing, the Commission concludes that the notified operation will result in the creation of a dominant position in the EEA market for semi-rigid coated aluminium containers.

4. COMMITMENTS PROPOSED BY THE NOTIFYING PARTY

(127) On 22 February 2000, the notifying party offered certain commitments to remove the competition concerns which the Commission had identified in its statement of objections of 14 January 2000. The commitments will be summarised and assessed in the following points, following the order of the relevant markets on which the Commission stated its objections as followed above in the assessment part of this Decision.

(128) The full text of the commitments is attached to this Decision and forms an integral part of it.

4.1.1.1. ALUMINA TRI-HYDRATE

(129) The ATH Business, according to the parties, means the ATH plant operated by Algroup at Martinswerk, Germany and all related human resources, tangible and intangible assets.

(130) Alcan undertakes to divest itself of the ATH Business within a period of nine months of the date of the Commission decision declaring the concentration compatible with the common market (hereinafter "the date of the Decision"). Should Alcan not be able to effect the divestment by the end of the ninth month from the date of the Decision, it may request the Commission to extend the nine months' period by another three months. Moreover, should Alcan not be able to effect the divestment by the end of the ninth or the 12th month from the date of the Decision, as the case may be, Alcan will give an irrevocable mandate to a trustee to effect the sale of the ATH Business within a period of three months, at no minimum price.

(131) In order to enhance the commercial viability of the ATH Business to prospective purchasers, the ATH Business will include the assignment to the purchaser of all existing contracts, and all contracts entered into between the date of these commitments (22 February 2000) and the conclusion of the sale of the ATH Business, which are related to, or associated with, the ATH Business, and sufficient sales staff and production and administrative personnel to operate the ATH Business as a going concern.

(132) Alcan is to be deemed to have complied with the undertaking if by the ninth, the 12th or the 15th month, as the case may be, it has entered into a binding contract for the sale of the ATH Business with a purchaser to be approved by the Commission.

Assessment

(133) The proposed undertaking removes the overlap of the merging parties' activities. The market test carried out by the Commission has indicated that the proposed ATH plant may be considered a viable business. Moreover, the modalities of implementation of the proposed undertaking, in particular the transfer of the business as a going concern, the hold-separate provisions and the divestiture period, will guarantee that the ATH Business will not suffer any damage during the divestiture period and that effective competition will be restored in the ATH market after the divestiture. If these conditions are fulfilled, the status quo ante prevailing before the concentration will be re-established so that the concentration, subject to the fulfilment of the proposed undertaking, will not lead to the creation of dominant position in the EEA market for ATH.

4.1.1.2. LITHOGRAPHIC SHEET

(134) The Litho Business, according to the parties, means the lithographic operations currently carried out by Algroup in the Star mill at Bridgenorth, United Kingdom, including the cast house, the hot mill, all cold mills and the litho finishing line.

(135) Alcan undertakes to divest itself of the Litho Business within a period of nine months of the date of the Decision. Should Alcan not be able to effect the divestment by the end of the ninth month from the date of the Decision, it may request the Commission to extend the nine months period by another three months. Moreover, should Alcan not be able to effect the divestment by the end of the ninth or the 12th month from the date of the Decision, as the case may be, Alcan will give an irrevocable mandate to a trustee to effect the sale of the Litho Business within a period of three months, at no minimum price.

(136) In order to enhance the commercial viability of the Litho Business to prospective purchasers, the Litho Business will include the assignment to the purchaser of all existing contracts, and all contracts entered into between the date of the commitments (22 February 2000) and the conclusion of the sale of the Litho Business, which are related to, or associated with, the Litho Business; and sufficient sales staff and production and administrative personnel to operate the Litho Business as a going concern.

(137) Alcan undertakes to offer to provide the purchaser of the Litho Business with technical assistance to continue to use the existing Algroup technology currently used by the Litho Business, for three years from the date of conclusion of the sale of the Litho Business.

(138) Alcan is to be deemed to have complied with the undertaking if within the nine, the 12 or the 15 months' period, as the case may be, it has entered into a binding contract for the sale of the Litho Business with a purchaser to be approved by the Commission.

Assessment

(139) The proposed undertaking removes the competitive overlap in the lithographic sheet market. The lithographic production of the merged firm is organised, for Alcan in the Norf mill (hot coils) and in one of Norf's satellite mills (finishing), and for Alusuisse in the Bridgenorth mill (hot coils and finishing). The parties propose to divest themselves from the hot mill and the related cast house, as well as all cold mills of Star. One of these cold mills currently produces mainly foilstock and the other litho stock. The Commission confirmed with third party operators in the market that provided the hot mills are offered to the purchaser, the latter will become an autonomous producer and supplier of lithographic sheet, to the extent that he will control the quality of the metal throughout the whole supply chain. Moreover, in order to make the divested package more attractive to prospective purchasers, the notifying party proposed to divest itself of a cold mill currently producing foilstock. This additional measure was proposed in order to alleviate concerns raised during the market test, namely that the divested hot mill would produce more hot re-roll band than could be absorbed by the cold litho mill. This could act as an economic obstacle in the profitable operation of the hot mill, as the purchaser would have to find outlets to dispose of this extra capacity. By acquiring the foilstock cold mill in addition to the litho mill, the purchaser of the divested assets will be able to cold roll the surplus output of the hot mill and sell foilstock, a commodity FRP, to the market, thus making profitable the operation of the hot mill in particular, and of the Litho Business in general. Under these circumstances, lasting and effective competition in the lithographic sheet will be restored as a result of the undertaking proposed by the notifying party. Furthermore, the implementation and accompanying measures, in particular the transfer of the business as a going concern, the hold-separate provisions, the divestiture period and the offer of technical assistance to the purchaser, will guarantee that the Litho Business will not suffer any damage during the divestiture period and that the purchaser will be able to compete effectively in the Litho market after the divestiture. If these conditions are fulfilled, the status quo ante prevailing before the concentration will be re-established so that the concentration, subject to the fulfilment of the proposed undertaking, will not lead to the creation of dominant position in the EEA market for lithographic sheet.

4.1.1.3. SEMI-RIGID ALUMINIUM CONTAINERS

(140) Alcan undertakes to sell to a third party, approved by the Commission, punching machines with an aggregate annual capacity of at least 3200 tonnes for the manufacture of smoothwall semi-rigid-containers and the accompanying lids (hereinafter, the "transferred semi-rigid aluminium containers business"). In addition, if requested by the third party purchaser, Alcan undertakes to grant a licence on normal commercial terms in respect of Alcan's existing technology (and any know-how associated with such technology) used in the lamination of aluminium foil for the manufacture of smoothwall semi-rigid-containers and the accompanying lids.

(141) Alcan undertakes to effect the transfer of the semi-rigid aluminium containers business within nine months of the date of the Decision to a third-party purchaser approved by the Commission. Should Alcan not be able to effect the transfer by the end of the ninth month of the date of the Decision, it may request the Commission to extend the nine months period by another three months. Should Alcan not be able to effect the divestment by the end of the ninth or the 12th month from the date of the Decision, as the case may be, Alcan will give an irrevocable mandate to a trustee to effect the transfer of the semi-rigid aluminium containers business within a period of three months, at no minimum price.

(142) In order to enhance the commercial viability of the transferred semi-rigid aluminium containers business to the prospective purchaser, it will include the assignment to the purchaser of all customer contracts which are related to, or associated with, the transferred semi-rigid aluminium containers business. The transfer of the customers' contracts is to include the relevant customer-specific moulds and other equipment related to these contracts.

(143) Alcan will offer to the purchaser a non-competition and a non-solicitation clause of a duration of three years (which can be extended at the request of the purchaser by another two years, such request being made to the Commission), pursuant to which Alcan or the merged entity will not seek or solicit customers of the transferred semi-rigid aluminium containers business who have accepted the assignment of their contracts to the purchaser, unless the contracts have been terminated by the purchaser.

Assessment

(144) The proposed undertaking intends to reinstate a new producer and supplier of semi-rigid smoothwall aluminium containers. It concerns the divestment of an annual capacity corresponding to the competitive overlap. According to the results of the market test, the business would be divested as a going concern if the new acquirer were to have access to the punching machines, customer contracts and the relevant contract-specific moulds and other equipment, as well as to Alcan's lamination technology. The transfer of the customer contracts will be effective provided that it is accompanied by non-competition and non-solicitation clauses. Alcan has complied with all of these requirements. If these conditions are fulfilled, a new competing force will be re-established in that market so that the concentration, subject to the fulfilment of the proposed undertaking, will not lead to the creation of dominant position in the EEA market for semi-rigid smoothwall aluminium containers.

Conclusion on the proposed commitments

(145) Based on the foregoing assessment of the proposed undertakings, the Commission concludes that they are sufficient to eliminate the competition problems identified in the markets for ATH, lithographic sheet and semi-rigid smoothwall aluminium containers,

HAS ADOPTED THIS DECISION:

Article 1

On condition that the undertakings summarised in recitals 127 to 145 and set out in detail in the Annex are fully complied with, the concentration by which Alcan Aluminium Limited ("Alcan") acquires, within the meaning of Article 3(1) (b) of Regulation (EEC) No 4064-89, control of Alusuisse Lonza Group AG ("Alusuisse") is declared compatible with the common market and the functioning of the EEA Agreement.

Article 2

This Decision is addressed to:

Alcan Aluminium Limited

1188 Sherbrooke Street West

Montreal, Quebec, Canada

Mr David Ausland

Executive Vice-President

(1) OJ L 395, 30.12.1989, p. 1; corrigendum OJ L 257, 21.9.1990, p. 13.

(2) OJ L 180, 9.7.1997, p. 1; corrigendum OJ L 40, 13.2.1998, p. 17.

(3) OJ C 82, 5.4.2000.

(4) Turnover calculated in accordance with Article 5 (1) of the Merger Regulation and the Commission Notice on the calculation of turnover (OJ C 66, 2.3.1998, p. 25). To the extent that figures include turnover for the period before 1.1.1999, they are calculated on the basis of average ECU exchange rates and translated into EUR on a one-for-one basis.

(5) Primary aluminium is a commodity which is traded at the London Metal Exchange (LME). A daily quotation, the LME price, is fixed there reflecting the supply and demand for primary aluminium.

(6) Refers to thermodynamic stability.

(7) Alcoa/Alumix (Case IV/M.675, Decision of 21 December 1995; OJ C 121, 25.4.1996, p. 14); Alcoa/Inespal (Case IV/M.1003, Decision of 24 October 1997; OJ C 29, 27.1.1998, p. 7); VAW/Reynolds Metals (Case IV/M.1110, Decision of 19 May 1998; OJ C 211, 7.7.1998, p. 17).

(8) There is another rolling mill joint venture between Alcan and ARCO in Logan, USA. However, this is not comparable to Norf in many respects. First, ARCO has divested its aluminium activities and the Logan mill is only managed by Alcan; secondly, the Logan mill is much smaller in scale and therefore less complicated to operate than Norf (Norf is the largest aluminium mill in the world); and thirdly, Logan can only produce can body and end stock and, therefore, even if it were jointly run, it would not entail such a complex management of two different competitive product mixes.

(9) OJ C 114, 24.4.1993, p. 5.

ANNEX

The full original text of the conditions and obligations referred to in Article 1 may be consulted on the following Commission website: http://europa.eu.int/comm/competition/index_en.html