EC, November 21, 2001, No 2003-215

THE COMMISSION OF THE EUROPEAN COMMUNITIES,

Having regard to the Treaty establishing the European Coal and Steel Community, and in particular Article 66 (2) thereof, Having regard to the Notice concerning the alignment of procedures for processing mergers under the ECSC and EC Treaties (1), Having regard to the notification submitted by the parties dated 18 June 2001 and to subsequent information, Having regard to the Commission letter of 19 July 2001 communicating to the parties the Commission's serious concerns, Having given the companies concerned the opportunity to reply to the Commission's objections, Having regard to the final report of the Hearing Officer in this case (2), Whereas:

(1) On 18 June 2001, the Commission received a notification, under Article 66 of the ECSC Treaty, of a proposed concentration whereby the undertakings Arbed SA (Arbed), incorporated in Luxembourg, Aceralia Corporación Siderúrgica SA (Aceralia), incorporated in Spain, which is controlled by Arbed (3), and Usinor SA (Usinor), incorporated in France, merge into a new entity, provisionally named Newco Steel (Newco), to be incorporated in Luxembourg, by way of a public offer to exchange the shares of Aceralia, Arbed and Usinor with those of the new entity.

(2) After examining the notification, the Commission has come to the conclusion that the proposed concentration falls within the scope of Article 66 of the ECSC Treaty, read in conjunction with Article 80 thereof. In so far as the proposed concentration concerns products or markets which do not fall within the scope of the ECSC Treaty (namely, tailored blanks, construction profiles, sandwich panels and steel tubes), Council Regulation (EEC) No 4064-89 of 21 December 1989 on the control of concentrations between undertakings (4), as last amended by Regulation (EC) No 1310-97(5) applies (6). This Decision concerns only those parts of the proposed concentration which fall within the scope of the ECSC Treaty.

I. THE PARTIES

(3) Aceralia is a company incorporated as a sociedad anonima under Spanish law. It is controlled by Arbed. It is the largest steel producer in Spain and manufactures and distributes flat steel products, long steel products and processed steel products. These activities fall under the ECSC Treaty. It also produces certain types of small carbon welded tubes and products for the construction industry and it carries out transformation activities. These latter activities fall under the EC Treaty.

(4) Arbed is a company incorporated as a société anonyme under the laws of Luxembourg. It is a major European steel producer with important operations in Belgium, Germany, Italy, Brazil and the United States of America. In particular, its main activities are the production and distribution of steel products (including heavy and light long products, flat products and stainless steel products), the trade of scrap and the distribution of some raw materials for the steel industry (especially ferro-alloys and non-ferrous metal). Arbed also has certain other activities in the engineering sector. These activities fall in part under the ECSC Treaty and in part under the EC Treaty. Aceralia together with its controlling undertaking Arbed achieved a worldwide turnover of EUR 13232 million in the year 2000.

(5) Usinor is a company incorporated as a société anonyme under French law. It is one of the largest steel producers in the Community and has operations in Belgium, Germany, Italy, Spain and the United States of America. It manufactures processes and distributes steel products (in particular, flat carbon steel products, stainless steel and other alloy steel products). These activities fall in part under the ECSC Treaty and in part under the EC Treaty. In 2000, Usinor generated a worldwide turnover of EUR 15733 million.

II. THE OPERATION

(6) The proposed concentration is a full legal merger with the complete amalgamation of all of the businesses of Aceralia, Arbed and Usinor. For this purpose, a new single corporate entity called Newco Steel (Newco) will be created, which will be structured as a fully integrated group. Newco, which will be incorporated as a société anonyme under the laws of Luxembourg, has been established with the purpose of making an exchange offer of its shares for the entire issued share capital of each of the parties. It will simultaneously launch three separate conditional exchange offers on all ordinary shares of Aceralia, Arbed and Usinor outstanding at the time of the offers. All three offers will be conditioned upon an acceptance of at least 75 % of the issued shares in Aceralia, Arbed and Usinor. On completion of the offers, Aceralia's shareholders will hold approximately 20,1 % of Newco, Arbed's 23,4 % and Usinor's shareholders will hold the remaining 56,5 %.

III. THE CONCENTRATION

(7) Because of their activities in the manufacture and sale of steel products, the notifying parties are undertakings within the meaning of Article 80 of the ECSC Treaty. Therefore, the notified transaction constitutes a concentration within the meaning of Article 66 (1) of the ECSC Treaty.

IV. PROCEDURE

(8) On 19 July 2001, the Commission decided to initiate an in-depth investigation into the proposed concentration, as far as the ECSC products are concerned, inasmuch as the operation gives rise to serious concerns of giving the parties the power to hinder effective competition or to evade the rules of competition instituted under the ECSC Treaty. The parties were advised of this further investigation by letter of 19 July.

(9) On 28 September 2001, the Commission adopted a Statement of Objections, pursuant to Article 66 of the ECSC Treaty and paragraph 4 of the Notice concerning the alignment of procedures for processing mergers under the ECSC and EC Treaties. The Commission arrived at the preliminary conclusion that the proposed concentration would give the parties the power to hinder effective competition or to evade the rules of competition instituted under the ECSC Treaty.

(10) Regarding the non-ECSC products covered by the proposed concentration, the Commission decided on 19 July 2001 that the transaction does not raise serious doubts as to its compatibility with the common market or with the functioning of the EEA Agreement (7).

V. ASSESSMENT UNDER ARTICLE 66 (2)

A. BACKGROUND: THE STEEL INDUSTRY

1. PRODUCTION PROCESS

(11) Before discussing in detail the relevant product markets it will be useful to establish the framework within which the industry operates. Steel is one of the most important materials used in our society. It is present in the buildings in which we live and work, the transport we use and it is essential in the production and distribution of the food and drink we consume. Furthermore it is a raw material that is essential to the vast majority, if not all, other manufacturing industries.

(12) Steel production is very capital intensive. The minimum economic scale is high, and the investments in this sector are very specific. As a result there are very high entry barriers. Vertical integration is, with few exceptions, the rule. Most producers control an important part of the production chain and many are integrated downstream into steel distribution and first transformation products such as tubes, panels and profiles.

(13) The first important differentiation in the steel industry is one based essentially on the chemistry of the steel. Three main types of steel can be distinguished, carbon steel, stainless steel and special or alloy steels. For the purposes of this decision there is no need to treat the alloy steels any further as no issues arise in relation to these steels.

(14) Carbon steel is the most familiar to us and is used in a very wide range of applications including car bodies, beverage cans, beams and reinforcing materials for the construction industry, and so on. Stainless steel contains significant proportions of chrome and nickel and is therefore much more expensive than carbon steel. Therefore it is primarily used in applications which require resistance to corrosion and/or resistance to high temperatures.

(15) The physical properties of steels can be modified by changes in the chemical analysis or by the treatment they receive during their production or during the subsequent manufacturing processes. For example both beverage cans which require a very ductile material and knives which require a very hard material (to keep their sharpness) are made from different carbon steels.

(16) Today there are two steel making processes that account for nearly all the world's steel production, the integrated route and the electric arc furnace (EAF) route. The integrated route is the most capital intensive and involves the production of liquid iron from iron ore, coke and limestone. The liquid iron is subsequently transformed into steel in an oxygen convertor. At this stage or in a separate vessel the final chemical composition of the steel is adjusted. The alternative production route is the EAF in which scrap (and sometimes directly reduced iron or pig iron) is melted. Additions are then made to obtain the desired chemical specification The liquid steel is then continuously cast into semi-finished products which are further processed by rolling into the various steel products.

(17) Generally the flat carbon steel products that give rise to the competition problems outlined in this Decision are produced by the integrated route as this can produce steel without the tramp elements that are contained in scrap. Lower quality flat products can be produced by the EAF route but for applications where the steel must be free from tramp elements the integrated route is obligatory.

(18) Whichever route is used, the steel is continuously cast into semi-finished products, billets and blooms for long products and slabs for flat products. The slabs are subsequently hot rolled into quarto plate and strip (on either wide or narrow strip mills). Wide strip may be subsequently slit to produce narrower widths and/or cut to length to produce sheets. Virtually all quarto plate, hot rolled narrow strip and steel sheets are sold in this state and are not further transformed by the steel producer.

(19) Some of the hot rolled wide strip is sold directly to customers, including steel service centres, who may cut or slit it. A large part of the hot rolled wide strip that is produced is further processed to produce cold rolled strip, which is thinner and has a superior surface finish. Again some of the cold rolled material is sold at this stage but a significant proportion is metallically coated, with tin or chrome for the can industry or with zinc. Zinc protects the steel from corrosion and may be applied by passing the steel strip through a bath of molten zinc (hot dip galvanised strip) or electrolytically (electro-galvanised strip). The main outlets for galvanised strip are the automobile and construction industries. With the exception of some, predominantly hot dip, galvanised material which is organically coated, most of the coated material is sold as packaging steel, hot dip or electro-galvanised strip.

(20) EAF based plants are usually located near sources of scrap. Among the integrated plants a distinction can be made between the generally more modern coastal plants and inland plants. The inland plants were based on the local availability of iron ore and coal. Today in Europe nearly all the coal and iron ore used in steel making is imported. This gives certain advantages to coastal plants, particularly those with deep water ports.

(21) Within the production tree, it clearly appears that products do not bring the same levels of value-added. This is then translated into differing profitability levels. In fact, downstream products prove to be more value-added and to have appreciably higher levels of earning before tax, depreciation and amortisation, as is apparent from Graphic 1.

GRAPHIC 1

Indicative price levels (hot-rolled strip = 100)

[GRAPHIC DELETED]*(8)

Source:

Parties.

(22) Community sales of flat products in 2000 are shown in Table 1.

TABLE 1

Community sales of flat products in 2000

<emplacement tableau>

Source:

Parties.

2. STEEL MANUFACTURERS

(23) The parties claim that the European steel industry is not very concentrated. While this may have been true in the past, the last ten years have seen a significant consolidation. Today there are five large steel companies: Usinor, TKS, Corus, Riva and Arbed/Aceralia. They are followed by a number of producers including Voest-Alpine, Rautaruuki, Salzgitter, Duferco and SSAB which are considerably smaller. Depending on the product market considered, the sales of the five largest producers account for between 50 % and 85 % of total consumption.

(24) The first group of steel manufacturers is made of large, integrated companies manufacturing steel products at all production levels. All of them produce flat products and TKS, Arbed/Aceralia and Usinor produce stainless (Corus has an interest in the stainless producer Avesta Polarit). In addition Corus and Arbed/Aceralia have very substantial long product operations. All of them are the result of recent concentrations. Usinor acquired Cockerill Sambre which had itself acquired Ekostahl. TKS is the result of successive concentrations in Germany notably Krupp and Hoesch and then the merger with Thyssen. Corus is the result of a merger between British Steel and Hoogovens. Arbed acquired Stahlwerke Bremen and then took a controlling interest in Aceralia which has since acquired Ucin and Aristrain. Riva acquired the privatised ILVA.

(25) These large producers operate from a number of sites in several different European and third countries. They have integrated sites where steel from coal and iron ore is produced and further processed. They are capable of providing a very wide range of flat products. Most of them produce stainless and long products as well. They are also active in steel distribution and in downstream products. For instance, Riva, the smallest in the group, has 44 production sites of various sizes including the massive Taranto integrated steel works.

(26) The second group of steel manufacturers concerns smaller European producers, which do not have the same scale nor product range as the leading players, but focus on niche markets (products and/or regions). This group of second ranked players includes the following flat product producers with integrated plants: Voest-Alpine, Rautaruukki, SSAB, and Salzgitter. As stated by one of these companies in one of the Commission's questionnaires "the Group regards itself not as a bulk producer of steel but rather as a specialist manufacturer and processor of high-quality steel products". There are also a substantial number of long producers mainly based on the EAF route.

(27) The proposed operation would create Europe's and the world's largest steel company. Graphic 2 gives an idea of the relative sizes of the main European steel producers on the basis of their sales of all carbon steel flat products, which are the subject of the concerns dealt with in this Decision.

GRAPHIC 2

Total sales of carbon steel flat products (year 2000)

>PIC FILE= "L_2003088EN.000501.TIF">

B. THE RELEVANT PRODUCT MARKETS

1. PRELIMINARY REMARKS

(28) The object of the proposed concentration is the production of carbon steel and stainless steel products, the further processing of both into hot-rolled flat products, subsequent processing by way of cold-rolling and the production of metallic and organic coated products, electrical steel sheets and steel for packaging. Furthermore, Newco will also be active in the field of steel distribution, through a network of steel service centres, stockholding centres and oxycutting centres.

2. STEEL PRODUCTS

(a) The distinction between carbon steel products, highly alloyed steel products and stainless steel products

(29) According to previous Commission Decisions (9), carbon steel products constitute a different product market from stainless steel products and highly alloyed steel products.

(30) Carbon steel products cannot be substituted by highly alloyed steel products because of 1. their different chemical composition, 2. their significantly different prices, and 3. their different fields of application (10). Furthermore, carbon steel products are produced in plants which are not equipped for the manufacturing of highly alloyed steel products.

(31) In turn, carbon steel products are a separate market from stainless steel products, in particular because of the different chemical composition of non-alloy steel, stainless and heat-resistant steel and other alloy steels (11). Stainless steel is an alloy containing 10,5 % or more of chromium, with or without other alloy elements, and less than 1,2 % of carbon (12). Production of stainless steel differs from production of carbon steel through the addition of chrome, nickel and other alloying elements to the steel melting process in order to achieve metallurgical differences in the end product as compared to normal steel (13).

(b) Flat carbon steel products are a separate product market from long carbon steel products

(32) The Commission has concluded on a number of occasions that flat steel products form a separate product market from long steel products (14).

(33) Steel produced in a steelworks is cast into shape and further processed by subsequent forming into the products desired by final consumers. Rolled steel products may be subdivided into flat products (hot-rolled wide strips, hot-rolled narrow steel, quarto plates and sheets) and long products (sections and steel beams, permanent way material, merchant bars and wide rods). Flat steel and long steel products are produced on different rolling mills and are bought for different purposes. Most rolling mills are purpose built for a particular end product and are hardly ever converted to other uses (15). Furthermore, there are significant differences in the applications and prices of flat and long products. The markets for flat steel products can therefore be differentiated from those for long steel products.

(c) Carbon steel flat products: the distinction between hot-rolled and cold-rolled products

(34) The Commission has found in previous Decisions that hot-rolled carbon steel flat products and cold-rolled carbon steel flat products constitute separate product markets (16).

(35) Only a limited proportion of hot-rolled flat products are sold for use as such. More than half is further processed by cold-rolling. Subsequent cold-rolling results in a considerable reduction in thickness, greater dimensional accuracy, a smoother surface and, following work hardening, greater strength (17). Hot rolled products do not have the same surface characteristics as cold-rolled products (18). The value added by cold rolling is substantial, in the region of 25 % to 30 %.

(36) A modern hot-rolled wide strip mill can roll to minimum thicknesses of 1,5 mm. Thinner strips and sheets, which are used especially by the motor industry and after coating by the packaging and construction industries, require subsequent further processing in cold rolling mills, where minimum thicknesses of 0,12/0,15 mm can be achieved (19).

(37) Within the segment of carbon steel flat products, a distinction can therefore be made between hot-rolled and cold-rolled flat products.

(i) Hot-rolled carbon steel flat products

(38) Within the market for hot-rolled carbon steel flat products, the parties distinguish the following products:

- hot-rolled wide strips, that are first rolled on wide strip mills and then coiled,

- hot-rolled sheets, of square or rectangular shape, that result from the cutting to length of strip,

- hot-rolled narrow strip, that have a width less than 600 mm after finishing and which can be produced on specialised mills or by slitting wide strip to the desired width,

- quarto plates, non-coiled products whose dimensions, in particular their thickness, are very different from those of all other hot-rolled carbon steel flat products.

(39) The parties submit that there is a single relevant market for all hot-rolled carbon steel flat products, with the exception of quarto plates, which have different physical properties in terms of shape, thickness and width and are produced with different equipment and processes, and therefore constitute a separate market.

(40) Quarto plate is plate (20) made on special quarto plate mills with four rolls (four high mills) which, because of its thickness, cannot be rolled up (21). As the parties describe in the Form CO, quarto plates have specific physical properties: they are not coiled and their dimensions are very different from the dimensions of all other hot-rolled carbon steel flat products: (a) their thickness can reach 400 mm, as opposed to a maximum of 25 mm for hot-rolled coils; (b) their width can reach 5000 mm, as opposed to a maximum of 2200 mm for hot-rolled coils. Furthermore, quarto plates are used in applications that differ from those for thin flat steel products, in particular in the industrial equipment, mechanical, energy, nuclear, chemical and petrochemical, gas treatment, shipyard, metalwork, boiler-making and tool making industries. In all these fields, quarto plates dimensions offer significant economical advantage to the steel users (compared with, for instance, narrow flat steel products). On the basis of the aforementioned considerations, the Commission concludes that quarto plates constitute a separate market.

(41) As far as the other hot-rolled carbon steel flat products are concerned, the question arises whether hot-rolled wide strip constitutes a separate product market. Hot-rolled wide strip is strip more than 600 mm in width produced from slabs in automated hot-rolled strip mills and rolled up into coils (22). Wide strip represents approximately 82 % of the sales of all hot-rolled flat carbon steel products (excluding quarto plates). In view of the high degree of supply side substitutability between hot-rolled wide strip, narrow strip and sheets, it is concluded that all hot-rolled flat carbon steel products (excluding quarto plates) form part of the same relevant product market. Hence, unless otherwise specified, any subsequent reference in this Decision to hot-rolled flat carbon steel products is to be understood as comprising all hot-rolled flat carbon steel products excluding quarto plates (i.e., wide strip, narrow strip and sheets).

(ii) Cold-rolled carbon steel flat products

(42) Cold-rolled carbon steel flat products constitute a separate product market from hot-rolled carbon steel flat products. The parties submit that within the market for cold-rolled carbon steel flat products no differentiation should be made between either coils and sheets. This submission is in line with previous Commission's Decisions (23).

(43) Further, according to Annex I to the ECSC Treaty, cold-rolled flat products less than 500 mm in width do not come under the ECSC Treaty, but under the EC Treaty. However, against the background of present-day production processes and supply conditions, this distinction is no longer relevant when it comes to assessing mergers from a competition point of view. Cold-rolled flat products both less and more than 500 mm in width are nowadays produced on the same rolling mills. Sheets of less than 500 mm in width can, moreover, be obtained by slitting wide strip. In accordance with earlier Commission Decisions (24), there can therefore be assumed to be, for the purpose of assessing the effects of the proposed concentration, uniform product markets irrespective of the products' width, not only for cold-rolled carbon steel flat products, but also for products in downstream markets resulting from further processing, such as the different types of coated flat products.

(44) Cold-rolled carbon steel flat products are mostly intended for further processing by the producers themselves, in particular into coated products, and thus used internally. For instance, approximately 70 % to 80 % of the cold-rolled carbon steel flat products manufactured by the notifying parties is supplied within the group for further processing. The Commission has found in previous decisions that (uncoated) cold-rolled carbon steel flat products are likewise a separate market from the market for coated products.

(d) Electrical steel sheets

(45) In line with a previous Commission decision (25), the parties submit that electrical steel sheets constitute a separate product market from cold-rolled carbon steel sheets, in view of their electro-magnetic properties (conductivity and electrical resistance) and applications (the construction of transformers, electric motors, series and switching units and power plant generators).

(46) In addition, the parties distinguish between non-grain oriented electrical steel sheets and grain oriented electrical steel sheets, due to differences both in the demand side and in the supply side. The Commission has concluded in the past that grain-oriented and non grain-oriented electrical sheets constitute separate markets (26). Both products are pickled and cold-rolled; however, they differ in the annealing process, by which the material is set to the required electromagnetic properties. In particular, the installations needed for the manufacturing of these two types of products are very different, and therefore there is no supply side substitutability. Furthermore, there are demand side differences: grain oriented is used almost exclusively for large transformers while the principal uses of non-grain oriented are motors, generators and small transformers.

(e) Further processing of flat carbon steel: coated products are a separate market from uncoated products

(47) Coating is a production process that intervenes subsequently to hot-rolling or cold-rolling. In particular, it is applied to more than two thirds of all cold-rolled carbon steel flat products produced in the Community, which are provided with a coating tailored to its end use. The Commission has established that coated and uncoated cold-rolled carbon steel flat products constitute separate product markets, according to the differences that exist between their respective properties, prices and applications (27).

(48) Uncoated cold-rolled strip has quite different characteristics from coated cold-rolled strip as regards surface finish and resistance to corrosion. Moreover, the value added by the coating is very substantial, though it varies for the various coating procedures, from [20 % to 30 %]* (hot dip galvanised sheet) to [75 % to 85 %]* (organically coated sheet), hence approximately 35 % on average (28).

(i) The distinction between metallic coated and organic coated products

(49) As the parties explain in their notification, the corrosion-resistance of flat carbon steel can be increased by coating it with zinc, tin or another metal (29) and/or by applying an organic topcoat (30).

(50) The parties submit that organic coated products constitute a separate market from metallically coated products, in view of their different characteristics and, in particular, the significantly higher price of organic coating (31). Furthermore, the parties argue that the relevant market may be wider than organic coated steel and include a variety of competing products (32), in particular organic coated aluminium.

(51) The proposed delineation of organic and metallic coating as separate markets is in line with previous Commission's decisions (33). The market investigation has not confirmed, however, the existence of a wider market comprising organic coated steel and organic coated aluminium.

(ii) Tinplate and galvanised products

(52) Metallic coated strip can be divided into two different broad classes: 1. steel for packaging (tinplate and ECCS); and 2. galvanised steel (hot-dip and electro-galvanised, including coatings of zinc alloys).

(53) These groups of products have very different prices and applications. Steel for packaging is used almost entirely for food and beverage cans. On the other hand, galvanised steel strip is widely used in the automobile, construction and engineering industries, where the zinc coating impedes corrosion of the steel substrate and thereby increases its life. These two groups of products can therefore be clearly distinguished.

(iii) Galvanised products: hot-dip galvanised and electro-galvanised

(54) There are two different production processes for galvanising: hot-dip galvanising and electro-galvanising.

(55) Hot-dip-galvanised products are obtained through dipping hot-rolled or cold-rolled steel into a bath of zinc at 460oC, coating all exposed surfaces with zinc (34). Electro-galvanised products, in turn, result from the application of an electrolytic coating to one or both sides of the strip, with either pure zinc or an alloy of zinc and nickel.

(56) The parties submit that hot-dip galvanised and electro-galvanised products form part of the same market, namely that of galvanised products (though they claim that the market might be even wider if account were to be taken of products competing with metallic coated steel)(35). They argue that from a demand perspective both types of products are substitutable, and that, in particular, substitutability has been improved recently due to improvements in the appearance and performance of hot-dip galvanised products; hence, in all major applications (automotive, domestic appliances and metal furniture) both hot-dip and electro-galvanised can be used interchangeably, subject only to minor process adaptations which do not restrict substitutability. Furthermore, corrosion resistance, surface quality and prices are comparable.

(57) The Commission has suggested in previous decisions (36) that hot-dip galvanised and electro-galvanised may constitute separate product markets, in view of their different product characteristics and prices, but has left open the precise segmentation of galvanised products.

(58) However, the market investigation in this case has shown that due to recent improvements in hot-dip galvanised products, in particular corrosion resistance, appearance and performance, substantial differences between hot-dip and electro-galvanised characteristics do not appear to exist. Customers have largely confirmed that they regard both product categories as technically interchangeable for the majority of applications. Particularly, the automotive industry, which represents the largest customer group of both hot-dip and electro-galvanised products, uses both products in its production processes, including the manufacturing of external body parts, which in the past were exclusively made of electro-galvanised steel.

(59) This can be illustrated by the evolution of demand for hot-dip and electro-galvanised products in the Community. For instance, in 1993, 70 % of electro-galvanised steel was sold to the automotive industry, compared to only 27 % of hot dip galvanised strip. However, by 2000, 44 % of hot-dip was used in the automobile industry. In 1999, of the total tonnage of metallic coated products used by the automotive industry, approximately 60 % was hot-dip and 40 % electro-galvanised. Table 2 analyses the evolution of demand for hot-dip and electro-galvanised steel by sectors. According to the forecasts of the industry, the use of hot-dip galvanised steel by the automotive industry will further increase in the coming years.

TABLE 2

Estimated evolution of Community demand for hot-dip and electro-galvanised steel

<emplacement tableau>

Source:

Parties.

(60) With regard to prices of hot-dip galvanised steel and electro-galvanised steel, the market investigation shows that prices largely depend on the particular specifications (in particular the thickness and chemical composition of the coating, and thickness, width, chemical specification rigidity etc., of the steel support) and applications of either product. Therefore, it is difficult to determine comparable average or standard prices representative of the whole range of either hot-dip and electro-galvanised products.

(61) In any event, according to official price-lists of the main producers of galvanised sheet (see Table 3) the average price differentials in the Community between these products are below 5 %. The market investigation has confirmed that the range of price differentials between hot-dip and electro-galvanised are not substantial.

TABLE 3

Price differences between hot dip and electro-galvanised by Member States

<emplacement tableau>

Source:

Official price lists of producers located in the countries mentioned.

(62) Accordingly, the Commission has concluded that the relevant product market encompasses all galvanised products, namely hot-dip galvanised steel and electro-galvanised steel.

(f) Steel for packaging: the distinction between steel for beverage packaging and steel for non-beverage packaging

(63) Steel for packaging consists of very thin (0,13 to 0,49 mm) flat carbon steel coils or sheets which are electrolytically coated with a fine layer of tin (in the case of tinplate) or chromium (in the case of electrolytic chromium coated steel (ECCS) or tin-free steel (TFS)). Two distinct types of steel for packaging can be distinguished according to their applications: steel for beverage packaging and steel for non-beverage packaging. This distinction is based on the following factors: 1. the quality standards for beverage cans are much higher than those for non-beverage packaging (tolerance to stresses during processing, quality of metal in relation to porosity, etc.); 2. the tinning lines have to be wider to allow for the optimisation of can production lines (1,2 m instead of a maximum 1,0 m), and 3. steel intended for beverage must be de-gassed under vacuum in order to produce ultra-low carbon steel. While production routes for the manufacture of steel packaging for beverage cans could theoretically be used for the manufacture of steel cans for non-beverage packaging, it would be uneconomic to do so.

(64) The parties submit that the relevant product market for metal for beverage cans should include not only steel but also aluminium, since both materials are used to produce beverage cans. According to the parties, European customers have a free choice as to whether to use steel or aluminium for cans since, while aluminium is more expensive as a raw material, less processing is required to bring it to beverage-can grade. Prices of aluminium and steel for packaging would therefore interact. Moreover, while traditional can-making production-lines are configured to using either aluminium or steel for packaging, they can be converted to use the other material (the parties calculate a cost of EUR [...]* million from steel to aluminium and of EUR [...]* million from aluminium to steel); furthermore, modern installations consist of swing-lines which enable steel for packaging and aluminium to be used interchangeably. Likewise, the parties submit that the customers, being large concentrated beverage can producers, have sufficient numbers of steel and aluminium dedicated lines to quickly adapt to changes in demand and switch between the two materials.

(65) The Commission has concluded in the past that steel for packaging is a separate product market (37). However, the Commission has not determined yet whether the market should be segmented more narrowly, namely whether steel (or metal) for beverage packaging and for non beverage packaging constitute separate markets. The exact delineation may also be left open for the purposes of the assessment of the notified operation, since it does not affect substantially the competitive assessment.

(66) As regards the submission of the parties that steel and aluminium would constitute part of the same relevant market for beverage packaging, the Commission concluded recently that steel and aluminium for packaging in general constitute separate relevant markets (38). While the end products are similar in both cases, there are significant differences in the both the prices of the two materials and the price developments. The Commission's market investigation in this case does not confirm a combined aluminium and steel for beverage cans market.

(67) Finally, the parties submit that in addition to aluminium and steel the market for beverage packaging could be even wider, including competing materials, such as plastic and glass. These competing materials do not constitute part of the same market as either steel or aluminium on the basis of the totally different characteristics of the materials, the very different equipment needed to fill glass or plastic bottles and their different prices.

(g) Stainless steel flat products: the distinction between hot-rolled and cold-rolled products

(68) Stainless steel can be distinguished from carbon steel and from alloy steels by its physical and chemical characteristics, particularly by its resistance to corrosion and high temperatures. This makes it uniquely suitable for certain applications in process engineering, automobile engineering, hygiene products and cutlery, among others (39).

(69) In previous Decisions, the Commission has identified separate relevant product markets for hot-rolled stainless flat products and cold-rolled stainless flat products (40). The parties point out that the differences between hot-rolled and cold-rolled stainless steel products tend to fade due to technical evolutions undertaken by the overall stainless steel business, but they recognise that the zone of overlap between hot-rolled and cold-rolled stainless steel still remains limited to date.

(i) Hot-rolled stainless steel flat products

(70) As in the case of hot-rolled flat carbon steel products, the Commission has concluded that, in view of the high degree of supply side substitutability between hot-rolled wide strip, narrow strip and sheets, all hot-rolled flat stainless steel products (excluding quarto plates) form part of the same relevant product market.

(ii) Cold-rolled stainless steel flat products

(71) Cold-rolled stainless flat products (strip and sheet) are made by cold-rolling hot-rolled strip. They constitute a separate relevant product market because they differ from hot-rolled flat products in terms of product characteristics (particularly the surface finish), uses and prices (41). Cold-rolled stainless flat products are used in a large variety of industries with the following requirements (42): 1. resistance to heat, rust and acids (corrosion); 2. the basic structural quality of steel such as deep drawing suitability/weldability; and 3. a surface which is highly developed in order to provide for smoothness and good appearance.

(72) In addition, cold rolled strip below 500 mm falls under the EC Treaty, while cold rolled strip above 500 mm. falls under the ECSC Treaty (43). This distinction is however no longer meaningful as modern cold-rolling mills are capable of producing a wide range of widths both above and below 500 mm, and narrower widths can likewise be obtained by slitting wider material. The Commission therefore considers that the distinction is artificial and that all cold-rolled stainless steel strip belongs to a single market (44).

(h) Conclusions

(73) The Commission therefore concludes that the following categories constitute the relevant product markets which will have to be taken into account for the competitive assessment of the notified operation:

(1) hot-rolled flat carbon steel products;

(2) quarto plates;

(3) cold-rolled carbon steel flat products;

(4) non grain-oriented electrical steel sheets;

(5) galvanised steel;

(6) organic coated carbon steel;

(7) steel for packaging or, alternatively, separate markets for either (a) steel for beverage packaging or (b) steel for non-beverage packaging;

(8) hot-rolled stainless steel flat products;

(9) cold-rolled stainless steel flat products.

3. STEEL DISTRIBUTION

(a) The distinction between steel distribution and production and direct sales of steel

(74) The parties consider that steel distribution is a separate market from production of steel and from direct sales by steel producers to consumers. The Commission has held in previous cases that steel distribution activities constitute a distinct product market (45). Steel products' distribution can be differentiated from the production and direct sales of steel products by a number of factors (46):

(1) the diversity of its customers: customers for steel products through distribution channels are normally different customers, with different needs, when compared to the types of customers for direct sales;

(2) the smaller size of orders: while sales from mills are typically measured in tens of tonnes, steel service centres' sales are measured in tonnes and stockholder sales are typically measured in hundreds of kilograms;

(3) the ability of distributors to respond rapidly to customer requirements: delivery times from steel mills are usually measured in weeks if not months, while stockholders work in hours or days; steel service centres offer normally on-time deliveries, with commitment on a day or even sometimes a time slot;

(4) the local nature of its business: distribution outlets generally serve customers within a limited distance from their location; and

(5) the large number of companies involved in the business: from integrated distributors owned by steel companies to small independent firms.

(75) Furthermore, the Commission has concluded that distribution operations have full function characteristics and are not sales agencies (47), which is also evidenced by the fact that there is a large number of independent (non-integrated) steel distributors operating in the Community. In effect, although many steel producers have wholly or partially owned steel distribution operations, there is an independent sector consisting of a large number of companies of varying size and financial strength ranging from small family companies to subsidiaries of large industrial companies (48).

(b) The distinction among different steel distribution channels

(76) The parties submit that steel distribution shall be further subdivided in three different markets: 1. steel service centres; 2. stockholding centres; and 3. oxycutting centres. These market definitions are in line with previous Commission decisions (49) and have been largely confirmed by the market investigation. These distributions channels will be further subdivided according to the groups of products sold.

(i) Steel service centres

(77) Steel service centres purchase strip mill products from the steel producers and afterwards slit and cut the material to customers' requirements (50). Their customers include major consumers, such as the automobile and white goods manufacturers, stockholders and customers of all sizes.

(78) Steel service centres are only active in the sector of flat steel products, both carbon and stainless. In line with previous Commission Decisions (51), this Decision will differentiate between the markets for the distribution of carbon steel and stainless steel products. This distinction has been confirmed by the notifying parties, which have submitted that the distribution channels for carbon and stainless flat steel products are different. The parties' activities overlap only with regards to carbon steel products and not for stainless products (except in Germany, where their combined market share is [< 15 %]*).

(ii) Stockholding centres

(79) Stockholders perform the traditional wholesaling function of buying in bulk from the manufacturers and reselling in smaller quantities (52). In particular, stockholding activity consists of holding a wide variety of steel products at convenient locations to supply the needs mainly, but not exclusively, of small and medium sized clients, with quantities of steel that are not large enough to warrant dealing directly with the steel producers or with steel service centres (53). Furthermore, stockists have steel available for direct delivery and can therefore provide a service not available from steel producers. Stockholders have also developed their in-house processing facilities in order to add value and improve their service to customers (54). Such processing includes cutting to length, slitting, shearing and polishing.

(80) Stockholding activities cover almost the full range of steel products (55). As in the case of steel service centres, a segmentation will be made between the distribution of carbon steel and stainless steel products, the parties' overlapping operations relating only to carbon steel products. Furthermore, for the purposes of the assessment, this Decision will distinguish between the two product categories of carbon steel products, namely, (a) flat products and (b) long products. This distinction by types of products distributed is in line with a previous Commission Decision (56).

(81) In their reply to the Statement of Objections, the parties contest the segmentation of the relevant product market into the stockholding of flat carbon products and long carbon products. The parties submit that the relevant market is the overall market for stockholding, without distinction between flat and long products, on the basis of the following arguments: 1. stockholding is a customer-oriented service, characterised by a very fragmented customer base, and therefore, in order to answer the demand of their customers, many steel stockholders would have to carry most steel product groups; and 2. there is a high level of supply-side substitutability and no barriers preventing a stockholder involved in flat products from supplying long products and vice-versa.

(82) Notwithstanding the parties' submission, in particular the fact that a number of stockholders are active both in the market for flat carbon products and in the market for long carbon products, the delineation of the relevant product markets must take into account the different characteristics of demand for every type of product and the differences between their average prices. In particular, flat carbon products have different customers from those of long carbon products, and also have particular characteristics regarding their distribution channels (steel service centres only deal with flat products and not with long products).

(83) Accordingly, the Commission considers as relevant the following product markets: 1. distribution of flat carbon steel products through stockholding activities; 2. distribution of long carbon steel products through stockholding activities; and 3. distribution of stainless steel products through stockholding activities.

(iii) Oxycutting centres

(84) Oxycutting centres handle a particular product, quarto plates with a thickness of 8 mm to 600 mm, which requires a specific finishing tool (oxyhydrogen blowtorch) that is very precise and powerful (57). Oxycutting centres work in close relation with their clients and deal with small quantities of steel which are cut to the customers' requirements.

(85) The Commission has concluded in the past that oxycutting activities constitute a separate market (58). Moreover, the market investigation has largely confirmed this conclusion.

(c) Conclusion

(86) The Commission therefore concludes that the following categories constitute the relevant product markets which will have to be taken into account for the competitive assessment of the notified operation:

(1) distribution of carbon steel flat products through steel service centres;

(2) distribution of stainless steel flat products through steel service centres,

(3) distribution of flat carbon steel products through stockholding activities,

(4) distribution of long carbon steel products through stockholding activities,

(5) distribution of stainless steel products through stockholding activities,

(6) distribution of quarto plates through oxycutting centres.

C. THE RELEVANT GEOGRAPHIC MARKETS

1. STEEL PRODUCTS

(a) Hot-rolled carbon steel flat products

(87) The parties submit that the relevant geographic market for all hot-rolled carbon steel flat products (including quarto plates) is at least a Community-wide market, but they suggest that it might even be wider.

(88) In previous decisions the Commission found that the relevant market for hot-rolled carbon steel flat products comprises at least the territory of the Community (59). Though in the past the production capacities for hot-rolled steel products belonged in most Member States to one (national) steel undertaking, which accounted as a rule for the highest shares of supply in its home market, steel producers have long penetrated their competitors' home markets, achieving sizeable market shares (60). Nowadays, the big steel producers are active in almost every Member State both through direct supplies to final customers and through affiliated distribution centres. Furthermore, transport costs within the Community are not significant, custom or other barriers to entry do not exist and customer preferences are not an important factor since the majority of the products are homogeneous (commodities).

(89) The exact market delineation can be left open for the purposes of this Decision, since even on the basis of the narrowest geographic market definition the proposed concentration does not raise serious competition concerns in this market.

(b) Cold-rolled carbon steel flat products

(90) The parties claim that the relevant geographic market for the production of cold-rolled carbon steel flat products is at least Western Europe, that is the Community and EFTA, again suggesting that it could even be wider. The parties point to the level of imports of cold-rolled carbon steel flat products in 2000 (almost 18 % according to the parties) and to the findings of a report provided by the parties' economic consultants, which purports to show, using econometric analysis, that the relative price of imports significantly affects the import penetration in the Community.

(91) In previous cases, the Commission has determined that the relevant geographical market for cold-rolled carbon steel flat products was at least Western Europe, that is the Community and EFTA(61). The enquiries in the present case tend to indicate that the dimension of the market is not wider than Western Europe. In any event, the precise delineation of the market can be left open, since the proposed concentration will not give the merging entity the power to hinder effective competition or to evade the rules of competition instituted under the ECSC Treaty irrespective of the exact definition of the geographic market.

(c) Non grain-oriented electrical steel sheets

(92) The parties submit that the relevant geographic market for non-grain oriented electrical steel sheets is Western Europe. The proposed delineation is in line with previous Commission's Decisions (62) and has been largely confirmed by the market investigation.

(d) Coated carbon steel products

(93) The parties submit that the relevant geographic market for galvanised and organic coated steel is at least Community-wide, if not broader. The Commission has concluded in the past that the markets for hot-dip galvanised steel and other coated products are Community-wide (63). Transport costs within the Community are not significant (for instance, they represent less than 5 % of the end value of organic coated steel), and a substantial part of galvanised coated steel (40 % to 50 %) and organic coated steel (over 30 %) is traded cross-border within the EU.

(94) The relevant geographic market does not however extend beyond the Community. The parties have stated that imports from third countries have increased during recent years. Nevertheless, third countries' imports still represent relatively low volumes and vary substantially between the different products: 6 % to 9 % for galvanised steel and 6 % for organic coated steel in 1999-2000. Further, the market investigation has indicated that a number of customers believe that steel producers outside the Community are not currently in a position to meet the requirements of customers in the Communityin terms of quality, reliability and service. The Commission therefore concludes, in line with its previous decisions, that the markets for coated carbon steel products comprise the territory of the Community.

(e) Steel for packaging

(95) The parties submit that the relevant geographic market for steel for packaging is at least Community-wide. The Commission has concluded in the past that the market has a Community dimension (64). Intra-Community trade flows are important (around 40 %) and transport cost are not particularly high (3 % to 4 %).

(96) The market investigation has confirmed likewise that the relevant market is the Community. Imports in 2000 represented around 10 % for all steel for packaging, the vast majority of which consisted of steel for non-beverage packaging (there were virtually no imports of steel for beverage packaging).

(f) Stainless steel flat products

(97) The parties submit that the relevant geographic market for hot-rolled and cold-rolled stainless steel flat products is at least Western Europe, i.e., that comprises the Community and EFTA. However, they claim that the market is developing into a world-wide market, due to the importance of transcontinental flows of trade, the levelling of trade and competition conditions world-wide, the existence of a global trend with respect to prices and the absence of significant transport costs.

(98) The Commission has defined as relevant in the past the market comprising Western Europe (65). While imports from third countries have increased in recent years, they still represent less than 8 % for hot-rolled stainless steel flat products and 6 % for cold-rolled stainless steel flat products. In any event, the question may be left open, since the operation does not give raise to serious competition concerns even on the basis of the narrower possible market definition.

2. STEEL DISTRIBUTION

(99) The parties submit that the geographic markets for steel distribution are national or regional, and that they have to be determined on a case by case basis, taking into account, inter alia, the following factors: 1. value of the material handled; 2. value added by the relevant service; 3. location of competitors; 4. delivery time; and 5. relevant transportation costs.

(100) In particular, the parties consider as relevant markets for the purposes of this operation:

(1) France;

(2) Benelux and the neighbouring north-western part of Germany (North Rhineland-Westphalia);

(3) Spain and Portugal; and

(4) the United Kingdom and Ireland.

(101) In previous decisions (66), the Commission has concluded that the geographic area serviced by an individual stockholding depot is determined by transport costs and delivery times. The value added by steel distribution operations is comparatively low, so transport costs are an important constraint on the ability of a distribution operation to compete over long distances (67). In particular, the market investigation shows that distribution outlets generally serve customers within a radius of approximately 200 to 250 kms. Distribution is therefore essentially a regional business. Whether this region includes one or more Member States depends upon a number of factors, including the size of the Member State, the type and value of the products sold and the location of competitors (68).

(102) Hence, the Commission has established in previous decisions that the activities of steel distributors are normally confined to an individual member State and cross-border transactions by stockholders, steel service centres and oxycutting centres are limited (69). The Commission's investigation in this case shows that there exist some cross-border flows between neighbouring areas, but a large number of sales are made in the country of location of both the distribution centre and the customer.

(103) As the parties set out in the Form CO, France has already been considered the relevant geographic market for steel distribution by steel service centres and stockholding centres (70) and oxycutting centres (71). The market investigation has confirmed that there exist only limited cross-border sales between France and its neighbouring countries.

(104) The parties consider as a relevant geographic market the area comprising the Benelux and the neighbouring north western part of Germany (North Rhineland-Westphalia), on the basis of 1. the absence of trade barriers; 2. the similarity of prices charged within the region, and 3. the existence of significant cross-border flows of products. In the past, the Commission has identified as relevant the market comprising the Netherlands, Belgium and the north-western area of Germany (72). The market investigation has confirmed the existence of important cross border trade flows in the considered area. In particular, the different distribution networks supply to customers in the area on the basis of the location of their distribution outlets. For instance, the parties have concentrated their principal steel service centres in this area in Belgium, and from there they can effectively and economically supply to customers in Luxembourg, the Netherlands and the relevant part of Germany. In effect, a significant proportion of sales made by the parties' Belgian steel service centres was made to customers located in the aforementioned countries. Furthermore, with regards to stockholding activities, [20 % to 30 %]* of the parties' sales to Belgian customers were made by their Dutch stockholders, while [30 % to 40 %]* of stockholding sales to Dutch customers were made by the parties' Belgian stockholders; on the other hand, [40 % to 50 %]* of the parties' stockholding sales to North Rhineland-Westphalia were made by Benelux stockholders. Finally, the market investigation confirms that prices charged throughout this region are very similar.

(105) In turn, the parties submit that Spain and Portugal together constitute a relevant geographic market. The market investigation, has confirmed the existence of cross border sales, normally from Spain into Portugal. The relevant geographic market appears therefore to be the Iberian Peninsula. In any event, the exact delineation of the relevant geographic market (i.e., the existence either of a single market for these two countries or rather of separate markets) can be left open, since the competitive concerns raised by the assessment of the notified operation would remain irrespective of the existing alternative geographic definitions.

(106) Finally, the parties submit that the United Kingdom and Ireland constitute a single relevant geographic market. Transportation costs, which are an important constraint for steel service centres, would substantially limit imports from outside this area. The exact delineation of the relevant geographic market can, however, also be left open since under any alternative definition (i.e., separate national markets for the United Kingdom and Ireland) the operation does not raise competitive concerns.

(107) The Commission therefore concludes that the relevant geographic markets for the assessment of the notified operation are:

(1) France;

(2) Benelux and the neighbouring north-western part of Germany (North Rhineland-Westphalia);

(3) the area comprising Spain and Portugal or, alternatively each country as separate relevant markets; and

(4) the area comprising the United Kingdom and Ireland or, alternatively, each country as separate relevant markets.

D. COMPETITIVE ASSESSMENT

1. STEEL PRODUCTS

1. Hot-rolled carbon steel flat products

(108) As mentioned in the section on product market definition, only a limited part of the hot-rolled flat carbon steel produced is sold for use as such: more than half of the production of hot-rolled carbon steel is further processed by cold-rolling. As is shown by Table 4, the production of hot-rolled in the Community amounts approximately to 75 million tonnes of which only 29 million tonnes is sold in the form of hot-rolled wide strip, hot-rolled sheets or hot-rolled narrow strip (quarto plates will be dealt with separately).

TABLE 4

Production and sales of hot-rolled flat carbon in the Community

<emplacement tableau>

Source:

Parties.

(109) Newco will have an important share of the production capacity of hot-rolled in the Community ([35 % to 40 %]* in 1997-1999). However, due to its integrated nature, a substantial part of this production is used in-house in downstream production.

Market structure

(110) On the basis of the narrowest possible definition of the relevant market for hot-rolled carbon steel flat products, namely a Community-wide market, the parties combined market share both in terms of total sales and merchant sales (sales to third parties) have accounted for between [25 % to 35 %]* during the years 1997-2000 (see Table 5).

TABLE 5

Parties' combined sales of hot-rolled carbon steel flat products

<emplacement tableau>

Source:

Parties

(111) In particular, the Community market for hot-rolled flat products totalled 29,2 million tonnes in 2000, with Usinor's total sales amounting to [...]* million tonnes ([15 % to 20 %]* of the market) and Arbed/Aceralia's to [...]* million ([10 % to 15 %]*). The combined figure ([...]* million tonnes) represents [25 % to 30 %]* of total sales in the Community in 2000. In 1999, the parties' combined share of the Community market in terms of total sales amounted to [30 % to 35 %]*.

(112) As far as merchant sales are concerned, Usinor totalled [...]* million in 2000 while Arbed/Aceralia accounted for [...]* million. As is clear from Table 6, in 1999, the notifying parties' merchant sales totalled [...]* and [...]* million tonnes respectively, which represents a combined market share of [25 % to 30 %]* ([10 % to 15 %]* + [10 % to 15 %]*).

TABLE 6

Share of Community sales of hot-rolled carbon steel flat products in 1999

<emplacement tableau>

Source:

Parties.

(113) The three main competitors of Newco in the market for hot-rolled carbon steel flat products are Corus (market share of [10 % to 15 %]* in total sales in 1999 and [10 % to 15 %]* in merchant sales in the said year), TKS ([10 % to 15 %]* and [15 % to 20 %]* respectively) and Riva (< 10 % and < 10 %). The second-tier European producers are active as well in this market: Voest-Alpine (< 5 % in total sales), Salzgitter (< 5 %), SSAB (< 5 %) and Rautaruukki (< 5 %). Imports from third countries have increased during the last years, representing around [15 % to 20 %]* in 2000.

(114) This market structure does not give rise to serious competition concerns after the merger either of 1. Newco unilaterally hindering effective competition or evading competition rules or 2. parallel anticompetitive behaviour by the main producers. In particular, Newco is not likely to be in a position to raise prices unilaterally across the whole range of hot-rolled flat products.

Imports

(115) The market for hot-rolled carbon steel is characterised by a high level of imports: 16,3 % in 2000, 13,8 % in 1999 and 17,5 % in 1998. The vast majority of hot-rolled carbon steel consumed in the EEA is commodity-grade steel; therefore, imports from third countries do not appear to suffer from quality disadvantages.

Conclusion

(116) In view of those considerations, the Commission has concluded that the notified operation will not give the parties the power to hinder effective competition or to evade the rules of competition instituted under the ECSC Treaty with regard to the market for hot-rolled carbon steel flat products.

2. Quarto plates

Market structure

(117) In 2000 the total market for quarto plates totalled 7.8 million tonnes. Usinor's share in terms of total sales (including captive sales) amounted to [...]* million tonnes ([15 % to 20 %]*) in 2000. In the same year, Arbed/Aceralia's share was [...]* million tonnes ([5 % to 10 %]*) giving a combined market share of [20 % to 25 %]*. This represents a slight increase on the combined market share from [20 % to 25 %]* in 1999. As is clear from Table 7, the parties' share of the merchant market (sales to third parties) amounted to [20 % to 25 %]* in 1999 (Usinor: [15 % to 20 %]*; Arbed/Aceralia: [5 % to 10 %]*), which approximately corresponds to the market share in total sales of that year ([20 % to 25 %]*).

TABLE 7

Parties' combined sales of quarto plates

<emplacement tableau>

Source:

Parties.

(118) Corus held a market share of [5 % to 10 %]* in terms of total sales in 1999, Rautaruuki [5 % to 10 %]*, Riva [5 % to 10 %]*, Salzgitter [5 % to 10 %]* and TKS [5 % to 10 %]*. Although there is a considerable distance between Newco and its competitors, the total amount of Newcòs market share is still moderate covering only about one-fifth of the market.

Capacity utilisation

(119) The Community capacity utilisation rate in 1999 was 68 %. This suggests that competitors would also have enough spare capacity to increase their production and sales should Newco decide to raise prices unilaterally. They can therefore act as credible alternative suppliers thus constraining Newcòs market behaviour.

Market entries

(120) The parties have identified three new entrants to the market in the last five years: (a) Vitkowice, a Czech hot-roller, as an importer; (b) Duferco, with a planned capacity of 950000 tonnes; and (c) Marcegaglia.

Imports

(121) Import levels in this market represented 14,6 % of sales in 2000 and 19,5 % in 1999. These comparatively high penetration rates further decreases the scope for Newco to act independently of its competitors.

Conclusion

(122) On the basis of that analysis, the Commission concludes that the notified operation will not give the parties the power to hinder effective competition or to evade the rules of competition instituted under the ECSC Treaty in the market for quarto plates.

3. Cold-rolled carbon steel flat products

(123) In a previous section (recitals 90 and 91), the exact geographic market definition for cold-rolled carbon steel flat products was left open, indicating that the narrowest possible definition was Western Europe. For the purposes of the assessment, figures for the Community will be used, since exhaustive data regarding an overall Western European market is not available. In any event, the global assessment would not change if the figures for Western Europe were considered, since the weight of the other countries (Norway, Switzerland, Liechtenstein, Iceland) is so small in relative terms that the overall Western European situation would not be substantially different from that of the Community.

(124) In 2000, the Community market for cold-rolled flat products totalled 13,1 million tonnes. Usinor's total sales amounted to [...]* million tonnes ([15 % to 20 %]* of the market) and Arbed/Aceralia's to [...]* million ([15 % to 20 %]*). The combined figure ([...]* million tonnes) represented [35 % to 40 %]* of total sales in the Community in 2000. In 1999, the parties' combined share of the Community market in terms of total sales amounted to 40 % (see Table 8).

TABLE 8

Parties' combined sales of cold-rolled carbon steel flat products

<emplacement tableau>

Source:

Parties.

(125) Approximately [...]* tonnes of the referred [...]* million of total sales were sold to other companies within the Usinor and Arbed's groups. Newco's market share in terms of merchant sales (sales to third parties) was [35 % to 40 %]* in 1999. Table 9 illustrates the parties' and competitors' sales and market shares regarding either total sales and the merchant market in 1999.

TABLE 9

Share of Community sales of cold-rolled carbon steel flat products in 1999

<emplacement tableau>

Source:

Parties.

(126) With regard to the likely evolution of the market, the forecast made by an independent consultant giving advice to the parties on the effects of the merger outlined that Newco's share of the total market for cold-rolled carbon steel flat products would be about [35 % to 40 %]* in 2006.

(127) In the Statement of Objections, the Commission took the preliminary view that the notified operation would give the parties the power to unilaterally raise prices. The Commission came to this conclusion on the basis of (a) the particularly high combined market shares of the parties to the transaction and the great distance to their closest competitors; (b) the limited extent of planned new capacity in the forthcoming years; (c) the unlikeliness of new entries; (d) the absence of competitive pressure by alternative materials; (e) insufficient competitive constraints stemming from imports; (f) the inability of especially the automotive industry to change suppliers in the short and medium term; and (g) the strong position of Newco in up- and downstream markets.

(128) In their response, the parties contest the findings of the Commission. In essence, the parties argue as follows:

- the parties have lost substantial market share in the last few years and have arrived at [30 % to 35 %]* in terms of total sales in the year 2000 (starting form [40 % to 45 %]* in 1997),

- there is substantial spare capacity in the market, spare capacity which is expected to grow in the future. In addition, based on the results of a report provided by the parties' economic consultants, the parties claim, that - due to the high fixed costs - steel producers are under high pressure to sell the "marginal" tonne. Because of this, the threat of customers switching even small volumes away from suppliers is sufficient to exert considerable price pressure,

- the market for cold rolled flat products faces substantial imports, the bulk of which (68 %) stems from central and eastern Europe, Turkey and the CIS. Spare capacity in these countries will not be absorbed by domestic demand in the short or medium term,

- imports from the countries mentioned do not suffer from quality disadvantages, as the vast majority of the cold rolled flat products consumed in the EEA is commodity-grade steel,

- imports exercise a real, lasting competitive restraint on the behaviour of domestic producers, even without long-term sales commitments or distribution networks,

- the customers - being highly concentrated - are in a position to exert substantial buyer power,

- vertical integration does not privilege Newco as competitors are equally integrated and customer bases differ in the distinct markets.

In sum, the parties' argue that market conditions in the market for cold rolled flat products are such as to assure effective competition also post-merger.

(129) The Commission has carefully examined the arguments raised by the parties, considering in particular several decisive elements for the maintenance of competition such as (a) a high rate of imports and (b) spare capacity in the market under consideration.

(130) In the first place, as set out in Table 10, imports amounted to almost 18 % in the year 2000 starting from 9,8 % in 1997. This represents the highest import- and import growth rate of all steel markets which have been under close consideration in the current proceedings. The increase of market shares appears to have occurred at the expense of parties, whose combined market share decreased from [40 % to 45 %]* in 1997 to [30 % to 35 %]* in terms of total sales. The market shares of the other players in the market have remained stable.

TABLE 10

Import rates in the market for cold rolled products

<emplacement tableau>

Source:

Form CO.

(131) As the parties point out, these imports originate mainly from Eastern European countries, so transport costs are unlikely to differ substantially from transport costs within the Community. In any event, the high level of imports as well as their high growth during the past years appears to suggest that cold rolled flat steel products from third country-suppliers are not hampered by competitive disadvantages regarding quality, price or logistics which would prevent them from entering the Western European market.

(132) Moreover, as can be seen from Tables 11 and 12, the market for cold rolled flat products has one of the highest rates of spare capacity amongst the markets which have been under close examination in the current proceedings.

TABLE 11

Capacity utilisation in 1999 (%)

<emplacement tableau>

Source:

Parties' notification.

TABLE 12

Evolution of capacity utilisation rates in cold-rolled products

<emplacement tableau>

Source:

Parties and EU-Data.

(133) The spare capacity in the market is not expected to be absorbed by additional demand in the forthcoming years. In fact, the market for total sales of cold rolled products has grown only modestly during the past four years (less than 4 % between 1997 and 2000). The market for merchant sales has even decreased as customers have shifted their demand to hot dip galvanised sheet. In consequence, the current capacity utilisation rate gives competitors in the Community sufficient scope to react on price increases by Newco by an extension of production thus exercising competitive pressure on Newco. This is all the more true, as the industry-wide capacity utilisation rate is lower than Newco's. Additional spare-capacity also exists in Eastern Europe, the main source of imports of cold rolled flat products.

(134) In sum, the Commission concludes that although the market structure post-merger suggests that Newco could have the power to act independently of its competitors, Newco will face the competitive pressure exerted by the high level of imports into the Community and will also have to take into account the existence of a substantial spare capacity in the industry when planning its strategic decisions. The remaining competitors (including importers) will be in a position to effectively constrain Newco's market behaviour.

Conclusion

(135) Those considerations have led the Commission to the conclusion that the proposed concentration will not give the parties the power to hinder effective competition or to evade the rules of competition instituted under ECSC-Treaty with regard to the market for cold-rolled carbon steel flat products.

4. Non grain-oriented electrical steel sheets

Market structure

(136) In 2000, total sales in the market for non grain-oriented electrical steel sheets were 1,4 million tonnes in Western Europe (the available data correspond to the Community, Norway and Switzerland). Parties' sales totalled [...]* million tonnes, comprising Usinor with [...]* million tonnes and Arbed/Aceralia with [...]* million (including captive sales). This represents a share of the market of [15 % to 20 %]* (Usinor [10 % to 15 %]*; Arbed/Aceralia [5 % to 10 %]*), down from [20 % to 25 %]* in 1999.

TABLE 13

Parties' combined share of non grain-oriented electrical steel sheets in the Community + Norway and Switzerland

<emplacement tableau>

Source:

Parties.

(137) TKS will remain the market leader with a market share of [35 % to 40 %]* in 1999. Corus ([15 % to 20 %]* market share) and Voest-Alpine ([10 % to 15 %]* market share) also hold strong positions in the market. Due to the strong position of the competitors, the risk of Newco unilaterally hindering effective competition or evading competition rules can be excluded.

(138) Moreover, the Commission has concluded that the proposed concentration does not create or strengthen the risk of parallel anticompetitive behaviour in this market. Market shares are not symmetric but spread in a range of more than 20 percentage points thus suggesting a lack of common economic interest. TKS is the clear leader of the market, well above its competitors. Voest-Alpine has been very active and its market shares increased by almost 30 % in the last years (from [10 % to 15 %]* in 1997 to [10 % to 15 %]* 1999). Imports represented a market share of about 10 %. The parties have suggested that this figure is likely to increase in the future as transport costs fall relative to the total product price.

Conclusion

(139) On that basis, the Commission believes that the proposed concentration will not give rise to competition concerns in the market for non grain-oriented electrical steel sheets.

5. Galvanised steel

(140) In 2000, the Community market for galvanised steel totalled 21,1 million tonnes. The market has grown significantly during the last years: the average annual growth rate since 1997 has been 9 % to 10 % (see Table 14).

TABLE 14

Sales of galvanised steel in the Community

<emplacement tableau>

Source:

Parties.

(141) In 2000 the parties sold [...]* million tonnes (Usinor [...]* million tonnes, Arbed/Aceralia [...]* million tonnes) or approximately [40 % to 45 %]* of the total market. However some [...]* million tonnes of this total was sold to other companies within the Usinor and Arbed groups. The parties' share of the merchant market (sales to third parties) in 1999 (figures are not available for 2000) was around [40 % to 45 %]* (Usinor [25 % to 30 %]* and Arbed/Aceralia [10 % to 15 %]*).

Market leader

(142) As a result of the proposed concentration, Newco will become the undisputed leader of the Community market for metallic coated products. In this position it will be able to unilaterally raise prices. Other producers will not have an incentive to compete strongly nor will they be able to do so.

(143) In effect, the structure of the market will allow Newco to act unilaterally. As can be seen from Table 15, the parties will be more than double the size of their largest competitor, TKS, and will be very much larger than any of the other players (four times larger than Corus and five times larger than Riva). In fact, Newco will be bigger than its four main competitors combined.

TABLE 15

Share of Community sales of galvanised steel in 1999

<emplacement tableau>

Source:

Parties.

(144) This large difference between the market share of Newco and the market shares of its competitors makes it more difficult for the other players to respond to Newco's price increases and/or output reductions in such a way as to make the unilateral action of the merged firm unprofitable. The merger will provide Newco with a larger base of sales on which to enjoy the resulting price rise while eliminating a competitor (Arbed/Aceralia) to which customers otherwise might have diverted their purchases. It should be noted that Arbed/Aceralia, is currently Usinor's second largest competitor (after TKS). In sum, customers are not likely to be able to find economical alternative sources of supply, as explained below.

Constraint of existing capacity

(145) In general capacity utilisation rates are high (86 % average Community capacity utilisation rate in 1999) in particular in the segment of hot-dip galvanised (90 % average in 1999) so the competing producers will have only limited spare capacity for the supply of Newco's customers.

Planned new capacity

(146) The parties, in their response to the Commission's Statement of Objections, argue that in the future there will be significant spare capacity in galvanised steel in the Community, because of capacity increases in hot-dip galvanised steel between 2001 and 2004 (hot dip galvanised steel represented 76 % of the total market in 2000 and is foreseen to represent over 80 % in 2003). To support the argument, they have produced data about new hot-dip capacities under construction or formally announced in Europe and the European periphery, as well as forecasts of growth in the demand, as estimated by Eurofer for the 2000-2005 period. They conclude that in 2004 cumulated demand will have grown by [ > 2000]* thousand tonnes against [ > 5000]* thousand tonnes for the cumulative capacity (73).

(147) During the investigation, the Commission has been confronted with different figures -from different sources- regarding the evolution of capacity and demand for the next years. On the basis of these different figures, it is difficult to predict the future development of capacity. According to the Commission's investment survey, third parties plan an additional 1,6 million tonnes of capacity between now and 2004. This is lower than the [...]* million identified by the parties in the Form CO(74) and than the 2,6 million tonnes forecast for 2001 to 2004 in the Eurofer survey. In turn, the latest forecast of the parties in their response to the Statement of Objections predicted an increase of [...]* million tonnes.

(148) There are also different estimates of demand growth during this period. The growth rate was estimated at [...]* % per year in the form CO. This figure is low compared to previous growth during the last years (over 10 % per year between 1997 and 2000). The last forecast of the parties in their response to the Statement of Objections result in an average annual growth rate of [...]* % only, as illustrated in Table 16.

TABLE 16

Evolution of Community demand for hot-dip galvanised

<emplacement tableau>

Source:

Reply to the Commission's Statement of Objections, p. 21.

(149) In any event, the parties have calculated the increase of capacity on a wider basis (including Eastern Europe and Turkey) than the relevant geographic market. Moreover, the parties have not taken into account the growth of demand in those countries. This suggests that the foreseen spare capacity may be lower than the parties' prediction. Table 17 summarises the data provided by the parties, in terms of capacity increase.

TABLE 17

Forecast of capacity increases in hot-dip galvanised

<emplacement tableau>

Source:

Reply to the Commission's Statement of Objections, p. 20.

(150) On the basis of the forecast of the respective evolution of Community capacity and demand submitted by the parties, it is possible to predict that the capacity utilisation rate is likely to remain relatively high. Furthermore, according again to the parties' own estimates, Newco will account for a high share of the installed Community production capacity ([40 % to 45 %]* for all galvanised steel and [40 % to 45 %]* for hot-dip galvanised steel by 2004) and therefore the combined amount of spare capacity controlled by Newco's competitors will be relatively small with regards to Newco's capacity and production figures.

(151) As to the increase of capacity in the periphery of the Community, and even without taking into account the eventual increase in demand in these countries, this additional capacity will not substantially affect the competitive situation in the Community in the short to medium term, since deliveries from these countries do not appear to meet the necessary quality requirements of the automotive industry, which constitutes the main customer group of hot-dip galvanised steel.

New entry

(152) New entry is also unlikely to provide competitive constraints. Even though capital cost and technology licences may not be significant barriers to entry, there are some supplementary constraints due to the specificity of the galvanised steel product market. First of all, galvanised steel being a down-stream product, any company willing to enter the market needs to be able to secure its supply of cold-rolled carbon steel. And the steel quality required tends to be speciality grades (high or ultra-high resistance, to meet customers' requirements) which are not commodity products. Secondly, access to customers, and in particular automobile ones, is a difficult process whereby, as will be explained later, the company has to go through some homologation process. As underlined in a report provided by the parties in response the Commission's Statement of Objection, "it is evident that the homologation process implies high time costs and quite considerable administrative expenses for the steel suppliers"(75).

(153) Moreover, new entries in a growing market are not likely to produce a major competitive impact. There have been three examples of new entry into the hot dip galvanised segment recently. Marcegaglia in 1999 (700000 tonnes a year capacity), Bancolor in 2001 (200000 tonnes a year) and a planned line for Duferco in 2002 (350000 tonnes a year). The scale of these new entries is such that even taken together they would result only in an increase in capacity of less than 5 %.

(154) Furthermore all of these increases were planned before the Newco project was announced and have been dealt with in the discussion above (recitals 146 to 151). The situation for potential market entrants will become less favourable with the creation of Newco. This undertaking will have the power to selectively target the (potential) clients of any newcomer, since it is the largest supplier of the automotive and construction sectors.

(155) As far as electro-galvanised steel is concerned, new entries are unlikely to occur either, given the decline in demand for electro-galvanised strip.

Alternative materials

(156) In general, alternative materials will not be able to replace galvanised steel. Once the choice of material has been made, be it in an industrial building components or a car or white goods parts, the die is cast. In the case of cars and white goods it is extremely difficult if not impossible to change from one material to another during the life of the product. Automobile manufacturers set up their production equipment, welding/assembly and painting in particular for specified material for each part. They cannot in nearly all cases credibly threaten to use an alternative material as this would involve major changes to the production processes.

(157) The parties, in their reply to the Commission's Statement of Objections, argue that competition would exist in the long-term, in particular between steel and aluminium, and at the moment of the conception of a new car. However, even after taking into consideration the points raised by the parties, it appears that steel and aluminium are not used in competition as substitutes. The parties explain that "aluminium usage was traditionally focused on a specific niche market for cars with a price premium"(76) and do not contradict that it is still the case today: car makers using aluminium to substantial levels are all in the upper segment (Audi, Jaguar). In practice aluminium and plastic cost between two and four times as much as steel for a given part and could not be used to counter price increases for steel. Consequently, threatening to use aluminium or plastics is not an option either.

Imports

(158) Imports have represented between 3 % and 9 % of total sales of galvanised products in the Community during the last four years. Imports do not appear to represent a serious competitive constraint for Community producers in this market. In particular, imports, largely being commodity-grade steel, are not an alternative for the qualities of metallically coated material required by the automobile industry, the largest customer group representing 51 % of galvanised steel demand in 1999. Few non-Community producers can actually provide these qualities. Only the Far Eastern suppliers, Japan and Korea can supply to the necessary quality standards. However, prices are higher than those in Europe and this additional cost together with the costs of transport and the special packaging result in a delivered price nearly double than that of material produced in Europe. These market conditions being such despite the difficult situation of the steel industry in the Far East.

(159) The existing price differences have not substantially been disputed by the parties. In their reply to the Statement of objections, they have, however, expressed the view that prices in the Community and the Far East move very closely together (notwithstanding the existing price differentials) and that there is a clear cross-sensitiveness between Community -prices and third country imports. Again it must be noted that imports concern mainly commodity-grade steel. Imports of high-quality steel, as requested by the automotive industry, appear to be very limited. According to the parties FIAT recently purchased 200000 tonnes of hot dip steel from Baosteel and PSA bought 50000 tonnes of hot dip steel purchased by PSA from Posco. As can be seen from these figures, imports do not amount to a level which could exercise considerable constraining effects on the important high-end of the product market [...]*.

(160) Consequently, imports do not represent a credible threat against the possibility for Newco to unilaterally increase price or reduce output.

Buying power

(161) Outside the automobile industry no customer will account for more than [0 % to 5 %]* % of Newco's sales and most of them will be very much smaller than 1 %. None of these customers will be able to exercise any pressure on Usinor as the importance of their individual requirements will form only an insignificant part of Newco's sales.

(162) About 50 % of the current demand for galvanised strip originates from the automobile industry. In the automobile industry the largest customer would account for about [5 % to 15 %]* of the combined companies' sales. However this automobile manufacturer currently sources a large majority of its requirements from either Usinor or Arbed. The second largest company accounts for less than 10 % of Newco's sales but it is likewise dependent on either Usinor or Arbed for a large part of its requirements. Given the long time scale necessary to introduce new suppliers, the limited number of alternative suppliers and the limited available capacity at these suppliers it appears unlikely that either of these companies would be able to exercise any form of buyer power in the short term. Other automobile customers are less dependent on Usinor and Arbed but account for even smaller proportion of these companies' sales.

(163) Furthermore, the automobile industry has special requirements, including just-in-time delivery, quality control for safety-critical parts right through the production process (including an audit trail) and specific physical and chemical characteristics set out in detail in the product specifications which may be unique to a given customer. Thus to service the automobile industry a steel producer must make a considerable investment in establishing quality control systems acceptable to the potential client and in providing the infrastructure necessary for just-in-time delivery. Stocks of steel at the automobile manufacturers plants are minimal: one major manufacturer holds an average of three days' supply. Each car part is specified in a particular material and then allocated to a specific supplier. The choice of supplier is very limited and the manufacturer must be certain that the steel maker will be able to consistently meet the specification. There is therefore a long process of testing and trials before a new supplier is accredited. In one case this took five years before the supplier was fully accepted. Other automobile manufacturers have indicated that the periods last from three months to one year and that the process involves quite considerable expenses.

(164) Therefore automobile companies cannot easily switch suppliers and in any case there is a limited number of potential candidates. For these potential suppliers there are considerable risks and costs involved in supplying test materials, setting up the necessary technical support teams and the infrastructure necessary to supply just in time. Considerable time and effort must be expended with no guarantee that sufficient orders will be forthcoming to offset the initial and on-going costs.

(165) This difficulty for customers to find economical alternatives of supply reinforces the likelihood of unilateral action by the merged firm, whether by raising prices or suppressing output.

(166) In their reply to the Commission's Statement of Objections the parties contested those findings of the Commission. They pointed to the following elements: 1. the automotive industry has considerable buying power given that it is highly concentrated, with the top five manufacturers having a combined market share of 77 % in the Community; 2. vehicle manufacturers can and do switch volumes in the short-term, since they "homologate" more suppliers than they purchase from; 3. given that even a loss of small volumes has a disproportionate impact on the parties' profitability, the threat of car manufacturers to switch even smaller volumes to competitors within homologated firms would constrain the parties in their ability to increase prices. The parties have used a report prepared by their economic consultant to argue that, due to the high fixed costs, there are strong incentives in the steel industry to sell the last tonne, even for downstream products such as metallic coated carbon steel. The idea is that selling larger amounts of downstream products increases capacity utilisation upstream and in particular at the level of the blast furnace; this subsequently reduces the cost of production per tonne, due to economies of scale. For instance, increasing capacity utilisation from 90 % to 100 % in an integrated blast furnace -would result in a cost saving of 35 % on the cost per tonne of producing electro-galvanised steel.

(167) After carefully considering these arguments, the Commission maintains its view that buyer power will not be sufficient to prevent Newco from acting unilaterally in the market for metallic coated products. As regards point 1. the concentration level of the automotive industry, the first five galvanised steel suppliers will have a combined market share of [ > 80 %]* post-merger, thus reaching a higher level of concentration than the automotive industry (which is not their only customer-group). Some individual automotive manufacturers source a very substantial part of their demand ([ > 50 %]*) from the merging parties whereas Newco's biggest customer would only account for [5 % to 15 %]* of its galvanised steel sales. It is therefore doubtful that their degree of concentration will be sufficient to balance Newco's market power.

(168) Regarding point 2. even a highly concentrated customer-group can exercise considerable buying power only if there are credible alternatives to shift their demand to. This, however, is not the case. The threat to shift larger volumes is not appropriate to discipline Newco's market behaviour. As stated, capacity utilisation in the industry is high and will remain at a high level in the foreseeable future so that alternative suppliers would not be able to serve considerable additional demand.

(169) As far as point 3. is concerned, the threat to shift small volumes (the last tonne) of demand in the short term is not a sufficient means to control Newco's bargaining power. Blast furnaces in Europe generally already operate at high levels of capacity utilisation (significantly over 90 %). The shift of small amounts of a customers' demand would not considerably impact the cost curve of Newco. This can be illustrated as follows: the parties produced 40 million tonnes of crude steel and about [5-10]* million tonnes of galvanised steel in 1999. Newco's biggest customer ([5-15]* % of the parties combined galvanised production) accounted for about [...]* million tonnes of galvanised steel. Even if this customer shifted its total demand of galvanised steel, that would only represent about [< 5 %]* of Newco's crude steel production. This proportion show that customers shifting only small amounts of their demand are not able to effectively constrain Newco's market behaviour.

Conclusion

(170) On that basis, the Commission has concluded that the proposed concentration will give the parties the power to hinder effective competition or to evade the rules of competition under the ECSC Treaty with regard to the market for galvanised products, including hot-dip galvanised steel and electro-galvanised steel.

6. Organic coated carbon steel

No risk of unilaterally hindering effective competition or evading competition rules

(171) The total value of the market for organic coated carbon steel in 2000 was 4,13 million tonnes. Usinor's share amounted to [30 % to 35 %]* with [...]* million tonnes and Arbed/Aceralia's share was [5 % to 10 %]* with [...]* million tonnes, giving a combined market share of [35 % to 40 %]*. In terms of merchant sales the parties aggregate market share in 1999 was [25 % to 30 %]* (Usinor [15 % to 20 %]*; Arbed/Aceralia: [5 % to 10 %]*).

(172) Newco's closest competitors will be Corus, which held a market share in 1999 of [20 % to 25 %]* of total sales ([20 % to 25 %]* of merchant sales), and TKS with a market share of [10 % to 15 %]* in terms of total sales ([< 10 %]* of merchant sales) in 1999 (see Table 18).

TABLE 18

Parties' combined market shares of organic coated carbon steel

<emplacement tableau>

Source:

Parties.

(173) After the merger, Newco will be the clear market leader in terms of total sales with a market share amounting to [35 % to 40 %]*. The high combined market share figures may give rise to concerns as to whether Newco would be in a position to act independently of its competitors. However, the combined market share of the parties in the merchant market (sales to third parties) is only [25 % to 30 %]*. The difference is due to the high degree of vertical integration in this sector, with both Usinor and Arbed/Aceralia using a large part of their production in further processing activities carried out by companies owned or controlled by their respective groups. The Commission has not identified any competition concerns in these downstream markets. Furthermore, Newco will face competition from important producers such as Corus and TKS, both with substantial spare capacity in this market.

No risk of parallel anticompetitive behaviour

(174) Equally, market characteristics are not conducive to parallel anticompetitive behaviour. The combined share of the three main players in terms of merchant sales represents less than 60 % of the total market. Further, capacity utilisation rates are not symmetric. Usinor and Arbed/Aceralia utilisation is comparatively high at [ > 80 %]*, while the average Community capacity utilisation rate for the total sector is 75 %. Finally, a number of second-tier producers are very active in this market, such as SSAB, Rautaruukki, Voest-Alpine and Salzgitter.

Market growth

(175) The market has been characterised by significant growth with supply increasing by 41 % in the last three years (which can be partly explained by substitution of standard cold-rolled flat carbon steel and galvanised steel by organic coated steel products). This trend may continue in the future.

Conclusion

(176) For those reasons, the Commission does not find that the merged entity will be in a position to hinder effective competition or to evade the rules of competition under the ECSC Treaty in the market for organic coated carbon steel.

7. Steel for packaging

(177) The exact product market definition has been left open between either (a) an overall market for steel for packaging or (b) narrower separate markets for (i) steel for beverage packaging and (ii) steel for non-beverage packaging. In each of the different possible market segmentation, the market share allocation are as set out in Table 19.

TABLE 19

Market shares of steel for packaging in 1999

<emplacement tableau>

Source:

Parties.

(a) Steel for beverage-packaging

Market structure

(178) The market size amounted to 0,6 million tonnes in 2000 and in 1999. In 2000 the parties' combined market share amounted to [40 % to 45 %]* (Usinor [35 % to 40 %]*; Arbed/Aceralia [5 % to 10 %]*). Corus's market share in 1999 was [30 % to 35 %]* and TKS held [25 % to 30 %]* in the same year.

No risk of unilaterally hindering effective competition or evading competition rules

(179) Due to the strong position of the competitors in the market, it is possible to rule out the creation or strengthening of a position where Newco could act independently of the other producers.

No risk of parallel anticompetitive behaviour

(180) The Commission has also examined whether the proposed transaction would lead to parallel anti-competitive behaviour by the main European producers. First, the transaction increases the asymmetry between the producers. More importantly, the three players will be faced with only few multinational customers from the packaging manufacture industry. In the sector of steel for beverage packaging, Usinor generates its turnover with only [...]* customers ([...]*). Arbed/Aceralia has only [...]* customers ([...]*). These undertakings are, in turn, under intense pressure from the international bottlers (such as Coca Cola, Pepsi Cola and the concentrating beer industry) In addition, although the Commission still considers that steel and aluminium (as well as other materials) for beverage cans are separate markets, it is nonetheless true that steel is under competitive pressure from aluminium and PET.

(181) This situation does not leave much scope for the three steel producers to raise prices in a parallel action. Moreover, the capacity utilisation rate in the market is rather low (75 % on average in the overall Community market (steel for beverage and non-beverage packaging). This significant spare capacity provides incentives to "cheat" on agreed prices, thus further destabilising tacit collusion.

(182) Therefore, the Commission has concluded that the proposed concentration will not lead to parallel anti-competitive behaviour of the main European producers.

(b) Steel for non-beverage packaging

(183) The market size amounted to 3,8 million tonnes in the year 2000 and 3,5 million tonnes in the year 1999. In a possible market for steel for non-beverage packing the parties' combined market share amounted to [35 % to 40 %]* (Usinor [20 % to 25 %]*, Arbed/Aceralia [10 % to 15 %]*) in the year 2000. In 1999, Corus's market share was [25 % to 30 %]*, and TKS and Riva held [10 % to 15 %]* and [10 % to 15 %]* respectively, with others producers accounting for [15 % to 20 %]*.

No risk of unilaterally hindering effective competition or evading competition rules

(184) In view of the strong position of the competitors in the market, it is possible to rule out the creation or strengthening of a position where Newco could act independently of the other producers.

No risk of parallel anticompetitive behaviour

(185) The Commission has also examined whether the proposed transaction would lead to parallel anti-competitive behaviour by the main European producers in the market for steel for non-beverage cans. The proposed transaction will also in this market increase the asymmetry between the market shares of the main players. The demand side is very concentrated and represents a strong countervailing element. Three undertakings ([...]*) account for 50 % of Community-consumption. [...]* is Usinor's largest customer and accounts for [20 % to 25 %]* of its sales and [15 % to 20 %]* of Arbed/Aceralia's sales. Arbed/Aceralias biggest client [...]* represents [20 % to 25 %]* of Arbed/Aceralia's turnover in the relevant market.

(186) As in the case of steel for beverage packaging, further competitive pressure stems from alternative materials for packaging (plastic, composite materials, aluminium, glass). For instance, in the field of foodstuff packaging, the new conservation methods, such as freezing, vacuum packing and controlled atmosphere cooking, tend to favour plastic and composite materials. Therefore, it is not in the interest of the steel manufacturers to raise prices and thus put at risk the competitiveness of their product.

(187) Another hurdle for tacit collusion is imports from third countries which represent [15 % to 20 %]* of total sales. As steel for non-beverage use appears to be a standardised commodity product, this rate could be increased in the case of unreasonable pricing of the European suppliers.

(188) Finally the significant spare capacity in the market, as mentioned in recital 181, is also an incentive for the competitors to undercut a supracompetitive price level for steel for non-beverage cans.

(189) Therefore, the Commission has concluded that the proposed transaction will not lead to parallel anti-competitive behaviour of the main European producers.

(c) Overall market: steel for packaging

(190) The market size amounted to 4,3 million tonnes in the year 2000 and 4,1 million in the year 1999. Newco's market share was [35 % to 40 %]* (Usinor [25 % to 30 %]*; Arbed/Aceralia [10 % to 15 %]*) in the year 1999. Competitors' market shares amounted to [25 % to 30 %]* for Corus, [10 % to 15 %]* for TKS, [< 10 %]* for Riva and [10 % to 15 %]* for other producers. The figures do not differ very much from those observed in the steel for non-beverage packaging segment. This stems from the fact that steel for non-beverage packaging accounted for almost 90 % of the overall market. Therefore, the results of the investigation of that market also apply here. In addition, the Commission did not identify competition problems in the smaller alternative market for steel for beverage packaging. In sum, also in a hypothetical overall market, the proposed transaction does not give rise to competition concerns.

Conclusion

(191) For those reasons, the Commission concludes that the proposed concentration would not hinder effective competition or allow the parties to evade the competition rules under the ECSC Treaty in the market or markets for steel for packaging.

8. Stainless steel flat products

(192) Stainless steel flat products include hot-rolled stainless and cold-rolled stainless products. As in a previous Commission decision in these markets (77), the effect of the proposed merger must be examined on two levels, to see whether 1. after the operation Newco would be able to determine prices or prevent competition on its own, and 2. the operation would create conditions which would make tacit collusion between the limited number of European stainless steel producers possible and effective.

No risk of unilaterally hindering effective competition or evading competition rules

(i) Hot rolled stainless steel products

(193) Regarding hot rolled stainless steel products, Table 20 gives the shares for hot rolled stainless steel products in terms of total sales in the Community + Norway and Switzerland. On this basis the parties' combined market share was [25 % to 30 %]* in 2000 and [25 % to 30 %]* in 1999. They would face competition from AvestaPolarit ([30 % to 35 %]* in 1999), TKS ([20 % to 25 %]*) and Acerinox ([10 % to 15 %]*).

TABLE 20

Market shares in 1999 in the Community + Norway and Switzerland (total sales)

<emplacement tableau>

Source:

Parties.

(194) In the market for stainless steel hot-rolled flat products, the merged entity would therefore not be the European leader. Moreover, Newco would not be able to benefit from its size and degree of integration, since, according to the parties, its competitors also belong to vertically integrated groups: AvestaPolarit is a joint-venture between Outokumpu Steel and Avesta Sheffield; Krupp Thyssen Stainless belongs to TKS, Acerinox controls the American undertaking North American Stainless, and one of its main shareholders is Nisshin Steel, Japan's largest stainless steel producer. In this situation the possibility that the proposed merger would enable the parties to unilaterally determine prices or prevent effective competition can be ruled out.

(ii) Cold rolled stainless steel products

(195) The parties' and their competitors' shares of the market for cold rolled stainless steel products are given in Table 21. The parties' combined share in terms of total sales (including captive sales) would be about [20 % to 25 %]* in 2000 and [20 % to 25 %]* in 1999 (well behind TKS which has a share of about [35 % to 40 %]* in 1999). In terms of merchant sales, the parties' combined market share would amount to [15 % to 20 %]*. Newco would also face competition from AvestaPolarit and Acerinox ([20 % to 25 %]* and [10 % to 15 %]* in terms of total sales respectively). Similarly the market shares and the existence of strong competitors rule out the possibility that Usinor/Arbed/Aceralia would be able to unilaterally determine prices or prevent effective competition.

TABLE 21

Market shares in 1999 in the Community + Norway and Switzerland (total sales)

<emplacement tableau>

Source:

Parties.

No risk of parallel anticompetitive behaviour

(i) Hot rolled stainless steel products

(196) After completion of the proposed concentration, there will only be five suppliers of stainless steel hot rolled products. The top three players will have > 80 % of the market for hot rolled stainless steel respectively (see Table 22).

TABLE 22

Market shares in 1999 in the Community + Norway and Switzerland (total sales)

<emplacement tableau>

Source:

Parties.

(197) The Commission examined whether the merger may lead to parallel anticompetitive behaviour. The Commission concluded that, for the following reasons, the operation is not likely to lead the stainless steel producers collectively to raise prices and/or restrict production or distribution or otherwise evade the ECSC competition rules.

Market growth

(198) In contrast to almost all other sectors in the steel industry, stainless steel products enjoy a growth rate of at least 6 % per annum. This rate of increase is expected to continue for at least the next four years. There are therefore incentives for existing players to compete for market shares by investing continuously. Significant capacity extensions have been announced recently by Avesta-Polarit (700000 tonnes in Tornio) and ALZ (600000 tonnes by 2003).

Increase in capacity

(199) While it is possible to increase capacity by eliminating bottlenecks, these possibilities are becoming rarer and investments in new plants are required to meet demand. The optimum scale of these investments is very large in proportion to the size of the Community market. Avesta Polarit's (the result of the merger of Avesta Sheffiels and Outokuumpu) new steel plant in Finland has a capacity of 1 million tonnes a year (700000 tonnes a year of hot rolled and 300000 tonnes a year of cold rolled) and AST's new capacity is for 600000 tonnes a year. This is to be compared with the total capacity of hot-rolled stainless steel (6,7 million in 2000). As new capacity comes on stream, the owner has to compete vigorously to load the new plant.

Cost structures

(200) The competitors present on the market have very different cost structures, as a result of a number of factors, including differences in terms of access to raw materials. Evidence supporting this conclusion has already been given in the Outokumpu/Avesta Sheffield Commission decision of December 2000 (recitals 31 to 35), where differences of up to 40 % in the conversion costs of Community producers were noticed. The parties also explained that alloying elements such as nickel and chrome account on average for some 30 % of the price of hot-rolled stainless steel. Some competitors are vertically integrated into mining activities, such as for instance Outokumpu Steel, Columbus (a large part of its shareholders are chromium producers), or a number of Japanese competitors that have financial links with nickel and chromium producers. By contrast, Arbed/Aceralia and Usinor have no mining activities relevant to this sector. Differences of costs structures also result from differences in terms of location of production facilities that result in different transportation costs. Transportation costs are low in absolute terms, but can make the difference between loss and profitability. Internal transportation costs are very limited for some competitors (e.g. Acerinox and Outokumpu) and very significant for other competitors (AvestaSheffield). The parties' transportation costs range between these extremes.

(ii) Cold rolled stainless steel products

(201) After completion of the proposed concentration, there will only be five suppliers of stainless steel cold rolled products. The top three players will have [ > 80 %]* of the market for cold rolled stainless steel, as shown in Table 23. However, even after the merger, TKS will have a notably larger market share than the merged entity.

TABLE 23

Market shares in 1999 in the Community + Norway and Switzerland (total sales)

<emplacement tableau>

Source:

Parties.

(202) The Commission examined whether the proposed concentration may lead to parallel anticompetitive behaviour. The Commission concluded that, for the same reasons as in hot-rolled stainless steel, the operation is not likely to lead the stainless steel producers collectively to raise prices and/or restrict production or distribution or otherwise evade the ECSC competition rules.

Conclusion

(203) In conclusion, the Commission's investigation concluded that the proposed merger would not give rise to any competitive concerns as regards either stainless steel hot-rolled or cold-rolled flat products.

2. STEEL DISTRIBUTION

1. Vertically integrated distributors/independent distributors

(204) Steel distribution constitutes a very important part of the global commercial policy of the vertically integrated steel producers. Arbed/Aceralia submit that roughly 40 % of the steel consumption in Europe goes through the different distribution channels.

(205) The main European steel producers all have their own networks of distribution centres, covering generally a substantial part of the European territory, in particular, the major industrial areas. These networks of distribution centres normally share a common commercial policy or sales strategy, are coordinated centrally or operate under one business management structure or a central sales department, though each centre has a certain degree of independence.

(206) In addition to these integrated distributors, there is an independent sector consisting of a large number of companies of varying size and financial strength. In 1999 the proportion of sales by the independent sector through steel service centres was as set out in Table 24.

TABLE 24

Share sales through steel service centres of independent distributors in 1999

<emplacement tableau>

Source:

Parties.

(207) Independent distributors can in theory supply the same services as integrated distributors, provided that they have invested in the necessary equipment. Nevertheless, a large part of their supplies come necessarily from the mills of the European steel producers, whose integrated distributors are their competitors in the distribution sector. In particular, the possibility of securing all the necessary supplies or the extended validity of prices will depend entirely on the individual relationships between independent distributors and producers. It should be noted that, with regard to the steel service centre's activities, the price of wide coils supplied by the mills determines in a high degree the slit coil price, as the parties have outlined.

(208) Moreover, in its reply to one of the Commission's questionnaires, one independent distributor pointed out that in difficult times integrated distributors fix prices for end customers that are substantially to the same as those that the producers charge for the same products to independent distributors. Likewise, another independent distributor, after affirming that large steel producers show clear preference for their integrated distributors, claimed that some of the pricing positions adopted by integrated steel service centres are not based on the need to achieve full cost recovery of the processing and logistics concerned.

(209) Further to their integration upstream in the manufacturing of the relevant products, distributors owned by steel producers have several other advantages with respect to independent distributors. In particular, the geographical location of the different centres of integrated distributors, spread throughout Europe, allows them to conclude global contracts with large customers, those with different production sites in Europe and who have one decision centre for purchasing. In effect, while individual negotiations take place between small customers and the sales organisation of each steel service centre, for large customers global supply contracts are negotiated, normally on a European basis, and afterwards the products and services are delivered through the most convenient distribution centre on the basis of the location of the customers' sites. This practice is particularly common in the automotive sector and in the domestic appliance industry.

(210) The market investigation confirms that a number of independent competitors have serious concerns about the operation. In particular, even if a small number of competitors believe that the proposed merger will bring about more efficiency for the merging parties and therefore more stable markets, others fear that steel distribution will be further dominated by producers, to the detriment of small independent distributors. This could result in a reduction in the number of suppliers to the distributors and an increase in prices. Some independent distributors submitted that the proposed concentration could bring about the disappearance of their supply sources, particularly taking into account that in the past they have faced serious uncertainties of continuity of supply from manufacturers outside the Community.

(211) Therefore, in line with previous Commission Decisions (78), it should be concluded that, though steel distribution is a regional or national service, in evaluating the impact of the merger in the distribution sector account should be taken of the implication of the vertical integration of steel manufacturers on the competition between distributors, in particular the supply of steel products to distributors at a European level.

2. Assessment of the different distribution markets

(212) The proposed operation will bring about particularly high market shares in the following markets:

(a) steel service centres in France;

(b) steel service centres in Spain/Portugal;

(c) stockholding activity in France; and

(d) oxycutting centres in France.

(213) Other affected markets for the purposes of this Decision are:

(a) steel service centres in Benelux and North Rhineland-Westphalia;

(b) steel service centres in the United Kingdom and Ireland; and

(c) stockholding activity in Benelux and North Rhineland-Westphalia.

(a) Steel service centres in France

(214) The total quantities of carbon flat products sold in France in 2000 by steel service centres amounted to 3.9 million tonnes (which represented about 41 % of the total supplies of flat carbon products), of which approximately [...]* million tonnes were sold by Usinor and [...]* million tonnes by Arbed/Aceralia. The parties' combined market share was [40 % to 45 %]* in 2000 ([40 % to 45 %]* in 1999), the proposed concentration bringing about an aggregation of [5 % to 10 %]* ([5 % to 10 %]* in 1999). Their main competitors' market shares in 1999 were [5 % to 10 %]* (Corus), [5 % to 10 %]* (Riva) and [5 % to 10 %]* (TKS).

(215) After the merger, Newco's market share will be almost five times bigger than that of its principal competitor in the French market, and will almost double the combined market share of its three main competitors. Furthermore, the market share of independent steel service centres in France is quite low if compared with other neighbouring countries. Independent steel service centres only account for [< 30 %]* of the market in France, while they account for [ > 50 %]* in Germany (excluding North Rhineland Westphalia), [ > 40 %]* in Italy and [ > 40 %]* in the United Kingdom. Moreover, this independent sector is highly fragmented into a large number of small suppliers.

(216) Account should also be taken of the parties' vertical integration and, in particular, of Newco's strong share in the upstream market for the production and direct sales of carbon steel flat products, both at the EEA level and in France. The parties' production of hot-rolled carbon steel in 1999 represented [40 % to 45 %]* of Community production, while their production of cold-rolled carbon steel flat products represented [45 % to 50 %]* of Community production. Moreover, the market investigation outlines that some independent distributors competing with Usinor's distribution network in France rely strongly from Arbed/Aceralia's supplies; the operation will bring about the disappearance of this alternative source of supply. In sum, the new entity will be in a position, after the merger, to determine prices, to control or reduce distribution or to hinder effective competition.

(217) In view of those considerations, the Commission has arrived at the conclusion that the notified operation will give the parties the power to hinder effective competition or to evade the rules of competition under the ECSC Treaty with regard to the market for the distribution of carbon steel flat products through steel service centres in France.

(b) Steel service centres in Spain/Portugal

(218) In the Iberian Peninsula, steel service centres account for about 44 % of the total supplies of flat carbon products, whilst stockholding and oxycutting centres supply aproximately 10 %. In particular, the total quantities of carbon flat products sold in Spain and Portugal in 2000 by steel service centres amounted to 4,2 million tonnes, of which approximately [...]* million tonnes were sold by Usinor and [...]* million tonnes by Arbed/Aceralia. The parties' combined market share was [45 % to 50 %]* in 2000 ([50 % to 55 %]* in 1999), the proposed concentration bringing about an aggregation of [10 % to 15 %]* ([10 % to 15 %]* as well in 1999). Their main competitors' market shares in 1999 were [10 % to 15 %]* (Grupo Ros), [5 % to 10 %]* (Gutser) and [5 % to 10 %]* (A. Gallardo).

(219) Further, if Spain and Portugal were considered to constitute separate geographic markets, serious competition concerns would arise in both countries. Newco's share would have represented [40 % to 45 %]* ([30 % to 35 %]* + [10 % to 15 %]*) in Spain in 1999, and [65 % to 70 %]* ([65 % to 70 %]* + [< 5 %]*) in Portugal.

(220) Again, the difference post-merger between Newco and its main competitors will be very substantial. Newco will be more than four times bigger than its nearest competitor. Newco's main competitors, moreover, will be independent distributors (rather than integrated distributors of other steel producers), which will depend to a large extent on the notifying parties for their supplies of steel products. Further, Newco will enjoy a particularly strong position in the distribution of carbon steel flat products through both steel service centres and stockholding centres in France, the closest market to the Iberian Peninsula, reducing the possibilities for alternative suppliers for Spanish and Portuguese customers.

(221) In their reply to the Statement of Objections, the parties submit however that, on the basis of national markets, there would be no material overlap in Portugal, since Arbed/Aceralia's market share is [< 5 %]*. Nonetheless, given the particular high market share of Usinor in Portugal ([65 % to 70 %]*) and the level of cross-border sales between Portugal and Spain, Arbed/Aceralia represent, through their operations in Spain, the main source of potential competition in the Portuguese market. In particular, given the different size of the Portuguese and Spanish markets, Arbed/Aceralia's sales of carbon flat products through steel service centres in Spain ([...]* kilotonnes in 1999) while representing only [10 % to 15 %]* of the Spanish market are substantially higher than the overall Portuguese market for the sales of carbon flat products through steel service centres (399000 tonnes in 1999).

(222) In view of those considerations, the Commission has arrived at the conclusion that the notified operation will give the parties the power to hinder effective competition or to evade the rules of competition under the ECSC Treaty with regard to the market for the distribution of carbon steel flat products through steel service centres in Spain and Portugal.

(c) Stockholding activities in France

(223) The parties' combined market share for the distribution of all carbon steel products was [25 % to 30 %]* in 2000 ([20 % to 25 %]* + [5 % to 10 %]*) and in 1999 ([15 % to 20 %]* + [5 % to 10 %]*). Their main competitors were KDI, whose market share in 1999 was [25 % to 30 %]*, and Descours et Cabaud, with [15 % to 20 %]*.

(224) However, Newco would have a substantially higher market share regarding the distribution of carbon steel flat products ([35 % to 40 %]* in 1999). Newco would be much bigger than its main competitors in the sector of the distribution of flat products, namely KDI (10 % to 15 %) and Descours et Cabaud (5 % to 10 %). The remainder stockholding market is extremely fragmented, with over 500 small independent suppliers. The customer base is fragmented as well, with the main customers being small and medium-sized enterprises.

(225) In addition, it should be noted that Newco would not only have a substantial share of the distribution of carbon steel flat products in France through stockholding centres, but a dominant position also on the distribution of carbon steel flat products in France through steel service centres (with a market share above 40 %). Moreover, Newco, being vertically integrated, would also enjoy a strong position in the upstream market for the production of carbon steel flat products in Europe (for instance, [45 % to 50 %]* of production of cold-rolled carbon steel flat products in the Community in 1999).

(226) In their reply to the Statement of Objections, the parties contest the Commission assessment, on the basis that the relevant market should include the stockholding of all carbon steel products, where the parties' combined market share represents [25 % to 30 %]* of the overall market, as mentioned in recital 223. However, this issue has been dealt with in the section on product market definition.

(227) In view of those considerations, the Commission has arrived at the conclusion that the notified operation will give the parties the power to hinder effective competition or to evade the rules of competition under the ECSC Treaty with regard to the market for the distribution of carbon steel flat products through stockholding centres in France.

(d) Oxycutting centres in France

(228) The total quantities sold in France in 2000 by oxycutting centres amounted to 200000 tonnes, of which approximately [...]* were sold by Usinor and [...]* by Arbed/Aceralia. The parties' combined market share was [35 % to 40 %]* in 2000 ([30 % to 35 %]* in 1999), the proposed concentration bringing about an aggregation of [5 % to 10 %]* ([5 % to 10 %]* as well in 1999). Their main competitors' market shares in 1999 were [5 % to 10 %]* (Oxymetal), [5 % to 10 %]* (Devillers) and [5 % to 10 %]* (Oxycentre).

(229) Newco's market share will be considerably bigger than that of its main competitors: Newco will be five times larger than its closest rival. An important part of the market (34 %) is highly fragmented, with more than a hundred players active. The customer base is quite fragmented as well: most of the actors are small and medium enterprises with a low plate consumption and little or no buying power. Imports from outside the EEA are insignificant and represent no competitive constraint. Further, oxycutting is a market in decline, due to the arrival of new technologies such as laser cutting and water jet, which renders unlikely the possibility of substantial new entrants in the market, as is evidenced by the fact that no significant entries have taken place during the last five years.

(230) In their reply to the Statement of Objections, the parties disagree with the Commission's conclusions, on the basis of the following arguments: 1. Arbed's sales are small; and 2. Usinor's position includes 100 % of Eurodecoupe's output whereas Eurodecoupe is in fact a DHS subsidiary over which Usinor does not have sole control.

(231) As far as the first argument is concerned, Arbed's sales account for [5 % to 10 %]* of the market for oxycutting centres in France, Arbed being currently the second main competitor of Usinor on the basis of the volume of sales, slightly behind the first one (Oxymetal with [5 % to 10 %]* market share in 1999) and above the other market players (Devillers, Oxycentre, Savoy Decoupe, Bretagne Oxy and UF Acier). With regard to the second argument, the parties themselves have confirmed that Usinor does have joint control over Eurodecoupe, and therefore all of the sales of Eurodecoupe must be taken into account for the purposes of the assessment of the proposed concentration.

(232) In view of those considerations, the Commission has arrived at the conclusion that the notified operation will give the parties the power to hinder effective competition or to evade the rules of competition under the ECSC Treaty with regard to the market for the distribution of quarto plates through oxycutting service centres in France.

(e) Other affected markets

(233) For the purposes of the assessment of the proposed concentration, other affected markets in the distribution sector are the following:

(1) Steel service centres in Benelux and North Rhineland-Westphalia, where the parties' combined share in the market for the distribution of carbon flat products in 2000 was [20 % to 25 %]* ([10 % to 15 %]* + [10 % to 15 %]*). Their main competitors are Corus, whose market share is [20 % to 25 %]*, and TKS, which also has a [20 % to 25 %]* market share.

If that regional market were to be further subdivided, Newco's market shares in 1999 would have been [20 % to 25 %]* in the area comprising Belgium and Luxembourg, [25 % to 30 %]* in the Netherlands and [15 % to 20 %]* in a global German market (including North Rhineland-Westphalia). In all three markets Newco would face competitors of comparable or even bigger size: Corus in Belgium and Luxembourg ([25 % to 30 %]*) and in the Netherlands ([25 % to 30 %]*) and TKS in Germany ([15 % to 20 %]*).

(2) Steel service centres in the United Kingdom and Ireland, where the parties' combined market share in 2000 was [15 % to 20 %]* ([5 % to 10 %]* + [10 % to 15 %]*). If national markets were to be considered, Newco's market share would be [20 % to 25 %]* in the United Kingdom ([5 % to 10 %]* + [10 % to 15 %]*) and < 5 % ([< 5 %]* + [< 5 %]*) in Ireland. In any event, in the market for the distribution of carbon flat products in the United Kingdom and Ireland, the parties' market share is clearly smaller than that of Corus ([30 % to 35 %]*).

(3) Stockholding activities in the Benelux and North Rhineland-Westphalia, where the parties' combined market share for carbon steel products in 2000 was [15 % to 20 %]* ([10 % to 15 %]* + [5 % to 10 %]*). In none of the alternative further subdivisions, whether by product markets (flat/long products) or by geographic markets (national markets), would the parties' combined market shares, taking into account those of the other market operators, raise serious competition concerns.

(234) The operation, therefore, does not raise serious competition concerns in the affected markets described in this section.

VI. UNDERTAKINGS SUBMITTED BY THE PARTIES

1. SUMMARY OF THE COMMITMENTS

1. Proposed divestitures

(235) In order to remedy the competition concerns identified by the Commission, the parties have offered to divest a number of undertakings and businesses in the sectors of steel production and steel distribution.

(a) Steel production markets

(236) The parties will divest their shares in the following undertakings:

(1) Finarvedi: Usinor agrees to divest, or procure to divest, its interest in the hot-dip production line of ISP Arvedi.

(2) Galmed: Aceralia and Usinor agree to divest, or procure to divest, their shares in Galvanizaciones del Mediterraneo, SA (51 % and 24,5 % of Galmed's capital, respectively). Galmed produces hot dip galvanised steel.

(3) Lusosider: Usinor agrees to divest, or procure to divest, its shares in Lusosider Projectos Siderurgicos SA (50 % of the capital). Lusosider produces cold rolled steel, hot dip galvanised steel and tinplate.

(4) Segal: Arbed's Sidmar and Usinor's Cockerill Sambre agree to divest, or procure to divest, their shares (each 33,33 % of the capital) in Société Européenne de Galvanisation SA Segal produces hot dip galvanised steel.

(237) Further, the parties will divest, or procure to divest, the following businesses:

(1) Beautor: Usinor agrees to divest, or procure to divest, its Beautor at La Fere, France. Beautor houses a cold rolling mill and an elecro-galvanised steel line.

(2) Dudelange: Arbed agrees to divest, or procure to divest, the hot-dip steel plant at Dudelange, Luxembourg.

(3) Straßburg: Usinor agrees to divest, or procure to divest, its site at Strasbourg, France (hot-dip steel and organically coated steel).

(238) The divestiture of the latter businesses will include all tangible and intangible tangible assets and the transfer of personnel currently employed at the plants.

(239) Where the production lines referred to in recitals 236 and 237 are not integrated with cold-rolling lines, the parties will offer, at the request of the purchaser, corresponding cold-rolling feedstock capacity or, alternatively, offer toll-rolling, in accordance with normal market conditions. At the request of the purchaser, transitional services/supply agreements will be offered. The units Galmed, Beautor and Strasbourg will be offered as a package, subject to the existing pre-emption right with regard to Galmed.

(b) Distribution markets

(240) In the field of steel distribution, the parties have submitted the following undertakings:

(1) Cofrafer: Arbed agrees to divest, or procure to divest, Cofrafer SA, a wholly-owned subsidiary, including its subsidiaries Flaberga Découpe, situated at Barberey St. Sulpice, France, Orn'oxycoupage, situated at Chailloue, France, and Surfaçage de Normandie, situated also at Chailloue.

2. Bamesa: Usinor agrees to divest, or procure to divest, all its shares in Bamesa Aceros, SL, or alternatively to transfer the remaining minority stake, if any, to an independent trustee, until such time as the shares can be sold (currently, Usinor owns directly and indirectly 49 % of Bamesa's capital, over which it has joint control).

2. Related commitments and procedure

(241) The parties have committed to comply with a number of related obligations designed to ensure the preservation of the full economic viability and competitiveness of the divested businesses, which include the appointment of a trustee responsible for the monitoring of the implementation of the commitments. Moreover, the commitments set out in detail the divestment procedure. The full text of the commitments is set out in the Annex to this Decision.

2. ASSESSMENT OF THE UNDERTAKINGS

1. Galvanised products

(242) In terms of market shares, the divestments offered would bring Newco appreciably closer to the level of Usinor in the pre-merger situation. In 1999 and in terms of total sales, Usinor had a market share of [30 % to 35 %]* in galvanised steel, against [10 % to 15 %]* for Arbed/Aceralia. After divestments, the combined market share of Usinor, Arbed and Aceralia would fall from [40 % to 45 %]* before divestment to [35 % to 40 %]*. More than [5 % to 10 %]* market share would be divested by the parties, which would correct their position in the market. The figures for 2000 are quite similar, as shown in Table 25.

TABLE 25

Newco's sales and proposed divestments

<emplacement tableau>

(243) As far as capacity is concerned, the proposed divestments would also substantially bring down Newco's market share, from [40 % to 45 %]* to [35 % to 40 %]* in 2002-2003, as Table 26 illustrates.

TABLE 26

Newco's capacities and proposed divestments

<emplacement tableau>

(244) With such market shares, the proposition that Newco could unilaterally increase prices and reduce output would become less realistic, especially due to the potential improvement in the competitive position of its direct competitors after the divestments.

(245) The proposed divestments will improve the market structure in two ways. If larger competitors like TKS or Corus were to buy some of the plants, this would reduce the gap in terms of size between themselves and Newco. If smaller competitors (like Salzgitter or Voest-Alpine) were to buy some of the plants, that would create a new competitive force with around [5 % to 10 %]* market shares. This would create a competitor of the same order of magnitude as Arbed/Aceralia.

(246) Moreover, the proposed divestments would resolve the problems identified in the Statement of Objections with regard to the automobile industry. More than 65 % of the divestments relate to this industry. The investigation showed that the accreditation process operates at the plant level. Potential buyers would therefore also gain access to some automobile customers, by buying the plants supplying these customers. As the parties explained in their reply, the supply of steel to the automobile sector is effectively a competitive bidding process. With the divestments proposed, automobile customers will be in a position to find alternative sources of supply to Newco, because new competitors or stronger competitors will be able to take part in the bidding process.

(247) Furthermore, following the divestments, joint ownership by the largest Community steel manufacturers would largely disappear in the market for galvanised steel products. This would diminish the incentives and possibilities for competitors to potentially co-ordinate their behaviour.

(248) Finally, since the divestments take place in the "home territory" of the merging firms (the area comprising France, Spain and Portugal and Benelux), it would also reduce incentives to develop subcontinental geographic markets for these products.

(249) Those considerations have been borne out by the market test of the proposed remedies undertaken by the Commission.

2. Steel distribution

(250) The proposed divestments would eliminate the competition concerns identified by the Commission with regard to several distribution markets in France and Spain/Portugal.

(251) In France, the divestment of Cofrafer would bring the total market share of Newco down from around [40 % to 45 %]* to around [35 % to 40 %]* in the market for steel service centres in 2000. As far as the markets for stockholding of flat carbon steel products and oxycutting are concerned, Newco's market shares after divestments will amount to around [30 % to 35 %]*. More importantly, the divestment will eliminate the overlaps in the markets for steel service centres, stockholding and oxycutting in France, since it includes all Arbed/Aceralia's steel distribution operations in these markets.

(252) In addition, the divestment of Bamesa would appreciably reduce the market shares of Newco in the market for steel service centres in Spain/Portugal. On the basis of market shares in 2000, Newco's share in Bamesa represents around [15 % to 20 %]* of the market. After divestment, Newco would go from having [45 % to 50 %]* share of the market to [25 % to 30 %]*. This compares to [10 % to 15 %]* for Grupo Ros, [5 % to 10 %]* for Gutser and [5 % to 10 %]* for A. Gallardo. If national markets were to be considered, Newco's market share in Spain would be reduced from [40 % to 45 %]* to [25 % to 30 %]*, and from [65 % to 70 %]* to [30 % to 35 %]* in Portugal. The divestment will therefore enable a competitor to appreciably increase its market share, outweighing the imbalance created by the merger between Usinor Arbed and Aceralia.

(253) Those considerations have been borne out by the market test of the proposed remedies undertaken by the Commission.

3. Conclusion

(254) The Commission therefore considers that the undertakings satisfactorily address the competition concerns identified by the Commission both in the markets for galvanised steel and steel distribution.

VII. CONCLUSION

(255) In the light of all those considerations, the Commission has reached the conclusion that, provided that the undertakings set out in the Annex to this Decision are fully respected, the proposed concentration would not give rise to competition problems and that in particular it would not give Newco Steel the power:

- to determine prices, to control or restrict production or distribution or to hinder effective competition in a substantial part of the market for the relevant products; or

- to evade the rules of competition under the ECSC Treaty, in particular by establishing an artificially privileged position involving a substantial advantage in access to supplies or markets.

(256) Since the requirements of Article 66 (2) of the ECSC Treaty are thus met, the proposed merger should be authorised, on condition that the parties fully comply with the following commitments (subject to any change pursuant to the review clause set out in the Annex and):

[...]* (See non-confidential version of commitments in Annex).

(257) Those aspects of the commitments constitute conditions, since only by fulfilling them (subject to any change pursuant to the review clause), can the structural change on the relevant markets be achieved. The remaining aspects of the commitments, concerning the implementing steps which are necessary to achieve the structural change that is sought, constitute obligations within the meaning of the fifth subparagraph of Article 66 (5) of the ECSC Treaty.

HAS ADOPTED THIS DECISION:

Article 1

The operation whereby the undertakings Arbed SA, Aceralia Corporación Siderúrgica SA and Usinor SA merge into the company Newco Steel is hereby authorised under Article 66 (2) of the ECSC Treaty, subject to full compliance with the commitments set out in the Annex and in particular with the following conditions:

[...]* (See non-confidential version of commitments in Annex).

Article 2

This Decision is addressed to the notifying parties.

(1) Commission Notice concerning the alignment of procedures for processing mergers under the ECSC and EC Treaties (OJ C 66, 2.3.1998, p. 36).

(2) OJ C 81 of 4.4.2003.

(3) Cases No. IV/M.0980 and IV/ECSC.1237 - Arbed/Aceralia.

(4) OJ L 395, 30.12.1989 p. 1; corrected version in OJ L 257, 21.9.1990, p. 13.

(5) OJ L 180, 9.7.1997, p. 1.

(6) Cf. Case No COMP/M.2382 - Usinor/Arbed/Aceralia.

(7) Case No COMP/M.2382 - Usinor/Arbed/Aceralia.

(8) Parts of this text have been edited to ensure that confidential information is not disclosed; those parts are enclosed in square brackets and marked with an asterisk.

(9) Case No IV/ECSC.1268 - Usinor/Cockerill Sambre. Case No IV/ECSC.1243 - Krupp Hoesch/Thyssen. Case No COMP/ECSC.1342 - Outokumpu/Avesta Sheffield. Case No IV/M.239 - Avesta/British Steel/NCC/AGA/Axel Johnson. Case No IV/M.484 - Krupp/Thyssen/Riva/Falck/Tadfin/AST.

(10) Case No IV/ECSC.1268 - Usinor/Cockerill Sambre.

(11) Case No IV/ECSC.1243 - Krupp Hoesch/Thyssen. See also Case No COMP/ECSC.1342 - Outokumpu/Avesta Sheffield.

(12) Case No IV/M.239 - Avesta/British Steel/NCC/AGA/Axel Johnson.

(13) Case No IV/M.484 - Krupp/Thyssen/Riva/Falck/Tadfin/AST.

(14) Case No IV/ECSC.1269 - Sollac/Aceralia/Solmed. Case No IV/ECSC.1243 - Krupp Hoesch/Thyssen. Case No IV/ECSC.1268 - Usinor/Cockerill Sambre. Case No IV/ECSC.1237 - Arbed/Aceralia. Case No IV/ECSC.1264 - Aceralia/Aristrain.

(15) Case No IV/ECSC.1243 - Krupp Hoesch/Thyssen.

(16) Case No IV/ECSC.1269 - Sollac/Aceralia/Solmed. Case No IV/ECSC.1243 - Krupp Hoesch/Thyssen. Case No IV/ECSC.1268 - Usinor/Cockerill Sambre. Case No IV/ECSC.1237 - Arbed/Aceralia. Case No IV/M.484 - Krupp/Thyssen/Riva/Falck/Tadfin/AST.

(17) Case No IV/ECSC.1269 - Sollac/Aceralia/Solmed.

(18) Case No IV/M.484 - Krupp/Thyssen/Riva/Falck/Tadfin/AST.

(19) Case No IV/ECSC.1269 - Sollac/Aceralia/Solmed.

(20) Plates, known according to their thickness as heavy plates or medium plates, are rolled square or rectangular pieces made by cutting hot-rolled wide strip. Universal plates are hot-rolled flat products cut into pieces between 150 and 1250 mm wide and more than 4 mm thick.

(21) Case No IV/ECSC.1243 - Krupp Hoesch/Thyssen. Case No IV/ECSC.1237 - Arbed/Aceralia.

(22) Case No IV/ECSC.1243 - Krupp Hoesch/Thyssen. Case No IV/ECSC.1237 - Arbed/Aceralia. In turn, narrow strip (or band steel), including slit strip, is produced either directly in widths of less than 600 mm or by slitting hot-rolled wide strip.

(23) Case No IV/ECSC.1268 - Usinor/Cockerill Sambre.

(24) Case No IV/ECSC.1237 - Arbed/Aceralia. Case No IV/M.239 - Avesta/British Steel/NCC/AGA/Axel Johnson. Case No IV/M.484 - Krupp/Thyssen/Riva/Falck/Tadfin/AST. Case No IV/ECSC.1268 - Usinor/Cockerill Sambre.

(25) Case No IV/M.484 - Krupp/Thyssen/Riva/Falck/Tadfin/AST.

(26) Ibidem.

(27) Case No IV/ECSC.1243 - Krupp Hoesch/Thyssen. Case No IV/ECSC.1269 - Sollac/Aceralia/Solmed. Case No IV/ECSC.1237 - Arbed/Aceralia. Case No IV/ECSC.1268 - Usinor/Cockerill Sambre.

(28) Case No IV/ECSC.1243 - Krupp Hoesch/Thyssen.

(29) Aluminium or zinc alloys (i.e., zinc/aluminium, zinc/nickel, etc).

(30) Paint, thin organic coatings (TOC), laminate film.

(31) Organic coating entails an added-value of 15 % to 35 % on the raw material used, depending on the exact application.

(32) Organic coated aluminium, non-coated flat carbon steel products and various other metallic and non-metallic materials.

(33) Case No IV/ECSC.1268 - Usinor/Cockerill Sambre. Case No IV/ECSC.1269 - Sollac/Aceralia/Solmed.

(34) In some cases, the metal is then given a further treatment which converts the coating to a zinc/iron alloy (8 % to 12 % iron), facilitating welding at the customer's premises (galvannealed).

(35) Such as aluminium, plastic, composite materials, glass and even other steel products (stainless steel, cold rolled steel, organic coated).

(36) Case No IV/ECSC.1237 - Arbed/Aceralia. Case No IV/ECSC.1243 - Krupp Hoesch/Thyssen. Case No IV/ECSC.1268 - Usinor/Cockerill Sambre.

(37) Case No IV/ECSC.1268 - Usinor/Cockerill Sambre.

(38) Case No COMP/M.1663 - Alcan/Alussuisse.

(39) Case No COMP/ECSC.1342 - Outokumpu/Avesta Sheffield.

(40) Case No IV/M.239 - Avesta I. Case No IV/M.484 - Krupp/Thyssen/Riva/Falck/Tadfin/AST. Case No COMP/ECSC.1342 - Outokumpu/Avesta Sheffield.

(41) Case No COMP/ECSC.1342 - Outokumpu/Avesta Sheffield. Case No IV/M.239 - Avesta/British Steel/NCC/AGA/Axel Johnson.

(42) Case No IV/M.484 - Krupp/Thyssen/Riva/Falck/Tadfin/AST.

(43) Annex I of the ECSC Treaty.

(44) Case No COMP/ECSC.1342 - Outokumpu/Avesta Sheffield. Case No IV/M.239 - Avesta/British Steel/NCC/AGA/Axel Johnson. Case No IV/M.484 - Krupp/Thyssen/Riva/Falck/Tadfin/AST. Case No IV/M.1203 - Usinor/Finaverdi.

(45) Case No IV/ECSC.1268 - Usinor/Cockerill Sambre. Case No IV/ECSC.1264 - Aceralia/Aristrain. Case No IV/ECSC.1243 - Krupp Hoesch/Thyssen. Case No IV/ECSC.1237 - Arbed/Aceralia. Case No COMP/ECSC.1342 - Outokumpu/Avesta Sheffield. Case No IV/M.73 - Usinor/ASD. Case No IV/M.239 - Avesta/British Steel/NCC/AGA/Axel Johnson. Case No IV/M.0503 - British Steel/Svensk Stal/NSD. Case No IV/M.578 - Hoogovens/Klöckner & Co. Case No IV/M.971 - Klöckner/Comercial de Laminados. Case No IV/M.918 - Klöckner/ODS. Case No IV/M.484 - Krupp/Thyssen/Riva/Falck/Tadfin/AST. Case No IV/M.1329 - Usinor/Cockerill Sambre. Case No IV/M.504 - Avesta (III).

(46) Case No IV/M.971 - Klöckner/Comercial de Laminados. Case No IV/M.918 - Klöckner/ODS.

(47) Case No IV/M.578 - Hoogovens/Klöckner & Co.

(48) Case No IV/M.971 - Klöckner/Comercial de Laminados. Case No IV/M.918 - Klöckner/ODS.

(49) Case No IV/M.918 - Klöckner/ODS. Case No IV/M.1329 - Usinor/Cockerill Sambre. Case No IV/M.0971 - Klöckner/Comercial de Laminados.

(50) Case No IV/M.971 - Klöckner/Comercial de Laminados.

(51) Case No IV/ECSC.1268 - Usinor/Cockerill Sambre. The distribution of stainless steel products has been analysed by the Commission, inter alia, in the following Decisions: Case No COMP/ECSC.1342 - Outokumpu/Avesta Sheffield; Case IV/M.504 - Avesta (III); Case No. IV/M.484 - Krupp/Thyssen/Riva/Falck/Tadfin/AST; Case No IV/M.239 - Avesta/British Steel/NCC/AGA/Axel Johnson.

(52) Case No IV/M.503 - British Steel/Svensk Stal/NSD.

(53) Case No IV/M.971 - Klöckner/Comercial de Laminados. Case No IV/M.918 - Klöckner/ODS.

(54) Case No IV/M.239 - Avesta/British Steel/NCC/AGA/Axel Johnson.

(55) Case No IV/M.73 - Usinor/ASD.

(56) Case No IV/ECSC.1268 - Usinor/Cockerill Sambre.

(57) Case No IV/M.1329 - Usinor/Cockerill Sambre.

(58) Case No IV/M.1329 - Usinor/Cockerill Sambre.

(59) Case No IV/ECSC.1264 - Aceralia/Aristrain. Case No IV/M.1329 - Usinor/Cockerill Sambre. Case No IV/ECSC.1243 - Krupp Hoesch/Thyssen.

(60) Case No IV/ECSC.1243 - Krupp Hoesch/Thyssen.

(61) Due to the lack of figures about Western Europe, the analysis has been conducted using EU data. This would not anyway change the result of the investigation, given the low impact of EFTA countries on the total Western European market.

(62) Case No IV/M.484 - Krupp/Thyssen/Riva/Falck/Tadfin/AST.

(63) Case No IV/ECSC.1269 - Sollac/Aceralia/Solmed. Case No IV/CECA.1268 - Usinor/Cockerill Sambre. Case No IV/ECSC.1237 - Arbed/Aceralia.

(64) Case No IV/CECA.1268 - Usinor/Cockerill Sambre.

(65) Case No IV/M.484 - Krupp/Thyssen/Riva/Falck/Tadfin/AST. Case No IV/M.239 - Avesta/British Steel/NCC/AGA/Axel Johnson.

(66) Case IV/M.503 - British Steel/Svensk Stal/NSD.

(67) Case No IV/M.760 - Klöckner/ARUS.

(68) Case No IV/M.0918 - Klöckner/ODS.

(69) Case No IV/M.760 - Klöckner/ARUS. Case No IV/M.1329 - Usinor/Cockerill Sambre.

(70) Case No IV/CECA.1268 - Usinor/Cockerill Sambre. Case No IV/M.1329 - Usinor/Cockerill Sambre.

(71) Case No IV/M.1329 - Usinor/Cockerill Sambre.

(72) Case IV/M.578 - Hoogovens/Klockner.

(73) Reply to the Commission's Statement of Objections p. 21.

(74) Volume II, Annex 4(1)(4) p. 26.

(75) The parties' economic consultants' report "Steel products for the automotive sector", 10 October 2001, p. 9.

(76) The parties' economic consultants' report "Steel products for the automotive sector", 10 October 2001, p. 29.

(77) See Decision of 4 December 2000 in Case COMP/ECSC.1342 - Outokumpu/Avesta Sheffield.

(78) Case No IV/M.484 - Krupp/Thyssen/Riva/Falck/Tadfin/AST.

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