EC, November 11, 1998, No M.1157

THE COMMISSION OF THE EUROPEAN COMMUNITIES,

Having regard to the Treaty establishing the European Community, Having regard to the Agreement on the European Economic Area, and in particular Article 57 (2) (a) thereof, Having regard to Council Regulation (EEC) No 4064-89 of 21 December 1989 on the control of concentrations between undertakings(1), as last amended by Regulation (EC) No 1310-97 (2), and in particular Article 8 (2) thereof, Having regard to the Commission decision of 14 July 1998 to initiate proceedings in this case, Having given the undertakings concerned the opportunity to make known their views on the objections raised by the Commission, Having regard to the opinion of the Advisory Committee on Concentrations (3) Whereas:

(1) On 3 April 1998, the Commission received a notification of a concentration by which Skanska AB (publ) (Skanska) of Sweden has acquired within the meaning of Article 3 (1) (b) of Regulation (EEC) No 4064-89 (the Merger Regulation) sole control of Scancem AB (publ) (Scancem) of Sweden by way of purchase of shares. The shares were acquired on the Stockholm Stock Exchange in the period between 9 and 14 October 1997.

(2) The notification was declared incomplete on 7 May 1998 according to Article 4 of Commission Regulation (EC) No 447-98 of 1 March 1998 on the notifications, time limits and hearings provided for in Council Regulation (EEC) No 4064-89 on the control of concentrations between undertakings (4). Complete information was obtained on 12 June 1998 and the notification thus became effective within the meaning of Article 4 (2) of Regulation (EC) No 447-98 on 15 June 1998.

(3) After examination of the notification, the Commission concluded that the notified operation falls within the scope of the Merger Regulation and could create or strengthen a dominant position as a result of which effective competition would be significantly impeded in the common market or in a substantial part of it and in the territory covered by the EEA Agreement. Therefore, on 14 July 1998, the Commission decided to initiate proceedings pursuant to Article 6 (1) (c) of the Merger Regulation.

I. THE PARTIES

(4) Skanska is active in building construction, production and distribution of building materials and real estate management. Its main activities are in Sweden, Denmark, Germany, Finland, Norway and the United States.

(5) Scancem is active in production and distribution of building materials, primarily cement and other mineral-based materials. Its main activities are in Sweden, Norway, Finland, the United Kingdom and Denmark.

II. THE OPERATION

A. BACKGROUND

(6) On 2 October 1995, Skanska and Aker RGI (5) "Aker" participated in a transaction whereby Euroc AB (a company active in the production of cement in Sweden and Finland), acquired Aker's cement and construction materials business. In the arrangements that led to this transaction Skanska and Aker both played key roles. Aker received, as payment, partly newly issued shares in Euroc, partly cash which was used to purchase all shares in Euroc owned by the Finnish company Partek (6). Following the transaction, Aker became the owner of 33,3 % of the shares and voting rights in Euroc. Given the existence of a pre-emption agreement from 1993 between Skanska and Partek relating to their shareholdings in Euroc, it was a precondition to Aker's purchase of Partek's shares that Skanska did not exercise its pre-emption right over these shares in accordance with the pre-emption agreement. Skanska consented to the sale of Partek's shares to Aker and, at the same time, increased its own shareholding in Euroc to the same level as Aker by purchasing most of the shares held by the Swedish company Robur (which is, among other things, one of the largest shareholders in Skanska). As a result, Skanska and Aker each took a shareholding of 33,3 % in the merged entity (which was subsequently renamed Scancem).

(7) The abovementioned transactions led to contacts between the Commission on the one hand, and Skanska, Aker and Euroc on the other. Skanska and Aker submitted that according to its statutes, all strategic decisions in Scancem were to be be taken by simple majority, both at the level of the shareholder's meeting and on the Board. Aker and Skanska stated that they had made no voting or similar arrangements, and that they had no strong common interest that de facto would lead them to act collectively in exercising their rights in relation to Scancem. Skanska and Aker concluded that there would be real scope for changing alliances and that there were no elements sufficient to support a finding that Skanska and Aker would jointly control Scancem and, therefore, the transactions were not notifiable under the Merger Regulation.

(8) The Commission's services pointed out that Skanska and Aker would have the means to control Scancem if they acted together, and that the available information provided strong indications that the acquisition by each company of 33,3 % of the shares in Scancem would result in de facto joint control. Elements which in the view of the Commission's services supported this were, inter alia, that the acquisitions had taken place by means of concerted action, and the existing evidence that Skanska and Aker had in principle agreed on a comprehensive shareholder's agreement which would have included agreement on the business of Scancem. It could therefore be concluded that it would make more sense for Skanska and Aker to act together on a durable basis, as opposed to trying to form shifting alliances with minor shareholders on a case-by-case basis.

(9) Skanska and Aker, however, did not share that view and thus maintained that the transaction did not have a Community dimension and should therefore not be notified to the Commission. In view of the parties' unwillingness to notify the transaction, they were informed that the Commission would carefully monitor Scancem to see whether appropriate proceedings were warranted.

(10) Until October 1997 the ownership situation in Scancem remained unchanged. On 8 October 1997 Aker exchanged its Scancem B shares (low voting power) for Scancem A shares (high voting power). Aker thereby controlled 41,2 % of the votes in Scancem. Skanska immediately responded to this by increasing its holding of shares in Scancem to 48,06 % of the votes (see recital 12).

(11) On 23 and 24 April 1998, the Commission carried out an investigation pursuant to Article 13 (3) of the Merger Regulation at the premises of Skanska and Scancem. The purpose of the investigation was to clarify, in part the transactions, agreements and other arrangements relating to the establishment of Scancem, in part other events and circumstances of relevance for Scancem's operations in order to assess the notifiability of the operation. However, Skanska has submitted undertakings in this case intended also to remove the competitive concerns relating to the 1995 transaction.

B. THE NOTIFIED TRANSACTION

(12) The notifed operation concerns Skanska's acquisition of additional shares in Scancem. Skanska thereby increased its shareholding in Scancem from 17751090 shares equalling 33,3 % of the shares and votes, to 208520188 shares equalling 39,16 % of the shares and 48,06 % of the votes. The acquisitions took place on the Stockholm Stock Exchange in the period between 9 and 14 October 1997.

(13) Skanska maintains that, in its view, the increased shareholding did not constitute a concentration within the meaning of Article 3 of the Merger Regulation.

III. THE CONCENTRATION

(14) According to paragraph 14 of the Commission's notice on the concept of a concentration (7), a minority shareholder may be deemed to have sole control on a de facto basis. This is the case when the shareholder is highly likely to achieve a majority at the shareholder's meeting, given that the remaining shares are widely dispersed. In such cases, the Commission has consistently based its assessment on participation in shareholder's meetings in previous years. (8)

(15) In the present case, such an assessment can be made on the basis of participation in Scancem's shareholders' meetings in 1996 and 1997. When these meetings took place, Skanska and Aker each had 33,3 % of the votes in Scancem. Of the remaining 33,3 %, which are dispersed among a large number of institutional and private investors, less than half (i.e. 39,4 % and 37,2 %, respectively) were represented. When assessing whether Skanska would be in a position to achieve the majority of the votes exercised at future shareholders' meetings, account has to be taken of the subsequent acquisitions of additional shares in Scancem made by Skanska and Aker. Skanska's acquistion is explained above. On 8 October 1997, Aker increased its voting rights to 41,2 % by exchanging B shares to A shares. The total amount of voting rights held by Skanska and Aker was thus 89,26 % whereas 10,74 % of the votes were dispersed among the remaining owners.

(16) These figures show that Skanska, following the notified transaction, could be almost entirely certain to achieve a majority of votes cast at future shareholders' meetings. Participation by the group "other shareholders" corresponding to that observed at the shareholders' meetings in 1996 and 1997 would mean that, in addition to the votes held by Skanska and Aker, at most another 4,2 % of the votes would be present at the meeting. Skanska would therefore hold at least 51,5 % of all votes present. The table below indicates the result of a holding of 48,06 % (9)

<emplacement tableau>

(17) Skanska has submitted that the assessment cannot be based on the above analysis, given that other outcomes are possible. In particular, Skanska has pointed out that other shareholders, given the present situation where Skanska and Aker recently increased their shareholdings, have an obvious interest in participating in future shareholders' meetings in order to protect their financial investments.

(18) The Commission, however, cannot accept Skanska's view. The incentive of the other shareholders to participate in the shareholders' meetings during the period where Skanska and Aker each controlled 33,3 % of the votes must be considered as the same, if not stronger, compared to the present ownership structure. This is due to the fact that each shareholder in the previous situation would have had a greater possibility to exercise its influence in the event that Skanska and Aker would not have had a common opinion. Despite Skanska's statement to the contrary, the decreasing incentive for "other shareholders" to participate in shareholders' meetings in Scancem is clearly supported by the facts. As has been indicated above, participation by this group fell from 37 % to 31 % after Skanska's acquisition in October 1997 (10).

(19) Finally, the Commission notes that if it had exercised all its 48,06 % of the votes at the shareholders' meeting on 22 April 1998, Skanska would have represented 52,1 % of all votes cast at that meeting.

(20) Based on the above, it can be concluded that the notified transaction results in Skanska acquiring sole control of Scancem within the meaning of Article 3 (1) (b) of the Merger Regulation.

(21) Given that the acquisitions took place in October 1997, the Commission will consider a possible application of Article 14 of the Merger Regulation.

IV. COMMUNITY DIMENSION

(22) Skanska and Scancem have a combined aggregate worldwide turnover in excess of ECU 5000 million (Skanska: ECU 6339,9 (11) million and Scancem: ECU 1857,6 million). Each of them has a Community-wide turnover in excess of ECU 250 million (Skanska: ECU 3884,3 million and Scancem: ECU 1092,9 million), and they do not achieve more than two thirds of their aggregate Community-wide turnover in one and the same Member State. Therefore, the operation has a Community dimension. It constitutes a cooperation case under the EEA Agreement, pursuant to Article 57 of that Agreement, and thus the case is to be assessed by the Commission in cooperation with the EFTA Surveillance Authority, in accordance with Article 58 and Protocol 24 of the EEA Agreement.

V. RELEVANT MARKETS

(23) In Sweden, the combined activities of Skanska and Scancem cover the whole value chain in the construction sector. Thus, Scancem's main activity is the production of key raw materials, primarily cement. Skanska and Scancem are active in the production of cement-based construction materials (e.g. concrete) and other construction materials (e.g. bricks, plasterboard, paint plaster and floor levelling products). Finally, Skanska is the largest construction company in the Nordic region.

(24) In view of this, the notified operation has a significant vertical impact on all the markets mentioned below. There are also significant horizontal overlaps between Skanska and Scancem on the Swedish concrete and aggregates markets.

(25) In the notification the parties indicated that there are no horizontal overlaps in Norway and Finland, since Skanska is not active in the production of construction materials in those countries. However, subsequent to receiving the statement pursuant to Article 18 of the Merger Regulation, Skanska has acknowledged that its Finnish subsidiary, Tekra OY, is active in the production of ready-mixed concrete and precast concrete products in Finland. Moreover, in the course of the investigation, it was indicated that the operation in those countries too may have vertical effects due to Scancem's position on the cement market and Skanska's construction activities, similar to those described below for Sweden.

A. RELEVANT PRODUCT MARKETS

(26) The notified operation affects the following markets:

- cement,

- aggregates

- ready-mixed concrete

- dry concrete

- pre-cast concrete products

- construction.

(27) In addition, Scancem's pre-existing activities in a number of other construction materials, such as bricks, plaster board, paint plaster and floor levelling products will have an impact on the operation.

(a) Cement

(28) Cement is a mineral-based binding material which, when mixed with water, sets to a hard mass. Cement is used as an intermediary product mainly in the production of concrete. The properties of cement vary considerably depending on the quantities of raw materials used and the production methods chosen (12).

However, all types of cement derive from a single intermediate product known as clinker which is obtained by burning a mixture of calcareous materials containing chalk and lime, with argillaceous products such as shales, slate and sand.

(29) There are two main types of cement: grey cement and white cement. The main difference lies in a very particular quality of chalk, which is used for the production of white clinker (13).

According to Skanska, white cement is used for decorative purposes. Scancem produces only grey cement but imports a very limited quantity of white cement which would only represent around 0,6 % of the Swedish cement sales in 1997. Therefore, only the market for grey cement will be taken into consideration in the present assessment.

(30) There are two basic methods for producing clinker: the "wet" process (the traditional method where the raw materials are ground and mixed with water before being burned) and the "dry" process (a method consuming less energy). The latter process is used by Scancem whose cement production process comprises broadly the following steps: limestone and marl are quarried, crushed, stocked and finally ground into a fine powder to be stored in silos. The powder, called "raw meal", is prepared in a dry condition by a so-called precalcining process (splitting calcium carbonate into calcium oxide and carbon dioxide). The alkali content is condensed out before the powder is burnt (at around 1400°C) in rotary kilns and transformed into clinkers which are cooled and crushed into the required sizes and stored in silos. Finally, the clinkers are ground and mixed with gypsum and additives to make cement.

(31) Cement has in earlier decisions been characterised as a banal semi-finished product for which competition among different brands is weak (14). At the same time, each type of cement must meet certain criteria with respect to composition, specifications and conformity set out in European standards, with national amendments. Most of Scancem's cement production is so-called standard cement, which is suitable for a wide range of purposes. One of Scancem's plants, Degerhamn, produces a particular grade of cement, which hardens more slowly than standard cement. This grade of cement produces concrete that is particularly suitable for certain construction works, such as bridges and roads, where a higher resistance to weather conditions is desirable. Furthermore, in Sweden a reduced chrome content in cement is required in order to reduce the risk of allergic reactions. In the notification Skanska has characterised the cement market as one overall market. The Commission has in earlier merger decisions concluded that various types of cement appear to be substitutable (15). This has largely been confirmed by the Commission's investigation in the present case. Based on the above, the grey cement market should be seen as one market.

(b) Aggregates

(32) According to Skanska, the term "aggregates" is typically used to refer to three primary raw materials in construction and civil engineering: gravel, crushed rock and sand. It has been submitted that gravel and crushed rock can be considered close substitutes. For environmental reasons, the Swedish Government has introduced legislation imposing a tax on the "use" of gravel. This has served to equalise the cost of using crushed rock and gravel (crushed rock otherwise being more expensive) (16). The two materials are used primarily in the preparation of foundations for buildings and civil engineering works and in the production of concrete and asphalt. Sand is unsuitable for use in foundations and is employed almost exclusively in the production of concrete. Skanska has submitted that despite the lack of substitutability between, on the one hand, gravel and crushed rock, and on the other hand, sand, it is customary to include all three types of aggregates in an analysis of the Swedish aggregates markets even though this may not be entirely justifiable on strictly economic grounds.

(33) In any event, given that sand accounts for less than 5 % of the Swedish aggregates market and that Skanska extracts and sells sand in very small quantities (less than 5 %), it will make no material difference in the context of the present assessment and the aggregates market will therefore be considered as a whole for the purpose of the present assessment).

(c) Concrete

(34) Concrete is produced by mixing cement with aggregates and water. It can be mixed at a specific concrete plant ("ready-mixed concrete"), and subsequently transported to the point of use in specific vehicles ("mixer trucks").

(35) Concrete can also be mixed on-site, at the point of use. The traditional form of on-site mixed concrete is where the builder simply adds sacked cement, aggregates and water together. This, however, is labour intensive and results in concrete of uneven quality. Such on-site mixing is therefore only used for projects where speed and quality are not important factors and the required volumes are relatively small (< 0,75 m3 according to Skanska). It appears that on-site mixing is of limited economic importance. For example, Skanska estimates that it corresponds to at most 1 to 2 % of the concrete used in its construction activities.

(36) A more sophisticated form of on-site mixed concrete is dry concrete, a product to which only water needs to be added. Dry concrete is prepared according to specific recipes, and therefore offers an even quality. Furthermore, dry concrete offers the builder a higher degree of flexibility, since it reduces the need to plan when the concrete will be required as well as the exact amounts needed. However, since dry concrete is significantly more expensive than ready-mixed concrete, builders are unlikely to use dry concrete if volumes exceeding 1,0 m3 are required.

(37) Finally, for certain applications concrete is also sold as precast elements or other precast products, which may also include steel reinforcements.

(38) In its notification Skanska submitted that the relevant product markets in the field of concrete should be considered as ready-mixed concrete, dry concrete and precast concrete products. Each of these markets will be further assessed below.

(i) Ready-mixed concrete

(39) Ready-mixed concrete is sold and delivered in a semi-wet form. It consists of cement (15 %), aggregates (85 %) together with water and additives. The product is used as a base for on-site moulded concrete constructions. Ready-mixed concrete is perishable and is normally delivered in large volumes by mixer-trucks.

(40) In its notification Skanska submitted that there is some substitutability, depending on the actual use, between ready-mixed concrete on the one hand and dry concrete and/or on-site mixed concrete on the other. As noted above, figures provided by Skanska indicate that on-site mixed concrete appears to be of limited economic importance. On the other hand, figures submitted by the parties indicate that there is a significant price difference between ready-mixed concrete and dry concrete, the latter being about 50 % more expensive. The fact that ready-mixed concrete is normally used when large amounts are required also indicates that dry concrete may more properly be seen as a complementary product.

(41) For the purpose of this assessment, it is therefore considered that ready-mixed concrete constitutes a separate market (17).

(ii) Dry concrete

(42) As stated above, dry concrete is a pre-mixed concrete, which only requires the addition of water. It is therefore easy to use and tends to be used primarily in applications where the required amounts are relatively small, but where an even quality is needed.

(43) In its notification Skanska stated that there is, depending on the specific applications, a competitive relationship between dry concrete and traditional on-site mixed concrete. Whereas it is correct that both products are used in applications where relatively small amounts of concrete are used, there are several factors that indicate that the competitive relationship between the two products is not very significant. First, dry concrete can be used in quality-sensitive applications, which is not the case for traditional on-site mixed concrete. Secondly, the price differences appear to be even greater than between dry concrete and ready-mixed concrete.

(44) In its reply to the statement pursuant to Article 18 (hereinafter referred to as "the reply"), Skanska argued that it has not been an active competitor for Scancem on the market for dry concrete, given that Skanska has focused its sales on construction companies, whereas Scancem's sales have been focused on the retail level. This argument, however, fails to take into account that the volume produced by Skanska through its sub-contracting agreements is sold through retailers. Moreover, Skanska has not argued, or provided any evidence to indicate that sales to construction companies should be seen as a market separate from that of sales to retailers. In conclusion, dry concrete should be considered as a market separate from that for other types of concrete.

(iii) Precast concrete products

(45) Precast concrete products comprise, among other things, ready-made pillars, beams, joists, road barriers, containers, railway sleepers, and blocks and elements for house construction. These products normally include supporting steel constructions. Moreover, the products are often non-standardised and produced according to customer's specifications. Precast concrete products are sold separately or as part of a package, which may also contain certain construction work.

(46) Following the notification, Skanska has revised its view on the relevant product market, and submitted that precast concrete products should be divided into products used for house construction and products used "in and on earth" (pipes, kerbstones, railway sleepers, etc.). In Skanska's view "house products" constitute a homogeneous range of precast products sharing common production equipment and know-how, which, in the company's view, is not the case for "in and on earth products" which require different and more specialised technology and equipment. Skanska has also submitted that "in and on earth products" should be further subdivided.

(47) However, for the purpose of this assessment the question whether such a division should be undertaken may be left open, since, as will be shown below, the market effects would remain essentially the same whether the assessment is based on an overall market for precast concrete products, or on a market for precast concrete "house" products.

(d) Other construction materials

(48) Scancem has substantial pre-existing activities relating to a number of other construction materials. Whereas these activities are not directly affected by the notified operation, they would indirectly strengthen the parties' combined position on other markets. These pre-existing activities of Scancem include plasterboards, which are used for the construction of walls, floors and inner ceilings. They also include products for levelling or smoothing new and reconstructed floors before a wooden textile or plastic floor is put in, as well as paint plaster, which is used for levelling walls prior to painting or hanging wall paper, and is used both in the construction of new buildings and in renovation projects.

(49) In a recent submission Skanska has suggested that some of these products (floor levelling products, mortar (a type of "glue" used in, inter alia, bricklaying) and plaster (used for facades)) should be assessed in conjunction with dry concrete. According to Skanska all these products make up one "family of products", since there is a high degree of supply-side substitutability between them. Skanska has, however not been able to supply any significant market data relating to the sale of this "family of products", nor has it supplied information about the alleged substitutability.

(50) It is, however, not necessary for the purposes of this assessment to reach a final conclusion on the exact scope of the market for plaster boards, floor-levelling products and paint plaster, as it would not affect the assessment of the operation.

(e) Construction activities

(51) In addition to the markets identified by Skanska in its notification, the investigation has shown that the notified acquisition would have an impact on the construction market, where Skanska is the largest supplier in Sweden.

(52) Industry participants, including Skanska, have submitted that the construction market should be divided into two separate relevant markets; one for construction of houses and one for infrastructure construction work (e.g. roads, railroads, tunnels and bridges). The reason stated for this separation is that different skills, personnel, machinery and other equipment are needed to be active on the two markets (18).

(53) Furthermore, a distinction can also be made according to the size of the construction project. The reason for this is that the competitive conditions vary significantly between small and large projects. Although there is no absolute limit between what should be qualified as a small and a large project, the Swedish Competition Authority has made a distinction between projects with a total value below or above SEK 40 million (approximately ECU 4,6 million). Under the Swedish Law on Public Procurement, SEK 40 million is the threshold value, above which projects must undergo a competitive tendering process. Moreover, in Sweden, the number of construction companies that are able to undertake projects with a value exceeding the abovementioned value is significantly lower compared to the situation below that level.

(54) It is therefore appropriate to assess the market position of Skanska on the basis of separate markets for house construction and infrastructure construction respectively. Moreover, on both these markets it is appropriate to make a distinction between projects with a combined value below or above SEK 40 million.

(f) Conclusion on product markets

(55) Based on the above, the Commission has come to the conclusion that there are separate markets for cement, aggregates, ready-mixed concrete and dry concrete. As regards precast concrete products it is not necessary to decide whether this should be seen as an overall market, or whether, as submitted by Skanska, it should be separated into "house construction products" and "in and on earth products". For the remaining products, the product market definition may be left open.

B. RELEVANT GEOGRAPHIC MARKETS

(56) The Commission has previously held that the cement market may be seen as a set of markets, centred around the various factories, overlapping one another and covering the whole of Europe. The size of each market and the extent of market overlap are determined by the distance from the factory at which cement may be sold (19).

(57) However, in its notification Skanska has submitted that the market for cement is mainly regional/national but has provided market share figures only on a national level. It may be observed that there are virtually no exports or imports between Sweden or Finland and any other Member State (20). In its reply, Skanska has indicated that the situation is different in Norway, where the largest source of imports is Blue Circle (United Kingdom). However, it should be noted that this company is an equal shareholder in Embra's import terminals in that country. In conclusion, the apparent absence of imports indicates that cement producers in other Member States have not sought to exercise competitive pressure in the Nordic area through export sales. This holds true also for Norway, where, outside the indicated structural investment by Blue Circle, there have been virtually no imports from other west European cement producers.

(58) Accordingly, the market where the current operation should be assessed is not Europe-wide but rather limited to each of the three Nordic countries or, at most, the Nordic area. Given that Scancem is the sole cement producer in all three Nordic countries, the lack of intra-Nordic trade does not necessarily indicate that the geographic market is not Nordic. In any event, the question whether the relevant market is national or Nordic does not need to be decided given that the result of the assessment would be the same regardless of the definition used.

(59) Skanska has submitted that the markets for precast concrete products, dry concrete, plaster boards, facing materials, floor-levelling products and paint plaster are mainly national in scope. This has essentially been confirmed by the investigation, which shows that generally there are relatively insignificant imports and exports of all the abovementioned products. In addition, Skanska has submitted that there is a specific certification procedure for concrete production in Sweden, which is handled by BBC, a trade association. Prior to certification the producer must, inter alia, verify the properties of the raw material.

(60) The markets for house and infrastructure constructions are national, at least as far as large projects are concerned. It is still rare for non-Swedish construction companies to take on construction projects in Sweden, other than as subcontractors or in consortia with Skanska or one of the other Swedish construction companies. For the purposes of this assessment it is not necessary to define the geographic scope of the market for small projects.

(61) In conclusion, the abovementioned markets cannot be seen as wider than national. At the same time the investigation shows that the supply structure, the price levels and other conditions of competition do not differ significantly within each country.

(62) In the notification Skanska sumbitted that the markets for ready-mixed concrete and aggregates are local, due to high transport costs and, with respect to ready-mixed concrete, that distances exceeding one hour of transport result in a lower quality. In a subsequent submission Skanska revised its view, stating that these markets nevertheless should be assessed at national level since the local segments overlap throughout the country. In Skanksa's view the competitive constraints faced by suppliers in local areas are dictated by competition in a nationwide network of interlocking segments. A number of third parties have indicated that they consider the markets for ready-mixed concrete and aggregates to be local or regional, however, other third parties have indicated arguments similar to those in Skanska's subsequent submission, i.e. that the market is national.

(63) From an economic viewpoint the catchment area for a ready-mixed concrete plant (or an aggregate supplier) is normally limited by a boundary within which the transport can be made in no more than about one hour. From the customers' viewpoint, a one-hour distance might offer a choice of one, two, or more suppliers, depending on his and their location. It is clearly impracticable to define exactly all the possible combinations and permutations, which are a function of the density and distribution both of the population of customers and of the network of suppliers. In addition, for any particular customer of ready-mixed concrete, or aggregates, the choice of suppliers for each construction project will be limited according to the location of both the construction site and the location of each producer. Since most purchasers of these products are construction companies with a relatively wide geographic area of activity (in most cases nationwide), the identity and number of economically feasible suppliers will change for every project.

(64) Furthermore, the degree of overlap which occurs between the suppliers' catchment areas, together with the distribution of the customer population, will not only determine the competitive interactions between geographically proximate ready-mixed plants, but will also, to some extent, have a "knock-on" or "chain-reaction" effect on more distant suppliers. Moreover, many important parameters of competition, such as location of the production facilities, range and quality of products offered, service level (opening hours, etc.), advertising, promotion and, to some extent, prices and rebates are not decided at local level, but at regional or national level. Such decisions are generally implemented nationwide. In view of the above, it is necessary and appropriate to aggregate what from the viewpoint of any single project may be a very local market.

(65) Aggregation of sales figures can be made on, in principle, an indefinite number of different levels. There are virtually no imports or exports of ready-mixed concrete or aggregates, which can be explained by their high transport costs and, regarding the former, its perishable nature. The widest possible geographic market is therefore national.

(66) Skanska and Scancem (as well as their main competitors) operate ready-mixed plants, and aggregate production, with a wide geographic spread. Although the available information suggests that the parties' market shares vary somewhat, if measured on national, regional or local level, it is not necessary for the assessment of the operation to decide finally on the exact scope of the relevant geographic market, since the assessment would not be significantly different whether made on national, regional or local level.

(67) Based on the above, an assessement of the competitive impact of the operation can be made on the basis of the parties' strengths at national level, supported by evidence concerning the markets for ready-mixed concrete and aggregates on a regional and local level (21).

VI. COMPATIBILITY WITH THE COMMON MARKET AND THE FUNCTIONING OF THE EEA AGREEMENT

A. RAW MATERIALS

(a) Cement

(i) Overview

(68) Scancem is active in the production of cement in Sweden, Norway and Finland. In these countries, its cement activities are organised in three national companies (22), which in combination make up one business unit (Cement Nordic). This business unit comprises the cement activities in Sweden (three plants), Norway (two plants) and Finland (two plants) and had a total turnover of ECU 367,5 million in 1997 (23). Scancem International comprises, inter alia, all the international cement production and export activities, the latter primarily to Africa and the USA, and had a turnover of ECU 439,2 million in 1997. Finally, the subsidiary Castle Cement is active in the cement production in the United Kingdom and had a turnover of ECU 225,3 million (market share of approximately [20 to 30 %] (24) in 1997. Skanska has no cement production but is, in its capacity as the largest Swedish producer of concrete, as well as the largest construction company, the most important purchaser of cement in the Nordic region.

(69) The cement markets in Sweden, Norway and Finland each have an annual turnover of between ECU 80 million and ECU 95 million. Scancem's market share is, according to its own figures as well as those of third parties, about [80 to 90 %]* in each of these countries. The market shares have remained stable [...]* over the last three years. There is no other producer of cement in these countries and the only competitor on these markets with a market share exceeding 5 % is the Norwegian company Embra, which imports cement from Lithuania.

(70) It appears from the notification that total sales of cement increased by more than 20 % in Norway and Finland between 1995 and 1997. During the same period sales in Sweden decreased by slightly less than 20 %. Despite these diverging trends, Scancem was able to maintain its sales prices in all three countries more or less unchanged over the period [...]*. The company's capacity utilisation increased over the period from [...]* in Norway and from [...]* in Finland, but decreased from [...]* in Sweden.

(ii) Supply structure

(71) Scancem is the only producer of cement in Sweden and has three cement production facilites which are all situated in the southern part of Sweden; in Slite, Degerhamn and Skövde, with a total production capacity of approximately [< 2]* million of clinker ([< 1, < 1 and < 1])* million t respectively) (25). The two former plants are located on the Baltic islands of Gotland and Öland.

(72) The fixed costs in cement production are high. According to figures provided by Scancem, its proportion of fixed costs in 1997 (excluding depreciation and financial costs) was [...]* in Slite and Degerhamn, whereas in Skövde it was [...]*. The company therefore has a strong incentive to run its production facilities on a very high capacity utilisation level. In Sweden, Scancem's capacity utilisation has normally been at least [...]*, save the Skövde plant, where the utilisation has [...]* over the last three years and is now [...]*.

(73) With respect to distribution, Scancem has a total of 18 terminals (including the terminals at the three production plants), located around the Swedish coast. Due to the location of Scancem's production plants, the cement is delivered to the terminals by sea (with the exception of the production at the Skövde plant, which is delivered by rail to the terminal located in Gothenburg). The average maximum capacity of each terminal is approximately 15000 t. The largest terminal is located in Stockholm with a maximum capacity of 33200 t. The parties have indicated capacity utilisation in terms of the highest volume filled in a terminal during one year. Using this measure, the capacity utilisation has been stated to be [...]* in most terminals.

(74) More than [a large majority]* of Scancem's cement sales are made in bulk, directly to the customer. The customers normally collect the cement directly from the terminal and transport the cement themselves by road or railway. The remaining sales, which consist of sacked cement, are made via third party distributors, for example specialised construction materials outlets and DIYs. Furthermore, Scancem's sales business (45 people in total employed with sales, marketing, research and development and technical advice to customers) is operated centrally from Stockholm, although there are also sales offices in Gothenburg and Malmö.

(75) According to the information available to the Commission, the Norwegian company Embra is the only company that presently sells imported cement in Sweden (26). Embra, owned by the Norwegian Selvaag group, is involved in cement and lightweight aggregate production and distribution, with activities in the Nordic countries, the Baltic States, Poland and Germany. Embra entered the Swedish market in 1992 and imports cement from a partly-owned plant, AB Akmenes Cementas, in Lithuania (27). According to its Annual Report, Embra sold 58000 of cement in Sweden in 1997. The company operates three terminals in Sweden: two in the southern part along the coast (Halmstad and Åhus) and one inland (Västeras). These terminals are significantly smaller than those operated by Scancem. Embra has made significant losses.

(76) Furthermore, Scancem has three cooperation agreements involving cement with FLS Industries/Aalborg Portland (the sole Danish cement producer). One agreement concerns cooperation with respect to the supply of cement to the Öresund Link bridge between Sweden and Denmark. The second agreement [...]*. The third agreement concerns cooperation for a special patented method for reducing the chrome content in cement.

(iii) Market size and market shares

(77) According to the figures submitted by Skanska, the cement market in Sweden had an annual turnover of around ECU 90 million (1315000 t) in 1997. These figures are generally in line with those submitted by other companies and have consequently been used for the purpose of the present assessment. As opposed to the growth in Norway and Finland (by 22 % and 24 % respectively), cement sales in Sweden decreased by 15 % between 1995 and 1997. According to Skanska the sale of cement is closely related to the level of construction activity and the proportion of concrete in building materials. According to the notification, Skanska foresees that the Swedish cement market will increase slightly by 3 to 5 % over the next years.

(78) According to the notification, Scancem has a market share of [80 to 90 %]* (value and volume) which has been confirmed by the Commission's investigation. Its share has remained fairly stable over the last three years ([...]* in 1996 and [...]* % in 1995). According to figures provided by the parties, Embra's market share is around [...]* of the total market and thus less than one tenth of Scancem's market share. Also, when comparing Embra's distribution system with that of Scancem, it is clear that Scancem's geographic coverage is far more comprehensive. On the basis of the information available to the Commission, the remaining market volume of cement in the Swedish market is assumed to originate from limited imports used for specific projects (28).

(79) In its reply, Skanska stated that Scancem's average sales prices in Sweden (adjusted for inflation) decreased by about [...]* from 1995 to 1997, and that its margin over the same period declined from [...]* (29).

(80) However, the development of these prices and margins are only of relevance for this assessment to the extent that they may provide an indication of market power (or the absence thereof). The fact that the company, in response to a significant decline in market volume and Embra's increased presence on the market, has lowered its prices by [...]* cannot be seen as an indication of the absence of market power. This can be demonstrated by an analysis of Scancem's activities in the three Nordic countries during the 1995 to 1997 period. As stated above, this period saw a significant downturn in Swedish cement sales, whereas sales in Finland and Norway rose by 22 % and 24 % respectively. It is also important to note that, during this period, Scancem's only significant competitor, Embra, was established in all three countries.

(81) The figures on profit margins for the individual countries, submitted in Skanska's reply, indicate, at most (30), that Scancem's profit margins are cyclical. Indeed, the indicated profit margins for Norway and Finland show [...]*. Therefore, Skanska's figures on decreasing profit margins for Sweden between 1995 and 1997 cannot be seen as indicative of Scancem not having market power. Rather, these figures would indicate that the profit margins are closely connected to the degree of capacity utilisation. This is to be expected in the cement industry, which is largely based on economies of scale. Therefore, it must be concluded, firstly, that Scancem's level of profit margins in the Nordic countries have not been significantly restrained by Embra or any other competitive action, and, secondly, that the development of these profit margins is not incompatible with the above conclusion as to Scancem's market power in the Nordic countries, including Sweden. The parties have presented no argument indicating that profit margins in Sweden will not increase again when the business cycle reverses and capacity utilisation is increased.

(82) In conclusion the existing facts indicate that Scancem, despite the varying trends in the business cycles, has been able to maintain its profit margins [...]* in all three Nordic countries. Moreover, these facts do not indicate that competitive constraints, for example by Embra, have had a significant impact on Scancem's pricing policy or profit margins.

(iv) Barriers to entry

(83) According to the notification, the establishment of a cement production plant is estimated to cost around ECU 200 million. For this reason the parties are of the opinion that potential competition is possible only through imports. Imports of cement are, however, dependent on investments in terminals and storage silos and a comprehensive distribution system, including shipments. Normally, one silo would only handle one type of cement at a time and according to Scancem, considerable time (up to several days) is needed to change to another type of cement. Therefore, the best use of capacity is achieved by using one silo for each cement type. In the course of the Commission's investigation it has been indicated from other market participants that the minimum up-front investments for an importer would amount to roughly ECU 2 million for one single 6000 t silo that will be able to handle cost-efficiently about 40000 to 60000 t of cement per year (or two silos of 3000 t, with a cost of ECU 1,2 to 1,7 million each) (31).

(84) Skanska has argued that the costs of entry are not prohibitive, and, consequently, that new entry would be possible. However, the above figures on the costs of entry also have to be contrasted with the fact that Scancem's average capacity at its terminals is approximately 15000 t and that the company for most terminals achieves a capacity utilisation of [...]*. Therefore, even if Skanska's argument were to be accepted, it is unlikely that such entry, if it were to occur, would significantly restrict the market behaviour of Scancem.

(85) The entry into the Swedish market in 1992 by Embra may serve to illustrate the difficulties faced by a cement importer, as well as the minor significance in relation to Scancem's dominant position of such entry. As stated above, Embra has invested in three distribution terminals in Sweden. The investigation indicates that the company, in order to reach its current level of sales, has had to offer significantly lower prices than those of Scancem. Consequently, Embra has made significant losses over recent years. For 1997 the result for the company's Swedish cement business shows a loss of SEK 2720000 on a turnover of SEK 38312000. Moreover, according to its annual reports this strategy only enabled the company to increase its market share in Sweden by one percentage point from 1995 to 1997.

(86) As has been described above Scancem's profit margins have not been significantly affected by Embra's efforts to increase its presence on the market, even though the company has implemented a strategy based on substantial investments and lower cement prices. This is consistent with the result of the investigation, which shows that many customers do not consider Embra to be a feasible alternative to Scancem, but rather as a supplement to their existing supplies from Scancem. It is therefore not only unlikely that further new entry will occur in the near future, but also that any such entry would have the effect of significantly restricting Scancem's market power.

(87) Therefore, the most significant barrier to entry is likely to be the very high and stable market shares of Scancem, and its installed production capacity, which is about twice as high as annual cement consumption. Moreover, as a result of the high proportion of fixed costs in cement production, Scancem has a strong incentive to defend its current market position in a market which is only forecast to grow moderately in the near future. It therefore must be concluded that there is no evidence that potential entry is likely in the near future on a scale that would be sufficient to neutralise Scancem's market power.

(v) Absence of buying power

(88) Scancem has approximately [100 to 200]* cement customers, of which approximately (< 100)* purchase 300 t or more per annum. The main customers are Skanska, Swerock, NCC, Färdig Betong and Strängbetong. These companies are all engaged in one or more building materials markets and are represented in most of Sweden. However, apart from Skanska, none of them has vertical links to a cement supplier.

(89) Skanska is currently the largest single customer and purchases more than Scancem uses internally in its construction materials business. Based on 1997 figures submitted by Scancem, Skanska and Swerock each accounts for [...]* of Scancem's domestic sales; corresponding to [...]* of Scancem's total Swedish sales (including export). NCC is the third largest customer and accounts for less than [...]* of Scancem's domestic sales; corresponding to less than [...]* of Scancem's total sales. The remaining customers, including Färdig Betong and Strängbetong, all account for far less than [...]* of Scancem's domestic sales and less than [...]* of its total sales. A common characteristic for all Scancem's cement customers is that they buy a very high proportion of their total cement requirements from Scancem (up to 100 %) (32). Each customer is therefore significantly more dependent on Scancem than vice versa. Therefore, based on their purchases, the vast majority of these customers have no or insignificant buying power.

(90) The Commission's investigation has shown that the vast majority of Scancem's customers only have one source of supply of cement, namely Scancem. Many customers have stated that they do not consider Embra to be a feasible alternative to Scancem. The relatively few customers who also buy cement from Embra consider those supplies to be a supplement to already existing deliveries from Scancem. Therefore, the majority of customers do not consider that there are any viable alternative sources of supplies of cement in Sweden.

(91) In this context it may also be noted that Scancem's contracts with customers purchasing more than 300 tonnes per annum are negotiated individually, normally for one to three years and often include also joint research and development projects concerning the cement. Consequently, these customers are not only locked into Scancem by the lack of viable alternatives, but also through their existing contracts.

(92) In conclusion, the available facts indicate that not even the largest customers are in a position effectively to constrain the market behaviour of Scancem.

(vi) Conclusion

(93) Scancem's stated market shares, in themselves, create a strong presumption that the company has a dominant position (33).

This is further strengthened by its apparent ability to maintain, largely independently of external market developments, its sales prices and profit margins, even during a period when, as mentioned above, total sales of cement in Sweden decreased by approximately 15 %.

(vii) Strengthening of a dominant position on the Swedish cement market

(94) The notified transaction would create vertical effects between cement and, on the other hand, concrete and other mineral-based construction materials. This relationship stems from the fact that cement (together with aggregates) constitutes the main raw material for the production of the latter products. The downstream effects on concrete and related products will be dealt with below.

(95) As stated above, Scancem had, prior to the notified operation, a dominant position on the Swedish cement market with a market share of [80 to 90 %]*. There are no other producers of cement and the remaining cement sales are made up of imports from countries outside the EEA. The notified operation could further foreclose the possibilities for importers to sell cement into Sweden.

(96) Between 80 and 90 % of all cement is sold to the concrete industry. Following the operation, Skanska would be by far the largest producer of different types of concrete and it must be presumed that Skanska would have no incentive to buy cement from Scancem's competitors. Therefore, this part of the market would effectively be foreclosed for cement importers.

(97) Furthermore, other producers of concrete products would also become less likely to purchase imported cement. The reason for this is that such concrete producers, following the operation, would be dependent on Skanska to a significantly higher degree than they previously were dependent on Scancem. Following the operation, this dependency may, in addition to that created by Scancem's dominant position in cement, come from the combined entity's strength in the fields of concrete and aggregates (the latter being the other main input in concrete production).

(98) Moreover, the fact that Skanska is the largest Swedish construction company, with important activities also in Norway and Finland, means that most concrete producers will have Skanska as one of their main customers. Other concrete producers are significantly small companies than Skanska (in terms of financial strength). None of them are vertically integrated upstream towards cement production, and most are not vertically integrated downstream towards the construction market.

(99) In view of the above, Skanska would be able to employ a number of strategies to foster loyalty among Scancem's cement customers. Indeed, if Skanska were to consider that any particular cement customer acted in a way which was incompatible with Skanska's interests, it could worsen the supply conditions relating to cement, focus its competitive activities in the concrete markets on that company and/or reduce or threaten to reduce its purchases from the company. This variety of measures was previously not available to Scancem.

(100) In this context it should also be mentioned that another negative consequence of the operation is that it creates a new link between Scancem and FLS Industries/Aalborg Portland (the Danish cement production monopoly), through AB Sydsten, which is currently a 50/50 joint venture between FLS and Skanska. Sydsten is a significant producer of ready-mixed concrete in southern Sweden, and could, in the absence of the notified transaction, have provided a means for entry on the Swedish market for FLS Industries.

(101) Based on the above, the Commission considers that Skanska, following the operation, would be in an even better position than Scancem to foreclose the markets from competition from imported cement, and that the vertical relationships created through the operation therefore would lead to the strengthening of a dominant position in the Swedish cement market.

(viii) Norway and Finland

(102) The market structure and size is very similar in Norway and Finland to that described above for Sweden. Scancem's market share is between [80 to 90 %]*, and Embra is the only competitor with a market share over 5 %. Moreover, the barriers to entry are similar to those described above, and customers are in a similarly weak position vis-à-vis Scancem, with no customer accounting for more than [...]* of Scancem's domestic sales.

(103) According to the notification, only Scancem is active as a producer of concrete and other construction materials in Norway and Finland (34). However, subsequent to receiving the statement pursuant to Article 18, Skanska acknowledged that its Finnish subsidiary Tekra OY is active in the production of ready-mixed concrete and precast concrete products. In addition, Skanska has recently, through the acquition of Polar OY, become the largest house-construction company in that country (35). The vertical effects arising from Skanska's position on the construction market would therefore, at least in Finland, be similar to those described above concerning Sweden.

(104) In conclusion, the available information strongly suggests that Scancem has a dominant position for cement also in Finland and Norway. However, for the purposes of this assessment the question whether the notified operation would strengthen this dominant position can be left open, since it does not materially affect the assessment of the operation.

(ix) Overall conclusion on cement

(105) Based on the above, the Commission has concluded that Scancem, prior to the notified operation, had a dominant position for the sale of cement in Sweden, Norway and Finland, and that this position, at least for Sweden, would be strengthened through the notified operation.

(b) Aggregates

(i) Supply structure

(106) According to Skanska, the market for aggregates has declined since 1990 to almost half its previous size. That development mirrors the reduction of activity in the Swedish construction industry over that period and the inception of a recessionary cycle. As a result, there currently exists substantial excess capacity in the market, with many quarries having substantial unutilised reserves of aggregates. As earlier indicated, around half the pits in Sweden are currently in use. This is also reflected in the fact that both Skanska and Scancem operate their business on a relatively low capacity utilisation amounting to around [...]* of maximum capacity. Also, the main competitors appear to have relatively large surplus production capacity.

(107) Skanska and Scancem have 40 (of which three are part-owned) and seven aggregate plants in Sweden, respectively. Skanska has a very strong geographic coverage while Scancem's plants are broadly concentrated on the Eastern coast. On a local level, Skanska is active in 19 out of 21 counties (län) while Scancem is active in eight counties which, however, in combination represent more than half of the total sales of aggregates in Sweden.The activities of Skanska and Scancem overlap in six out of 21 counties.

(108) The largest competitors in Sweden are Ballast/NCC, Swerock and the National Road Administration (Vägverket). Furthermore, there exists a large number of small, local competitors.

(109) With respect to Skanska's aggregates activities it should be noted that six of its plants are operated as joint ventures with different partners: Jordbrokrossen (JM Bygg 50 %). Olundakrossen (Uppsala kommun 50 %), Krossbolaget (Swerock 35 %), Härryda Kross (Fraktkedjan 50 %), and Sydsten (FLS Industries 50 %) which owns half of Forserumsten HB. Furthermore, Skanska and Sabema (part of Scancem) and Färdig Betong each own one third of Strömstadbetong & Co. KB, which produces and sells ready-mixed concrete and aggregates. Finally, Scancem is party to a joint venture with NCC (15 %) and Färdig Betong (40 %) (36).

(ii) Market size and market shares

(110) The Swedish Geological Institute (Sveriges Geologiska Undersökning) annually compiles a report on production volumes in Sweden by county (län) and by local area (kommun). The statistics are based on information gathered by the county administrative boards regarding pit and crushing operations in Sweden. Based on these calculations, the total value of the Swedish market for aggregates is estimated in the notifiction to be between ECU 326 and 370 million. Skanska, Scancem and other major aggregate suppliers have submitted their own estimates for the total volume and value of the Swedish aggregate market. The most recent figures provided by Skanska indicate for 1997 that sales amounted to 62629000 t equivalent to approximately ECU 366 million. These figures are broadly in line with those submitted by the other companies and have consequently been used in the following estimates. Based on the figures in the notification, the total aggregates market declined by almost one third from 1995 to 1997 while the parties' market shares have increased slightly in the same period.

(111) Based on the most recent submissions from the parties, Skanska and Scancem have market shares on a national basis of approximately [10 to 20 %]* and [< 10 %]*, respectively, and thus a combined market share of approximately [20 to 30 %]* (37).

The second largest supplier of aggregates is Ballast/NCC, which enjoys a slightly weaker market position than the combined entity while Swerock and Vägverket are not even half the size. The rest of the market consists of a large number of small and local suppliers.

(112) The combined position of the parties would, however, be much stronger in some areas if market shares were to be measured on a more regional or local level. One competitor has provided a breakdown of the Swedish market into five regions of similar size. According to this competitor, the combined market shares of Skanska and Scancem would be 35 to 40 % in the Stockholm area and 30 to 35 % in western Sweden. In these areas the combined entity would be the market leader. If market shares are measured at an even more local level, figures provided by Skanska and Scancem indicate that their combined market share (measured by value) would exceed 40 % in Stokholm and 25 % in two counties (Västra Götaland and Gävleborg).

(113) The figures provided by Skanska and Scancem indicate that their market shares on a national basis have increased slightly over the last three years despite the general decline in the aggregates market. Furthermore, it should be noted that, according to figures submitted by Skanska, their average sales prices [...]* (38).

It should be noted that Skanska's market shares fell slightly in 1997.

(iii) Conclusion

(114) Despite the fact that Skanska/Scancem would have a combined share of no more than around (20 to 30 %)*, the combined entity would still be the overall market leader on a national basis and enjoy substantial market shares in some local markets. Skanska has submitted that a combined Skanska/Scancem would face material competition from local competitors, who, while they do not account for a material share of the market as a whole, are nevertheless capable of providing strong competition on a local level. While the Commission recognises the existence of a large number of local competitors it should be borne in mind that the aggregates market appears to be undergoing a change in the proportions of gravel and crushed rocks used which may make the competitive conditions more difficult for the small competitors.

(115) As mentioned above, the Swedish Government has introduced legislation imposing a tax in order to limit the use of gravel and consequently encourage the use of crushed rock. Given that the investment costs are considerably higher for the production of crushed rock than for gravel (39), this will favour the large and financially strong suppliers, such as Skanska/Scancem (40).

However, as a number of small suppliers of aggregates currently have significant overcapacity in rock crushing plants, the large companies will only be favoured if smaller suppliers must invest in new rock-crushing capacity.

(116) Skanska states in its reply that aggregates producers face competition from "on-the-line" aggregates (obtained as a result of the construction activity itself) which can either be used by the contractor for its own construction purposes on site or, less frequently, sold on the market. Also, Skanska expects that recycled aggregates will become an important competitive constraint in the future. Skanska has, however, not been in a position to quantify the possible importance of these products and therefore their possible impact is unclear. Moreover, it should be noted that none of the respondents in the Commission's investigation have attached importance to these products.

(117) Most third parties have estimated Skanska/Scancem's combined market share to be roughly the same as indicated above, and have also confirmed that the entity would be considered to be the market leader. The Commission's investigation has however not produced evidence that the notified operation would lead to the creation or strengthening of a dominant position for Skanska/Scancem on the Swedish market for aggregates (whether assessed at national, regional or local level). Nevertheless, it has to be borne in mind that aggregates, together with cement, are the main inputs used in the production of many of the concrete products where the parties would substantially increase their combined market position. The parties' activities in the market for aggregates therefore contribute to demonstrating their overall strength and strong position as a vertically-integrated producer, and support the conclusions in the present assessment regarding the vertical effects of the notified operation on the markets for cement and concrete.

B. CONSTRUCTION MATERIALS

(a) Concrete - overview

(118) The Swedish market shares of Skanska and Scancem (1997, based on value) in the field of concrete are set out in the table below.

<emplacement tableau>

(b) Ready-mixed concrete

(i) Supply structure

(119) According to statistics from the Swedish Ready-Mixed Association (Svenska Fabriksbetongföreningen), the ready-mixed industry has undergone significant concentration since 1989. Over that period the number of ready-mixed producers in Sweden has declined from 87 to 39, a decrease of 55 %. However, according to the same statistics the number of ready-mixed plants only decreased by 6 % over the same period.

(120) Among the remaining 39 ready-mixed producers there are only five companies who achieve sales volumes exceeding 100000 m3/year (about 5 % of the total volume). A clear majority of the remaining producers have sales below 10000 m3/year (< 1 % of the total volume).

(121) Skanska and Scancem have both invested in very large ready-mixed production capacity (in combination they are able to produce close to three times the amount sold in Sweden in 1997). Consequently both companies operate their business on a low capacity utilisation. Although the figures vary, both over time and geographically, Skanska's capacity utilisation is in general around [...]* or lower. On the other hand, its capacity utilisation necessary to reach break-even is only around [...]*. The figures submitted by Skanska indicate that it operates all production units on or above the break-even level. Scancem has submitted similar figures on capacity utilisation and break-even levels. Also the main competitors appear to have significant surplus production capacity, although generally less than Skanska and Scancem. However, in this group of companies, at least some of the production plants are currently operating at a level below break-even.

(122) Skanska and Scancem operate 39 and 20 ready-mixed plants respectively in Sweden. They both have a strong geographic coverage of the Swedish market. On a local level Skanska is active in 13 of 21 counties, which together represent [70 to 80 %]* of total sales of ready-mixed concrete in Sweden. Scancem is active in seven counties. However, since Scancem is particularly strong in the areas surrounding Sweden's three largest cities (Stockholm, Gothenburg and Malmö), Scancem's area of activity represents [50 to 60 %]* of total ready-mixed sales. The activities of the two companies overlap in all local areas.

(123) The other three main producers of ready-mixed concrete in Sweden are NCC, Färdig Betong and Swerock. On a national basis, all three companies have sales significantly smaller than those of Skanska and Scancem. NCC, Sweden's second largest construction company, is a vertically-integrated construction materials company. It, however, has no structural links to any cement producer. Färdig Betong is a family-owned company, whose only activity is the production of ready-mixed concrete and aggregates. Swerock is also a dedicated construction materials company. The future ownership structure in Swerock is currently unknown, as one of its two previous owners has undertaken vis-à-vis the Swedish Competition Authority to divest its interest in the company (41).

(124) Out of Skanska's 39 ready-mixed plants, eight are operated by two joint ventures, Sydsten AB (Skanska 50 % and FLS Industries 50 %) and Strömstadsbetong (Skanska 33 %, Scancem 33 % and Färdig Betong 33 %). In addition, Skanska has concluded a long-term agreement with Färdig Betong, according to which the latter will supply ready-mixed concrete to Skanska in areas where Skanska does not have its own production. Scancem has a joint venture for the production and sales of ready-mixed concrete with Swerock (Essbetong KB). Also the other main ready-mixed producers are involved in various joint venture arrangements with one another.

(ii) Market size and market shares

(125) The Swedish Ready-Mixed Association produces annual statistics on the volumes of ready-mixed sales in Sweden. These figures indicate that the Swedish market has undergone a significant reduction in size since 1989. Indeed, the statistics show that total ready-mixed concrete volumes are down by more than 50 % over the period, although an increase of about 10 % was recorded between 1996 and 1997. According to the notification, the explanation for this is that the ready-mixed market is mature and follows closely the state of the construction market, which has seen a similar decline over the period.

(126) The abovementioned statistics are, however, only available on a national level, and do not indicate the value of the market. However, Skanska, Scancem and the three other largest ready-mixed suppliers have submitted their own estimates for the total volume and value of the Swedish market. The most recent figures provided by Skanska indicate for 1997 that the total volume of the market was 1827000 m3, and that the value was approximately ECU 150 million. These figures are generally in line with those submitted by the other companies, and have consequently been used in the calculations below (42).

(127) As indicated above Skanska and Scancem have national market shares of approximately [20 to 30 %]* and [20 to 30 %]* respectively in the field of ready-mixed concrete (43).

Thus, their combined market share [40 to 50 %]* is about 50 % larger than that of the next competitor (Swerock), and several times larger than the next two competitors (NCC and Färdig Betong).

(128) The combined position of Skanska and Scancem would be even stronger if market shares were to be measured on a more regional or local level. One of the competitors has provided a breakdown of the Swedish market into five regions of similar size. According to this competitor, the combined market shares of Skanska and Scancem would be below 20 % in eastern Sweden, 25 to 30 % in western Sweden, 35 to 40 % in south and north Sweden and > 50 % in the Stockholm area. According to this competitor, Skanska and Scancem would, by far, be the largest ready-mixed supplier in all areas except eastern Sweden (where it would be the second largest).

(129) If market shares are measured at an even more local level, figures provided by Skanska and Scancem indicate that the combined entity's market share (measured by value) would exceed 50 % in five counties (Stockholm, Halland, Gävleborg, Skåne and Norrbotten). Moreover, in four counties the market share would be between 40 and 50 % (Jönköping, Västra Götaland, Uppland and Kopparberg). Finally, in another three counties the market share would be between 20 and 40 %. The areas where the combined entity, following the concentration, would achieve market shares above 50 % correspond to approximately 43 % of the Swedish market. The company would have more than 40 % in areas representing 65 % of Sweden.

(130) Since ready-mixed concrete is normally sold to construction companies on a project basis, a certain volatility in the market shares should be expected, in particular at the local level. Nevertheless, the figures provided by Skanska and Scancem indicate that the combined market shares for the two companies remained relatively stable between 1996 and 1997. Thus on a national level the companies' respective market shares in 1996 were [20 to 30 %]* and (10 to 20 %)* respectively (44). Similarly, on a local level, their combined market share (measured by value) exceeded 50 % in four counties (Stockholm, Gävleborg, Skåne and Norrbotten). Moreover, in five counties the combined market shares would be between 40 and 50 % (Jönköping, Halland, Västra Götaland, Uppland and Västernorrland). Finally, as in 1997, their combined market shares were between 20 and 40 % in another three counties.

(131) It should also be noted that, according to the figures submitted by Scancem, its average sales price (ECU/m3) [...]* between 1995 and 1997, and that this [...]* was made [...]* in total sales volumes over the same period of [...]*. Similarly, Scancem's profit margin (45) [...]* from [...]* in 1995 to [...]* in 1997. In its reply, Skanska submitted that these prices [...]* may be affected by inflation and variations in product and/or customer mix, and that the indicated figures on profit margins are unreliable, since they are based on information in Scancem's annual report. However, Skanska has not provided any evidence that the conclusions drawn from the figures it has previously provided are not correct, and has not provided new figures on which these calculations should be based. Therefore its submission on theoretically distorting factors is of little practical relevance to the assessment of the case.

(iii) Vertical effects

(132) The most significant costs for a ready-mixed producer are cement and aggregates. As mentioned above, Scancem has a dominant position in Sweden for cement. According to figures submitted by Skanska, cement constitutes betwen [20 to 30 %]* of its total costs for production of ready-mixed concrete. Figures submitted by Scancem indicate that cement makes up a smaller proportion of its total production cost for the majority of its production (46).

(133) In comparison, figures submitted by the main competitors indicate that cement constitutes between 35 and 40 % of their total costs.

(134) Although it cannot be excluded that this cost comparison may be affected by extraneous factors, such as capital costs, etc., the indicated figures should be seen as a strong indication of the importance of the cement price for the profitability and competitiveness of a producer of ready-mixed concrete. In this respect, it should be noted that the possibility of purchasing cement from importers (i.e. Embra) does not appear to have improved the cost situation of Skanska's competitors in reaady-mixed concrete. It therefore has to be concluded that there is significant scope, given Scancem's dominant position in cement, for Skanska/Scancem to raise the costs for its competitors on the ready-mixed market.

(135) In its capacity as Sweden's largest construction company, Skanksa is also among the most important customers of its competitors on the ready-mixed market. In 1997 Skanska purchased goods from NCC with a total value exceeding [50 to 60 %]* of the latter's turnover on the ready-mixed market. Similarly, Skanska's total purchases from Färdig Betong and Swerock represented [10 to 20 %]* of those companies' turnover in ready-mixed. Although the percentages are lower for Färdig Betong and Swerock, it must be remembered that these companies are relatively small and only active in the field of construction materials production.

(iv) Contestability

Actual competition

(136) Following the concentration, the main competitors of Skanska and Scancem have significantly smaller market shares than the combined entity, whether assessed at national, regional or local level. Nevertheless, in its reply Skanska has expressed a view that Färdig Betong, NCC and Swerock would be able to constrain its market behaviour after the concentration, and that its links with the two former companies are of limited economic importance.

(137) However, the vertical integration resulting from the operation would mean that Skanska, as the market leader, would be both the dominant supplier of their main raw material, cement, and a very important customer of all ready-mixed producers. Skanska would thus be in a position to affect both the cost level of these competitors, by raising the cement prices, and to affect their levels of sales by reducing or threatening to reduce its purchases from them. This in itself provides an indication that Skanska after the concentration would be in a position to influence the market behaviour of its competitors to the extent that it could act largely independently of them. The fact that some of the production plants run by the largest competitors are not able to reach break-even in current market conditions further indicates that they would be vulnerable to any attempts by Skanska to use its market power against them.

(138) The Commission cannot accept Skanska's submission that the structural links between Skanska and Färdig Betong (Strömstadsbetong, a production and sales joint venture, and the long-term delivery agreement) are of limited economic importance compared to the overall value of the Swedish market. Skanska has compared the value of the respective links to the value of the overall market. However, it has not indicated why, in assessing the competitive position of Färdig Betong, this comparison would provide any meaningful information. The Commission considers that the effects of existing links must be assessed on the basis of their importance to the companies involved. In that respect, it must be concluded that the economic importance for Färdig Betong is significant, as these links with Skanska refer to about [...]* of the former's overall production of ready-mixed concrete. In a market where margins are relatively low, Färding Betong is bound to realise that any action that might threaten the continuation of its links with Skanska could also threaten the company's profitability. The same applies to Swerock, which has a production and sales joint venture with Scancem, which also corresponds to about [...]* of Swerock's total production of ready-mixed concrete.

(139) In conclusion, it is very unlikely that Färdig Betong, NCC or Swerock would be in a position to significantly restrict Skanska if it were to attempt to, for example, raise prices for ready-mixed concrete above the competitive level. On the contrary, given their position, it appears much more likely that they would be inclined to follow such attempts.

Potential competition

(140) Skanska has submitted that the barriers to entry into ready-mixed production are low, given that the investment necessary to start production on a viable scale is relatively low. Even if this were to be accepted, it must however be remembered that there already exists a significant overcapacity on the Swedish market, and that the market is not forecast to grow significantly in the near future. An entrant would therefore have to take significant sales from the existing players in order to establish itself. An entrant would also have to consider that Skanska, given its control over the main raw material, cement, would be in an excellent position to affect its possibilities of making a sufficient return on the investment. Moreover, given the already exiting overcapacity, and the fact that a ready-mixed plant cannot readily be used to produce other goods, any investment in new production capacity would largely be a sunk cost.

(141) Skanska's submission that entry is likely also goes against the facts. As stated above, since 1989 there has been significant exit from the market, which is reflected in the decrease by over 50 % of the number of companies active on the market for ready-mixed concrete. The parties have provided no arguments explaining why new entry would be likely in a market where, over a long period, the trend has been that a large number of, predominantly, smaller producers have left the market.

(142) Finally, Skanska has not indicated the identity of any particular company which it considers to be a likely entrant. In this respect it should be noted that Scancem, in the neighbouring geographic areas, has a market share of [50 to 60 %]* in Finland and about [20 to 30 %]* in Norway (on a regional level it has indicated its market share to be between [40 to 60 %]* in seven out of 10 regions in Norway, and in three out of six regions in Finland (47). In conclusion, there is insufficient evidence to conclude that potential entry is likely in the near future on a scale that would be sufficient to neutralise the market power that Skanska would hold following the notified operation.

Absence of buying power

(143) In total, Skanska has between 8000 and 10000 ready-mixed concrete customers in Sweden. The corresponding figure for Scancem is about half of that. Clearly the vast majority of these customers buy very small amounts and have no, or at least insignificant, buying power.

(144) On a local level, the largest customer of Scancem is in most cases Skanska. There are some local examples where a third party buys more than 10 % of Scancem's or Skanska's local production (NCC and Peab, Sweden's third largest construction company). However, since both NCC and Peab (through its 50 % holding in Swerock) are active in production of ready-mixed concrete, it must be presumed that in the local areas where they purchase from Skanska or Scancem, they have a limited ability to switch supplier. In this respect it should also be mentioned that Skanska, as one of the larger customers of NCC and Peab/Swerock, would be in a position to retaliate effectively if any of these companies were to significantly reduce, or threaten to reduce their ready-mixed purchases from Skanska following the concentration.

(145) Finally, Skanska has submitted that NCC and Peab would have the possibility to build dedicated ready-mixed plants at their construction sites. However, there are only a handful of examples where Skanska or any other consctruction company over the last three years has built such a dedicated plant. This appears to be a feasible alternative only in relation to the very largest construction projects, such as the Öresund Bridge, which require very large amounts of ready-mixed concrete. It is therefore unlikely that such dedicated plants could be relevant for the majority of projects, which do not involve such quantities, and that the potential impact on the overall sales of ready-mixed concrete is limited.

(146) It therefore must be concluded that there is no such buying power on the side of the ready-mixed customers as would be sufficient to neutralise the market power of Skanska following the concentration.

(v) Conclusion

(147) On the basis of the above considerations, the Commission has concluded that the notified concentration would create a dominant position on the Swedish market for ready-mixed concrete (whether assessed at national, regional or local level). This results partly from the horizontal overlaps between Skanska and Scancem, and the fact that their combined market share would be significantly higher than those of all competitors. It also partly results from the vertical effects of the concentration, which mean that the competing ready-mixed producers would be largely dependent on Skanska/Scancem for their supplies of the main raw material, cement, and also for sales of ready-mixed concrete to Skanska in its capacity as Sweden's largest construction company. Finally, it is unlikely that potential competition would play any significant role in the foreseeable future, or that the customers of the combined entity would be in a position to effectively neutralise its ability to act independently on the market.

(c) Dry concrete

(i) Supply structure

(148) The supply-side of the Swedish market for dry concrete is concentrated, with four companies supplying more than 90 % of the market (Skanska, Scancem, Finja Betong and STO Scandinavia (STO). Finja Betong is a relatively small company, whose main activity is the production of dry concrete products, STO is a subsidiary of STO AG of Germany. Scancem is the producer that has the largest number of production plants (12), whereas Skanska operates two plants. Finja operates one plant and [...]* production is partly produced by Skanska on a subcontracting basis, partly imported from Germany.

(149) Skanska has indicated that its total capacity in dry concrete production is almost as large as the total sales volume in Sweden, and that its capacity utilisation is approximately [...]*. Scancem's capacity is larger than that of Skanska, and its utilisation varies significantly between its different plants, but is generally higher than that of Skanska.

(150) As mentioned above, Skanska has a subcontracting agreement with [...]*, according to which Skanska produces [...]* dry concrete products for sale in Sweden. The production is done in accordance with [...]* proprietary recipe. This agreement was concluded in 1993, when the company entered the Swedish market. Skanska also has a subcontracting agreement with one of the smaller Swedish suppliers of dry concrete, [...]*. According to this agreement, [...]* is allocated a specific geographic area where it may sell the products produced by Skanska.

(151) In addition, in its reply, Skanska stated that it also produces dry concrete on a subcontracting basis for three other smaller competitors, [...]*, but has provided no further details concerning these additional contractural arrangements. Also Scancem is producing on a subcontracting basis for two smaller Swedish suppliers. In its reply, Skanska argued that only sales under own brands should be attributed to the parties when calculating market shares. This cannot be accepted. First, no objective reason has been provided why the parties, in the absence of the subcontracting agreements, would not be able to supply the same volumes under their own brands. Secondly, no evidence has been submitted to indicate that the parties' subcontracting partners would be able to supply these volumes without recourse to the production facilities of the parties. As such it must be concluded that the parties produce these volumes because they have the production facilities most suitable for this activity.

(ii) Market size and market shares

(152) According to Skanska there are no statistics available for the total supply of dry concrete in Sweden. However, Skanska has estimated that the market in 1997 had a total value of approximately ECU 14 million, and a volume of 85000 t. The Commission's investigation generally supports these estimates.

(153) Scancem is clearly the largest supplier of dry concrete. Measured in value, its market share was (50 to 60 %)* in 1997. Measured in volume its market share was even higher ([60 to 70 %]*). Skanska's market share was (< 10 %)*

(48), measured by value, and (10 to 20 %)* measured by volume. Thus, the combined market share would be [50 to 60 %]* by value and [70 to 80 %]* by volume.

(154) Although the parties have provided no explanation for this discrepancy between value and volume shares, it appears likely that this may be explained by the fact that the [...]* product is a dry concrete that has a significantly higher value/weight ratio than the parties' products. Consequently, the [...]* product is several times as expensive as those of the parties. Measured in value, [...]* market share in Sweden is only about one fifth of the parties combined market share. The other main competitor, Finja Betong, has a market share which is about one third of that of the parties.

(155) The figures provided by Skanska indicate that the market shares of Scancem and Skanska remained relatively stable between 1995 and 1997. Thus, the companies' respective market shares, measured in value, in 1996 were [50 to 60 %]* and [< 10 %]*. The corresponding figures for 1995 were [50 to 60 %]* and [< 10 %]*.

(156) According to the figures submitted in the notification, Skanska and Scancem increased their average sales price (ECU/t) by [...]* and [...]* respectively between 1995 and 1997. It is likely that the higher increase for Skanska results from its production of STO's more expensive products. Skanska has contested these figures, and has stated that the parties' prices have fallen since 1995. In that respect, it should be noted that the indicated price increases follow from the parties' own value and volume figures indicated in the notification, and that neither party has provided any indication that these figures were incorrect (with the exception of the minor adjustment indicated by Skanska for its 1997 value figure). Skanska's statement that prices have fallen must therefore be considered unsupported by the facts.

(iii) Vertical effects

(157) According to Skanska the cost of cement constitutes about [...]* of variable costs in production of dry concrete, and [...]* of total costs. Scancem has supplied similar figures.

(158) Thus, cement accounts for a significant part of the production costs in dry concrete. Skanska would therefore, through the notified operation, gain an opportunity to affect the profitability and competitiveness of other producers of dry concrete products, by raising their costs for this raw material. Already prior to the operation Skanksa had a significant influence over, and insight into, the production volumes and costs of STO through the existing subcontracting agreement. The same is true for the smaller suppliers, for which Skanska also produces dry concrete. The control over Scancem would provide Skanska with the means to influence these companies, even if they were able to terminate their current subcontracting agreements with Skanska. For all subcontracting partners, the concentration would also mean that there would be one less potential producer of their product.

(159) The abovementioned ability to raise the costs of its competitors also applies in relation to Finja Betong. In addition, Skanska in its capacity as Sweden's largest construction company is a large customer of Finja Betong. In 1997 Skanska's total purchases from the company can be estimated to have equalled about [30 to 40 %]* of Finja Betong's 1997 turnover on the dry concrete market.

(iv) Contestability

Actual competition

(160) Following the concentration, all other suppliers of dry concrete would have significantly smaller market shares than the combined entity. In addition, the vertical integration resulting from the operation means that Skanska, as the market leader, would be both the dominant supplier of their main raw material, cement, a very important customer, and in some cases a subcontractor for significant parts of their production. Skanska would thus be in a position to both affect the cost level of these competitors, for example by raising the cement prices, and to affect their levels of sales by reducing its purchases from them. Given the smaller size of all competitors it must be presumed that they would be vulnerable to any attempts by Skanska to use its market power against them. The structural link between Skanska/Scancem and a number of competitors through various subcontracting agreements further reduces the likelihood of competition. In conclusion, it is very unlikely that any competitor would be in a position to significantly restrict Skanska if it were to attempt, for example, to raise prices for dry concrete above the competitive level. On the contrary, given their position, it appears much more likely that they would be inclined to follow such attempts.

Potential competition

(161) Skanska has submitted that the barriers to entry into dry-concrete production are low, given that the investment necessary to start production on a viable scale is relatively low. Even if this were to be accepted, it must however be remembered that there already exists a significant overcapacity on the Swedish market, not least for Skanska and Scancem. Skanska has not submitted any studies indicating that the market is likely to grow significantly in the near future. An entrant would therefore have to take significant sales from the existing players in order to establish itself. The very high and stable market share of the parties would in itself represent a major barrier to entry. An entrant would also have to consider that Skanska, given its control over the main raw material, cement, would be in an excellent position to affect its possibilities of making a sufficient return on the investment. Finally Skanska/Scancem have significant market shares on a number of neighbouring product markets (e.g. ready-mixed concrete, mortar and paint plaster and floor-levelling products), as well as on neighbouring geographic markets (Norway and Finland), which reduces the likelihood of entry from companies active on any of these neighbouring markets. In conclusion, there is insufficient evidence to conclude that potential entry is likely in the near future on a scale that would be sufficient to neutralise the market power that Skanska would hold following the notified operation.

Absence of buying power

(162) In total, Skanska has 100 dry concrete customers in Sweden. Scancem has about 800 customers, and has indicated that it has no single customer which buys more than [...]* of its yearly sales.

(163) Skanska has submitted that certain Scancem customers belong to "buying groups" who purchase dry concrete together. The largest such "buying group" is stated to purchase [...]* of Scancem's sales. However, it has provided no information on how Scancem's negotiations and contracts with those buying groups are concluded, nor has it explained why Scancem's accounting system indicates each member of these groups individually, rather than as one group. Skanska has indicated that large DIY retailers, such as Beijer, would have significant purchasing power. This, however, appears very doubtful since the company buys less than [...]* of Skanska's sales and less than [...]* of Scancem's sales. In fact, apart from its subcontracting customers, no customer of Skanska bought more than [...]* of its total sales in 1997.

(164) It therefore must be concluded that there is no such buying power on the side of the dry concrete customers, as would be sufficient to neutralise the market power of Skanska following the concentration.

(iv) Conclusion

(165) Based on the above, the Commission has concluded that the notified concentration would create a dominant position in Sweden for the sale of dry concrete. This results partly from the horizontal overlaps between Skanska and Scancem, and the fact that their combined market share would be significantly higher than those of all existing competitors. It also partly results from the vertical effects of the concentration, which mean that the competing dry concrete producers would largely be dependent on the combined entity for their supplies of the main raw material, cement, and also for sales of dry concrete to Skanska in its capacity as Sweden's largest construction company. Prior to the concentration, Skanska had a relatively low market share. Nevertheless, the concentration would have the effect not only of removing Skanska as a competitor of Scancem, but also, through Skanska's link to STO and its other subcontracting partners, of lessening the likelihood of competition from the latter companies significantly. Finally, it is unlikely that potential competitors would play any significant role in the forseeable future, or that the customers of the combined entity would be in a position to effectively neutralise its ability to act independently on the market.

(d) Precast concrete products

(i) Supply structure

(166) As mentioned earlier, Skanska has submitted that precast concrete products should be divided into house products and products used in and on earth, since the former constitute a homogeneous range of precast products sharing common production equipment and know-how, whereas in and on earth products require different and more specialised technology and equipment.

(167) The suggested division follows that made in the statistics of the National Association for Concrete Products (Betongvaruindustrins Riksförbund), and appears to be supported also by competing suppliers.

(168) Scancem is mainly active in house products, which, according to Skanska is the relevant market for assessment of the notified operation. Nevertheless, this section will indicate the position of the parties on the overall market for precast concrete products as well as in the house product segment and the in and on earth segment, since, as stated above, the question of which definition to use is not decisive for the assessment.

(169) Already prior to the notified transaction the supply-side of the Swedish market for precast concrete products was heavily concentrated, with three companies supplying 55 to 75 % of the market (Skanska, Scancem and Strängbetong, a relatively small company whose only activity is the production of precast concrete products). Skanska is the producer which has the largest number of production plants (18), whereas Scancem operates six such plants and Strängbetong seven.

(170) Skanska has indicated that its capacity utilisation for the last three years has fluctuated between [...]*. Scancem and Strängbetong have also provided similar indications.

(171) Scancem has created a production and sales joint venture with Strängbetong in Estonia (Swetrak AS). This company is currently expanding its activity also into Lithuania.

(172) Apart from the three largest producers, the remaining part of the supply side is fragmented, and consists mainly of companies whose production is dedicated to one or a small number of products (more often in the in and on earth segment). The majority of these suppliers operate only one production plant. One of the smaller suppliers on the market is Swerock. This company is involved in a 50/50 production joint venture with Scancem (Gottåsa AB), and also in a 50/50 R & D joint venture with Skanska (AB G-stöd).

(ii) Market size and market shares

(173) According to the parties, precast concrete products may, depending on the product type, be measured in volume, weight or length. There are therefore no available figures on the total volume of the market. However, the National Association for Concrete Products publishes annual data on the total value of the market. Skanska has used these statistics to calculate the total value of the market (49).

(174) The overall Swedish market had a total value of ECU 202 million in 1997. Skanska had a market share of [20 to 30 %]*. Scancem's market share was [10 to 20 %]*. Thus, the combined market share of the parties was [40 to 50 %]*. Both Skanska and Scancem have increased their market share over the last three years. For Skanska the increase represents more than [...]* of its 1995 sales. For Scancem the increase is significantly bigger.

(175) In 1997 the largest competitor, Strängbetong, had a market share on the overall market which was similar to that of Scancem. It is, however, noteworthy that the company lost a significant part of its market share between 1996 and 1997. Other competitors, including imports, all have market shares below 5 %.

(176) The house products segment represents approximately half the overall market and had a total value in 1997 of ECU 106 million. Skanska's share of this segment was [20 to 30 %]*. Since Scancem's turnover stems largely from house products, its share of this segment was [20 to 30 %]*. Thus, the combined market share of the parties was [40 to 50 %]*. Whereas Skanska has lost [...]* of its 1995 market share, Scancem has more than [...]* its share in the same period. Therefore, the combined market share of the parties has increased by [...]* since 1995.

(177) Like Scancem, Strängbetong has a higher share in the house products segment than on the overall market, and its market share for 1997 was similar to that of the parties. However, in this segment too Strängbetong has lost a significant part of its market share, even if its loss in house products was less pronounced than on the overall market.

(178) Finally, Skanska has submitted that in 1997 it had a market share of [20 to 30 %]* of the in and on earth segment (although its share for specific products, such as wells and sedimentation tanks, concrete pipes and retaining walls is stated to be between ([40 to 60 %]*). Scancem's share of the in and on earth segment was [< 10 %]*. In this segment, the parties' market shares have developed in the opposite way, as described above for house products. Since 1995 Skanska's market share has increased by [...]*, whereas Scancem's share has decreased by almost [...]*.

(iii) Vertical effects

(179) Apart from being a necessary ingredient, cement is also one of the most important cost items in production of precast concrete. According to figures provided by Skanska, cement is the single largest variable cost and accounts for approximately [...]* of total costs (both in relation to the house products segment and the overall market). The fact that cement accounts for a lower proportion of total costs than in, for example, ready-mixed production is due, inter alia, to the fact that precast concrete products are significantly more labour-intensive, include other materials, such as supporting steel constructions, and therefore include a value added element.

(180) The fact that cement accounts for roughly [...]* of the total production costs nevertheless demonstrates its importance for the profitability and competitiveness of a producer of precast concrete products. Given, in particular, Strängbetong's declining market position, it has to be concluded that there is significant scope, in view of Scancem's dominant position in cement, for the combined entity to raise the costs for its main competitor.

(181) In its capacity as Sweden's largest construciton company Skanska has recently become one of the most important customers of Strängbetong. Until 1997 Skanska purchased relatively small amounts from Strängbetong. However, during the first seven months of 1998 Skanska's total purchases from the company represented about [...]* of Strängbetong's 1997 turnover on the market for precast concrete products (and even more of its turnover in the house products segment).

(iv) Contestability

Actual competition

(182) The only significant competitor on the Swedish market for precast concrete products, whether assessed as an overall market, or on the basis of the house products segment, is Strängbetong. This company, however, has a market share which is less than half the combined market share of Skanska and Scancem. Other competitors are much smaller. In addition to its smaller and declining market share, Strängbetong would also continue to be dependent on Skanska/Scancem for its supply of cement. The fact that Skanska at this critical point in time has become one of its most important customers will further reduce Strängbetong's incentives to provide active competition. The company is bound to realise that Skanska could employ a variety of means to use its market power in order to make it align with Skanska's market behaviour. Consequently, Strängbetong would not be in a position to significantly restrict Skanska if it were to attempt, for example, to raise prices for precast concrete products above the competitive level. On the contrary, given its position, it would have a strong incentive to follow such attempts.

Potential competition

(183) Skanska has submitted that the barriers to entry in precast concrete production are low, given that the investment necessary to start production is relatively low. As an example of this, Skanska has indicated that two companies, Ulricehamn Betong and Kynnisrud AB, started up production of house products in 1990 and 1997 respectively. Skanska has argued that this clearly shows that entry is possible.

(184) However, for the purposes of this assessment, the relevant question is not only whether new entry is possible, but also whether it is likely to be on a scale sufficient to restrict Skanska from behaving largely independently of its competitors following the concentration. In this respect, it should be noted that none of the indicated new entrants have been able to take a substantial part of the market, and that the limited positions that they have achieved have been taken from other small producers, or possibly from Strängbetong. As mentioned above, Skanska and Scancem have been able to increase their market shares since 1995. In addition, there is no reason why these companies, or any other smaller producers, would not be subject to the same competitive restrictions as described above concerning Strängbetong.

(185) Skanska has also proposed that it would also be subject to potential competition from companies in neighbouring areas: Spenncom (Denmark) and Spenn Gruppen (Norway) (50). However, according to Skanska, neither of these companies have made any deliveries into Sweden in the last three years. In addition, the Norwegian company is a cement customer of Scancem, and, at least for the Danish company, Skanska is one of the larger customers. Consequently, both companies are subject to restrictions similar to those described regarding Strängbetong. In assessing the likelihood of potential entry, it is important to note that both Skanska and Scancem have significant spare capacity, and that the market is not forecast to grow significantly in the near future. In conclusion, there is insufficient evidence to conclude that potential entry is likely in the near future on a scale that would be sufficient to neutralise the market power that Skanska would hold following the notified operation.

Absence of buying power

(186) In total, Skanska has [ > 3000]* customers for precast concrete products in Sweden, of which [< 500]* are precast house products customers. The corresponding figure for Scancem is more than half of Skanska's total. Clearly, the vast majority of these customers buy very small amounts and have no, or at least insignificant buying power. In 1997 no third party bought more than [...]* of Skanksa's production (overall or in house products). Scancem had one customer who bought more than that amount. Based on the same figures no customer would, following the notified transaction, buy more than [...]* of the combined production. It therefore cannot be concluded that there would be such buying power on the side of the customers, as would be sufficient to neutralise the market power of Skanska following the concentration.

(v) Conclusion

(187) Based on the above, the Commission has concluded that the notified concentration would create a dominant position in Sweden for the sale of precast concrete products, whether assessed on the overall market for such products, or on the basis of precast concrete house products. This results partly from the horizontal overlaps between Skanska and Scancem, and the fact that their combined market share would be approximately double that of the only significant competitor on the market. It also partly results from the vertical effects of the concentration, which mean that competing producers would be largely dependent on Skanska for supplies of the main raw material, cement, and also for sales of precast concrete to Skanska in its capacity as Sweden's largest construction company. Finally, it appears very unlikely that potential competition or purchasing power would be able to effectively neutralise the market power of the combined entity.

(e) Other construction materials

(188) In addition to the activities described above in the field of concrete, Skanska and, in particular, Scancem are active in the Swedish market as suppliers of a number of other construction materials.

(189) In the notification Skanska submitted that both companies were active on the market for facing materials. In this area, the parties are active primarily as suppliers of bricks, mortar, concrete rendering and precast wall elements. In addition to these product types, Skanska also included products such as wood, glass and tin plates in the overall market for facing materials. Skanska has not been able to provide any evidence regarding substitutability between this diverse group of products. Indeed, it has not even been able to provide information on the total value of the proposed market. Since it appears doubtful that the proposed market can be seen as the relevant market from the point of view of competition policy, Skanska's estimate that it has a share of [10 to 20 %]* of all facing materials, and that Scancem has a share of [30 to 40 %]*, appears to be of little value. However, since the main overlap between the parties has been covered above under the various concrete markets, it is not necessary for the purposes of the present assessment to pursue this matter further.

(190) Scancem is active in a number of construction materials where Skanska has no activities. Scancem's market share in Sweden for bricks was [30 to 40 %]* (51) in 1997 and for plaster boards it was [60 to 70 %]*. Moreover, in paint plaster, the company had a market share of [70 to 80 %]*. Finally, in floor-levelling products, Scancem's market share is stated in the notification as being 15 to 30 %, depending on whether products used for new floors are separated from products used for repair works. In reaching these figures for levelling products, the parties have included an estimated value attributed to manual techniques, such as grinding. If such alternative techniques are excluded, information from third parties indicates that Scancem's market share in floor-levelling products is between 70 and 85 %.

(191) In the absence of any significant overlap between the activities in the abovementioned areas the Commission's investigation does not indicate that the notified operation would lead to the creation or strengthening of a dominant position. Nevertheless, the pre-existing position of Scancem in the mentioned areas would provide additional leverage against customers on the concrete markets, since most of these customers will also use bricks, plaster board, paint plaster and/or floor levelling products in their construction activities.

C. THE CONSTRUCTION MARKETS

(192) Skanska has submitted that the Swedish construction market in 1997 had a total value of ECU 21370 million, of which the house construction market accounted for 72 % and the infrastructure construction market for the remaining 28 %.

(193) The Swedish Competition Authority recently made an in-depth investigation of the Swedish construction markets (52). According to its decision, the market situation concerning small construction projects (house and infrastructure) did not raise any competitive concerns. Projects of this size require limited resources, and are less demanding in terms of, for example, technical know-how and financial resources. There are therefore, in addition to the large, national construction companies, a large number of regional or niche companies who are able to compete effectively on these markets. The Commission's investigation has not resulted in any findings that would contradict this conclusion.

(194) However, it follows from the abovementioned decision that the market for large house construction projects is significantly more concentrated, with between 80 and 90 % of all projects attributable to only three companies. On this market Skanska and NCC each have market shares of approximately 35 %, although Skanska in total is a larger company (53).

(195) The situation is similar on the market for large infrastructure projects, where four companies supply about 85 % of the total turnover. Skanska is the largest supplier, with a market share of approximately 30 %. The market shares of NCC and Peab are 25 % and 10 % respectively. Finally, the National Road Administration (Vägverket), whose construction activities have been separated from its administrative tasks, has a market share of 20 %.

(196) In the abovementioned decision, the Swedish Competition Authority concluded that, on the markets for large house and infrastructure projects, the market characteristics indicated that Skanska and NCC would hold a collective dominant position. The Authority based this conclusion on the similar size and structure of the two companies, as well as on the existence of several links through jointly-owned companies. However, the Authority subsequently removed this objection, following a commitment by NCC to terminate all existing links with Skanska.

(197) Most third parties have estimated Skanska's market share to be within the same range as indicated above, and have also confirmed that the company is slightly stronger on the infrastructure market than on the house construction market. The Commission's investigation has however not produced evidence that the notified operation would lead to the creation or strengthening of a dominant position for Skanska on these construction markets. Nevertheless, the abovementioned facts clearly demonstrate the concentrated nature of the Swedish markets for, in particular, large construction projects, as well as the powerful position of Skanska on these markets. As such, it lends support to the above conclusions regarding the vertical effects of the notified operation on the markets for cement and concrete.

VII. OVERALL ASSESSMENT

(198) As has been described above, the notified transaction would create a dominant position on the Swedish markets for ready-mixed concrete, dry concrete and precast concrete products. Furthermore, it would strengthen Scancem's dominant position on the Swedish cement market. These negative effects would follow partly from the superior market power of the combined entity, as evidenced by its high market shares in relation to its smaller and weaker competitors. However, these effects would be substantially reinforced by the vertical effects, resulting from the combined activities of Skanska and Scancem on the raw material markets (cement and aggregates), the construction materials markets (concrete and other construction products) and, finally, on the construction market itself. The result of the operation is therefore to create a combined entity with strong or dominant positions covering the whole value-chain of the construction sector. Seen in contrast with the fact that most of Skanska's competitors on the various levels are not at all vertically integrated, this strengthens the overall dominance of the company, and further reduces the possibility that its competitors would be able to engage in effective competition with Skanska. The combined entity's wide-ranging activities would also serve as a significant barrier to entry on all levels of its activities.

(199) For the above reasons, the Commission has come to the conclusion that the notified concentration would be incompatible with the common market and the functioning of the EEA agreement, since it would create or strengthen a dominant position in the Swedish markets for cement, ready-mixed concrete, dry concrete and precast concrete products, as a result of which effective competition would be significantly impeded in the common market within the meaning of Article 2 (3) of the Merger Regulation.

VIII. UNDERTAKINGS SUBMITTED BY SKANSKA

(200) In order to resolve the competitive concerns raised by the Commission, Skanska, in a letter of 12 October 1998 (54), offered to enter into the folowing undertakings:

(a) Cement restructuring

(201) Skanska will ensure that Scancem, within a period of [...]* months [...]* of the Commission's decision, divests, as a single undertaking, all assets used in the production, marketing, sale and distribution of cement in Finland to a viable purchaser, possessing the financial resources and proven expertise to enable it to maintain and develop the business as an active competitive force. Moreover, the purchaser will be independent of and unconnected to Skanska and Scancem.

(b) Scancem share disposal

(202) Within a further period of [...]* months [...]* of the Commission's decision, Skanska will divest its entire shareholding in Scancem to a purchaser who is independent of and unconnected to Skanska. This purchaser will be independent of and unconnected to the purchaser of the assets included in the cement restructuring undertaking.

(c) Mechanisms for the undertakings

(203) Pending the divestiture, Scancem will continue to be managed as an entity separate and distinct from Skanska. In particular, none of Scancem's activities in cement, ready-mixed concrete, dry concrete or precast concrete products will be sold or otherwise transferred to Skanska or entities which are dependent on or connected to Skanska.

(204) Skanska will, within two weeks of the Commission's decision, propose for its approval the name of an investment bank or similar institution unconnected to Skanska to act for it as a trustee (to be paid by Skanska). The trustee, who will be appointed on an irrevocable basis, will have a mandate:

(i) to monitor Skanska's implementation of, and compliance with, the undertakings;

(ii) to report to the Commission, on a quarterly basis (or at the Commission's request), on:

(a) relevant developments in Skanska's negotiations with third parties in relation to the cement restructuring and the Scancem share disposal,

(b) the time-frame within which an agreement with a third party is expected,

(c) Skanska's compliance with the provision requiring Scancem to be managed as an entity separate and distinct from Skanska (the hold separate provision);

(iii) to provide the Commission with a report containing sufficient information for it to assess whether a prospective purchaser satisfies the purchaser standards indicated above, and, in relation to the cement restructuring, whether the proposed transaction includes all relevant assets;

(iv) to supervise Skanska's voting in relation to all its Scancem shares pursuant to the following rules:

A. In the period allowed for the cement restructuring, Skanska will be entitled to use its full voting rights to elect the board of Scancem (which will include at least one member elected by shareholders, but not nominated by Skanska, and not connected to Skanska in any respect), and on matters reasonably necessary to implement this undertaking.

B. Skanska will not, except as provided in A, vote in relation to its Scancem shares in excess of a number of votes equal to the number of votes held by the second largest shareholder (parity), and will instruct its board members not to exercise their board votes in excess of parity in support of any resolution which the trustee (after consultation with the Commission) determines would cause:

(a) a material change in the business direction of Scancem; or

(b) any breach by Skanska of the terms and conditions of these undertakings.

C. Skanska may ask the Commission to review a determination by the trustee that voting in excess of parity would be inconsistent with point B.

D. At the end of the period allowed for the cement restructuring, Skanska will ensure that an extraordinary general meeting of shareholders is called for the purpose of electing board members. Moreover, from the end of that period Skanska will not vote in relation to its shares in excess of parity.

(205) In order to comply with the undertaking indicated under (iv), Skanska will provide the trustee and the Commission with relevant information pertaining to any resolution proposed for a general meeting of shareholders, demonstrating that it will not be inconsistent with Skanska's undertakings, and will provide the trustee with information about action to be taken by the board, in order to enable the trustee to determine if such action would be inconsistent with Skanska's undertakings. Skanska will provide such information at least 10 days prior to the scheduled vote. If the trustee (or the Commission in the case of point C) determines that any such vote or action is inconsistent with Skanska's undertakings, Skanska will not vote in excess of parity.

(206) Skanska will only continue to negotiate with a prospective purchaser if the Commission does not, within [...]* of receipt of a report from the trustee, indicate in writing why it does not consider that prospective purchaser to satisfy the purchaser standards indicated above, and/or that the proposed cement restructuring does not include all relevant assests.

(207) Skanska reserves its right, where more than one purchaser satisfies the purchaser standards, to select the purchaser of its choice. It also reserves its right to ask the Commission to approve an amendment to the undertakings set out above, which would be substantially as favourable for competition.

IX. ASSESSMENT OF THE UNDERTAKINGS

(208) Through its submission of the proposed cement restructuring and the share disposal as a package, Skanska has proposed to resolve the concerns relating to the notified transaction as well as that relating to the transaction of 2 October 1995, which the Commission is investigating with a view to establishing whether it gave rise to a notifiable concentration and with regard to its effects on competition.

(209) The cement restructuring relates primarily to Scancem's cement operations in Finland. In terms of the company's total production and sales for the Nordic area, the Finnish activities constitute roughly one third. They also include two of the company's seven cement production plants, and a comprehensive network of cement terminals. It is notable that Finland, in comparison with Sweden and Norway, has the lowest cement prices, and that the Finnish cement production facilitites, at least for large parts of Sweden, are within an equal (or even shorter) distance by sea as the two Scancem plants on Gotland and Öland.

(210) The cement restructuring will therefore contribute to improving competitive conditions for the sale of cement in the Nordic area. It will mean that cement purchasers in Finland, Sweden and Norway will have an additional independent source of supply, where cement of the same quality can be obtained. The undertaking will therefore also indirectly improve the competitiveness of the downstream markets for ready-mixed concrete, dry concrete and precast concrete products, where producers who are competing with Scancem and Skanska would be able to buy their main raw material from a source independent of those companies.

(211) While the cement restructuring does not directly address the competition concerns arising from the 1997 transaction, that undertaking resolves the Commission's doubts with respect to the 1995 transaction and would therefore, in combination with the Scancem share disposal, obviate the need to pursue that investigation further, provided that the undertakings are fully implemented. After having heard Scancem and Aker, the Commission has reached the conclusion that Skanska will be able to implement the proposed undertakings.

(212) However, this decision only concerns the 1997 transaction, under which Skanska increased its voting rights in Scancem from 33,3 % to 48,06 %. The Scancem share disposal would remove the Commission's concerns arising from the 1997 transaction relating to the Swedish markets for cement, ready-mixed concrete, dry concrete and precast concrete products, since it would retain Scancem as a legal entity separate from Skanska, and thereby remove the horizontal overlaps indicated above, as well as the vertical relationships between the two companies.

(213) The time limits proposed by Skanska cannot be considered as unreasonable taking into account the facts that the cement restructuring will require significant changes to the existing structure of the company and that Skanska should be allowed to withdraw from Scancem in the manner that it has proposed. In this respect it should be noted, first, that not to allow Skanska to remove the concerns relating to the transaction of 2 October 1995, prior to the disposal of the Scancem shares, would render the sales of its Scancem shares very difficult, and, secondly, that Skanska, following its divestiture of the Scancem shares, would largely be in a position similar to that described above for other customers of Scancem. The fact that Skanska, for the purpose of implementing its undertakings, will need to exercise a certain degree of control over Scancem for a limited period cannot, in view of the indicated oversight by the trustee and the overall beneficial effects of the undertaking, be considered to give rise to concerns from a competition point of view. The Commission therefore considers that the proposed undertakings provide a proportionate and proper remedy in relation to the notified operation, as well as to the concerns in relation to the transaction of 2 October 1995.

(214) As for the remaining mechanisms for the implementation of the undertakings, the Commission considers that obligations should be imposed on Skanska to ensure the effectiveness of the undertakings. These obligations concern the method whereby the Commission can approve proposed purchasers in relation to the Scancem share disposal. In order to deal with these matters in a timely fashion, and in order to reduce any hardship for Skanska, the method for verifying the identity of proposed purchasers should be based on a procedure of non-opposition by the Commission as indicated in point 215.

(215) Whenever Skanska wishes the Commission to approve a proposed purchaser, Skanska (or the trustee) must provide the Commission with sufficient information to show that the purchaser criteria indicated in its undertaking are satisfied. If the Commission does not, within [...]* of the submission of the request, object to the sale or require that further information be submitted, the sale shall be free to proceed. If the Commission has to request additional information, the receipt of such information shall constitute the starting point for the [...]* period.

(216) Furthermore, Skanska has stated that it will propose within two weeks of this decision, for the Commission's approval, an investment bank or similar institution unconnected to Skanska to act for it as a trustee. If the Commission has not objected thereto, within two weeks of having received all relevant details, the appointment may proceed.

(217) Subject to the above-indicated obligations, the Commission considers that the undertakings proposed by Skanska would remove the competitive concerns in relation to the notified concentration, as well as those relating to the transaction of 2 October 1995.

X. FINAL CONCLUSION

(218) The notified concentration relating to Skanska's increase of its voting rights in Scancem from 33,3 % to 48,06 % and thus the acquisition of sole control over Scancem should, on the condition that the abovementioned undertaking and obligations are adhered to, in so far as they relate to that increase in voting rights, be declared compatible with the common market and the functioning of the EEA Agreement.

HAS ADOPTED THIS DECISION:

Article 1

Subject to the disposal by Skanska of its shares in Scancem so as to reduce its voting rights to 33,3 % of the total voting rights in Scancem in accordance with the terms contained in Skanska's letter to the Commission of 12 October 1998, as clarified by its letters of 3 November 1998 and 9 November 1998, and compliance with the obligations set out in Articles 2 and 3, the concentration notified by Skanska AB (publ), on 15 June 1998, relating to its acquisition of sole control over Scancem AB (publ), is declared compatible with the common market and the functioning of the EEA Agreement.

Article 2

In relation to the disposal of shares as set out in Article 1, the purchaser must be independent of and unconnected to Skanska. The Commission must be in a position to assess the choice of purchaser and evidence sufficient to show that those criteria are fulfilled must be produced by Skanska to the Commission before the sale is made. If the Commission does not, within [...]* of the submission of the request for approval, either formally indicate its disagreement with the choice of the purchaser or require that further evidence be provided, the sale to the chosen purchaser may proceed. If the Commission requests additional evidence, the receipt of that information shall constitute the starting point for the [...]* period.

Article 3

Within two weeks of the adoption of this Decision, Skanska shall propose to the Commission the name and terms of appointment of an investment bank or similar institution, which is unconnected to Skanska, and which should fulfil the duties described in point 204 of the preamble to this Decision, in so far as they relate to the disposal of shares as set out in Article 1. If the Commission has not objected thereto within two weeks of having received all relevant details, the appointment may proceed.

Article 4

This Decision is addressed to: Skanska AB (publ) S - 18225 Danderyd.

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

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

(3) OJ C 201 16.7.1999.

(4) OJ L 61, 2.3.1998, p. 1.

(5) The Aker group is one of the largest industrial groups in Norway with activities in cement and building materials (through Scancem) and offshore oil and gas engineering.

(6) Prior to the transaction, Skanska and Partek each owned 25 % of the voting rights in Euroc and were thus the main shareholders.

(7) Commission notice on the concept of a concentration under Council Regulation (EEC) No 4064-89 on the control of concentrations between undertakings (OJ C 66, 2.3.1998, p. 5).

(8) For example, Commission decisions in Cases No IV/M.343 - Société Générale de Belgique/Générale de Banque of 3 August 1993 and No IV/M.754 - Anglo American Corporation/Lonrho of 23 April 1997.

(9) The total number of votes for Skanska is stated as the number held after all acquisitions in the period 9 to 14 October 1997. The column "votes cast/ (adjusted)" indicates, first, the actual number of votes cast at the respective meetings, and, secondly the number of votes that would have been cast if Skanska and Aker had used all their present shares in those meetings. For the group "other shareholders", the actual participation has been used for the adjustment (1996: 39,4 %, and 1997: 37,2 %). For 1998, actual figures have been used, but adjusted to show the situation as if Skanska had voted in respect of its total holding; due to this proceeding the company exercised only part of its voting rights.

(10) The table above shows a decreasing trend in the participation at the shareholders' meetings in the group "other shareholders", both in actual numbers (1996: 13,1 % of all votes cast, 1997: 12,4 % and 1998: 3,3 %) and as a percentage of the votes held by shareholders in this group (1996: 39,4 %, 1997: 37,2 % and 1998: 30,8 %).

(11) To this should be added the turnover of JM Byggnads och Fastighets AB (ECU 461 million). Skanska controls 57 % of the votes in JM, which is one of Sweden's five largest construction companies.

(12) See Commission Decision of 30 November 1994, Cases IV/33.126 and 33.322 - Cement, Case No IV/M.460 - Holdercim/Cedest and Case No IV/M.1030 - Lafarge/Redland.

(13) See Commission Decision of 30 November 1994, Cases IV/33.126 and 33.322 - Cement.

(14) Case IV/M.460, Holdercim/Cedest.

(15) See Case IV/M.460, Holdercim/Cedest.

(16) By way of illustration, in 1995 there were around 5000 pits in Sweden out of which 3785 were gravel pits and 487 were crushed rock pits. Around 2700 of these pits were in use in 1995. From the period 1984 to 1995 the share of gravel decreased from 82 % to 51 % of the total aggregates production while the share of crushed rock increased from 13 % to 37 %. The objective is to lower the share of gravel to 30 % before 2000, see the Swedish Competition Authority's decision in Case No 292-97 - NCC/Siab, 22 July 1997.

(17) See Commission Decision in Cases No IV/M.460 - Holdercim/Cedest and No IV/M.1030 Lafarge/Redland.

(18) See also the Swedish Competition Authority's decision in Case No 292-97 NCC/Siab, 22 July 1997.

(19) See Commission Decision of 30 November 1994, Cases IV/33.126 and 33.322 - Cement.

(20) Imports into these countries emanate from the former eastern block whereas Scancem's exports are made primarily to the USA and Africa. Scancem imports white cement from Denmark. Independent imports from other Member States are less than 1 % of Swedish sales.

(21) Case No IV/M.460 - Holdercim/Cedest and Case No IV/M.1030 - Lafarge/Redland.

(22) Prior to the abovementioned 1995 merger between Euroc and Aker's cement business, the Swedish and Finnish cement activities were held by Euroc, the Norwegian cement activities by Aker and the United Kingdom and other international activities by Scancem JV, a joint venture between the two companies.

(23) The parties have throughout the procedure supplied revised figures, e.g., relating to their turnover. Unless otherwise stated, all figures indicated below represent the latest submitted figures.

(24) 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.

(25) Scancem considers that it is only the production capacity of clinker that can be measured with any reasonable degree of accuracy.

(26) The parties submitted in the notification that Nordic Industrial Cement (NIC) is one of their main competitors in Sweden. This company was, however, acquired by Scancem in March 1998, and has been wound up. The parties have also submitted that Cem-Ox Import & Export AB is in the process of establishing a cement terminal at Oxelösund in Sweden. They, however, have been unable to verify whether it has actually started any business activities.

(27) In 1997 Scancem [...]*.

(28) Skanska has indicated that the Öresund Link contractors (where Skanska is involved) have imported cement from Spain, in connection with the construction of the bridge between Sweden and Denmark. However, as this import related to precast concrete segements, it is unclear why it would be included in official import statistics as cement imports.

(29) The profit margins in Finland have been indicated as [...]* in 1995 and [...]* in 1997. For Norway, the corresponding figures have been indicated as [...]* and [...]*.

(30) The company has not provided any indication as to how these figures have been calculated.

(31) To these start-up costs it would be necessary to add costs for arranging transport from the cement production plant to the terminals (for which Scancem owns three purpose built vessels), and costs for product approval costs, working capital, staff, marketing costs, etc.

(32) In its Reply, Skanska has submitted examples where, prior to the establishment of Scancem in 1995, it sought quotes from other European cement producers or traders. However, as none of these contacts appears to have resulted in it buying any cement from these sources, it is unclear how this, in the view of Skanska, is relevant to the present assessment.

(33) It should be noted that the Stockholm City Court (Stockholms Tingsrätt), in a judgment of 19 March 1998, stated that Scancem has a dominant position on the Nordic cement market (Case No T 8-1002-97, p. 17).

(34) Scancem's market share in Finland is indicated as [50 to 60 %]* for ready-mixed concrete and about [20 to 30 %]* for precast concrete products.

(35) Skanska press-release, 3 August 1998.

(36) It should be noted that in accordance with a decision from the Swedish Competition Authorities, No 292-97 - NCC/Siab, NCC/Siab will divest its interests in this joint venture.

(37) In order to assess the true competitive strength of the parties, these figures include the total turnover of Sydsten AB in which Skanska has joint control. Skanska had originally only included its share part in this company (i.e. 50 % on the basis that Skanska holds 50 % of the shares in that company). However, as long as the test of control is fulfilled, it must be concluded that Skanska has decisive influence over the whole production of the company).

(38) Scancem has been unable to provide information on prices for 1995 to 1996.

(39) Based on the information submitted by Skanska, the cost of providing a suitable rock-crushing plant is up to ECU 1 million while in the case of a plant for extraction of gravel, the cost would be around ECU 0,3 million.

(40) Case No 292-97 - NCC/Siab, 22 July 1997.

(41) Case DNo 292-97 - NCC/Siab, 22 July 1997.

(42) The local estimates submitted by the five companies show a higher degree of variation (as an example, Skanska's estimate for Stockholm is more than [...]* lower than that of Scancem (by volume) and more than [...]* lower by value. However, since the relative position among the main ready-mixed suppliers remains unchanged regardless of which figures are used, those submitted by Skanska have been used below.

(43) In order to assess the true competitive strength of the parties, these figures include the total turnover of Sydsten AB and Strömstadsbetong AB, in both of which Skanska has joint control, as well as Essbetong KB, which is jointly controlled by Scancem. Skanska had originally only included the parties' share part in these companies (e.g. 50 % of Sydsten's turnover, on the basis that Skanska holds 50 % of the shares in that company). However, as long as the test of control is fulfilled, it must be concluded that the parties have decisive influence over the whole production of these companies.

(44) Skanska has been unable to provide comparable data for 1995.

(45) Measured as result after financial costs/turnover.

(46) Neither Skanska, nor Scancem, have explained the reasons for the indicated local variations.

(47) In a submission, subsequent to the statement pursuant to Article 18, Skanska has indicated that its Finnish subsidiary, Tekra OY, is active in ready-mixed concrete. In addition to further reducing the likelihood of entry from Finnish companies on the Swedish market, this also indicates that, contrary to what was disclosed in the notification, the notified operation may have a significant horizontal impact in Finland.

(48) In the reply, Skanska has indicated that its sales figure for 1997 (in value) was about [10 to 20 %]* lower than previously indicated, but has provided no evidence of this. However, as the revised figure would only mean that Skanska's market share (insignificant change)*, this would in any case not materially affect the assessment.

(49) According to Skanska, the statistics overstate the market shares of the parties since, in its view, approximately 10 % of all sales are not reported. Even if Skanska's view were to be accepted, the figures would nevertheless to the same extent overstate also the figures of the competitors. Therefore, the relative positions of the main suppliers would remain unchanged.

(50) As noted above in relation to ready-mixed concrete, Skanska has in a submission subsequent to the statement pursuant to Article 18 acknowledged that its Finnish subsidiary, Tekra OY, is active in precast concrete in Finland. Also, Scancem has a market share of about [20 to 30 %]* in Finland.

(51) Skanska has no production of bricks, but sells such products, in order to complement its sales of other products. Skanska estimates that its sales correspond to [< 10 %]* of total Swedish market.

(52) Decision No 292-97 NCC/Siab, 27 July 1997.

(53) According to its 1996 Annual Report, Skanska is in total more than twice as large as NCC. Moreover, according to the same source, Skanska ranks as the ninth largest construction company in Europe.

(54) On 3 and 9 November 1998 Skanska submitted clarifications of its originally submitted undertakings.