EC, August 18, 2006, No M.3848

THE COMMISSION OF THE EUROPEAN COMMUNITIES,

Having regard to the Treaty establishing the European Community, Having regard to the Agreement on the European Economic Area, and in particular Article 57 thereof, Having regard to Council Regulation (EC) No 139-2004 of 20 January 2004 on the control of concentrations between undertakings (1), and in particular Article 8(1) thereof, Having regard to the Commission's decision of 31 March 2006 to initiate proceedings in this case, Having regard to the Commission's decision of 3 May 2006 to extend the procedure, After consulting the Advisory Committee on Concentrations (2), Having regard to the final report of the Hearing Officer in this case (3), Whereas:

(1) On 24 February 2006, the Commission received a notification of a proposed concentration pursuant to Article 4 of Regulation (EC) No 139-2004 ("the Merger Regulation") by which the undertaking Sea-Invest NV ("Sea-Invest", Belgium) acquires within the meaning of Article 3(1)(b) of the Merger Regulation joint control of the undertakings Europees Massagoed-Overslagbedrijf BV and Erts- en Kolen Overslagbedrijf BV ("EMO-EKOM", the Netherlands) by way of purchase of shares. After the concentration, EMO-EKOM will be jointly controlled by Sea-Invest, ThyssenKrupp Veerhaven BV ("TKV", the Netherlands), H.E.S. Beheer NV ("HES", the Netherlands) and Manufrance BV ("Manufrance", the Netherlands).

(2) By decision dated 31 March 2006, the Commission found that the notified operation raised serious doubts as to its compatibility with the Common market and the functioning of the EEA agreement. The Commission accordingly initiated proceedings in this case pursuant to Article 6(1)(c) of the Merger Regulation.

I. THE PARTIES

(3) Sea-Invest is a privately owned Belgian company which provides terminal services (4) in a number of ports in Belgium, France, Germany and South Africa. Its core business is the handling of dry bulk and other non-containerised cargo. Sea-Invest has a [...]*% controlling share (5) in Antwerp Bulk Terminal ("ABT") in Antwerp which provides terminal services for coal, iron ore and other dry bulk. Sea-Invest also controls the company Sea-Rail NV which provides inland transport of minerals by rail from the ABT terminal in Antwerp to the industrial site of the customer. Sea-Invest also controls Ghent Coal Terminal ("GCT") and Compagnie Belge de Manutention ("CBM") terminal in Ghent and the Sea-Bulk terminal in Dunkirk, which provide terminal services for dry bulk. Furthermore, Sea-Invest owns the company Kleimar SA, which operates a fleet of approximately 30 deep sea dry bulk vessels.

(4) The target EMO-EKOM provides terminal services for coal and iron ore in the port of Rotterdam. Its services include, in particular, loading, unloading, ground storage services and client-specific processing services such as washing, screening, crushing and blending. EMO operates the terminal and EKOM is the owner of the facilities (6). The existing (indirect or direct) shareholders in EMO-EKOM are TKV, RAG Logistic GmbH ("RAG") (which is selling its shareholding to Sea-Invest), HES and Manufrance.

(5) TKV is ThyssenKrupp Steel's seaport forwarder for the ports of Amsterdam, Rotterdam and Antwerp. It operates push boat and barge services via the river Rhine to the group's blast furnaces in Duisburg. TKV also operates a stevedoring terminal, EECV, in the port of Rotterdam, which handles coal and iron ore. The capacity of EECV is entirely dedicated to the ThyssenKrupp group.

(6) In addition to its shareholding in EMO-EKOM, the Dutch company HES holds stakes in a number of companies which handle dry bulk in the ports of Rotterdam, Amsterdam and Zeeland. Its wholly owned subsidiary European Bulk Services BV ("EBS") handles coal, iron ore and other dry bulk in the port of Rotterdam. It also has shareholdings in the Rotterdam Bulk Terminal BV ("RBT"), which handles iron ore, coal and other dry bulk in the port of Rotterdam, the OBA Group BV ("OBA"), which handles coal and other dry bulk in the port of Amsterdam (7) and Overslagbedrijf Terneuzen BV ("OVET"), which handles coal, iron ore and other dry bulk in the ports of Zeeland (8). HES' shareholdings in RBT (45%) and OBA (50%) give HES joint control, whereas the shareholding in OVET (33,3%) is a non-controlling interest.

(7) In addition to its shareholding in EMO-EKOM, Manufrance holds controlling interests of 66,6% in OVET and (indirectly via OVET) of 50% in OBA. Manufrance is a subsidiary of ATIC Services, a joint venture between the Total group, the EDF group and the Arcelor group (9). ATIC Services holds interests in companies offering services related to coal trading, inland-waterways logistics, maritime transportation and quality control of coal and iron ore cargoes.

II. THE CONCENTRATION

(8) Sea-Invest envisages acquiring the 50% share currently held by RAG in the holding company Transport- en Handelsmaatschappij Steenkolen Utrecht BV ("SNV"). SNV is one of the three parents of EMO-EKOM. Currently, shares in EMO-EKOM (that is to say, both EMO and EKOM) are held as follows: SNV - [...]*%, HES - [...]*%, Manufrance - [...]*% (10). In the Supervisory Board of EMO-EKOM, which adopts the strategic decisions, SNV has [...]* and HES and Manufrance each have [...]* (11). Decisions of the Supervisory Board require a majority of at least [...]* votes out of [...]* (12). The managing directors of EMO-EKOM can only be appointed by Unanimity (13). Following an amendment of the shareholders agreement in 1993, certain important decisions such as decisions on investments or loans of certain size and on the business plan require a majority of [...]* votes out of [...]* (14). The annual business plan includes the annual budget and a long term strategy. Consequently, SNV, HES and Manufrance can each block these strategic decisions and therefore have joint control over EMO-EKOM.

(9) SNV (a holding company whose only purpose is to hold shares in EMO-EKOM) will be jointly controlled by Sea-Invest and TKV, which holds the remaining 50% share in SNV. According to the statutes of SNV, no decision can be taken if the shareholders do not agree on a common position (15). If no decision is taken, SNV cannot vote in the board of EMO-EKOM. Decisions in the board of EMO-EKOM require at least [...]* positive votes. Consequently, either Sea-Invest or TKV will be able to prevent the adoption of decisions in the board of EMO-EKOM, via SNV. It follows that Sea- Invest will have joint control over EMO-EKOM together with TKV, HES and Manufrance.

III. COMMUNITY DIMENSION

(10) The undertakings concerned (16) have a combined aggregate world-wide turnover (17) of more than EUR 5,000 million (18). Sea-Invest and TKV each have Community-wide turnover in excess of EUR 250 million (19), and the undertakings concerned do not achieve more than two-thirds of their aggregate Community-wide turnover within one and the same Member State. The notified operation therefore has a Community dimension.

IV. RELEVANT MARKETS

Relevant product market

(11) The Commission has in the past suggested that the market for terminal services (cargo handling and storage in ports) could be further subdivided according to the three main types of cargo: (i) parcel goods (in particular containers), (ii) dry bulk goods and (iii) liquid bulk goods (20). The market investigation in this case confirmed this subdivision. A previous decision also made reference to indications that the market for terminal services for dry bulk cargo could be further subdivided according to the type of commodity handled (21). A further distinction which has been made by the Commission concerns the split between terminal services for hinterland traffic as opposed to terminal services for transhipment traffic (22).

Terminal services for coal and iron ore

(12) In general, dry bulk is usually split by the industry into several categories: coal and iron ore (sometimes referred to together as major bulk), agri-bulk (for example, grains or cereals) and other dry bulk (a wide range of various bulk commodities such as zinc concentrates and other non-ferrous concentrates, cement clinkers, pig iron, kaolin, phosphates and other minerals, cokes, pet-cokes, anthracites etc.).

(13) Sea-Invest submits that within the group of dry bulk the handling of coal and iron ore belongs to the same product market.

(14) Coal and iron ore are major bulk products which are handled in large quantities and for which speed of discharging is crucial while handling and storage does not require special attention. Terminal companies are able to switch between handling coal and iron ore and many of them do handle both products at the same terminal and use basically the same equipment. When changing from discharging iron ore to coal or vice versa, the terminal only needs to change the grab of the crane and run the conveyor belts empty or possibly clean them (in particular when changing from iron ore to coal). These changes can be made within a very short time (1-2 hours) and the terminals handling both products actually make these changes as part of their day-today operation. Both iron ore and coal are stored in the open air and basically in the same area, only in separate stockpiles. Further, the customers of terminals for handling coal and iron ore are, in both cases, usually the end users of these commodities, namely large steel companies (iron ore and coking coal) and electricity companies (steam coal). Both these commodities are also transported in larger seagoing vessels with higher draught which are able to transport large volumes of coal and iron ore. Also because of the large volumes transported, the preferred means of inland transport for both commodities are, in particular, barges and, to a lesser extent, rail.

(15) According to Sea-Invest, other dry bulk products and agri-bulk do not form part of the same terminal services market as coal and iron ore. Sea-Invest argues that discharging of other dry bulk products usually takes more time than for coal and iron ore, handling often requires special care, storage is not always in open air but in warehouses and the terminal company often provides value added services (such as packaging). It also argues that the equipment for handling coal and iron ore is different from the equipment for handling other dry bulk and that no significant switching of terminals away from other dry bulk to coal and iron ore or vice versa has been observed so far.

(16) HES considers, however, that the relevant product market should not be limited to coal and iron ore, but should be extended to include all dry bulk with the exception of agri-bulk. It argues that terminal companies typically handle several types of dry bulk and that it is possible to change from handling one type of dry bulk to another within a short time and without substantial additional costs.

(17) The market investigation indicated that there are some terminals, usually smaller ones such as EBS in Rotterdam, which handle both coal/iron ore and other dry bulk and are able, at least to some extent, to switch between all these commodities. Further, there are some examples of terminals handling only other dry bulk having extended their handling services to coal (and to a lesser extent also iron ore) in the past. However, at the same time, the market investigation confirmed that, in those few instances where switching between coal and iron ore and other dry bulk is considered feasible, this requires considerable investments into equipment specifically designed to handle the normally larger volumes of coal and iron ore. Further, a number of customers indicated that the mixed handling of coal and iron ore, on the one hand, and other dry bulk products, on the other, would create risks of contamination of the products concerned.

(18) Important customers for the handling of coal and iron ore, in particular, steel and electricity companies, confirmed that they do not consider other dry bulk terminals as a viable alternative, mainly because of the lack of sufficient storage capacity and the inability to quickly load and unload large volumes of cargo. The market investigation also revealed that, for efficient handling of large volumes of coal and iron ore, the terminals need cranes with high capacity grabs and a system of conveyor belts. In contrast, the cranes for other dry bulk are usually smaller and are not able to handle large volumes of coal and iron ore sufficiently quickly. The conveyor belts are usually not used for other dry bulk as these commodities are often more fragile than coal and iron ore and are difficult to transport by conveyor belts. This seems to also be the case for cokes, pet-cokes and anthracites which, even though they are products close to coal, are much more fragile and valuable and need to be handled differently from coal.

(19) Furthermore, other dry bulk commodities are transported and handled in significantly smaller volumes compared to coal and iron ore (23). Much smaller sea-going vessels are used for other dry bulk than for coal and iron ore. These vessels are able to enter fully loaded into draught restricted ports. As the volumes of individual shipments are smaller, the inland transportation of other dry bulk products is also more flexible and not limited to barge and rail transport as in the case of coal and iron ore. The customers of stevedoring terminals for other dry bulk are also different and traders and logistics companies play a greater role.

(20) Even though some smaller terminals do switch between handling these two groups of products, as set out above, (24) there are substantial differences between handling coal and iron ore, on the one hand, and other dry bulk, on the other, which challenge effective and immediate supply-side substitution. This conclusion was confirmed during the in-depth investigation, which included on site visits to the terminals of ABT in Antwerp, EMO-EKOM and EBS in Rotterdam.

(21) Customers using the terminal services for coal and iron ore are electricity companies, steel companies and, to a lesser extent, traders. Despite certain differences between these customer segments described below, (25) the market investigation did not produce any evidence that it would be appropriate to define separate markets for each customer group, in particular taking into account high supply side substitutability.

(22) With respect to agri-bulk, it was confirmed by the market investigation that terminals handling coal, iron ore and other dry bulk would not generally switch to handling agri-bulk or vice versa, in particular for food safety reasons.

(23) In view of the above, it is concluded that there is a separate product market for terminal services for coal and iron ore.

Terminal services for hinterland and transhipment traffic

(24) The Commission has, in the past, made a distinction between terminal services for hinterland traffic (from deep-sea ship directly to inland barge, train or truck) and terminal services for transhipment traffic (from deep-sea ship to relay/feeder vessel) (26), in relation to container transport. This distinction also seems applicable in relation to the transport of coal and iron ore.

(25) Most of the coal and iron ore handled by the terminals in the ARA range (namely, Antwerp, Rotterdam, Amsterdam, Zeeland) goes to hinterland customers. However, a limited volume of coal and iron ore handled in these ports is transhipped by coasters to another final discharge port. In many instances this final discharge port is a draught restricted port in the United Kingdom.

(26) Although the main handling techniques and the handling terms and conditions are similar, whether the coal or iron ore is transported to the hinterland or transhipped, the range of ports and terminals that may be potential substitutes for transhipment traffic is not identical to the range of ports and terminals that may be substitutes for specific hinterland traffic. Terminal handling for transhipment purposes can, in principle, take place in a larger group of ports than terminal handling for hinterland transport to a specific destination. (27) For these reasons, and in line with previous Commission decisions, separate product markets can be distinguished for terminal services for hinterland traffic for coal and iron ore and terminal services for transhipment traffic for coal and iron ore.

Relevant geographic market

(i) Terminal services for hinterland traffic for coal and iron ore

(27) In its decision pursuant to Article 6(1)(c)of the Merger Regulation ("the Article 6(1)(c) decision"), the Commission took the position that the relevant geographic market would either extend to all ports in the ARA range or, alternatively, should be limited to the port of Antwerp, on the one hand, and to the Dutch ports (Rotterdam, Amsterdam, Zeeland), on the other. A final position on this issue was left open.

(28) The in-depth market investigation confirmed that, in view of the limited substitutability between the ABT terminal in Antwerp and terminals in Ghent and Dunkirk, on the one hand, and the EMO-EKOM terminal in Rotterdam as well as other terminals in Rotterdam, Amsterdam and Zeeland, on the other, it is more appropriate to define the Dutch ports as a separate market from the ports of Antwerp, Ghent and Dunkirk.

Terminal services as part of the overall logistic chain

(29) The coal and iron ore terminals in the ARA range have three main groups of customers: steel producers, electricity producers and traders. When deciding on the use of the terminal, the customer has to take into consideration the whole logistic chain in order to optimise the total logistic costs connected with importing coal and iron ore. This imposes significant constraints on its ultimate choice for a specific terminal. The most important cost elements of the logistic chain besides terminal handling and storage are sea freight and inland transport (28). Terminal handling and storage fees represent only around 10-15% of the total logistic costs. Therefore, the Commission analysed these individual components of the total logistic costs and their influence on the choice of the terminal and the possibility of switching between terminals.

(30) Apart from the total logistic costs, speed and security of supply are also of relevance for the customers of coal and iron ore terminals. Further, the logistics are also influenced by the fact that steel mills and power plants normally need a certain predetermined mix of different qualities of coal and iron ore, which come from different origins.

(31) All of the above elements (individual costs elements as well as other factors which are specific for each customer) need to be considered in their totality. A change in one of the elements often automatically entails changes in other elements of the total logistic chain. These changes and factors are usually specific for each customer. Therefore, the following sections analysing individual components of the logistic chain are followed by an analysis of the possibilities of customers of ABT and EMO-EKOM to switch between Antwerp and Rotterdam, taking into account all relevant factors as a whole.

Differences in sea freight costs

(32) The sea freight costs are, for the majority of customers, the most important factor influencing the choice of a particular terminal. It is, in general, more economical to use as large vessels as possible for the maritime leg of the transport of coal and, even more so, for the transport of iron ore. However, the choice of vessel is limited by the draught of the ports. The market investigation confirmed that Antwerp is more draught restricted than Rotterdam and other ports in the ARA range. The maximum draught (29) in Antwerp is 15.56 meters while maximum draught in Rotterdam is 23 meters, in Amsterdam 17.20 meters and in Zeeland 16.50 meters. As a result, very large vessels (exceeding around 140.000 dwt) cannot use Antwerp when fully loaded. In contrast, vessels exceeding 140.000 dwt bring the vast majority of all coal (around 79%) and iron ore (around 93%) discharged in the port of Rotterdam.

(33) The investigation further confirmed that Rotterdam is the main port for iron ore as the terminals in Rotterdam handle almost 80% (30) of all iron ore discharged in the ARA range. Iron ore is transported in even bigger sea going vessels than coal (usually significantly exceeding 140.000 dwt (31)), for which Antwerp is not an alternative due to its restricted draught. Therefore, the importance of large vessels and the resulting difference in sea freight costs between Antwerp and Rotterdam are even greater than in the case of coal.

(34) The conclusions stemming from the below described quantitative analysis of sea freight costs are therefore even more valid for iron ore and the sea freight costs implications for iron ore do not have to be quantitatively analysed separately. These conclusions were also confirmed by the information from the iron ore customers received in the market investigation.

(35) This situation will not change in the future. Even though it is envisaged to deepen the river Schelde which leads to the port of Antwerp, this will not affect the ABT terminal, which is situated behind the locks. As there are no plans to deepen the locks and the canal behind the locks, the ABT terminal will not benefit from the deepening of the river Schelde which is mainly aimed at creating better conditions for the new container terminals situated before the locks.

(36) Draught restrictions in the port of origin are also relevant. Coal from Australia, South Africa and Columbia can generally be transported in very large vessels, whereas coal from the US, Russia and Poland, in particular, can only be transported in smaller vessels due to draught restrictions in the ports of loading. The market investigation showed that at least 70% of the total volumes of coal discharged in the ARA range is transported from regions where ports of loading are not draught restricted (in particular Australia, South Africa and Colombia).

(37) In order to quantify the importance of the sea freight costs for coal and how they differ for Antwerp and Rotterdam, the Commission analysed the sea freight costs data submitted by the customers in the market investigation for different origins of coal for the ports of Antwerp and Rotterdam.

(38) On average, the sea transport costs for coal range from around 8 EUR/tonne for coal from Russia to around 18 EUR/tonne for coal from Australia and represent, on average, around 60% of the total logistic costs (32). The difference between the sea transport costs for the ports of Antwerp and Rotterdam was confirmed, in particular, for the non-draught restricted ports of origin such as in the two most important countries of origin - Australia and South Africa - which mainly use large vessels which often cannot enter the port of Antwerp fully loaded. In order to be able to use Antwerp, smaller vessels (with less than around 140.000 dwt) must be used. The sea transport costs per tonne of coal are higher for those vessels. Alternatively, large vessels can be used if they are only partially loaded in the port of loading or are lightened before going to Antwerp by discharging part of the coal in a non-draught restricted port such as Rotterdam. However, even in these alternative scenarios, the costs of sea transport per tonne of coal are higher due to the inefficiently used capacity of the vessel or the necessity to make a double call in two different ports.

(39) On average, the sea transport costs for Antwerp as compared to Rotterdam are [1.5- 2]* EUR/tonne higher for coal from Australia and [1-1.5]* EUR/tonne higher for coal from South Africa. However, as there are significant differences in the absolute values of sea transport costs provided by different companies, the Commission also analysed separately the average cost differences based only on figures from those customers that provided their costs estimates for both Rotterdam and Antwerp. This calculation reduces the influence of the possible use of different methodologies of cost calculations by different customers. On such a basis, the cost difference between Antwerp and Rotterdam is [1-1.5]* EUR/tonne for coal from Australia and [0.5-1]* EUR/tonne for coal from South Africa. In any case, the difference in cost is significant compared to the total stevedoring fee, which ranges between [1-4]* EUR/tonne of coal.

(40) For coal imported from draught restricted ports in the US and Russia, the cost difference is lower ([0-0.5]* EUR/tonne for coal from the US and [0-0.5]* EUR/tonne for coal from Russia). This reflects the fact that vessels importing coal from these locations are not so large and can also enter the port of Antwerp. However, coal imported from these origins represents less than 30% of total coal discharged in the ports of the ARA range.

(41) The quantitative analysis of the sea transport costs confirmed the importance of draught and the relative significance of the resulting cost disadvantage of the ABT terminal in Antwerp as compared to the EMO-EKOM terminal in Rotterdam. In the case of coal transported from non-draught restricted origins, which represents more than 70% of coal discharged in the ARA range, the difference in sea transport costs between Antwerp and Rotterdam amounts, on average, to around 50% of the handling tariff. This in itself makes switching in reaction to a 10% price increase unlikely for the large majority of coal volumes.

(42) Taking into account the draught available in Amsterdam and Zeeland, similar conclusions may be drawn as regards the difference in sea freight costs of these ports in comparison with Antwerp. It was confirmed by the market investigation that the draught conditions in Amsterdam, in particular, are sufficient for the large coal vessels. As regards the Zeeland ports, the fact that they are located closer to the sea than Antwerp further contributes to the difference in sea freight costs resulting from their more advantageous draught conditions.

(43) As regards the terminals controlled by Sea-Invest in Ghent and Dunkirk, the maximum available draught in the port of Ghent is 12.50 meters and in the port of Dunkirk, up to 18 meters. Therefore, the conclusions concerning the difference in sea freight cost in comparison with Rotterdam as well as Amsterdam and Zeeland also relate to the Sea-Invest terminal in the port of Ghent taking into account its even lower available draught. Only the Sea-Invest terminal in Dunkirk has a sufficient draught to accommodate fully loaded vessels comparable to those discharged in the Dutch ports. However, its substitutability with Rotterdam, Amsterdam and Zeeland is limited by other factors described below, (33) in particular its limited inland transport connections.

Differences in inland transport costs

(44) The inland transport costs may also have a significant effect on the choice of a particular terminal. The Commission analysed the inland transport costs for coal (34) on the basis of the data submitted by the customers in the market investigation concerning costs of transport from Antwerp and Rotterdam to various locations in Belgium, the Netherlands, Germany and France.

(45) Due to the large volumes of coal and iron ore involved, the only efficient inland transport means are inland waterway transport (barges) and rail transport (35). In general, the inland transport costs for barges are lower than in case of the rail transport. However, this depends on the situation of each customer and, in particular, on the access of the customer's plants to the waterway system or suitable rail connections.

(46) The market investigation showed that the average inland transport costs for coal range from around 2 EUR/tonne for Belgium and the Netherlands to around 10 EUR/tonne for southern Germany. On average, inland transport costs account for around 25% of the total logistic costs (36). The comparison of average inland transport costs for different regions from Antwerp and Rotterdam shows that inland transport of coal from Antwerp is in general significantly cheaper than from Rotterdam in particular for Belgium and, to some extent, for Northern France. Rotterdam, on the other hand, is in general cheaper than Antwerp for the Ruhr area and the Saarland area of Germany, Southern Germany and to some extent for the Netherlands. The cost difference between Antwerp and Rotterdam for these regions ranges between 20% and 50% of the average stevedoring fee. Therefore, these differences in inland transport costs significantly limit the economic incentive of customers in these regions to switch between Antwerp and Rotterdam in reaction to a 10% increase in the stevedoring tariff. The results of the analysis are rather mixed only for North-East France where the inland transport costs from Antwerp and Rotterdam seem to be more balanced.

(47) However, the absolute amount of the inland transport costs is only one of the criteria relating to the inland transport which limit the customers' choice of terminal. There are a number of other factors to be taken into account, such as how many different plants the customer has in different locations, the timing of the supplies and the need for storage, access to and suitability of barge or train connection, the need to ensure security of supply (for example, in the periods of low water in the river Rhine) or the existence of long-term contracts with the providers of the inland transport. These factors are specific to each customer and are discussed in more detail below (37).

(48) The in-depth market investigation did not reveal any evidence of projects which could increase the possibility of switching between the ports of Antwerp and Rotterdam in the foreseeable future. The majority of coal and iron ore customers replied that they are not aware of any infrastructural project which could change the attractiveness of any of the ARA terminals in the foreseeable future. Works are underway to enhance rail transport from Rotterdam and Amsterdam (the so called "Betuwe Lijn") as of 2007. However, those works are intended, in particular, to improve the connection with Germany but not with destinations which currently form the hinterland of Antwerp, such as the rest of Belgium or Northern France. As regards the possible improvement of rail connections between Antwerp and Germany (the so called "Iron Rhine"), there is still no agreement between the authorities involved as to whether it will be undertaken at all. In any case, if the work were to be undertaken, according to the information obtained during the market investigation it would not be completed before 2015 (38). This was also confirmed by the replies in the market investigation as none of the coal and iron ore customers identified Iron Rhine as an infrastructural project that could change the attractiveness of the ABT terminal in the foreseeable future.

(49) Therefore, the analysis of the data on the inland transport costs in general supports the statements of Sea-Invest and of a number of respondents in the market investigation that, to a large extent, Antwerp and Rotterdam serve different hinterlands and that the choice of a terminal is significantly predetermined by the location of the customer's plant(s).

(50) The market investigation also confirmed that the inland transport connections of Amsterdam are comparable to those of Rotterdam, in particular as regards the Netherlands and Germany (access by barge via the river Rhine). In contrast, the investigation confirmed that in view of their inland transport connections, Ghent and Dunkirk are even less suitable substitutes for the Dutch ports than Antwerp, in particular for the German hinterland.

Comparison of stevedoring tariffs

(51) Further, the Commission analysed in detail the prices paid for stevedoring services at ABT and EMO-EKOM by customers located in different geographic areas. This analysis clearly showed for both ABT and EMO-EKOM that there are no specific geographic regions where customers pay significantly lower or higher prices as compared to other regions. This indicates that there are no particular regions where the competitive pressure of other terminals is particularly strong and leads to lower terminal prices than in other areas. Therefore, even for customers in the regions for which the difference in inland costs is not so significant (in particular North-East France), the competition between EMO-EKOM and ABT does not seem to be stronger than in regions with high differences in inland transport costs from Antwerp and Rotterdam.

(52) The analysis of stevedoring prices also demonstrated that these prices differ significantly between individual customers of both ABT and EMO-EKOM. However, despite the relative disadvantages of the ABT terminal, in particular as regards the available draught, prices are on average higher in the case of ABT than in other coal and iron ore terminals in the ARA range. The general difference in the overall price level is thus another element supporting the definition of separate markets for Antwerp, on the one hand, and other terminals in the ARA range, on the other.

Possibilities of switching for customers

(53) During the in-depth investigation the Commission sent detailed requests for information to all customers who used the stevedoring services at the ABT or the EMO-EKOM terminal for appreciable volumes of coal or iron ore. It also sent detailed questionnaires to important customers for the handling of coal and iron ore of the other terminals in Rotterdam, Amsterdam, Zeeland, Ghent and Dunkirk. In these questionnaires the Commission asked for the location of the customers' plants or, in the case of traders, for the location of their clients' plants. Further the Commission asked for the volume which was transported to each final destination, the terminals used, the mode of inland transportation and the origin of the raw material. This allowed the Commission to understand better the logistic chain of each customer.

(54) For each destination, the Commission asked for the possibilities of switching from ABT to another terminal or from EMO-EKOM to another terminal, for the volume which could switch within one and within three years and for estimated cost differences in case of switching.

(55) In addition, the Commission contacted the customers who currently use either the ABT terminal and another terminal in Rotterdam, Amsterdam or Zeeland or the EMOEKOM terminal and another terminal in Antwerp, Ghent or Dunkirk. The Commission met these customers and/or conducted phone interviews with them. The Commission also contacted customers who currently do not use terminals in both Antwerp/Ghent/Dunkirk and Rotterdam/Amsterdam/Zeeland, but indicated some possibilities of switching, in their replies to the questionnaire. The contacts allowed the Commission to learn more about the organisation of the transport logistics of the customers of the terminals, how the decision to use a particular terminal is made and the reasoning behind the attribution of volumes to ports and terminals.

(56) The replies to the questionnaires, the interviews with the customers and the analysis of the costs of the whole transport chain explain the constraints on switching. All customers emphasised that their choice of terminal is very much determined by the sea freight costs and the inland transport costs. Further, there are a number of other factors determining and limiting the choice of the stevedoring terminals for each individual customer. These factors are, for example, participation in part cargo services (39), additional services (such as blending of coal with biomass, magnet separation of scrap from coal etc.) and the storage capacity of terminals, the risk of congestion at terminals, how many plants the customer has and their location, the timing of delivery, access to and suitability of barge or train connection, the need to ensure security of supply (for example, in the periods of low water in the river Rhine) and the existence of long-term contracts with the providers of the inland transport.

(57) As to the specific situation for each of the customer groups, the following observations can be made.

Steel producers

(58) Steel producers buy most of their coal and iron ore volumes directly and not via traders. Some of them have captive terminals (40). For iron ore, switching between Antwerp and other ARA ports does not occur. In view of the draught advantage and better hinterland connections, steel producers in the German Ruhr area mainly use terminals in Rotterdam. They do not use terminals in Antwerp and do not intend to switch to Antwerp. Steel producers in the Saar area, in North Eastern France and Belgium have their iron ore handled by the EMO-EKOM terminal and, to a limited extent, by the EBS and OVET terminals. Only one steel mill in Belgium uses the ABT terminal for substantial volumes of iron ore because of good hinterland train connections from Antwerp. This steel mill does not use any other terminal for the handling of iron ore and does not intend to switch. The terminals in Ghent and Dunkirk mainly handle iron ore for steel mills close to the terminal.

(59) As regards coal, steel mills in the German Ruhr area only receive deliveries via Rotterdam, Amsterdam and Zeeland. They do not intend to switch to Antwerp. Steel mills from the Saar area mainly use terminals in Rotterdam or Amsterdam. Antwerp only handles limited volumes of coal, which is shipped in part cargo services. In the case of part cargo services, where the discharge terminal is not fixed by the shipping line, it is the customer representing the largest volume who decides on the terminal to be used. The other customers have to follow. They do so because taking part in a part cargo service reduces the costs of transport as compared to individual shipment.

(60) Steel mills in Belgium and North-Eastern France use several terminals in the ARA range, Ghent and Dunkirk for handling of coal, but the investigation has revealed that they attribute volumes to certain terminals for specific reasons which strongly limit their possibilities of switching. In case of difficult waterway access, train transport from Antwerp is the cheapest mode of inland transport. Antwerp also has inland transport cost advantages for steel mills with an appreciably shorter barge transport time than from Rotterdam or Amsterdam. Coal is transported to these steel mills via Rotterdam only if it comes from non draught restricted ports of origin such as ports in Australia or South Africa. These vessels are often lightened in Rotterdam to the extent that the vessel can enter the port of Antwerp and the remaining volume is handled in Antwerp. Handling more volume in Rotterdam than necessary for the lightening of the vessel is avoided due to the disadvantage of Rotterdam in terms of inland transport costs. Some of the volume is also transported in part cargo services.

(61) For the other steel mills, transport of coal via Rotterdam or Amsterdam leads to an overall cost advantage. These steel mills receive only limited amounts of coal via Antwerp which is delivered in part cargo services from draught restricted ports of origin such as US ports. To the extent that the terminals in Ghent and Dunkirk handle coal for steel producers, the coal is delivered to steel mills which are located very close to the respective terminal.

Electricity companies

(62) Some electricity companies purchase and import coal directly and others use traders. Dutch electricity companies only receive deliveries of coal for Dutch power plants via Dutch ports and do not intend to switch. The Belgium electricity company Electrabel applies a refined logistical system to deliver coal to its power plants. It mainly uses the terminals in Antwerp and Ghent for its Belgian power plants. The main reason is the better hinterland connections. Its power plant in the Netherlands receives deliveries both via the ABT terminal and via terminals in Rotterdam and Amsterdam, but the volumes are attributed to each terminal according to logistical needs. Some German electricity companies only use terminals in Rotterdam and Amsterdam, whereas others also have limited amounts of coal delivered via Antwerp through traders. In this case it is the trader and not the electricity company who chooses the terminal.

(63) Power plants in North-Eastern France receive coal via both Antwerp and Rotterdam, but there are specific reasons for the attribution of volumes. For power plants which have no direct waterway connection, inland transport by train from Antwerp is the cheapest alternative. These plants receive only limited volumes of coal via Rotterdam or Amsterdam, mainly coal from non draught restricted ports which is delivered in part cargo services. Power plants in North-Eastern France which are connected to a waterway receive most of their coal via Rotterdam or Amsterdam. Only small volumes are delivered via Antwerp in the form of part cargo services from draught restricted ports of origin such as US ports.

(64) According to the electricity companies the current attribution of volumes to terminals reflects the optimal organisation of the overall logistics chain.

Traders

(65) Traders sell coal ore and, to a limited extent, iron ore at the wholesale level. They usually sell coal or iron ore under "FOB barge" conditions (41) from certain ports and do not provide the inland transport. There are independent traders and traders which belong to an industry or electricity group. The latter do in-house business, but might also act as traders for third parties. For the in-house business, traders can be in charge of the whole logistics chain.

(66) Traders usually use several terminals in the ARA range. However some of them exclude Antwerp as an alternative. Other traders also use Antwerp and to a limited extent Dunkirk, but rarely Ghent. Their flexibility in the use of the terminal and the possibilities of switching depend, in particular, on the requirements of the customers and the delivery conditions. Some customers, such as certain German utilities, require delivery to Rotterdam or Amsterdam, thereby excluding Antwerp. Cost differences for the other elements of the logistic chain can also determine the choice of the terminal. Coal from the non draught restricted ports in Australia and South Africa is nearly only handled in Rotterdam.

(67) Nevertheless, some traders indicate that they can switch limited volumes between Antwerp and the other ports in the ARA range. However, they cannot transfer all their volume. Switching from Antwerp to other ports in the ARA range, even though already very limited in scope, is normally easier than switching from Rotterdam, Amsterdam or Zeeland to Antwerp, Dunkirk or Ghent, in particular due to draught restrictions and hinterland connections.

Quantification of volume which can switch

(68) Based on the in-depth market investigation, the possibilities of substitution for hinterland traffic which remain in view of the constraints described can be quantified as follows: The volumes for which ABT or terminals in Ghent or Dunkirk can substitute EMO-EKOM are marginal in relation to EMO-EKOM's total volume (less than 5%). The volumes for which EMO-EKOM or other terminals in Rotterdam, Amsterdam or Zeeland can substitute ABT are small in relation to ABT's total volume (less than 10%). These numbers are based on customer data which cover nearly all the volume handled by EMO-EKOM and ABT. Furthermore, only very limited volumes could switch from the other terminals in Rotterdam, Amsterdam or Zeeland to Antwerp, Ghent or Dunkirk. The same is true of the volumes which could switch from terminals in Ghent or Dunkirk to terminals in Rotterdam, Amsterdam or Zeeland.

Most of the limited volume which can switch is attributable to traders.

Switching in the past

(69) The obstacles to switching are illustrated by the fact that only limited switching has occurred in the past. Customers of terminals only reported a few instances of switching between Antwerp/Ghent/Dunkirk, on the one hand, and Rotterdam/Amsterdam/Zeeland, on the other, during the last five years. These instances of switching only involved small volumes. They were, in general, not motivated by differences in the terminal handling fees, but by less congestion, lower sea freight or inland transport costs or changes in the other parts of the logistic chain (for example, switch from rail to barge inland transport).

(70) However, customers reported switching of appreciably higher volumes between terminals in Rotterdam and Amsterdam. These instances of switching were often motivated by differences in the terminal handling fees.

Conclusion

(71) All this supports the conclusion that Rotterdam/Amsterdam/Zeeland is a separate market from Antwerp/Ghent/Dunkirk. Conditions of competition are not the same in these two areas. Nevertheless, there is some limited fringe competition between these geographic markets. This fringe competition mainly targets volumes of traders. For ABT, this fringe competition seems to be more relevant than for EMO-EKOM. However it still represents less than 10% of total volume handled by ABT. It can be left open whether the ports of Ghent and Dunkirk are in the same geographic market as Antwerp for terminal services for coal and iron ore or if they form a separate market (which is in any case different to Rotterdam/Amsterdam/Zeeland). The proposed concentration does not lead to any competition concerns under either analysis.

(ii) Terminal services for transhipment traffic for coal and iron ore

(72) In previous decisions in the field of container liner shipping the Commission has defined the geographic market for transhipment traffic as Northern Europe. Such a market includes all main deep-sea ports in the Gothenburg - Le Havre range including UK and Irish ports (42). The market investigation has shown that these ports can also compete with each other for terminal services for the transhipment of coal and iron ore and that the range of substitutable ports is clearly wider than in the case of coal and iron ore for hinterland traffic, mainly due to different logistical requirements. Therefore, the geographic market for terminal services for transhipment traffic for coal and iron ore includes at least the deep-sea ports in the range Gothenburg - Le Havre and the UK deep sea ports.

V. COMPATIBILITY WITH THE COMMON MARKET AND THE EEA AGREEMENT

A. Terminal services for coal and iron ore for hinterland traffic

1. Article 2(3) of the Merger Regulation

a) Unilateral effects

(i) Strengthening of Sea-Invest's dominance

(73) It follows from the market definition that Sea-Invest is dominant on the Antwerp market (market share 100%) for terminal services for hinterland traffic for coal and iron ore (43). As set out in the geographic market definition. (44) Sea-Invest's ABT terminal only faces fringe competition on this market. Whether the proposed concentration will enable Sea-Invest to strengthen ABT's dominance in the port of Antwerp depends on whether Sea-Invest will have the possibility and incentive to use its powers of joint control to restrict or eliminate the remaining fringe competition.

Powers of Sea-Invest

(74) As described in detail in section II, Sea-Invest's joint control over EMO-EKOM stems from its shareholding in SNV, which is a [...]*% shareholder of EMO-EKOM. This joint control provides Sea-Invest, via the SNV representation in the EMO-EKOM Supervisory Board, with a right of veto over all issues for which the EMO-EKOM Board of Directors needs approval from the Supervisory Board. Sea-Invest does not have the power to force the EMO-EKOM Supervisory Board to adopt certain decisions. For this it would need the agreement/cooperation of the other shareholders.

(75) The possession of a right of veto in respect of proposals concerning important policy matters in principle provides Sea-Invest with an important tool to block proposals which may have a direct bearing on the fringe competition, or to discipline EMOEKOM by randomly blocking proposals, should EMO-EKOM act against the interests of Sea-Invest. The right of veto would seem especially important in relation to the adoption of decisions concerning major investments.

(76) Another option for Sea-Invest would be to use its right of veto to block proposals concerning other issues, such as the adoption of the business plan or the appointment of the management, should it wish to discipline EMO-EKOM for undesired behaviour with respect to the fringe competition. This could either concern proposals which target the fringe competition or randomly chosen proposals.

(77) Further, Sea-Invest's participation as shareholder in the Supervisory Board of EMOEKOM provides it with a certain degree of general information about EMO-EKOM's business policy and strategic planning. The general pricing level of EMO-EKOM - although not prices of individual customers - is also discussed in Supervisory Board meetings. This may allow Sea-Invest to assess the competitive pressure which EMOEKOM exercises on its ABT terminal in Antwerp.

Alternatives of fringe customers

(78) However, none of these strategies could strengthen Sea-Invest's position because the fringe customers can switch to alternative terminals. It follows from the in-depth investigation that the fringe customers that can be identified are nearly all traders. In comparison with other types of customers, traders have more flexibility in their choice of the terminal. Most of these fringe customers can switch volumes of coal from ABT to Rotterdam or Amsterdam. However, it was confirmed by the in-depth market investigation that none of the fringe customers which can switch volumes from ABT in Antwerp to Rotterdam would solely depend on EMO-EKOM. On the contrary, most of these fringe customers identified terminals other than EMO-EKOM, such as OBA and Rietlanden in Amsterdam and the other terminals in Rotterdam, as the best possible alternative to ABT. Similarly, the fringe customers which are able to switch volumes from Rotterdam to the ABT terminal in Antwerp also confirmed that they had more and better alternatives than just ABT, in particular other terminals in Rotterdam and Amsterdam.

(79) It follows from the above that even if the proposed concentration would allow Sea- Invest, as an indirect shareholder in EMO-EKOM, to determine the business strategy of EMO-EKOM after the concentration, it would not be able to eliminate the fringe competition. The in-depth market investigation confirmed that a possible shift of volumes by the fringe customers to the other Dutch terminals is not just a theoretical possibility. ABT would still have to face the fringe competition of the other terminals in Rotterdam and Amsterdam. The market investigation indicated that these terminals have sufficient capacity to absorb the coal tonnages of the fringe customers that could switch away from ABT. In this respect it is important to note that the OBA and Rietlanden terminals in Amsterdam are currently both undertaking a substantial expansion of their coal handling capacities. Even though these expansions are mainly being carried out in anticipation of the expected growth in demand in Germany for imported coal, they would still further enhance the ability of these terminals to absorb the volumes that may be shifted by the fringe customers from ABT. A response to possible price increases of the ABT terminal is further facilitated by the fact that the volumes which the fringe customers can actually switch are limited.

(80) It should be underlined that the total volumes of coal and iron ore represented by these fringe customers are fairly limited. The volume which can switch represents less than 10% of ABT's total volume. It could be argued that even though the customers can only switch part of their volume, the total volume of each customer handled at the ABT terminal should be taken into consideration, because terminals usually do not price discriminate between volumes of one and the same customer. According to the results of the in-depth investigation, the sum of the total volume of the customers which can partly switch represents less than 20% of ABT's volume. In any case, if the volume which can switch is small in relation to the captive volume of the customer, the terminal could still choose to give up the marginal volume and to keep the captive volume at a higher price.

Lacking incentives of Sea-Invest

(81) Blocking investments in EMO-EKOM to make ABT more attractive always risks encouraging switching from EMO-EKOM to the other Dutch terminals. In addition, blocking investments which might increase EMO-EKOM's profitability only makes sense if the benefits outweigh the costs. In this respect the limited importance of the volumes represented by the fringe customers as compared to the volumes of the other customers plays an important role.

(82) As an example, the blocking of an investment to improve the current rail loading facilities of the EMO-EKOM terminal would not make sense. Such an investment which would be linked to the opening of the new Betuwe Lijn rail connection between the Netherlands and Germany would allow EMO-EKOM to benefit from the expected growth in demand in Germany for imported coal. It would therefore increase EMOEKOM's profitability, which is also in Sea-Invest's interest. Further, an important risk would be that the fringe customers would rather switch from EMO-EKOM to OBA or Rietlanden than to ABT. The costs of such a veto for Sea-Invest would therefore outweigh its benefits.

(83) Blocking the adoption of the business plan, which includes the adoption of the annual budget and business strategy, may have negative consequences which by far exceed the limited benefits which Sea-Invest may hope to obtain from an EMO-EKOM policy which leaves the fringe customers to ABT. For example, considering that EMO-EKOM's primary growth potential concerns the German market and not the Belgian or French market, any veto by Sea-Invest to block a business plan which favours EMO-EKOM's orientation towards expansion in this area would affect the profitability of EMO-EKOM. It is very unlikely that the loss of profitability would be outweighed by the possible benefits of reduced fringe competition. In any case, these benefits would be uncertain in view of the switching possibilities of the fringe customers. The same applies to the use of the veto rights with respect to the appointment of EMO-EKOM's management to influence its policy towards fringe customers.

(84) As to Sea-Invest's access to information concerning EMO-EKOM, it seems that EMO-EKOM has a policy of only providing its shareholders with necessary information. This can be explained by the fact that EMO-EKOM runs its day to day business to a large degree independently of its shareholders. This was confirmed during the in-depth market investigation. For instance, information on prices charged to individual customers is not available to shareholders. In any case, in view of the switching possibilities of the fringe customers the use of this information by Sea- Invest could not result in a restriction of fringe competition. For the same reason, a possible strategy of Sea-Invest to share fringe customers between itself and EMOEKOM would not work.

(ii) Unilateral price increase of EMO-EKOM

(85) Another possible theory of harm concerns the unilateral effect of the proposed concentration which might provide EMO-EKOM with the ability and incentive to increase prices. Such a unilateral effect would, however, only be possible if prior to the concentration, Sea-Invest represented such a competitive constraint for EMOEKOM that it prevented EMO-EKOM from increasing its prices and if the concentration eliminated or substantially weakened the competitive constraint exercised by Sea-Invest on EMO-EKOM. The in-depth market investigation, however, demonstrated that neither of these two conditions is fulfilled.

(86) First, it is clear from the analysis of the possibility of customers switching between Antwerp (or even Ghent or Dunkirk) and other ARA terminals that only very marginal volumes of coal or iron ore could be switched from EMO-EKOM to Sea-Invest terminals. Therefore, EMO-EKOM and Sea-Invest are in different geographic markets and any possible constraints exercised by Sea-Invest terminals on EMO-EKOM would be limited to less than 5% of the total volumes of fringe customers handled by EMO-EKOM. Therefore, for customers representing more than 95% of volumes handled at EMO-EKOM, Sea-Invest terminals are not an alternative and cannot be regarded as a competitive constraint.

(87) Further, customers of EMO-EKOM (including those customers that might possibly switch certain volume to ABT) most often identified other terminals in Rotterdam or Amsterdam, in particular OBA or Rietlanden, as possible substitutes. Taking into account the expansion plans at both OBA and Rietlanden and the very limited volumes of the fringe customers able to switch from EMO-EKOM to Sea-Invest terminals, the Amsterdam terminals would have sufficient capacity to accept these customers. In the event of a price increase at EMO-EKOM, it would thus be possible and more reasonable for the fringe customers to switch to terminals other than Sea- Invest terminals. Therefore, the in-depth market investigation confirmed that Sea- Invest is not the closest competitor for EMO-EKOM and that the competitive constraints exercised by Sea-Invest terminals on EMO-EKOM are negligible. Even after the concentration, other terminals in Rotterdam, Amsterdam and Zeeland, in particular OBA and Rietlanden, will exercise a competitive constraint on the pricing of EMO-EKOM.

(88) Second, even if Sea-Invest terminals exercised any competitive constraints on EMOEKOM, it is highly unlikely that the acquisition of joint control would substantially weaken these constraints. The customers' replies in the market investigation suggested that EMO-EKOM does not have any spare capacity. Even though EMOEKOM is expanding its handling capacity, this expansion is aimed at accommodating the increased tonnages resulting from the expected growth in demand in Germany for imported coal. In contrast, ABT seems to already have spare capacity and the tonnages handled by ABT are likely to decrease even further in the future with closures of steel mills using ABT. Furthermore, due to, in particular, hinterland connection limitations, ABT will not profit in any appreciable way from the increased imports of coal to Germany. Therefore, Sea-Invest would have no incentive to relax ABT's competition vis-à-vis EMO-EKOM as it would need to attract coal and iron ore tonnages to Antwerp in order to ensure an efficient level of utilisation of its coal and iron ore handling capacities.

(89) Taking into account the above, the proposed concentration does not lead to any competition concerns due to unilateral effects strengthening the dominance of Sea- Invest or enabling EMO-EKOM to increase its prices.

b) Coordinated effects

(90) The Article 6(1)(c) decision also identified, as a possible competition concern, the coordination of behaviour between Sea-Invest and EMO-EKOM as well as other terminals (OBA, EBS, RBT, OVET) controlled solely or jointly by the shareholders of EMO-EKOM. The decision stated that such coordination could be brought about by the structural links created between Sea-Invest and EMO-EKOM, HES and Manufrance.

(91) However, the competition concerns in the Article 6(1)(c) decision related to the coordinated effects of the proposed concentration in the event that the geographic market covered the whole ARA range. Taking into account the separate geographic markets defined above (45) on the basis of the in-depth investigation, terminals controlled by Sea-Invest, on the one hand, and EMO-EKOM and other terminals controlled by HES and Manufrance, on the other, are active on separate markets. Therefore, anti-competitive coordination of their behaviour due to the proposed concentration is unlikely with respect to the very limited competitive constraints between ABT and other terminals.

(92) To the extent that there is some remaining fringe competition between ABT and other terminals in the ARA range, the in-depth investigation produced no evidence that the proposed concentration might lead to coordinated effects. A number of customers and competitors indicated in the market investigation that there is currently competition between all Dutch ARA terminals of HES and Manufrance. They observed that the management of the terminals is rather independent as regards the operational running of their terminals, including pricing policy towards individual customers or expansion plans. Furthermore, as Sea-Invest exercises only very limited competition constraints on these terminals in the market for handling coal and iron ore, its acquisition of joint control over EMO-EKOM does not seem to give any specific incentive to the other EMO-EKOM shareholders to coordinate their behaviour after the proposed concentration.

(93) The coordination is also made less likely by the fact that the shareholders of EMOEKOM would be rather heterogeneous as regards the scope of their activities and vertical integration. While Sea-Invest and HES are not vertically integrated with the customers for the terminal services, Manufrance (46) is connected with several major customers and TKV is directly part of the ThyssenKrupp group with significant activities in the steel sector. Further, Sea-Invest is active in a number of ports in Belgium, France, Germany and South Africa while HES' handling activities, in particular, are concentrated in the Dutch ports. Therefore, the terminal services market in the ARA range is key for HES' business but relatively less important for Sea-Invest taking into account the number of other geographic areas where it is active. The proposed concentration whereby another customer-shareholder (RAG) is replaced by Sea-Invest, a company focused on terminal services, may increase the weight of mainly stevedoring companies in EMO-EKOM. However the group of shareholders would still remains insufficiently homogenous for coordination to be likely.

(94) The in-depth investigation did not confirm that coordination in the market for coal and iron ore terminal services would be likely with regard to the other criteria defined by the Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings (47) and based on the relevant case law, namely transparency, availability of a deterrent mechanism and reaction of customers and competitors. The relative transparency of the market mentioned in the Article 6(1)(c) decision was not confirmed in the in-depth investigation. Even though the general price level in other terminals may be known to the terminal operators, the investigation showed that prices for individual customers differ significantly even within a single terminal and are not known to the other terminals. Therefore, the transparency of the market as regards price conditions agreed upon in negotiations with individual customers is limited.

(95) Further, due to specificities of the situation of each individual customer, the economic environment in the market for terminal services for coal and iron ore seems to be rather complex, thus making coordination more difficult. Due to particularities of each customer, in particular, as regards geographic location, the terminal services cannot be regarded as homogenous. As was demonstrated in the section on geographic market definition, the suitability of each terminal for customers is largely determined by their geographic location. For each individual customer, terminal services of terminals in different ports are thus, in most cases, not easily substitutable.

(96) The in-depth investigation also did not provide any evidence of the possibility of a credible deterrent mechanism in the event of deviations from co-ordination. The possibility to use veto rights in EMO-EKOM as a retaliation mechanism suggested in the Article 6(1)(c) decision was not upheld by the in-depth investigation. As explained above, (48) the veto rights in EMO-EKOM relate to some strategic decisions and in particular to new investments. Limitation of the investment aimed at expansion of EMO-EKOM by Sea-Invest would necessarily be to the advantage of other terminals in Rotterdam, Amsterdam and Zeeland (including the terminals of HES and Manufrance), which would then capture larger shares of the increased coal imports for Germany. On the other hand, ABT in Antwerp would not gain any substantial volumes due to only marginal substitutability of Antwerp with Rotterdam. Therefore, it is not clear how a threat to veto investment decisions in EMO-EKOM could be used as a deterrent mechanism, at least by Sea-Invest.

(97) As regards the possible reaction of outsiders to coordination between the shareholders of EMO-EKOM, the in-depth market investigation demonstrated the importance of Rietlanden as an independent competitor which has been growing rapidly, apparently also due to its attractive pricing policy. Taking into account the expansion of Rietlanden (with the construction of its new terminal in Afrikaharbour in Amsterdam) which will in the coming years significantly increase its capacity, Rietlanden can be considered as a significant competitive constraint. Further, the customers of coal and iron ore terminals are, in particular, large and sophisticated steel and energy companies. Although they have limited possibilities to switch their whole volumes, their buyer power is enhanced by the relative importance of these customers for each terminal as the five most important customers on average represent around 80% or more of coal and iron ore volumes handled by each of the terminals involved. Therefore, both independent competitors and large customers represent important constraints able to jeopardise the outcome expected from the coordination.

(98) Taking into account the above, the proposed concentration does not lead to competition concerns due to coordinated effects between Sea-Invest, EMO-EKOM, HES and Manufrance.

2. Article 2(4)

(99) Another potential competition concern is coordination in the sense of Article 2(4) of the Merger Regulation between the shareholders of EMO-EKOM in the other dry bulk market. However, EMO-EKOM does not handle other dry bulk and is therefore, on the basis of the market definition retained above, (49) not active in this market. Furthermore, considering the differences between the two markets set out in recitals 17-20 above it is not certain that the handling of other dry bulk is a neighbouring market which is "closely related" to the market for handling of coal and iron ore in the sense of Article 2(5) of the Merger Regulation.

(100) In any case, there are no indications that the entry of Sea-Invest as shareholder in EMO-EKOM could lead to coordination with HES and Manufrance in the other dry bulk market. The fact that Sea-Invest will meet HES and Manufrance in the Supervisory Board of EMO-EKOM and discuss EMO-EKOM's strategy in the market for handling of coal and iron ore with them does not appreciably increase the risk of collusion in the other dry bulk market. Further, there are a number of independent terminal operators handling other dry bulk in Rotterdam and Amsterdam. Consequently, even if the concentration increased the risks of coordination, this would not afford Sea-Invest, HES and Manufrance the possibility of eliminating competition in the other dry bulk market.

B. Terminal services for coal and iron ore for transhipment traffic

(101) Due to the severe draught restrictions and the inland situation of the port of Antwerp, Sea-Invest's ABT terminal in Antwerp does not handle any coal or iron ore for transhipment out of this port. While Sea-Invest's terminal in Dunkirk (Sea-Bulk) does not handle any iron ore for transhipment purposes, it does handle a limited volume of coal for transhipment to one customer in the United Kingdom. This customer is Sea- Invest's only transhipment customer for the United Kingdom. EMO-EKOM, which is not involved in transhipment to the United Kingdom, has only very small transhipment volumes overall. Other terminals which are currently active in transhipment to the United Kingdom are OBA and OVET.

(102) Considering the broad definition of the relevant geographic market, there is currently a small overlap between the transhipment activities of Sea-Invest and EMO-EKOM. The potential effects of this overlap are reduced by the fact there are many deep water ports on the continent which could enter this market. In the case of Sea-Invest's customer in the United Kingdom, it is clear that at least Centre Multivra du Havre ("CIPHA") in Le Havre and Rietlanden in Amsterdam could provide competing services. However, some deep water ports in the United Kingdom could also, in principle, provide the transhipment services concerned. The United Kingdom has at least four ports that can host very large vessels (Hunterston, Port Talbot, Redcar and Immingham), one of which is located on its east coast (Immingham).

(103) Considering the broad scope of the relevant geographic market and the presence of multiple alternatives for the provision of transhipment services, the joint market position of Sea-Invest and EMO-EKOM on the transhipment market is not of such a nature as to be able to significantly impede effective competition.

C. Conclusion

(104) For the reasons set out above it must be concluded that the proposed concentration would not significantly impede effective competition in the Common market or in a substantial part of it, and that it would not result in coordination of the competitive behaviour of undertakings that remain independent in a manner contrary to the criteria of Article 81 of the Treaty. The concentration should therefore be declared compatible with the Common market and with the EEA Agreement,

HAS ADOPTED THIS DECISION:

Article 1

The notified operation whereby Sea-Invest NV acquires joint control within the meaning of Article 3(1)(b) of Regulation (EC) No 139-2004 of the undertakings Europees Massagoed- Overslagbedrijf BV and Erts- en Kolen Overslagbedrijf BV is hereby declared compatible with the Common market and the functioning of the EEA Agreement.

Article 2

This decision is addressed to:

Sea-Invest Arnold & Porter LLP

Skaldenstraat 1 Rue des Colonies, 11

9042 Ghent 1000 Brussels

Belgium Belgium

M. Luc Gyselen

Fax : +32 2 517 66 03

Notes

1 OJ L 24, 29.1.2004, p. 1

2 OJ C ...,...200. , p....

3 OJ C ...,...200. , p....

4 Terminal services are also called stevedoring services. ???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.

5 The other [...]*% are held in nearly equal shares by the Belgian railway SNCB and the private equity company GIMV.

6 The term "EMO-EKOM" will be used throughout the text, even if in some instances only EMO or EKOM might be meant.

7 Including IJmuiden.

8 These are the ports of Vlissingen and Terneuzen.

9 However, according to Manufrance none of these shareholders are controlling ATIC Services.

10 See Article 1 of the second annex of 15 December 1993 to the shareholders agreement of 12 June 1991.

11 See Article 2 of the shareholders agreement of 12 June 1991.

12 See Article 12(7) of the Statutes of EMO-EKOM.

13 See Article 15(7) of the Statutes of EMO-EKOM.

14 See Article 2(4) of the second annex of 15 December 1993 to the shareholders agreement of 12 June 1991. The binding nature of this annex was disputed between the parties. It was clarified that it can be enforced in court.

15 See Articles 18(4), 25(2), 30 (1) and 30(4) of the Statutes of SNV.

16 The substitution of a shareholder changes the quality of control in joint venture. The undertakings concerned are therefore the existing and new controlling shareholders and the joint venture itself. See Commission notice on the concept of undertakings concerned under Council Regulation (EEC) No 4064/89 on the control of concentrations between undertakings (OJ C66, 2.3.1998, p.14) paragraph 44. SNV is disregarded as a pure holding company which serves as an acquisition vehicle, see paragraph 28 of the Notice on the concept of undertakings concerned.

17 Turnover calculated in accordance with Article 5(1) of the Merger Regulation and the Commission Notice on calculation of turnover under Council Regulation (EEC) No 4064-89 on the control of concentrations between undertakings (OJ C66, 2.3.1998, p.25).

18 Sea-Invest EUR [...]* million, EMO-EKOM EUR [...]* million, TKV (including ThyssenKrupp) EUR [...]*, HES EUR [...]*, Manufrance (including ATIC Services) EUR [...]*.

19 Sea-Invest EUR [...]* million; TKV EUR [...]*.

20 Case No JV.55 Hutchison/RCPM/ECT; Case No M.3576 - ECT/PONL/Euromax.

21 Case No M.3884 - ADM Poland/Cefetra/BTZ.

22 Case No JV.55 Hutchison/RCPM/ECT; Case No M.3576 - ECT/PONL/Euromax.

23 According to the results of the market investigation, the total volumes of coal and iron ore represent more than 70% of all dry bulk (excluding agri-bulk) handled in the terminals of the ARA range, that is to say, less than 30% of dry bulk volumes were made up of a large number of different other dry bulk commodities.

24 See recital 17.

25 See recitals 57 ff.

26 Case No JV.55 Hutchison/RCPM/ECT and Case No M.3576 - ECT/PONL/Euromax.

27 Case No JV.55 Hutchison/RCPM/ECT.

28 Other cost elements are port dues or towage and piloting costs, however their relative importance in the total costs of the logistical chain is marginal (in general less than 5%) and, in particular, their difference between different ports in the ARA range compared to the total logistic costs is negligible (for example port dues in Antwerp for dry bulk vessels are cheaper than in Rotterdam by 0.2 EUR/gross ton of the vessel, which according to Sea-Invest translates approximately into around 0.1 EUR/ton of the dry bulk material transported).

29 It should be noted that the figures give maximum draught, but that the situation may differ in different parts of the port and that some terminals in these ports may in fact be more draught restricted because of, for example, location behind the locks or further away from the sea.

30 This share represents only the iron ore handled by terminals active in the open market and therefore excludes captive terminals including the EECV terminal in Rotterdam. If the captive terminals were also included, the share of Rotterdam would be even higher.

31 According to the information provided by the port of Rotterdam, almost 93% of the iron ore volumes discharged in Rotterdam came in vessels exceeding 140.000 dwt and almost 57% in vessels exceeding 200.000 dwt.

32 However, the exact share varies depending on the specific combinations of origins and final destinations of the coal.

33 See recitals 44 ff.

34 Taking into account only limited inland cost data relating to iron ore, the Commission was not able to conduct the same detailed quantitative analysis for iron ore. However, the investigation did not bring any evidence that the situation for iron ore would be significantly different than for coal as also for iron ore the main inland transport means are barges and rail and the inland transport possibilities are thus largely the same.

35 The only exceptions are plants located in the close vicinity of the terminal, where the coal and iron ore is transported directly by a conveyor belt.

36 However, the exact share varies significantly depending on the specific combinations of origins and final destinations of the coal.

37 See recital 53.

38 See the estimate by Interregio, a public platform for cooperation between the provinces of Noord Brabant and Limburg on their website: http://www.interregio.nu.

39 Also called "parcel services". A part cargo service carries cargo of several customers who do not import sufficient quantities to make it economical to charter a vessel independently. A part cargo service is usually semi-scheduled in that it will normally sail between a specific pair of load and discharge ports for which it serves a specific demand.

40 For example, ThyssenKrupp in Rotterdam, Corus in Amsterdam, Arcelor in Dunkirk and Ghent.

41 "FOB" is an International Commercial Term which stands for "free on board". "FOB barge" means that the seller is required to deliver the goods on board of a barge designated by the buyer.

42 Case No JV.55 Hutchison/RCPM/ECT; Case No M.3576 - ECT/PONL/Euromax.

43 Sea-Invest is also dominant if the ports of Ghent and Dunkirk are included in the relevant geographic market because the terminals handling coal and iron ore for non-captive customers are controlled by Sea- Invest. The following analysis would therefore be the same if Ghent and Dunkirk were in the same geographic market as Antwerp.

44 See recitals 27 ff.

45 See recitals 27 ff.

46 Its parent company ATIC Services is also active in coal trading, inland waterway logistic, maritime transportation and quality controls.

47 OJ C 31, 5.2.2004, p. 5.

48 See recitals 74 ff.

49 See recital 17.

Opinion of the Advisory Committee on Concentrations given at its 142nd meeting on 20 July 2006 concerning a draft decision relating to Case COMP/M.3848 - Sea-Invest/EMO-EKOM

Rapporteur : Portugal

1. The Advisory Committee agrees with the Commission that the notified operation constitutes a concentration with a Community dimension within the meaning of Article 1(2) and Article 3(1)(b) of the EC Merger Regulation.

2. The Advisory Committee agrees with the Commission's definitions of the relevant product markets as stated in the draft decision.

3. The Advisory Committee agrees with the Commission's definitions of the relevant geographic markets as stated in the draft decision.

4. The Advisory Committee agrees with the Commission's assessment that the transaction does not lead to any competition concerns due to unilateral effects, strengthening the dominance of Sea-Invest or enabling a price increase by EMO-EKOM, or coordinated effects in terminal services for hinterland traffic for coal and iron ore.

5. The Advisory Committee agrees with the Commission's assessment that the variations in the joint venture do not lead to any competition concerns due to coordinated effects in accordance with Article 2(4) of the Merger Regulation.

6. The Advisory Committee agrees with the Commission that the concentration as notified does not raise serious doubts as to its compatibility with the Common market in accordance with article 10(2) of the Merger Regulation.

7. The Advisory Committee agrees with the Commission that the notified concentration should be declared compatible with the Common Market and with the functioning of the EEA Agreement in accordance with Articles 2(2) and 8(1) of the Merger Regulation and Article 57 of the EEA Agreement.

8. The Advisory Committee asks the Commission to take into account all other points raised during the discussion.

<emplacement tableau>

FINAL REPORT OF THE HEARING OFFICER IN CASE COMP/M.3848 - Sea-Invest / EMO EKOM

(pursuant to Articles 15 and 16 of Commission Decision (2001-462-EC, ECSC) of 23 May 2001 on the terms of reference of Hearing Officers in certain competition proceedings - OJ L162, 19.06.2001, p.21)

On 24 February 2006 the Commission received a notification of a proposed concentration pursuant to Article 4 of Council Regulation (EC) No 139-2004 (the Merger Regulation) by which the undertaking Sea-Invest NV acquires, by way of purchase of shares, joint control of the undertakings Europees Massagoed-Overslagsbedrijf B.V and Erts- en Kolen Overslagsbedrijf BV (EMO-EKOM). Following the transaction EMO-EKOM will be jointly controlled by Sea-Invest NV, Thyssen-Krupp Veerhaven BV, H.E.S. Beheer N.V and Manufrance BV

By decision dated 31 March 2006 the Commission found that the transaction raised serious doubts as to its compatibility with the Common market and the functioning with the EEA Agreement. Accordingly, the Commission initiated proceedings in accordance with Article 6.1(c) of the Merger Regulation. Subsequently, on 3 May 2006, the Commission decided to extend the procedure with 20 working days pursuant to Article 10.3(2) of the Merger Regulation.

On 4 April 2006 Sea-Invest was upon request granted access to certain "key documents" in the Commission's file in accordance with chapter 7.2. of the Commission Note on "Best Practices on the conduct of EC merger control proceedings".

Following an in-depth market investigation the relevant Commission services considered that the serious doubts had been removed and that the proposed transaction would not significantly impede effective competition in the Common market or a substantial part of it. Accordingly, no Statement of Objections was sent to the parties.

No queries or submission have been made to the Hearing Officer by the parties or any other third party. The case does not call for any particular comments as regards the right to be heard.

Brussels, 26 July 2006.

(signed)

Serge Durande