EC, March 18, 2008, No M.4786
COMMISSION OF THE EUROPEAN COMMUNITIES
Judgment
DEUTSCHE BAHN/ TRANSFESA
Dear Sir/Madam,
Subject: Case No COMP/M.4786 - DEUTSCHE BAHN/ TRANSFESA
Notification of 12 February 2008 pursuant to Article 4 of Council Regulation No 139/20041
1. On 12 February 2008, 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 Deutsche Bahn AG ("DB", Germany) acquires within the meaning of Article 3(1)(b) of the Merger Regulation control of the whole of the undertaking Transportes Ferroviarios Especiales, S.A. ("Transfesa", Spain) by way of purchase of shares.
I. THE PARTIES
2. DB is the state-owned German-based railway company engaged in, inter alia, rail passenger transport and (through its subsidiary "Railion") in rail freight transport mainly in Germany, The Netherlands and Denmark, as well as in freight forwarding (by all modes of transport), logistics and ancillary services worldwide (inter alia through its subsidiary "Schenker"). DB via its subsidiary ATG is active in finished vehicle logistics services and via SAR also in car components logistics. Moreover, DB has recently acquired EWS, which is active in rail freight transport in the UK and France.
3. Transfesa is a Spanish-based logistics operator which delivers a full range of freight forwarding and logistics services, mainly within and to and from Spain. Transfesa is primarily active in rail and road freight forwarding and logistics services for automobile parts and components and finished vehicles […]. In addition, Transfesa is active in rail and road freight forwarding and some additional logistics services concerning chemicals and other industrial freight as well as courier and less-than-truck load services by road (through its subsidiary "TDL"). Transfesa does not currently provide rail transport or traction2 itself yet but has acquired the necessary licences […]. It already owns and leases a significant number of wagons. Transfesa also operates the only two axle changing stations between France and Spain.
II. THE OPERATION AND THE CONCENTRATION
4. Pursuant to a Share Purchase agreement entered into by DB and Transfesa on […] 2007, DB intends, through its 100% subsidiary […], to purchase all of the 50.4 % shares in Transfesa which are currently held by Transfesa’s controlling shareholder, the Fernandez family. The other shareholders Renfe and SNCF, who each hold 20.36% (the remainder of the shares being held by the company itself), do not have any legal veto or nomination rights which would exceed the normal role of a minority shareholder or give them joint control over Transfesa.
5. The transaction therefore leads to the acquisition of sole control by DB within the meaning of Article 3(1)(b) of Council Regulation (EC) No 139/2004.
III. COMMUNITY DIMENSION
6. The transaction has a Community dimension pursuant to Article 1(2) of the Merger Regulation. The undertaking concerned have a combined aggregate worldwide turnover in excess of € 5,000 million (DB € 30,297.0 million, Transfesa € 291 million) and a Community-wide turnover in excess of € 250 million (DB € […] million, Transfesa € […] million). [DB and Transfesa do not achieve more than two thirds of their Community-wide turnover within one and the same Member State.]
7. The notified transaction therefore has Community dimension.
IV. RELEVANT MARKETS
8. The proposed transaction affects freight forwarding, logistics and transport services, in particular in relation to the automotive industry, in the EEA, mainly in Spain and Germany.
A. Relevant product markets
A.1. Freight forwarding
9. Freight forwarding has previously been defined as “the organisation of transportation of items (possibly including activities such as customs clearance, warehousing, ground services, etc.) on behalf of customers according to their needs”.3 It has been segmented into (i) domestic and international freight forwarding and (ii) freight forwarding by air, land and sea.4 Further subdivisions for freight forwarding services have up to now been left open.
10. For the purposes of this case the Commission assessed whether a further subdivision of land freight forwarding according to the different modes of transport, in particular rail and road, would be appropriate. The notifying party (DB) reasons that such a distinction between the different modes would not reflect market conditions as freight forwarders could generally organize all modes of transport, especially as regards land transport.
11. As to the further segmentation between freight forwarding by rail and road, the market investigation confirmed that at least a partial substitutability exists between freight forwarding services by rail and road. In any case, since the concentration does not give rise to any competition concern under any alternative market definition, it is not necessary to conclude whether further segmentations of freight forwarding are appropriate.
A.2. Contract logistics
12. Contract logistics services have previously been defined as: “the part of the supply chain process that plans, implements and controls the efficient, effective flow and storage of goods, services and related information from the point of origin to the point of consumption in order to meet customer's requirements”.5 The Commission has pointed out that the important characteristic of general contract logistics is the management of goods for customers as opposed to bare transportation, freight forwarding and warehouse services.6 Contract logistics could therefore constitute a separate market, which is distinct from freight forwarding.7
13. Within contract logistics, the Commission has previously considered a distinction of separate product markets for domestic and cross-border contract logistics services and for lead and traditional logistics providers, but has concluded that such a distinction was unwarranted.8 The Commission has also looked at whether possible further segmentations of the market by industry sector or type of goods handled would be appropriate.9 The exact market definition for contract logistics has, however, been left open.
14. The market investigation confirmed the distinction of contract logistics as a separate market. As it would not change the results of the competitive assessment in the present case, it is not necessary to decide whether contract logistics should be further segmented.
A.3. Alternative market definitions proposed by the notifying party
- Overall market for freight forwarding and logistics
15. DB submits that the relevant product market for this case is the overall market for freight forwarding and logistics that would include freight forwarding services and all other logistics services with the exception of transport of parcels. In particular, all warehousing activities as well as all contract logistics services would be included in the relevant market. DB claims that such an overall market for freight forwarding and logistics would have to be further subdivided into domestic and international services.
16. As mentioned above, the respondents to the market investigation see freight forwarding as a separate market within the overall logistics sector and consider that this market does not include all logistics services as proposed by DB. Since the transaction would not raise any competitive concern under the alternative market definition proposed by DB, it is in any case not necessary to conclude whether the relevant market is an overall market for freight forwarding and logistics.
- Segmentation according to TOP100 Study
17. Alternatively, DB proposes to segment the market for freight forwarding and logistics according to a subdivision introduced by a study on logistic services Die TOP 100 der Logistik.10 This study segments the domestic logistics market into 12 categories, according to the type and volume of the goods transported, and the international logistics market into 3 categories, according to the mode of transport. The parties' activities overlap in 6 of these segments, namely those for national bulk logistics, national direct-load cargo (non-specialised), national tank container and silo transport, other national direct-load cargo using specialised equipment, industrial contract logistics and cross-border transport and forwarding (land-based).
18. In any case, the appropriateness of such a segmentation can be left open as it would not lead to any horizontally affected markets.11
-Segmentation according to type of goods
19. Moreover, DB proposes to further divide the TOP 100 Study's segments for the domestic and cross-border logistics markets by the type of goods that are transported. Following this subdivision, DB finds that their activities mainly overlap in the finished vehicle logistics (a sub-segment of “direct load cargo using specialised equipment”) and in the car components logistics (a sub-segment of “direct-load cargo (non-specialised)”). These two segments which have also been dealt with in previous Commission decisions will be described below.
A.4. Finished Vehicles Logistics (FVL)
20. In previous decisions12, the Commission has found indications for a separate FVL market: FVL services providers are capable of responding to the multiple requirements of car manufacturers all along the car supply chain and thus have a special know-how. A possible FVL market has been limited to services related to passenger cars and light commercial vehicles, excluding services related to trucks and other heavy commercial vehicles (which cannot be transported with the wagons and trucks operated by the parties).
21. DB submits that FVL should not be distinguished as a relevant market. If FVL were to be seen as a distinct market, DB argues that there is no need to differentiate between domestic and cross-border FVL activities or between FVL activities by different modes of transport (rail, road, sea). DB therefore submits that a hypothetical FVL market should not be subdivided into further segments.
22. While not deeming it appropriate, DB considers a potential segmentation of FVL introduced by a decision of the German Bundeskartellamt. The Bundeskartellamt took the view that the FVL business consists of three market segments13: (i) the collection of finished cars (from plant to compound), (ii) compound services (storage and post- production services), and (iii) regional delivery of finished cars (from compound to dealer). The Bundeskartellamt argued that these three services are generally sourced separately and require different kinds of skills and thus should constitute separate markets. However, DB’s view is that this is not generally the case any more, but that many customers source or at least try to source two of these services (e.g. collection and compound or regional delivery and compound) together. It also claims that suppliers try to sell such services together and that most of the major players offer all of these services.
23. The market investigation indicated that FVL could be considered a separate market from general freight forwarding and contract logistics because of the specific demand of customers and that from a suppliers view point dedicated equipment (special wagons or trucks) and specialised know-how is needed. The majority of FVL services providers indicated that a segmentation of FVL between domestic and international services was not appropriate because the same assets and know-how are needed for both types of FVL services.
24. As to a further segmentation of FVL between services by road, rail and sea, the market investigation showed a partial substitutability between the different modes of transport. The respondents as well as DB have provided several examples of switching occurring in the past years for significant volumes. Most car manufacturers use both rail and road but have different modal splits according to their specific needs (with regard to the situation of the individual plant, transported volumes, distance and time constraints). Car manufacturers monitor the price differences between modes and can modify their modal split in consequence. Similarly, FVL service providers would in most cases be active in more than one mode of transport. The decisive factor for the choice of mode for the individual transport would be related also to the mentioned specific needs of the customer and related to the price competitiveness of the offer. However, the investigation showed that for the biggest customers, a minimum level of FVL by rail is necessary because of space constraints within big automotive plants. Current switching possibilities for large volumes seem to be restricted given constraints in existing capacity both for road and rail wagons due to a lack of investment in recent years. Moreover, as rail transport involves significant fixed costs (traction, fees for using tracks), while truck costs are largely variable, the attractiveness of rail vis-à-vis truck transport may be higher for big volumes and long distances.
25. In any case, since the concentration does not give rise to any competition concerns under any alternative market definition, it is not necessary to decide whether FVL is a distinct market and whether further segmentations of FVL are appropriate.
A.5. Car components logistics
26. The Commission has in a previous decision considered logistics services for automotive components (including parts, such as body parts, engine parts, tyres etc. for cars, commercial vehicles, trucks and motorbikes, customers being car and motorbike manufacturers) as a segment of general contract logistics, but not as a separate product market.14
27. DB argues that car components logistics should not be distinguished from general direct- load cargo logistics, because components are transported in the same railway wagons and trucks as other direct-load cargo products.
28. While this view was supported by the majority of competitors, respondents mentioned the existence of specialised assets (mega trailers) that give competitive advantages for the transport of car components in comparison with traditional cargo logistics equipments. If car components logistics were to be considered as a separate market, the market investigation indicated that there is partial substitutability between car components logistics by road and rail. The investigation showed that most car manufacturers are able to switch at least parts of their car components between the two modes of transport. Further, parts of the specialised assets (in particular swap bodies) can be used for road as well as for rail transport.
29. In any case, since the concentration does not give rise to any competition concerns under any alternative market definition, it is not necessary to decide whether car components logistics is a distinct market and whether further segmentations of car components logistics are appropriate.
A.6.The Provision of Transport Services
30. In previous cases the Commission has found that the provision of transport services could be considered a relevant market distinct from freight forwarding services. Indeed, providers of freight forwarding and transport services do not primarily compete with each other and freight forwarders are generally considered to offer a distinct service to customers. Freight forwards often also sub-contract the actual transportation to specialist transport providers.15
31. In addition to the distinction between freight forwarding and the actual transport of goods, the Commission has found that not all modes of transport are generally substitutable for all types of goods in view of the geographic situation of the customer and the specific characteristics of the goods to be transported.16
32. In this regard, the Commission has found in previous cases that there are indications that a distinct market for the transport of goods by rail (so-called rail freight services) could be distinguished, in particular for certain goods. However, the Commission has not taken a definitive position on such a segmentation as often this has to be decided on a case-by-case basis depending on the type of transported goods, and other factors such as the volumes and weight transported, the distance to be transported etc.17
33. As to a possible narrower definition of the rail freight services market, the Commission has not excluded a subdivision between domestic and international services, between single wagon and block train services18 or according to type of goods to be transported.19
34. In DB's view, the provision of transport services should not be considered a separate market in its own right, since this activity is included as an input into the overall freight forwarding and logistics market. However, in order to abide by the Commission's previous decisions, they provided information also on separate markets for road and rail transport and for domestic and cross-border transport in their competitive assessment.
35. The market investigation in the current case has largely confirmed that transport services should be considered a distinct market from freight forwarding and other logistics services. Whether this market should be further segmented on the basis of the type of mode used can be left open in this case, since the proposed concentration does not raise competitive concerns under any alternative market definition.
A.7. Axle-changing services
36. As there is a difference between the gauges of the Spanish and Continental European rail networks, rail freight services from and to Spain are assured by using two different possibilities: using axle-changing or transhipment. Transfesa operates the two existing axle-changing stations whereas the different transhipment stations are operated by the Spanish rail operator Renfe and the French rail operator SNCF.
37. DB takes the view that axle changing stations and transhipment facilities are substitutable to one another. Moreover, it argues that transhipment from trains to trucks also exerts a competitive constraint on axle-changing stations. Therefore, DB does not regard the axle changing stations as essential facilities.
38. The market investigation indicated that there is partial substitutability between axle- changing and transhipment stations. Some FVL service providers which currently use axle-changing have indicated that they were considering switching to transhipment and vice-versa.20 In general, axle-changing is faster and more efficient but also more expensive than transhipment. Moreover, if switching to axle-changing requires an investment in axle-changing wagons, the opposite move does not require specific investments from the transporters point of view.
39. In any case, whether the market for axle-changing services should be defined as a separate market can be left open since the concentration does not raise vertical competitive concerns under any alternative market definition.
B. Relevant geographic markets
B.1. Freight forwarding
40. Although there are indications that the market for freight forwarding services may be wider than national21, the Commission has so far left the relevant geographic market definition open.
41. DB suggests that the geographic market for domestic freight forwarding services is national. Similarly, DB takes the view that there are still national markets for the provision of international freight forwarding services, because most players, in accordance with the customers’ preferences, have a particular focus on outbound freight forwarding services from their home country. Only few clients seem to tender contracts for international freight forwarding at least for certain goods on a wider than national basis. If one were to distinguish freight forwarding services according to the mode of transport, DB submits that these narrower hypothetical freight forwarding markets would also be national.
42. DB submits that it does not make sense to use an origin and destination (“O&D”) approach in order to define the markets for international freight forwarding, because the required know-how does not differ regarding different town-to-town or country-to- country routes. DB takes the view that it applies to all hypothetical subdivisions of the freight forwarding markets (distinction freight forwarding/contract logistics, segmentation by mode of transport, FVL, car components logistics). In particular, it is submitted that in spite of the fact that there are special gauges for the Iberian peninsula it would not be appropriate to consider a distinct market for rail freight forwarding from and to Spain.
43. The market investigation indicated that the provision of freight forwarding services could have a national or wider (EEA wide) dimension. Most respondents take however the view that due to the specificities of the Spanish railways there is a separate market for rail freight forwarding from and to Spain.
44. In any case, since the concentration does not give rise to any competition concerns under any alternative geographic market definition, the geographic dimension of the freight forwarding market can be left open.
B.2. General Contract Logistics Services
45. As concerns the relevant geographic market for contract logistics and any sub-segments thereof, the Commission has previously found so far that such markets would still be national in scope,22 but has also acknowledged that there is an increasing demand for cross- border logistic services where the location of the provider is less important. The precise scope of the relevant geographic market for general contract logistics has, however, been left open.
46. The market investigation in the present case was not conclusive in this respect. However, since the concentration does not give rise to competition concerns under any alternative geographic market definition, the geographic dimension of the general contract logistics market can be left open.
B.3. Alternative market definitions proposed by the notifying party
47. DB submits that the markets for domestic freight forwarding and logistics services as well as international freight forwarding and logistics services are national in scope. If one were to segment these markets according to the TOP 100 Study segmentation, the parties submit that all the segments of the TOP 100 Study would also be national.
48. In any case, since the concentration does not give rise to any competition concerns under any alternative geographic market definition, the geographic dimension of the alternative market definitions proposed by the parties can be left open.
B.4. Finished Vehicles Logistics (FVL)
49. In previous decisions, the Commission considered the FVL segment as national but did not conclude on the geographic dimension.23
50. DB submits that the FVL market would have an EEA-wide dimension, because all major players are active across Europe and would be able to serve any customer in Europe, and all major contracts are tendered EU-wide. If one were to segment the FVL business, the parties submit that (i) the collection segment is EEA-wide, and (ii) the dimension of the compound services and delivery segments can be left open (EEA, national or regional).
51. The market investigation confirmed that the market for FVL services may be wider than national and that wagons can in principle be used across most EEA countries. At the same time, economies of scale are more easily exploited by concentrating on particular routes and regions, to ensure an efficient usage of capacities.24 Most respondents also stressed the specificities of the Spanish railways, implying that the provision of FVL from and to Spain could be regarded as a possible separate market. Given the particularities of the FVL market, the investigation also provided indications for a possible further segmentation of inbound and outbound FVL services to and from Spain (see further below).
52. In any case, since the proposed concentration does not give rise to any competition concerns under any alternative geographic market definition, the geographic dimension of FVL services can be left open.
B.5. Car components logistics
53. The Commission in previous decisions considered logistic services for automotive components as a segment of general contract logistics, and therefore the relevant geographic market as national, although it left the relevant market definition open. 25
54. DB submits that the car components market would have a national dimension as part of the national direct cargo market.26
55. The market investigation provided indications that the geographic market for logistic services for automotive components is wider than national. The majority of companies providing freight forwarding services in this market would even perceive it as EEA- wide. Some car manufacturers pointed out that the region where the components are to be collected plays an important role. The fact that both the country of origin and the country of destination, constitute a decisive factor when selecting a service provider for most car manufacturers, may also be in favour of a wider than national geographic market for logistic services for automotive components.
56. In any case, since the concentration does not give rise to any competition concerns under any alternative geographic market definition, the geographic dimension of car components logistics can be left open.
B.6. Transport
57. As for the geographic market definitions of rail freight services the Commission has previously considered that in view of different technical and regulatory requirements not only the markets for domestic but also for international rail transport (traction) services (which would include traction) could be considered national in scope.27 In certain situations the Commission also found that an O&D-based approach might best reflect the competitive situation. It has also considered a wider than national market (cross-border regional).28 The Commission has indicated that the relevant geographic market definition of other modes of transport, in particular road transport, might be different and wider than national in scope.
58. DB considers, referring to Case COMP/M.4746 Deutsche Bahn/EWS that the relevant geographic market for rail transport services (domestic and international) is national in scope. The market investigation in this case largely confirmed this view.
59. As regards road transport services, DB considers that the relevant geographic markets are wider than national.With regard to road transport, the Commission has found indications that the market for road transport services might be wider than national and even EEA- wide.
60. However, given that the proposed concentration does not give rise to any competition concerns under any alternative geographic market definition, the geographic dimension of the alternative transport markets can be left open in this case.
B.7. Axle-changing services
61. For the purposes of this case the geographic dimension of a hypothetical market for axle-changing services would be limited to the French-Spanish border.
V. COMPETITIVE ASSESSMENT29
A. Horizontal overlaps
62. The proposed transaction leads to overlaps in DB's and Transfesa's activities only with regard to land freight forwarding services, mainly to automotive customers, and in particular for outbound FVL services from Spain (see further below).
A.1. Freight forwarding
A.1.1 Domestic and international markets for land based freight forwarding
63. On the national markets for domestic freight forwarding services, the proposed operation would lead to minor overlaps for land based freight forwarding in Germany and Spain, where both DB and Transfesa are active in domestic freight forwarding. Similarly, on a national market for international freight forwarding services, minor horizontal overlaps would occur for land-based freight forwarding services. On all alternative markets, the parties' combined market shares would be below 15% in all Member States. Thus, on such hypothetical market definitions the proposed transaction would not lead to horizontally affected markets and hence does not raise serious doubts as to its compatibility with the common market. .
A.1.2 Domestic and international markets for freight forwarding by rail
64. On a hypothetical national market for domestic and international freight forwarding by rail the proposed transaction would lead to affected markets in 3 Member States.30
65. In Germany the parties would have an estimated combined market share of around [40- 50]% for domestic and [40-50]% for international freight forwarding services by rail (with an increment due to Transfesa of [<1]% and below [0-5]% respectively). In The Netherlands the combined share in international freight forwarding by rail would be around [20-30]% (Transfesa [0-5]%, not active in domestic freight forwarding by rail). In Italy the parties' combined share in international freight forwarding by rail would be [10-20]% (Transfesa around [0-5]%, not active in domestic freight forwarding by rail). For a hypothetical EEA-wide market DB estimates a combined market share smaller than [30-40]% (with Transfesa smaller than [0-5]%).31
66. As can be seen from the market shares, the increment on these hypothetical markets would be small. Also the market investigation did not raise any concerns with regard to the parties' position in freight forwarding by rail, nor were concerns raised with regard to the smallest possible geographic dimension which would be from and to Spain (for FVL services see A.4 below). In Germany where DB alone has already significant market shares, its position is mainly due to the fact that DB, as the former incumbent for rail freight transport, has many direct contacts to customers and therefore an important position in rail freight forwarding as well. As DB already provides traction services for Transfesa's customers the concentration mainly leads to a vertical integration of these indirect customer relationships.32
67. On the basis of the above, the proposed concentration does not raise serious doubts due to the overlap in the parties' activities on a hypothetical market for rail based freight forwarding services.
A.2. General Contract logistics Services
68. The proposed transaction does not lead to any significant overlaps in the parties activities for general contract logistic services. Both in Spain, the only country where Transfesa is active in general contract logistics services, and on a hypothetical EEA-wide market, the combined market share of the parties would remain below 15%. The proposed transaction does therefore not raise serious doubts for contract logistic services.
A.3. Alternative market definitions proposed by the parties
69. Under the parties’ proposed market definition of an overall market for freight forwarding and logistics, the combined market share of the parties would remain below 15% post-merger in all countries where the parties' activities overlap, therefore not leading to any affected markets. The result would not be altered on any alternative segmentation of the product market according to the TOP 100 study. The proposed transaction does therefore not raise serious doubts on such alternative market definitions.
A.4. Finished Vehicles Logistics (FVL) services
70. The parties' activities overlap in FVL services. The provision of FVL services is one of Transfesa's main business activities where it generates around […]% of its overall turnover.
- Compound services and delivery
71. If the FVL market were to be further subdivided by distinguishing between the collection of finished vehicles, compound services and delivery, no affected market would exist with regard to compound services and delivery. Indeed, there would be no overlap on any national markets because Transfesa is only active in compound services and delivery in Spain (9 of 73 compounds, [5-10]% of deliveries), where DB is not active.33 On hypothetical EEA-wide markets for compound services and delivery, the combined market shares of the parties do not give rise to any competitive concern.
- Collection of finished vehicles by road and rail
72. As for finished vehicles collection, based on the number of transported units by road and rail the parties would have a combined market share of around [10-20]% (DB [10-20]%, Transfesa [0-5]%) on an EEA-wide market. If the market were to be considered as national they would have the following market shares with only very small increments: Spain [10- 20]% (DB [0-5]%, Transfesa [10-20]%), Germany [20-30]% (DB [20-30] %, Transfesa [0- 5] %), Poland [20-30]% (DB [20-30]%, Transfesa [0-5]%). According to DB the market is rather fragmented with several competitors of similar market shares, including CAT (former Renault-subsidiary) with an estimated market share of [10-20]%, STVA/SNCF with [5-10]%, GEFCO/PSA, Altmann, Mosolf, SITFA with market shares between [5- 10]% each, and around a dozen other players with market shares between [0-5]%.
- EEA-wide market for FVL collection by rail
73. DB has put forward that one would consider a separate market for FVL collection by rail it would be an EEA-wide market. While DB does not provide market shares for FVL by rail, it is submitted that the parties operate around [30-40]% of the specialised wagons for FVL in the EEA ([20-30]% for DB and [5-10]% for Transfesa). The market investigation largely confirmed these estimates based on the total capacity for specialised FVL wagons.
74. According to DB, post-transaction the merged entity will continue to compete with a number of FVL providers active in the provision of FVL services by rail, including FVL service providers with own special wagons such as PSA's subsidiary GEFCO with around [20-30]% of EEA capacity, SNCF's subsidiary STVA with [10-20]%, SITFA with [10- 20]%, Altmann with [5-10]%, Pecovasa with [0-5]% and Mosolf with [0-5]%. This has also been largely confirmed by the market investigation. .34
75. As mentioned above, in previous decisions the Commission found that rail freight service markets still tend to be national35. In certain situations the Commission also found that an O&D-based approach might best reflect the competitive situation.36 In DB/EWS it was found that due to technical and procedural barriers and the need for specially trained staff to provide cross-border services, the relevant geographic market for rail freight transport seems to be national. It may become international on certain routes which are part of a corridor37.
76. Similarly also in the case of a market for FVL collection by rail the geographic dimension can only be decided upon on a case-by-case basis. Wagons can in principle be used across most EEA countries. However, economies of scale are more easily exploited by concentrating on particular routes and regions, to ensure an efficient usage of capacities. For instance, the operation of an FVL collection by rail in one direction (i.e. from Spain to Germany) would be more economically viable if comparable traffic can also be found in the opposite direction (i.e. from Germany to Spain). Otherwise, transportation cost increase significantly due to the need for running empty capacity on the return. As such providers aims at providing similar volumes on both directions. However, the Commission's investigation has also shown that this is not always the case.
77. During the market investigation concerns were raised that the availability of specialised wagons for the transport of FVL by rail in the EEA would be a key element for customers' ability to switch FVL provider. It was argued that there is a general shortage of such capacity within the EEA and that this applies in a similar way to specialised trucks for transport by road. However, customers' concerns about limited capacity were not backed by freight forwarders active in the provision of FVL. While confirming the current scarcity of wagons and trucks and the waiting times for new investments in capacities (approximately 1-2 years for wagons and trucks), they stressed that the industry is cyclical and over- and under-capacity in this industry fluctuates over time.
78. As in the DB/EWS decision38, with regard to a subdivision of the market for transport of goods according to different transport means (rail, road, barge, sea etc), the market investigation showed that also such a division has to be assessed on a case-by-case basis.
79. While there may be strong substitutability for some types of contracts, as evidenced by customers switching between rail and road, according to the market investigation, this may be different if there is a need to transport large quantities over long distances. In the latter case, point-to-point transport by block trains may be more efficient. By way of contrast, road may have a particular advantage for transport which involves high frequency, short distances and local distribution39.
80. In spite of capacity constraints, customers confirmed that they would be able to switch at least parts of their volume to another supplier or, more easily even, to road. The amount that may be switched and the implication on cost would be a complex calculation based on a variable of factors for each individual customer. While for some customers switching would not be possible for certain parts of their volumes and involve significant cost (mainly high volumes from a specific plant and long-distance transport), other customers took the view that switching to another rail provider or to road would raise their costs only slightly or even decrease costs for volumes which for other than cost reasons are transported by rail. DB submits not only that rail can fully be substituted by road which, if considered as a separate market, would still act as a very competitive constraint but it also provides examples of customers who switched significant volumes from rail to road. This has been confirmed by switching examples provided by third parties. Altogether it can be concluded that the possibility of road transport exerts an important competitive constraint on rail FVL.
81. It also appears that car manufacturers have some bargaining power over their FVL providers and can thus offset possible increase in prices by switching to road and/or move capacity to other providers as indicated above. This is illustrated by the fact that Transfesa achieves […]% of its FVL turnover with only three large customers ([…]% with 5 customers). For DB the three major FVL customers accounted for […]%, the Top 5 for […]%. The relationship with some customers may also have taken years to build up, which means that equipment and transport needs have been built up by the car manufacturer and the FVL provider together. Following from this is a certain mutual dependence between car manufacturers and FVL providers, especially when a FVL provider has done substantial investments needed by a given customer that cannot, or is less suited to the needs of other customers. In such a situation, the FVL provider and its customer would incur high costs in case of switching.
82. Another possible reaction of customers would be the sponsoring of entry or investment in new capacity: a number of customers confirmed that they would be prepared to consider a longer contract period (up to 5 years) to raise incentives for freight forwarders to invest in the necessary equipment.
83. In view of the existence of a number of actual or potential competitors (freight forwarders), the competitive constraints from road and the fact that the market is characterised by a limited number of large customers with very specific needs and considerable know-how in logistics, the proposed transaction does not raise serious doubts on an EEA-wide market for finished vehicles collection by rail.
- - Spanish market for FVL collection by rail
84. Due to the specificities of the Spanish rail network (different gauge to the Continental European rail network) and the subsequent need to use different wagons within Spain than in the rest of the EEA, and that both parties are active in Spain, the Commission also considered the possible effects of the proposed transaction on a hypothetical market of FVL services relating to Spain (inbound, outbound and domestic services) overall and in particular for outbound FVL by rail (the narrowest possible market).
85. Both Transfesa and DB are providing FVL services in Spain. Transfesa provides domestic and cross-border (inbound and outbound) FVL services both by rail and road and with its own rail wagons. DB is only active in organising the outbound collection of finished vehicles by rail relying on a sub-contractor for the Spanish leg.
86. On the overall market for the collection of finished vehicles in Spain (including domestic and cross-border operations by road, rail and sea) the parties estimate their total market share to be around [10-20] % with only a marginal increment of <1% added by DB. This is based on the number of transported units.
a) Outbound FVL by rail
87. Due to DB's limited activities, the narrowest hypothetical market, where the parties' activities overlap is the outbound collection of finished vehicles by rail from Spain. In terms of destinations/routes the parties' overlap is limited to the routes from Spain to Belgium and from Spain to The Netherlands. DB's activities on these routes are limited to only […] vehicles based on the annualized volume of current contracts. Transfesa serves significantly more destinations (also for inbound routes), including the more important (higher volume) routes to Germany, France and Italy.
88. In 2007 DB organised the outbound transport of around […] vehicles and Transfesa of […] vehicles. DB submits that they are unable to provide market shares for a hypothetical market for the collection of finished vehicles in Spain by rail only, although it is acknowledged that to their knowledge only DB and Transfesa currently offer outbound FVL rail only services. There is, however, an important difference between the activities of the parties. Whereas Transfesa provides these outbound services using their own wagons, DB does not own any wagons compatible with the Iberian gauge. It therefore sub-contracts for the Spanish leg of the transport with the Spanish freight forwarder/wagon supplier, […], which in turns provides the wagons.
89. As regards the FVL market by rail, according to the information submitted by DB, Transfesa owns around [50-60]% of the approximately […] rail wagons with interchangeable axles and around [40-50]% of rail wagons with only the Iberian gauge, which can be used for FVL in Spain. Transfesa is not a rail operator, and traction (the actual transport by rail) in Spain is provided by the incumbent rail operator RENFE for the FVL movements organised by both Transfesa and DB.
90. At the border, Transfesa is using the axle changing station to cross the Spanish/French boarder, while DB is using transhipment.
91. As DB uses transhipment at the Spanish border instead of Transfesa's axle-changing facilities and DB does not own any wagons compatible with the Iberian gauge, the proposed transaction has no material impact on the availability of any material assets to provide FVL rail services in Spain. Furthermore, according to DB, and largely confirmed by the Commission's investigation, in addition to Transfesa other companies like Renfe, STVA, Pecovasa, GEFCO and LTF have the special wagons used for the transport of finished vehicles by rail in Spain. It has also been confirmed in the market investigation that there is some spare capacity in the Spanish market for such wagons.
b) Competitive constraint by road
92. In support of the competitive constraint stemming from other modes of transport, DB submits that […] customers of Transfesa and DB use at the same time transports by road and other modes for their outbound cross-border traffic from Spain. The combined activities relating to outbound collection of finished vehicles by rail should therefore be seen in the context of the overall number of vehicles transported from Spain: in 2007 the parties organised outbound FVL for only around […] vehicles out of a total of around 1.8 million vehicles40 transported from Spain to Europe.
93. DB further submits that Transfesa's major Spanish customers of FVL services use competitors' services in road transport ([…]: Setram; […]: Sintax and Toquero; […]: Tradisa and Capsa; […]: CAT). Furthermore, to support their claim of substitutability with and the competitive constraint exercised by other means of transport DB provided recent examples showing that Transfesa and another competitor lost actual contracts and/or lost the tender for substantial volumes of outbound traffic contracted by major car manufacturers to other means of transport, in particular to road and sea. […].
94. The Commission's market investigation has also shown that significant volumes have been shifted from rail to other modes of transport, and that this has been done for smaller and larger volumes and for shorter and longer distances indicating that other modes of transport provide a competitive constraint to FVL services by rail.
c) Conclusion on the Spanish rail FVL market
95. Based on i) the limited geographic overlap in the parties' activities and in particular DB's limited presence on the routes from Spain; ii) the considerable difference between the size and scope of the parties' FVL operations in Spain, iii) the reliance of DB on a domestic provider for rail-based FVL and the resulting absence of any effect on the available capacity for rail-based transports in Spain; and iv) the evidence of substitutability with other modes of transport, the Commission does not consider it likely that the competitive constraint exercised by DB on Transfesa would be higher than the competitive constraint exercised by other competitors. In view of this the proposed concentration does not raise serious doubts on a hypothetical market for FVL services relating to Spain.
A.5. Car components logistics
96. For car components logistics where Transfesa generates […]% of its overall turnover the parties submit that they do not have data on the total market volumes. They submit however that to their best estimates the parties do not have a combined market share of 15% or more, basing themselves on the TOP 100 study.
97. During the market investigation concerns were raised that also for car components logistics the merged entity might possess an important competitive advantage consisting of special assets.
98. Respondents to the market investigation mentioned that not only for FVL but also for the transportation of car components, specialised wagons are used, which are particularly adapted to the transport of higher volumes and often to the customers' needs. One automotive customer sponsored the investment into such wagons tailored for its needs in the beginning of a long-term relationship. Several respondents claimed that the merged entity would have an important share of specialised wagons which would make switching to other providers difficult.
99. Car components can in principle be transported either by swap bodies/trailers which can also be used for other types of direct cargo and for road transport or by special fixed covered wagons for rail. DB submits that these loading units and wagons are not specially designed for the transport of car components and can also be used for other goods. While it acknowledges that there are wagons used for car components, which are higher than normal and can therefore transport higher volumes, it also argues that there are other alternatives such as maritime containers and 'normal' wagons with sliding walls which are substitutable with these special assets.
100. This is in line with the result of the market investigation that specially adapted wagons make the transport of car components particularly efficient but that other options like swap bodies can also be used. It was also mentioned that containers may be used, although these cannot be loaded from the side and therefore have certain disadvantages. However, they would still represent an alternative in case of capacity constraints. Moreover, it appears that such specialised wagons would be needed only for parts of the overall car components logistics chain, while for other parts the general cargo equipment would be sufficient. Several customers also indicated that there are in principle no special assets required for providing car components logistics. Overall, it can be concluded from the market investigation that specialised wagons used to transport car components by rail, as the parties provide, rather represent a competitive advantage lowering unit costs for transport by rail than an asset which is essential for providing car component logistics.
101. According to DB there are important other providers of specialised equipment for car components such as Renfe and SNCF, rental companies like GE Capital/Tip, Transwaggon, AAE and Volkswagen for swap bodies and special trailers, which are mainly used for road, but could also be used for rail. Also freight forwarders such as Ewals and Schnellecke are mentioned. It also claims that swap bodies/trailers and special wagons could be bought or rented on the market. Estimated delivery times for newly manufactured wagons would be 1- 2 years, for swap bodies only 3-4 months. Rented wagons would be available within weeks.
102. The market investigation confirmed that there are other providers who have own wagons, which are particularly adapted to the transport of car components. Wagons would also be available from specialised rental companies. While some respondents indicated that renting may be subject to higher rates because of current capacity constraints, a majority of respondents considers that the rental of special wagons is possible under current market conditions. In addition, the time required to purchase such wagons would not be different from other types of wagons. Some automotive customers also confirmed that they would consider sponsoring investments by agreeing on contracts for up to 5 years duration.
103. In view of the above it seems unlikely that the merged entity would have special assets which would provide it with such a competitive advantage as to raise competition concerns. They would not only be in competition with other specialised providers of car components logistics but also of general contract or cargo logistics which could enter this segment while using these assets also for other types of cargo.
104. Moreover, as for FVL there is a competitive constraint from road transport. According to DB the modal split as regards car components is around 20% for rail compared to 80% for road. Some customers state that they can rather easily and quickly switch between rail and road services, others can switch only part of their volumes but would rely on rail for large volumes where the use of block trains makes rail more efficient. Cost and time for switching very much depend on the situation of the individual customer (including the organisation of the specific plant, geographic situation, the type of component, time constraints). It was also mentioned that for car components logistics just-in-time delivery is particularly important. Given that some respondents consider that road may have advantages in speed and reliability, they might see road transport as more appropriate for deliveries where these factors are particularly important. On the basis of the Commission's investigation it appears that for car components the competitive constraint by road is at least as significant as for FVL logistics.
105. Due to the competitive constraint of other competitors, general cargo logistics and of road transport, the proposed concentration does not raise serious doubts on the hypothetical market for car components logistics by rail.
A.6. Rail freight transport
106. Transfesa currently is not active in the provision of rail freight transport services since it does not yet provide rail traction but only wagons for the transport of certain goods of its customers (mainly finished vehicles and car components).41 Transfesa has obtained a licence for railway services in Spain […]. […]. Transfesa also offers transport services by road but its turnover is very small (€ [less than 10] million) on this market and does not request further investigation. DB currently is not active in rail freight transport or traction in Spain.
107. Therefore, even if DB, through EWS, could be considered as a potential entrant into the Spanish rail market this would not lead to an overlap on the (merchant) market for rail traction or rail transport in the near future. There is however a vertical relationship between the rail transport services of DB (via Railion) and the freight forwarding services (and its sub-segments) in which both DB and Transfesa are active.
B. Vertical links
108. As can be seen from the Table below DB's market shares of above 25% in rail freight transport lead to vertically affected markets in three Member States; Germany, the UK and The Netherlands. 42
Table
Rail freight | Domestic | International |
Germany | Approx. [80-90]% | Approx. [80-90]% |
UK | Approx. [50-60]% | Approx. [90-100]% |
The Netherlands | Approx. [90-100]% | Approx. [70-80]% |
Customer foreclosure
109. The Commission examined to what extent the merged entity would have the incentive and ability to use its market strength to foreclose other rail transport companies from freight forwarding and logistics customers who purchase rail transport services.
110. The parties' combined market shares on freight forwarding markets downstream of these rail activities would be higher in rail-based freight forwarding than in overall freight forwarding or freight forwarding by other means of transport. The vertical link between traction and freight forwarding is also most apparent in rail-based freight forwarding services. Transfesa is not active in domestic rail-based freight-forwarding in the UK and the Netherlands and has only negligible activities (less than 1% market share) in Germany. It is therefore with respect to cross-border rail-based freight forwarding, where the discussion on customer foreclosure is most relevant
111. In the UK, the parties would only account for around [5-10]% of the cross-border freight forwarding market. However, the investigation has shown that due to the lack of market power downstream, the merged entity's competitors in the UK would still have access to a sufficiently large customer base.
112. Due to the strong presence of DB, in Germany and The Netherlands, the parties' market shares in cross-border rail-based freight forwarding would be around [40-50]% and [20- 30]% respectively. DB was already vertically integrated in these countries, and the increment of around [0-5]% (The Netherlands) and [0-5]% (Germany) added by Transfesa is unlikely to significantly change the incentives of the merged entity post merger. Furthermore, as Transfesa already purchases rail transport/traction in Germany to a large extent from DB, the merger would not materially change the situation for DB’s competitors in rail transport.
Input foreclosure
113. It was further examined whether post merger competitors from DB and Transfesa in the various freight forwarding markets could suffer from an input foreclosure from traction services by DB.
114. It could be conceived that DB might serve Transfesa’s competitors only at a higher price or with a lower-quality service. However, even though theoretically possible, such quality foreclosure seems unlikely.
115. Although traction could be considered as an important input for some of Transfesa's rail-based activities (rail freight forwarding or FVL by rail), it should be noted that DB is already a vertically integrated player and is present in downstream markets in Germany and in The Netherlands and the transaction is unlikely to change its incentives in these Member States.
116. DB (through EWS) is also present in the provision of traction in the UK, but it is not active in FVL. The importance of rail traction as an input is most pronounced for long- haul, large volume transports, for example cross-border transports of finished vehicles. In terms of viable modes of such transports in the UK, the well-developed port and shipping network and the fact that short sea shipping is accessible by short-distance transports should be assessed in view of the bottleneck on cross-border rail transport, namely the Channel Tunnel. The competitive constraint stemming from other modes of transport therefore would be likely to make a foreclosure strategy more difficult in the UK. Since the merged entity is active in the downstream markets in other parts of the EEA, its incentives to foreclose customers in the UK may also be weakened by the presence of such customers elsewhere in the EEA. Finally, no customers have raised any concerns relating to input foreclosure in the UK.
Access to axle changing stations
117. During the market investigation it has been raised that the transaction would provide DB with control of the two axle changing stations at the border between France and Spain, which might be considered as an essential facility for rail transport and rail based freight forwarding between these countries if transhipment was not to be considered as a full substitute.
118. Transfesa manages the axle changing stations but those are located on land is owned by SNCF. […]
119. Furthermore, the operational agreements linked with the rental agreements provide that Transfesa has to grant access to the axle changing services to third parties (including SNCF) at the price set by SNCF and in a non-discriminatory basis. […].
120. In view of this contractual situation it can be concluded that the merged entity will not have full control of the axle changing stations and that SNCF's countervailing influence will prevent any foreclosure strategy to the detriment of third parties.
Conclusion on vertical effects
121. It results from the above that the proposed concentration has no vertical effects that raise serious doubts about its compatibility with the common market.
VII. CONCLUSION
122. For the above reasons, the Commission has decided not to oppose the notified operation and to declare it compatible with the common market and with the EEA Agreement. This decision is adopted in application of Article 6(1)(b) of Council Regulation (EC) No 139/2004.
1 OJ L 24, 29.1.2004 p. 1.
2 According to the notifying party 'traction' is defined as the provision of a locomotive and a locomotive
driver for the physical movement of railway wagons (be it loaded or unloaded). 'Transport' is defined as
the physical movement of goods (via any means of transport) including the provision of those services
necessary for the physical movement. In the case of rail transport, this includes the provision of traction,
the provision of rail wagons and hiring the infrastructure capacities such as train pass, energy and access
to marshalling yards. This is in line with the definition of traction services in a previous Commission
decision (Commission decision GVG/FS of 27.08.2003, OJ L 11/17 of 16.01.2004, para. 51-56), which
however clarified that traction also includes the ancillary service of locomotive and driver back-up,
maintenance and repair services. Traction services can only be provided by licensed railway companies.
According to this decision the market for traction is therefore a separate market from the market for the
renting or purchasing of locomotives.
3 Case COMP/M.4045 DB / BAX Global, Case COMP/M.3971 Deutsche Post / Exel, Case COMP/M.3603
UPS / Melto, M. 3496 TNT Forwarding Holding / Wilson Logistics, Case COMP/M.3155 Deutsche Post /
Securicor, Case COMP/M.2908 Deutsche Post / DHL and Case COMP/M.1794 Deutsche Post / Air
Express International.
4 Case COMP/M.4746 Deutsche Bahn / EWS, Case COMP/M.4045 DB / BAX Global, Case
COMP/M.3971 Deutsche Post / Exel and Case COMP/M.1794 Deutsche Post / Air Express International.
5 Case COMP/M.3971 Deutsche Post / Exel and Case COMP/M.3496 TNT Forwarding Holding / Wilson
Logistics.
6 Case COMP/M.2831 DSV / TNT Logistics / DSV Logistics, Case COMP/M.1895 Ocean Group / Exel,
Case IV/M. 1405 TNT Post Group / Jet Services and IV/M 1500 TPG / Technologistica.
7 Case COMP/M.3971 Deutsche Post / Exel, Case COMP/M.2831 DSV / TNT Logistics / DSV Logistics,
Case COMP/M.1895 Ocean Group / Exel and Case IV/M. 1405 TNT Post Group / Jet Services.
8 Case COMP/M.3971 Deutsche Post / Exel, Case COMP/M.1895 Ocean Group / Exel.
9 Case COMP/M.3971 Deutsche Post / Exel, Case COMP/M.2831 DSV / TNT Logistics / DSV Logistics,
Case COMP/M.2411 Autologic / TNT / Wallenius / CAT and Case COMP/M.1895 Ocean Group / Exel.
10 Die TOP 100 der Logistik, 2006, Deutscher Verkehrs-Verlag; also proposed in Case COMP/M.2905 DB /
Stinnes.
11 There would be vertically affected markets due to DB's strong presence in rail transport. However, the
arguments introduced below to exclude the threat of vertical foreclosure with respect to freight
forwarding by rail, FVL and component logistics also apply for the narrower segments of the TOP100
segmentation.
12 Case COMP/M.2722 Autologic / TNT / Wallenius / CAT, Case COMP/M.2411 Autologic / TNT /
Wallenius / CAT.
13 Az. B9 – 2013/94 (26 June 1994) – ATG/Menke/Silcock&Colling, WuW/E BKartA 2659.
14 Case COMP/M.2411 Autologic / TNT / Wallenius / CAT.
15 Case COMP/M.2905 Deutsche Bahn/Stinnes.
16 Case COMP/M.4294, Arcelor/SNCFL/CFL Cargo, Case COMP/M.3150 SNCF/Trenitalia.
17 Case COMP/M.3971 Deutsche Post/Exel, Case COMP/M.4746 Deutsche Bahn/EWS.
18 Case COMP/M.4746 Deutsche Bahn/EWS.
19 Case COMP/M.2905 Deutsche Bahn/Stinnes.
20 However, this could also be due to a cellophane fallacy. Supposing transhipment constitutes a largely
inferior alternative (and, hence, a separate market), a monopolist in axle-changing would maximize its
profit by raising the price of axle-changing so much that customers are starting to consider switching to
transhipment. Hence, switching considerations will occur even if axle-changing is a very weak substitute.
Therefore, it is not possible to conclude from the consideration of switching alone whether axle-changing
constitutes a separate market. Economically, considerations of switching would only indicate that
transhipment and axle-changing are in the same market if axle-changing was provided by competing
providers.
21 Case COMP/M.4045 DB / BAX Global, Case COMP/M.3971 Deutsche Post / Exel and Case
COMP/M.1794 Deutsche Post / Air Express International.
22 Case COMP/M.3971 Deutsche Post / Exel and Case COMP/M.1500 TPG / Technologistica.
23 Case COMP/M.2722 Autologic / TNT / Wallenius / CAT, Case COMP/M.2411 Autologic / TNT /
Wallenius / CAT.
24
25 Case COMP/M.2411 Autologic / TNT / Wallenius / CAT.
26 Case COMP/M.2411 Autologic / TNT / Wallenius / CAT.
27 Case COMP/M.2905 Deutsche Bahn/Stinnes, Case COMP/M. 4294 Arcelor / SNCFL / CFL Cargo, Case
COMP/M. 4746 Deutsche Bahn/EWS.
28 Case COMP/M.3150 SNCF/Trenitalia regarding the tunnel of Fréjus between France and Italy.
29 The parties have submitted market shares based mainly on data from the TOP 100 Study.
30 The market shares are based on volumes by value and the respective market shares of the parties in a
market for freight forwarding (without contract logistics and terminal services, non-integrated
warehousing, transhipment and other additional logistical services as well as courier, express and parcel
services). It covers only outsourced services and includes (transport) services provided by third parties or
the freight forwarder itself. The market shares are based on the country where the customer is invoiced,
which does not necessarily correspond with the origin or destination countries of the freight forwarding
service provided.
31 Horizontal overlaps between the parties freight forwarding services by rail would also occur in France,
the UK, Poland (only international freight forwarding by rail, as Transfesa is not active in domestic in
these countries) and Spain (only international as DB is not active in domestic) with combined market
shares of below 5% (no overlap in Portugal, where DB is not active in rail freight forwarding).
32 In The Netherlands where DB is also the incumbent rail freight provider its market shares in freight
forwarding services are lower due to the importance of incoming freight from the ports where DB is more
often subcontracted by other freight forwarders.
33 DB has a […] shareholding of around [10-20]% in the Spanish FVL provider Pecovasa who operates [0-5]
compounds with regional delivery.
34 These capacity calculations include the whole capacity of GEFCO, which, while active on the merchant
market, currently uses a high percentage of its capacity for its parent company. Although the capacity which
GEFCO uses currently in house for PSA is not part of the merchant market it would for the purposes of this
case act as a competitive constraint as it could, in case of a price increase for rail based services, in principle be easily and timely be put on the market. The amount of capacity that GEFCO would be able to switch would depend on the switching possibilities and margins for its current use of in-house capacity. This indicates that barriers to entry are not insurmountable.
35 Case No. COMP/M.4294, Arcelor/SNCFL/CFL Cargo
36 Case No COMP/M.3150 SNCF/Trenitalia.
37 Case No COMP/M.4746 – DB / EWS.
38 Para 13.
39 In spite of capacity constraints, customers confirmed that they would be able to switch at least parts of their volume to another supplier or to road. While for some customers switching would not be possible for certain parts of their volumes and involve significant cost (mainly high volumes from a specific plant and long-distance transport), other customers took the view that switching to another rail provider or to road would raise their costs only slightly or even decrease costs for volumes which for other than cost reasons are transported by rail.
40 Parties' submission based on the IEA (Spanish Automotive Studies Institute). 1.7 million units refer to outbound traffic to EEA Member States
41 For definitions of rail transport and traction see Fn. 2.
42 Transfesa's activities in Denmark, where DB is also the incumbent, are negligible. DB' rail transport shares in France and Italy are very small (<1 % for domestic and international transport in France; <1% for domestic transport within Italy and <[0-5]% for international transport in Italy).